/VIKNkGING IRRIGATED
    AGRICULTURE TO
    IMfeOVE WATER QUALITY
Proceedings of
National Conference on
Managing Irrigated Agriculture to
Improve Water Quality

-------
/MKNKGING IRRIGATED
AGRICULTURE TO
IMfeOVE WATER QUALITY
Proceedings of
National Conference on
Managing Irrigated Agriculture to
Improve Water Quality

Sponsored by:
U.S. Environmental Protection Agency
and
Colorado State University

May 16-18, 1972

-------
   The opinions expressed herein are solely that of the
authors and do not necessarily represent official policies
           of the representative organizations.
         Printed in the United States of America

      Library of Congress Catalog Card No.  72-83080
             Graphics Management Corporation
               1101 Sixteenth Street,  N.W.
                  Washington, D.C.  20036

-------
                                        Contents
The Need for Implementing Irrigation Return Flow Quality Control
   James P. Law, Jr. and JefferyD. Denit, Environmental Protection Agency and Gaylord V. Skogerboe,
     Colorado State University	1

Salinity Control Needs in The Colorado River Basin               '
   Myron B. Holburt,  Colorado River Board of California 	  19

Economic Impact of Salinity Control in The Colorado River Basin
   L. Russell Freeman, Environmental Protection Agency 	  27

Soil, Water and Cropping Management for Successful Agriculture in Imperial Valley
   Arnold J. Mackenzie, USDA Agricultural Research Service	  41

Irrigation Return Flows in Southern Idaho
   David L. Carter, Agricultural Research Service, USDA	  47

Salinity Problems in the Rio Grande Basin
   John W. Clark, New Mexico State  University	  55

Hydro logic Modeling for Salinity Control Evaluation in the Grand Valley
   Wynn R. Walker and Gaylord V. Skogerboe, Colorado State University  	  67

Sediment Control in Yakima Valley
   B. L. Carlile, Washington State University	  77

Treatment of Irrigation Return Flows in the San Joaquin Valley
   Louis A. Beck, California Regional Water Quality Control Board	  83

Examination of Agricultural Practices for Nitrate Control in Subsurface Effluents
   Donald R. Nielsen and John C. Corey, University  of California	  99

Grand Valley Salinity Control Demonstration Project
   T. John Baer, Jr., Colorado State Legislator	109

                                                   iii

-------
 iv        MANAGING IRRIGATED AGRICULTURE

 Salinity Control Measures in The Grand Valley
   Gaylord V. Skogerboe and Wynn R. Walker, Colorado State University  	123

 Management of Irrigation Water in Relation to Degradation of Streams by Return Flow
   H. R. Haise and F. G. Viets, Jr., U.S. Department of Agriculture	137

 Water Quality Aspects of Sprinkler Irrigation
   J. Keller, J.  F. Alfaro, and L. G. King,  Utah State University	147

 Subirrigation Studies in the High and Rolling Plains of Texas
   C. W.  Wendt, A. B.  Onken and O. C. Wilke,  Texas A&M University	157

 Irrigation Return Flow Studies in The Mesilla Valley
   P. J. Wierenga and T. C. Patterson, New Mexico State University	173

 Irrigation Scheduling in Idaho
   Marvin E. Jensen, Agricultural Research Service	181

 Irrigation Scheduling in the Salt River Project
   E. Win Kyaw and David S. Wilson, Jr., Salt River Project	187

 Irrigation Scheduling Studies for Water Quality Improvement
   RayS. Bennett and James H. Taylor, Colorado State University  	195

 The Role of Modeling in Irrigation Return Flow Studies
   Arthur G. Homsby and James P. Law, Jr., Environmental Protection Agency  	203

 Modeling Subsurface Return Flows
   Gordon R. Dutt, University of Arizona	211

 Modeling Salinity in the Upper Colorado River Basin
   M. Leon Hyatt, National Field Investigations Center	215

 Hydrologic Modeling of Ashley Valley, Utah
   Robert F. Wilson, Engineering and Research Center  	229

 Modeling Subsurface Return Flows in Ashley Valley
   Larry G. King, R. John Hanks, Musa N. Nimah, Satish C. Gupta andRusselB. Backus, Utah State University  . .241

 Surviving with Salinity in the Lower Sevier River Basin
   W. Roger Walker and Wynn R. Walker, Colorado State University	257

 Water Right Changes to Implement Water Management Technology
   G. E. Radosevich, Colorado State University	265

Institutional Influences in Irrigation Water Management
   Warren L. Track, Texas A&M University	281

Sociological Considerations in Irrigation Water Management: Facing Problems of Water Quality Control
   Evan Vlachos, Colorado State University  	285

-------
/1/IKNKGING IRRIGATED
    /IGRICULTURE

-------
     The  Need  for Implementing Irrigation

             Return  Flow  Quality  Control

                              JAMES P. LAW, JR.
                      Irrigation Return Flow Research Program
                         Environmental Protection Agency
                       Robert S. Kerr Water Research Center
                                 Ada, Oklahoma

                          GAYLORD V. SKOGERBOE
                        Agricultural Engineering Department
                             Colorado State University
                              Fort Collins, Colorado

                                       and

                              JEFFERY D. DENIT
                  Agricultural Pollution Control Research Program
                         Environmental Protection Agency
                                Washington, D.C.
ABSTRACT
  The'practice of irrigation increases salinity
and  has  other  detrimental effects on  water
quality. Major water pollution problems result-
ing from  irrigation occur in nearly all of the
western river basins. Control measures need to
be evaluated and implemented that will be ef-
fective in alleviating the deleterious effects of
irrigation  return flows.  These  will involve
physical  changes  in  irrigation systems,  im-
provements in present management and cul-
tural practices, and I or changes in the institu-
tional influences upon the  system.  Research
investigations will be required to further eval-
uate  the effectiveness of new and improved
water pollution control programs in river basin
areas. Demonstration and educational activities
will play a major role in gaining local accep-
tance and  support of recommended control
programs. To implement irrigation return flow
quality  controls, economic  incentives  to the
irrigator must be provided along with changes
in the  institutional constraints presently im-
posed on water management reform.

-------
      MANAGING IRRIGATED AGRICULTURE
INTRODUCTION
  The water quality problems associated with
irrigation  return  flows  are  of special concern
because  irrigated  agriculture  is  the  largest
consumer  of our  water resources. Additionally,
it  is  of major  economic importance  to the
Nation  as  the source of a  significant part of
the food and fiber produced annually. Irriga-
tion return  flows constitute a large portion of
the  flow  in many streams  of the Western
United  States. Due to the consumptive nature
of  irrigation,  some degree  of salt  concentra-
tion has been accepted  as the  price for irriga-
tion  development.  Nevertheless,   there  are
many areas  where water quality degradation
has  been  unduly  serious and excessive.  As
pressures on water resources increase, there is
mounting  concern  for   proper and adequate
control  of such serious water quality deteriora-
tion. The need for more precise information as
a basis  for wise action has been brought sharp-
ly into focus. The exact  role of irrigation return
flows in both surface and groundwater quality
problems  is now,  and  must  continue  to  be
examined  more  closely  to develop and imple-
ment measures to control or alleviate the un-
necessary detrimental effects.
  Legal authority to implement water pollution
control  was initiated essentially  with the pas-
sage of the Water Quality  Act of  1965 (P.L.
89-234), and the  Clean  Water Restoration Act
of  1966 (P.L. 89-753)  which together  greatly
expanded  earlier  basic laws and directed that a
"national  policy  for the prevention,  control,
and abatement of  water pollution" be  estab-
lished. No segment of the national economy was
exempted  from the implied intent  of this Na-
tional policy. The Water Quality Act of 1965
further  provided for all states to establish water
quality  standards for their interstate and coastal
waters.  The setting of these standards required
crucial  decisions  by each state  regarding the
uses of their water  resources, quality criteria to
support  these  uses,  and specific  plans for
achieving   these  levels  of quality.  The  chief
purpose of the national program has been to
enhance the quality and value of polluted water
and to  protect the quality of clean water. Al-
though the major efforts to date have been in
the treatment and control of municipal and in-
dustrial wastes,  agriculture  is by  no means
 exempted from the fight against water quality
 degradation.

 Irrigation Return Flow System
   The  relationships of  the  irrigation  return
 flow system to  a river  basin are very complex,
 as shown by the model  in Figure 1. The primary
 sources of  irrigation return flow_aTeshown to
 EenSypasT water,^ caharseepage, deep percpla-
 tionor grojandwater  flow,  and tailwater  or
^rface_return flpw. Bypass water  is required
 to maintain hydraulic head and adequate flow
 through the canal system. It is usually returned
 directly to the river and few pollutants are add-
 ed to the river by this route. Canal seepage,  on
 the other hand, contributes to high water tables,
 aggravates   subsurface  salinity,   encourages
 phreatophyte growth,  and generally increases
 saline drainage  from  irrigated areas.  Canal
 seepage amounts to a significant fraction of the
 total diversion in many project areas. Tailwater
 and deep percolation are the major contributors
 to irrigation return flow  from irrigated  crop-
 land.  These sources  convey dissolved  salts,
 plant nutrients, eroded  sediments,  pesticides,
 and  other  pollutants  to  the  stream drainage
 system.
   The diversion-return  flow cycle shown in Fig-
 ure  1 is typical of many western rivers, and may
 be repeated many  times  over in a river basin.
 The  flow at any point along the river may be
 composed  of undiverted river flow, natural
 inflow, irrigation  return  flow, municipal and
 industrial  effluents,  and  return flow of other
 used water. The proportion of each depends on
 such factors as the number of diversions, the
 extent and diversity of uses, the location along
 the  river, and the amount of each of the various
 inflows to the stream.
   Many  of the components and processes  in-
 volved in the irrigation return flow system can
 be subjected to some degree of manipulation or
 control for the purpose of improving the quality
 of irrigation return flows, thereby reducing the
 pollutant   load  presently  reaching  receiving
 streams.  Additional investigations and further
 research will be required in many areas to de-
 fine  pollutant sources  and develop technology
 needed  to  implement  control measures. The
 major goal of this conference is to accumulate
 present  knowledge and  technology in such a

-------
                                              INFLOW TO
                              PRECIPITATION^  (~ CANALS
      IMPLEMENTING QUALITY CONTROL

             EVAPOTRANSPIRATION
                FROM CROPS
       UPSTREAM
                                                                              OTHER
                                                                           EVAPOTRANSPIRATION
                                                                           FROM IRRIGATED LAND
 SURFACE RUNOFF
FROM NON-IRRIGATED
      LAND
        IND. 8 MUN.
         WASTES
                                            APPLIED TO
                                          IRRIGATED LAND
                          DIVERTED FOR
                             IRRIGATION
              GROUNDWATER
              CONTRIBUTION
                                                                                   IRRIGATION
                                                                                   RETURN FLOW
                                                                          DOWNSTREAM
                   NATURAL
                   INFLOW
                                           Figure 1:  Model of the Irrigation Return Flow System
     manner  as  to  make it  available  for imple-
     mentation, and at the same time delineate those
     areas where technology is lacking.

             Projected Impact Of Irrigation
                  On Water Resources
       Since  1890,  the  total  irrigated  land in the
     United  States  has  increased  from  about four
     million  acres (1) to an  estimated  48  million
     acres in  1970, or an increase of about 12-fold.
     More recently, the value of supplemental  irri-
     gation  in the  more humid  Eastern  United
     States has been recognized, with an estimated
     five  million acres being  irrigated in  1970 (2).
     There is reason to believe  that the acreage of
     irrigated land will continue  to increase.
       The Economic Research Service (3) has made
     protective estimates  of irrigated acreage in the
     United States to the year  2020. A breakdown of
     these projections by major  river basins (Figure
     2) is shown in Table 1. The projected increase
     in irrigated  acreage  in the  United States from
     1969 to  2000 is roughly 30  percent,  with the
     projected  increase for the  same period in the
     western  states being about 25 percent.
       Soils  in many of  the irrigated  areas  in arid
     regions contain  large quantities of salt and are
     classed as saline. Some are high in exchange-
able sodium and are referred to as sodic soils.
It has been estimated (1) that crop production
is reduced  on  one-quarter of the irrigated lands
in the Western United States due to saline soils.
Salinity presents a  potential hazard to about
half of the irrigated acreage in the West. Cali-
fornia, which  has the greatest  acreage of irri-
gated land, also has  the largest area  of  salt-
affected soils.
  More than one-third of the irrigated  lands in
the states  of Colorado, Hawaii, Nevada, South
Dakota, and Utah are affected by  highly saline
soils. Impaired  crop  production due  to  salt-
affected soils is not restricted to the arid regions
of the United  States. It has been estimated that
one-third   of  the  world's  irrigated   land  is
plagued by salt problems (4).
  The  National Academy of Sciences (5)  has
projected  water demands in the United States
for  the year 2000 as compared to the data for
1954. Total diversion, consumptive use,  and
return  flows  were  projected  for irrigation,
municipal, manufacturing,  mining, and power
plant cooling. The figures projected for irriga-
tion are particularly  pertinent to this  discus-
sion. In 1954;, the gross withdrawal for irriga-
tion averaged  176 billion gallons (540,OOC acre-
ft.)  daily,  and consumptive use was 104 billion

-------
      MANAGING IRRIGATED AGRICULTURE
                                  ARKANSAS'-WHITE-RED  / f\
                       Figure 2: Major Hydrologic Regions in the United States
gallons (319,000 acre-ft.) daily. It was projected
for the year 2000 that gross withdrawal  would
be   184.5  billion  gallons  (565,500  acre-ft.)
daily and consumptive use 126 billion gallons
(387,000  acre-ft.) daily. The projected  increase
in irrigation withdrawal amounted to about a 5
percent   increase,  while  the  increase  in  irri-
gated acreage  from 1954 to 2000 is expected to
double.  Thus,  the  projections for water with-
drawals  reflect an  expected major increase in
irrigation  efficiency.   Further  projections (5)
were made  for  estimated  annual  irrigation
water requirements based upon   current  effi-
ciencies  of  application   and   conveyance of
irrigation water  as compared  to future  esti-
mated attainable  efficiencies.  These projected
estimates are  shown  in  Table 2  for the years
1980 and 2000 in the western regions along  with
totals for the United States. When the projected
increase  in irrigated acreage is considered along
with the  projected   water  requirements at
future attainable efficiencies of water-use  and
conveyance, it is shown  that a decrease in  sur-
face water withdrawals by the  year 1980  (nega-
tive values in Table 2) is  a distinct possibility.
To meet the long-term projections for irrigation
water demand  will require at least a 2-fold in-
crease in water-use and conveyance efficiencies.
These projections  are based on  the  fact  that
present  water resources will not be able to meet
future irrigation  water requirements  at current
efficiencies.  If irrigated  agriculture  continues
to grow along with  other  economic sectors,
then  increased efficiencies  will  be  a  neces-
sity.  Many problems will have to be  solved be-
fore  such  widespread increases  in  irrigation
efficiencies are actually attained. Major among
these are changes in interpretations of water
law and other  institutional constraints to water
management reform.  Improved water  quality
management is critical to prevent further water
pollution effects  and  will get more  critical as
expansion occurs.


Irrigation Return Flow Water Quality Problems
   Usually, the  quality of water coming from the
mountainous watersheds  in  the  West is excel-
lent. At the base of the mountain ranges, large
quantities of water are diverted to valley crop-

-------
                                                        IMPLEMENTING QUALITY CONTROL
                                           TABLE 1
                 Projected estimates of agricultural irrigation in the United States,
                      from 1969 to 2020 (All data are in thousands of acres.)

North Atlantic
South Atlantic — Gulf
Great Lakes
Ohio Basin
Tennessee Basin
Upper Mississippi
Lower Mississippi
TOTAL-EASTERN REGIONS
Souris — Red — Rainy
Missouri Basin
Arkansas — White — Red
Texas Gulf
Rio Grande
Upper Colorado
Lower Colorado
Great Basin
Columbia — North Pacific
California
TOTAL-WESTERN REGIONS
MAINLAND UNITED STATES
7969
519
1,747
143
99
18
121
972
3,619
20
6,985
5,357
5,890
2,020
1,700
1,430
2,240
5,815
8,050
39,507
43,126
1980
730
2,480
230
150
30
210
1,400
5,230
90
8,050
5,600
6,510
2,050
1,900
1,820
2,340
7,350
9,050
44,760
49,990
2000
990
3,520
350
250
50
310
2,070
7,540
230
8,950
6,400
7,350
2,180
2,150
2,190
2,510
7,810
9,600
49,370
56,910
2020
1,120
4,150
470
340
70
410
2,570
9,130
250
9,600
6,690
7,770
2,200
2,250
2,400
2,570
8,490
1 1,540
53,760
62,890
            Source: Economic Research Service, USDA (Ref. 3).
 lands. Much of the diverted water is lost to the
 atmosphere  by  evapotranspiration  (perhaps
 one-half to two-thirds of the diverted  water),
 with  the  remaining  water supply being  irriga-
 tion return flow. This return flow will either be
 surface runoff,  shallow horizontal  subsurface
 flow, or will  move  vertically through the soil
 profile  unitl  it  reaches a  perched water table
 or the groundwater  reservoir, where it will re-
 main in storage or be transported through the
 groundwater reservoir until it reaches  a river
 channel.
  That portion  of the water supply which has
 been  diverted for irrigation but lost  by eva-
 potranspiration  (consumed) is essentially salt-
 free. Therefore,  the  irrigation return flow will
 contain most of the salts originally in the water
 supply. The surface irrigation return flow will
 usually contain  only  slightly higher salt con-
 centrations than the original water supply, but
 in some cases,  the  salinity may be increased
 significantly.   Thus,   the  water   percolating
through the soil  profile contains the majority of
salt  left behind by  the  water returned to the
atmosphere as vapor  through the phenomena
of   evaporation   and   transpiration.  Conse-
quently, the percolating soil  water contains a
higher concentration of salts. This is referred to
as the "concentrating" effect.
  As the water moves through the soil profile,
it  may  pick  up  additional  salts  by  dissolu-
tion. In addition, some salts may be precipitated
in the soil, while there will be an exchange be-
tween some salt ions in the  water and in the soil.
The  salts picked up  by the water in addition to
the salts which were in the water applied to the
land are  termed  salt "pickup." The total  salt
load is the sum of the original  salt in the applied
water as  the result  of the concentrating  effect
plus  the salt pickup.
  Whether irrigation return flows come from
surface runoff  or have returned  to the system
via the soil profile,  the water can  be expected
to undergo  a variety of quality changes due to
varying  exposure conditions. Drainage  from
surface sources consists mainly (there will  be

-------
                                                             TABLE 2

                              Estimated annual irrigation water requirements for years 1980, and 2000
                                    (All units in thousands of acre-feet.) (Excerpted from Ref. 5)
Total groundwater and surface-water
requirements
At estimated

Water Resource Region
Western:
Upper Missouri
Upper Arkansas, White, Red
Western Gulf
Upper Rio Grande-Pecos
Colorado
Great Basin
Pacific Northwest
Central Pacific
South Pacific
Subtotal (Western Regions)
United States Totals
At current
19802

20,747
4,876
11,789
7,079
20,335
9,481
16,837
35,031
4,486
130,661
139,472
efficiency1
2000

32,040
6,488
15,689
8,053
20,708
9,742
21,053
35,623
4,635
154,031
189,457
efficiency
1980

16,346
4,035
9,226
5,522
16,071
7,449
13,680
30,223
3,671
106,223
113,650
2000

22,331
4,922
10,914
5,315
14,696
6,727
15,132
27,241
3,371
110,649
137,246
Surface-water
requirement at
estimated future
efficiency
1980

14,058
1,130
4,613
2,761
12,214
6,481
12,586
16,623
2,019
72,485
75,819
2000

19,205
1,378
6,003
2,923
11,904
5,852
13,921
16,345
2,023
79,554
94,283
Additional surface-
water requirement
at estimated
future
efficiency
1980

- 1,1793
- 199
- 715
- 198
- 2,172
- 1,735
- 2,301
- 1,339
- 503
-10,341 ^
- 9,461 ^
2000

3,968
49
675
- 363
-2,482
-2,364
- 966
-1,617
- 499
-3,2723
9,003
2
>
>
o
N-«
o
                                                                                                                                          tn
                                                                                                                                          a
                                                                                                                                          70
                                                                                                                                          O
                                                                                                                                          m
'Assuming no increase in efficiency of application and transmission of irrigation water.
2Based on adequate irrigation of all land under irrigation.
3Negative values indicate a net decrease in water requirements resulting from estimated increased efficiency
in use and transmission of irrigation water.

-------
                                                       IMPLEMENTING QUALITY CONTROL
some precipitation  runoff)  of  surface  runoff
from irrigated  land.  Because  of its limited
contact and  exposure to the soil  surface,  the
following changes in quality  might be expected
between  application and runoff:  (a) dissolved
solids  concentration  only   slightly  increased;
(b)  addition  of  variable   and  fluctuating
amounts of pesticides; (c) addition of variable
amounts of fertilizer elements; (d) an increase
in sediments and other  colloidal material; (e)
crop residues and other debris floated from the
soil surface; and (f) increased bacterial content.
   Drainage water that has moved through the
soil  profile will experience  changes  in quality
different from surface runoff.  Because of its
more intimate contact with the  soil and  the
dynamic soil-plant-water regime, the following
changes  in quality are  predictable: (a)  con-
siderable  increase  in dissolved  solids  con-
centrations; (b)  the distribution of various  cat-
ions and anions  may be quite  different; (c)
variation in  the total salt load depending on
whether  there has been deposition or leaching;
(d) little or no  sediment or  colloidal material;
(e)  generally,   increased  nitrate  content;  (f)
little or no phosphorus content; (g) general
reduction  of oxidizable  organic substances;
and (h) reduction of pathogenic organisms  and
coliform bacteria. Thus each  type of return flow
will affect the receiving water in proportion to
respective  discharges  and the relative quality
of the receiving  water.
   The quality of irrigation  water and return
flow is determined largely by the  amount  and
nature of the dissolved and suspended  mate-
rials they contain. In natural waters, the mate-
rials are  largely dissolved inorganic salts leached
from rocks and minerals of  the soils contacted
by the  water.  Irrigation, municipal and in-
dustrial  use and reuse of  water concentrate
these  salts  and  add additional  kinds  and
amounts of  pollutants.  Inorganic salts, pesti-
cides,  sediments,  detergents, salts  of  heavy
metals, and  other organic  compounds render
water  less fit  for irrigation and other bene-
ficial uses.

Major Areas Having Water Quality Problems
   The more serious water quality  problems re-
sulting from irrigation return flow occur in the
Western  states,  where  salinity  and sediment
erosion and transport are the greatest offenders.
Plant  nutrients  and  pesticide  residues  follow
closely behind, but are not necessarily confined
to the more arid regions of the West. The major
areas  in  the Western United States  having
serious  water   quality   problems  resulting
from irrigation return flows are shown in Figure
3. Some of our major river basins are char-
acterized by a high degree of utilization of the
water  resources  and  are  also  experiencing
deleterious water quality effects.
  Water users in the Lower  Colorado  River
Basin,  especially Mexico and the  Imperial and
Coachella valleys, experience difficulties at the
present time due to high salt concentrations in
the  river.  Salt  concentrations in  the  Lower
Colorado River  are projected to increase by the
year 2000 due  to anticipated  water resource
development  projects.  Projects  are presently
nearing  completion  for  exporting additional
quantities of high quality water from Colorado
River  watersheds to  satisfy  water demands in
the more populous regions of  Utah, Colorado,
and New Mexico. The authorized Central Ari-
zona Project will divert large quantities of flow
from the Colorado River to Salt  River Valley,
which  is within  the basin, but the return flows
to the  Colorado River will be small. The irriga-
tion return flows will have long-term effects on
the  salt  balance in the Salt River Valley.  In
addition, large quantities of water will be divert-
ed  for use by present and future  power plants
in the four corners region (Utah, Colorado, New
Mexico,  and Arizona). A  salinity control pro-
gram,  including  a combination of  controlling
mineralized  springs   and  irrigation   return
flows,  would negate a large portion of the dam-
age  which  will  result from  anticipated  water
resource   development  projects,   which  will
export good  quality  water  to adjacent river
basins, thereby  leaving  less water  within  the
Colorado  River Basin for  diluting irrigation
return  flows.
  The  Rio Grande Basin is another example of
an  area  already experiencing serious  water
quality problems,  with  the  outlook for even
more  serious   problems.   Rapid  population
growths  in Albuquerque, El Paso,  and  Juarez
foretell  of  immediate  difficulties. Whereas
studies have been made in the Colorado River
Basin  which predict future water  quality prob-

-------
      MANAGING IRRIGATED AGRICULTURE
 San Joaquin
    Valley
 Santa Ana Basin
 Coachella  Valley
    Imperial Valley
                   '-COLUMBIA
                 NORTH PACIFIC;
                                                 MISSOURI
GREAT BASIN ,
                 Grand
                  Valley

                 Lower/
        j'Gunr/ison
                              ARKANSAS-WHITE-RED
                             COLORADO
                       RIO
                       RANDE
                                                            EXAS-GULF
                 Mexicali Valley
                                                   Lower
                                                 Rio Grande
              Figure 3:  Major Water Quality Problem Areas in the Western United States
lems due  to  basin  development,  such studies
have not been undertaken in the Rio Grande
River Basin. Future water quality problems in
this  basin could easily  result  in  international
problems  somewhat  similar  to those recently
experienced in the Lower Colorado River.
  Another major irrigated area presently ex-
periencing water quality problems from  irriga-
tion  return flows is the San Joaquin Valley.
Water  deliveries  from portions  of the recently
constructed  California  State   Water  Project
have begun. New lands are being  placed under
irrigation  and natural salts (including nitrates
and boron) in the soil root zone must be leached,
with subsequent higher water tables in irrigated
lands at lower elevation. A drain  is being con-
structed to carry the return flows to the ocean
by way of San Francisco  Bay.  Because of al-
ready serious quality  problems in the Bay, it
is likely that this drainage water will have to be
treated  for nitrate  removal  before being  re-
                          leased. The magnitude  of the potential prob-
                          lems can only be estimated at the present time.
                            The Yakima Valley in central  Washington
                          is an area which experiences considerable local
                          difficulty due to irrigation  return flow.  Even
                          though the net quality degradation in the Co-
                          lumbia River is low due to dilution, there is a
                          real need for an irrigation return flow control
                          program to provide for competitive water uses
                          (fishing,  recreation, municipal, and industrial)
                          within the  Yakima  River Basin,  as well as to
                          provide economic equity among the agricultural
                          water users.
                            Numerous other areas in the West experience
                          some degree of water quality problems due to
                          irrigation return  flows,  including  the  Pecos
                          River in New  Mexico  and  Texas, the  South
                          Platte River  in Colorado, the Platte  River in
                          Nebraska, the Sevier and Bear Rivers in Utah,
                          the Humboldt River in Nevada, the Santa Ana
                          Basin in California, and  other areas. With the

-------
                                                      IMPLEMENTING QUALITY CONTROL    9
passing of time, the water quality problems in
these  areas  will become more  serious because
of increasing water demands.
  Although  salinity problems due to irrigation
return flow  are not significant  in much of the
Pacific  Northwest,  water quality problems re
suiting  from pesticides, nitrogen, phosphorus,
nematodes, and sediments are  very  important.
Also,  there  is considerable potential for in-
creasing the irrigated acreage in  the Columbia
River Basin because  of abundant water sup-
plies.  For example, Idaho could increase its ir-
rigated cropland from 3.66 million acres (1969)
to 8 million  acres. Such an increase in irrigated
acreage could  be expected to have a dramatic
effect  upon the  water quantity  and quality
of the Snake River.
  The  Navajo  Indian  Irrigation  Project  in
northwestern  New  Mexico, presently  under
construction,  will eventually  irrigate  110,000
acres of land not previously irrigated. An initial
block of 10,000 acres is  scheduled for irrigation
in the spring  of 1975.  Of the approximately
500,000 acre-feet of water  to  be diverted an-
nually, half is expected to  be return flow. There
are estimates  that the  quality  of these return
flows will double in salt  concentration due to
evapotranspiration, but  there are no estimates
of salt pickup. Presently,  the irrigation system
for the initial block is  being  re-designed to
provide more  efficient water utilization. If this
results in less  water being diverted, then the
total salt load returning to the  San Juan River
will not be  as great as would have occurred
under the original plan  of development. In any
event, precise predictions of salt load returning
to the San Juan River  are not available.  As a
consequence,  downstream  damages  in  the
Lower Colorado River  to Arizona,  California,
and  Mexico  are  unknown,  with  only  very
crude estimates being possible.
  The Navajo Indian Irrigation  Project  is but
one example  of  our present-day problems in
predicting the effect of irrigation development
upon   the  quality  of  downstream  receiving
waters.  As our water resources  become  more
fully  utilized,  the  importance  of irrigation
return flow quality will  be  of even greater
significance  in the  overall  water management
and  development  in a  basin.  If  large-scale
schemes  for transporting water  to  the  Inter-
mountain West from the Columbia River Basin
or Canada were to become a reality, what would
be the effect of irrigating new desert lands in
Utah and Nevada,  or other adjacent areas?
Questions similar to this can be raised regard-
ing the effect of any projected changes to pres-
ent water resource schemes. Intelligent answers
are needed  to recommend  and initiate water
pollution  control programs  based on wise use
of technology.

        Irrigation Return Flow Quality
              Control Measures
   Irrigation return flow is an integral part of
the hydrologic  cycle,  therefore, control mea-
sures for managing the return  flow from ir-
rigated areas must be  a coordinate  part of the
overall water resource  development and man-
agement.  Irrigation return flow quality control
is one component in the management of water
resources that has been long neglected. In too
many instances, economic pursuits have by far
superseded   considerations   for   maintaining
environmental  quality.  Those measures  which
have  a potential for controlling the quality of
irrigation return flows may  involve physical
changes in the system, improvements in present
management  and cultural  practices,  and/or
changes in the institutional influences upon the
system.
  The irrigation return flow system can be
subdivided  into  three  major subsystems: (a)
water delivery; (b) the farm; and (c) water re-
moval (Figure  4).  The water  delivery  sub-
system includes  the  transport of  water  and
pollutants from the  headwaters  of the water-
shed to the point of diversion,  thence to the
individual farm. The farm subsystem  begins at
the point where water is delivered to the farm
and  continues to the  point  where  drainage
water is removed from the farm. Also, the farm
subsystem is defined vertically as beginning at
the soil surface and terminating  at the bottom
of  the  root zone. The water  removal  sub-
system includes both surface runoff  from the
farm  and water moving below the root zone,
where  quality  problems  are  minimized  by
having highly efficient water delivery and farm
subsystems.  Minimizing the quantity of surface
runoff will assist in alleviating quality problems-
due to sediments, phosphates, and pesticides;

-------
10     MANAGING IRRIGATED AGRICULTURE
                                                                          Open Drain
                                                                       (Surface Removal!
                                                            (c) Water Removal Subsystem
                                       %       &_   _£
                                    J  mi Hum it r
(a)Water Delivery Subsystem
                 Figure 4:  The Water Delivery, Farm, and Water Removal Subsystems
whereas  minimizing  deep  percolation  losses
will  reduce  quality problems due to salts,  in-
cluding nitrates, in  areas where salt  pickup
occurs.

Water Delivery
  The importation  of high quality water from
adjacent  river basins,  weather modification to
increase  precipitation  and  runoff  from  the
watersheds,   bypassing   mineralized  springs,
evaporation  reduction  from  water  and  soil
surfaces,   and   phreatophyte  eradication  are
some of  the available measures  for  improving
the quality of water within a river basin. Con-
sequently, they play a role in the management
of the  irrigation return  flow  system.  More
feasible approaches may  be found in the con-
trol  of losses  from  storage and conveyance
systems.
  Canal and Lateral  Lining -- Seepage losses
from irrigation canals may be considerable and
result  in  low  water-conveyance efficiencies.
Historically, the economics of canal  lining has
been justified primarily on the basis of value of
the  water  saved. The  possibility  that canal
seepage may greatly  increase the total  contri-
bution of dissolved  solids  to  receiving waters
has  only recently been  given serious attention.
Average seasonal canal losses have been found
to vary from 13 percent of the diversion on the
Uncompahgre Project, Colorado, to 48 percent
of the diversions on the Carlsbad Project, New
Mexico (1). On the basis of a conservative esti-
mate  that 20 percent of the total  water diverted
for irrigation in the United  States is  lost by
canal   seepage, the loss to  the intended users
would be 24 million acre-feet per year. This in
itself  is significant, but  if soils along the canals
are high in residual salts, the  salt pickup con-
tribution from this source could easily exceed
that leached from the irrigated land to main-
tain a salt balance.  Because  of enhancement
in both quantity and quality, the value  of  im-
proved water quality is another benefit to be
utilized in the economic justification  of canal
lining.
  Evaporation losses  from  canals  commonly
amount to a few percent of the diverted water.
The installation of a closed conduit conveyance
system has the  advantage  of minimizing both
seepage  and evaporation losses.  Either lined
open  channels or closed  conduits  will  reduce
evapotranspiration losses due to  phreatophytes
and other  vegetation along canals. The closed
conduit system uses less land and provides for
better water control than a canal  system. Water

-------
                                                     IMPLEMENTING QUALITY CONTROL     11
quality  improvements  may  very well prove to
be the greatest economic justification for closed
conduit systems  because of  minimal seepage
losses  and  considerable flexibility  in  water
control.
  Project  Efficiency — Because the amount of
water passing critical  points in  the  irrigation
delivery  system  must  be known  to provide
water control and improve water-use efficiency,
provisions for effective flow measurement must
be made.  Many present day systems employ no
flow measuring devices and, in some  cases, the
individual  farmer  operates  his  own turnout
facility  with no  close control of  the amount
diverted to the farm.
  Economics play a major role in existing proj-
ect  irrigation  efficiencies,  and  a close  cor-
relation exists between water abundance and/
or cost of  water and project efficiency.  For
example, where water  is scarce or high in cost,
the efficiencies are  found to be higher. Project
management,  as well as farm  management,
involves balancing the immediate cost of water
against  the  higher  labor and investment costs
required to  use it more efficiently.  The costs of
inefficient  water use  may  not be recognized
immediately, but may be reflected in reduced
yields due to nutrient losses or increased sa-
linity,  or in extra drainage facilities required
later to control rising water tables.

Farm Water Management
  The  most significant  improvements in  con-
trolling irrigation return flow quality will po-
tentially  come   from  improved   farm water
management. Because of the nature of irrigated
agriculture, whereby  salts  must  be  leached
from the  root  zone, an optimum  solution will
require improvements  in  on-the-farm water
management.  In  order to attain high irrigation
applicatio'n efficiencies, the timing and amount
of water  being delivered to the  farm must  be
controlled.  At  the  same time, the  fanner  must
be  capable of controlling the water  supply as
it moves across the farm by regulating the water
delivery rate  and  measuring the quantity  of
water applied.
  Application Methods — The effect of methods
of application on the quality and quantity of re-
turn flow has not received detailed study.  Con-
ventional surface  and  sprinkler  methods are
most commonly used because of their low initial
cost and ease of adaptability to a wide range of
field and surface conditions. New and unique
approaches to application methods need to be
found. Two that appear to offer great promise in
the control of both quantity and quality of re-
turn flows are subsurface application (6) and
drip or "trickle" methods (7).
  With subsurface irrigation, water can be ap-
plied  to the crop  in small amounts and at
frequent intervals so  that evaporation and the
concomitant increase in  salt concentration are
reduced. The water content  of the soil is main-
tained below field capacity so that some pre-
cipitation can  be stored,  introducing an ad-
ditional dilution factor. Comparable crop yields
have been produced with 40 to 50 percent less
water  than is required with furrow irrigation.
Thus,  limited water supplies can be extended
or  the acreage which can be  irrigated with a
given water supply can be  increased.  Applica-
tion rates can be  closely controlled  and the
method can be readily automated.
  The major advantages  of  drip irrigation in-
clude  increased  crop yield,  reduced  salinity
damage, and  shortened growing season  with
earlier harvest. The method involves  the slow
release of water  on the surface  near  the  base
of the plants. Evaporation losses are greatly re-
duced and moisture release  is  confined  to the
area of the plant root system.
   Both of these methods need to be evaluated
as  to their potential  for  reducing  salinity  in
return  flows,  increasing  yields  with limited
water supplies, and reducing fertilizer nutrient
losses. Liquid fertilizers can be applied by either
of  these methods  in small  controlled  doses
throughout  the growing season. The  potential
economic advantages of saving both water and
fertilizer nutrients and reducing labor costs of
operation make these methods attractive.  Eco-
nomic evaluation might show  that these bene-
fits would  largely  offset the  initial cost  of
installation.
   Irrigation Scheduling  —  Historically, irriga-
tion has been practiced  more as an art  than a
science.  When left  to his  own  discretion, a
farmer may  delay  irrigation until the crop is
stressed and then apply more water than actual-
ly needed, resulting in poor water management
and reduced yields. This two-fold problem is

-------
12    MANAGING IRRIGATED AGRICULTURE
presently being  overcome  by irrigation  sched-
uling based on climatic, soil, and crop data (8).
Commercial firms gather the data  and notify
the irrigator when to irrigate a certain crop and
the amount to be applied. The farmer pays a fee
for the service and is thus relieved of the re-
sponsibility  of deciding when is the best time
to irrigate.
  Experience in southern  Idaho and  the  Salt
River Project  in Arizona during the past few
years clearly indicates increased yields due only
to scientific scheduling of irrigation.  To date
there has been little reduction in water use, al-
though  it seems likely that this would occur
with time as more experience is gained with the
scheduling program. This approach offers great
promise as a water management tool and may
provide the  knowledge and experience needed
to overcome  present  antiquated institutional,
political, and legal constraints to water manage-
ment reform.

  Tailwater  Recovery  — In water-short areas,
tailwater recovery  is practiced.  In  addition
to increasing water-use efficiencies, the practice
serves as a control of sediment with its adsorbed
pollutants. Recovered water is readily returned
to the irrigation supply after settling of suspend-
ed solids rather  than  released to surface drain-
age systems (9).  Improved  irrigation practices
would be needed in  order to  minimize the
quantity of  tailwater  and sediment  load from
cropland.  In extreme  cases, enforceable regu-
lations may be  required to  effectively control
tailwater losses and protect downstream water
users.
  Fertilizer Nutrients  — Nitrogen-use efficiency
may be improved by the use of slow-release fer-
tilizers or by adding liquid fertilizers frequently
at low levels through the irrigation  water  sup-
ply.  The present higher cost of these  products
is the chief deterrent to increased acceptance. If
regulations were imposed to require their use in
areas where  nitrogen problems occur, increased
sales volume would act to lower the cost  of pro-
duction and a more favorable pricing schedule
would result.  An added advantage  of slow-
release  fertilizers is   fewer  applications  per
season.  Further evaluation  of  these  products
with  regard to  improved quality of irrigation
return flow is required.
  Cultural Practices — In areas of slowly per-
meable  soils and saline water supply, cultural
practices become significant if crops are to be
grown  successfully.  Under these conditions,
management alternatives  are: (a) use of more
salt tolerant crops; (b) special deep tillage may
be required; (c) leaching in the off-season or
alternate years; (d) careful  seed-bed  prepara-
tion and seed placement;  and (e) close control
of timing and amount of water applied. Special
practices such as mulching  and reduced tillage
may be effective in reducing soil water evapora-
tion. These special cultural  practices have been
aimed more at crop production under less-than-
ideal conditions than toward improved quality
of return flow, although the two could go hand-
in-hand. Soil  structure,  texture  and stratifica-
tion are the principal properties that control
water storage in the  soil. Deep tillage may be
required to  disrupt  slowly permeable  layers,
permitting  greater  water  storage capacity and
deeper  root  penetration.  Cultural  practices
which can  play a significant role in water qual-
ity management need further evaluation.
  Control  of Leaching — Another possible con-
trol measure would be to reduce deep percola-
tion losses, which  would minimize salt  pickup
from underlying geologic  formations. The most
drastic  action would be  to  eliminate irrigation
in areas of high salinity such as those where the
soils  are  formed  from,  and  overlying,  salty
shales.  Preventing  or reducing  the  amount  of
water penetrating to deeper saline strata would
also reduce the salt load in return flows. The use
of natural  or artificial barriers  below the root
zone, coupled  with drainage systems to inter-
cept drainage water before it reaches the deeper
strata,  would  be  effective.  Regulating  the
amount of  water applied to the land and thereby
regulating  the  amount of drainage  constitutes
a combination of controlled leaching and dilu-
tion.  Further  study is needed to evaluate this
approach  to the control  of salinity in return
flows.

Water Removal
  The water removal subsystem transports sur-
face runoff and subsurface drainage from ir-
rigated  lands. The tailwater may be returned
to the delivery system, become water supply for
an adjacent farm, or be transported back to the

-------
                                                      IMPLEMENTING QUALITY CONTROL    13
river. Subsurface drainage may seep into open
drains or to low  lands  along the  river where
salt damage often occurs. There are essentially
three management alternatives for preventing
or minimizing  the quantity  of pollutants dis-
charged to the  river: (a) divert the drainage to
some discharge  point or sink area away from the
river; (b) treatment for  removal of pollutants
prior to reuse or discharge; and (c) provide di-
lution  to  minimize  the  detrimental  effects.
Economic considerations are of prime impor-
tance in each of these alternatives.
  Diversion  — Subsurface  return flows are
often collected  by tile drainage systems under-
lying irrigated  areas. Being thus collected into
a common sump or open drain, they may easily
be diverted to some point away from the river.
The  proposed drainage system for the San Joa-
quin  Valley  and  the  present system for the
Wellton-Mohawk  district in  southern Arizona
are examples of return  flows being  removed
without returning  to the river or supply canal
system. Return flows diverted to suitable sink
areas may provide wildlife  habitat.  Either of
these alternatives prevents the addition of pol-
lutant loads to  the river system. Other consider-
ations,  such as water  rights of downstream
users,  may  be  important  in  selecting suitable
diversion schemes for irrigation return flows.
  Treatment — The present  high cost of desa-
lination processes is the  chief deterrent to their
use for reclaiming  saline  water. Only in extreme
cases would  they  prove economically feasible.
Recent   cooperative  research  at   Firebaugh,
California,  conducted by the Envirnomental
Protection Agency, Bureau of Reclamation, and
California Department of Water Resources has
developed methods for removing nitrates from
irrigation return flows.  In these studies, both
algae  stripping and  bacterial denitrification
proved feasible  to treat  agricultural tile drain-
age  prior to  its  release  into San  Francisco
Bay  via  the  San Joaquin drainage canal sys-
tem.
  Dilution —  Dilution  of return  flow  occurs
naturally in nearly all  headwater areas. Drain-
age waters and other return  flows are usually
diluted in the main stream so that the resultant
quality is acceptable downstream for irrigation
and  other beneficial uses. Multipurpose  reser-
voirs have provided storage capacity  for low-
flow augmentation which can be beneficial for
quality control, although such benefits are usu-
ally  incidental  to other primary requirements
for regulated flow.
  There are  limitations to the effectiveness of
dilution as a means  of  managing return flows.
The  major physical limitation is the amount of
high quality  water that can  be stored and re-
leased at the  desired  times to  achieve maximum
benefits from dilution.  The  legal implications
of obtaining water rights for the purpose of dilu-
tion  could be exceedingly complex. In general,
such practices would result in diminished sup-
plies for other purposes and this would occur in
areas where  water rights  already often exceed
the supply. Presently, it  appears  that stream-
flow regulation for quality control of irrigation
return flows  cannot  be  relied upon as a useful
management technique.  Reservoir  storage for
quality control has not received favorable sup-
port from a national policy standpoint.

      Research And Development Needs
             For Implementation
   In  the  preceding discussion  of  potential
water  quality control measures, the need for
further study and  evaluation of their  effects
on irrigation return  flow quality was  indicated.
Yet there  remain certain areas where advanced
technology and knowledge are required for the
development and implementation  of effective
water quality control programs. The needs  out-
lined below  are broader in scope and purpose
than  those  relating to  the  specific  control
measures.  The fact  that  certain research  and
development needs are  not discussed here only
means that the priority is considered less im-
portant from the  standpoint of immediate needs
for getting control programs underway.

Soil-Plant-Salinity Relationships
   In assessing  the effects of increased salinity
of water supplies, it becomes  essential that crop
damage functions be determined and  related to
field  conditions.  At the  present  time, there
exists  rather limited knowledge  of precise sa-
linity effects  upon crop  growth. Our knowledge
of crop utilization functions as affected by both
water quantity and stage of plant growth is also
weak.  Experimental  designs  should  incorpor-
ate both water quantity and quality as variables

-------
14    MANAGING IRRIGATED AGRICULTURE
in crop production and crop damage functions.
The salt  tolerance of  various  crops  under a
variety  of farm  water management  practices
should be investigated. The studies should in-
clude short-term effects due to salinity, such as
the ability of a plant  to withstand high salinity
concentrations  for  short durations.  Other cul-
tural  practices designed to  reduce  soil water
evaporation should be incorporated  in dem-
onstration projects. Evaluation of results should
be included in programs  to monitor the long-
term  effects  of recommended  irrigation  and
agronomic practices.  Rather than using  only
crop yields as a measure of success, monitoring
must   include  water ..quantity and  quality
changes in the root zone, as well as below the
root zone.
  There remains a  need to develop economical
slow-release fertilizers which will release nutri-
ents at a rate to match plant needs, thus maxi-
mizing plant-use efficiency and  minimizing the
quantity of fertilizer  constituents appearing in
irrigation  return flows.  The use of slow-release
fertilizers  should  be  incorporated  in experi-
mental designs on irrigation practices. Demon-
stration projects using such fertilizers could be
easily accomplished in most areas where nutri-
ents in irrigation return flow are a problem.

Leaching and Salt Balance
  Since crop salt tolerance and soil salt balance
are intrinsically interrelated, better techniques
for determining  optimum  leaching  require-
ments are needed. The  basic problem  is in de-
veloping a knowledge of transport phenomena
on a field basis, which can then be incorporated
into the development of criteria for determining
leaching requirements. The  transport phenom-
ena will  involve  a more  detailed analysis of
leaching based  upon an ionic evaluation of salt
movement through the  soil. For example,  cer-
tain salts, such as gypsum,  are not  really dele-
terious to plant growth and, if these salts are
precipitated within the soil, they present no real
problem. A more difficult problem results when
salts such as sodium are not being leached from
the root zone. Criteria for leaching should  take
into account the type  of salts being leached, as
well as total  dissolved solids, as a means of
evaluating leaching  requirements  and  salt
balance in field situations.
  Much  of our  present knowledge regarding
leaching  requirements and  the movement  of
water and salts in the soil has been developed
in laboratory-packed  columns. The results ob-
tained under these conditions can be unrealistic
when compared with  undisturbed soil  profiles.
There is a need to translate this type of informa-
tion to actual field conditions, and additional
studies in this area are  needed.  The relation-
ships between soil physical  characteristics and
quality of the water applied must be evaluated
to determine actual leaching requirements.

Prediction of Subsurface Return Flow
  The greatest single  technological  need at the
present time for the control  of irrigation return
flow quality is the development of prediction
techniques which will  describe the quantity and
quality of subsurface  return  flow. The critical
problem  is  in defining the  variability in sub-
surface return flows for large areas, such as an
irrigated  valley or large basin area.
  General   models should   be developed  to
describe  subsurface return flow which can be
adapted to numerous  areas. Conceptual models
must be developed and the sensitivity of various
parameters  which will be used in  the models
must be  determined. In studying large areas, a
balance must be reached between the sophisti-
cation of the model and the cost of collecting
field  data.   These  models must be capable of
predicting changes in the quantity and quality
of subsurface return  flows under a variety of
water management alternatives. To fully evalu-
ate chemical quality changes, the models should
be capable  of handling  precipitation  and ex-
change  reactions  which take  place as water
moves through the soil profile. These transfor-
mations alter the ionic balance of the chemical
constituents in solution and are very important
in describing the quality of irrigation  return
flow.
  Another  facet of predictive methods that is
not  generally available is predicting the long-
range quality considerations related to  irriga-
tion return  flow management and receiving
stream  quality.   Consequently, the problems
resulting from the development of new irriga-
tion projects, particularly those involving lands
not previously irrigated, are usually confronted
after-the-fact.

-------
                                                      IMPLEMENTING QUALITY CONTROL    15
Economic Evaluation
  There is a pressing need to delineate the wide
variety of benefits and damages that occur as a
result of  water quality changes. In assessing
costs and benefits associated with irrigation re-
turn flow, the development of crop production
and crop damage functions due to water quality
would   provide   necessary   information   for
making  more  accurate  economic evaluations
and  would  enhance the  decision-making  pro-
cess. Rather than thinking just in terms of crop
production,  or  crop damage,  resulting  from
water quality  degradation  economic studies
should  encompass  the  decreased  utility  of
return flows to downstream water users, due to
deteriorating water quality.
  Economic studies are  needed to  define  the
local, state,  regional,  and  national   benefits
which would accrue from the implementation,
either in an irrigated valley  or  river basin,  of
an irrigation return  flow  control program. For
example, a control measure  implemented  in a
particular valley  has  certain  benefits to  the
local area,  including non-agricultural  sectors.
The  degree  of quality  improvement  achieved
by this control measure will have a  number of
benefits  to  downstream  water  users. Also,
benefits may accrue to upstream users result-
ing from water exchanges or because of institu-
tional arrangements (e.g., if   standards fix the
allowable water quality leaving a region, then
the improvement of irrigation return flow qual-
ity in an existing  area may be required before
new agricultural lands can be placed under  ir-
rigation within the region).

Institutional Changes Needed
  Coupled with economic uncertainties, institu-
tional  considerations,   particularly  the inter-
pretation of western water laws,  represent the
greatest hindrance to promulgating incentives
for efficient  water use. First, a study should  be
undertaken  to  delineate the  changes in water
law interpretation required to provide efficient
water use   incentives.  Next, the procedures
required to  have such interpretations  included
in the water law structure of each state should
be determined.  Then,  as a minimum,  such
changes in interpretation  should be attempted
in the states  having water  quality  problems
resulting from  irrigation return  flow.  This
would be particularly beneficial in a state where
a control program would soon be getting under-
way.
    Studies should be undertaken which would
evaluate the possibilities of incorporating water
quality  into  a water right  (e.g., California
Porter-Cologne Act). For the most part, present
water  rights  pertain only to the  quantity  of
water.  Since  the  quality  of water can  be a
limitation  upon  its  use,  both  quantity  and
quality should be specified in a water right.  A
variable water right based on  quality may be
feasible. For  example, a  greater  quantity of
low quality water will be required to produce
the same results  that can  be obtained with a
lesser amount of high quality water.
  The  effects  of placing  certain regulations
upon   an area in order to  control irrigation
return flow quality  should be evaluated.  For
example,   regulations regarding  the   use  of
fertilizers should be studied. Such studies must
consider requirements for  administering fertil-
izer regulations, as well as gaining knowledge
on local and  downstream  benefits, along with
local  damages. Other regulations that  need
evaluation  are limiting water use (e.g., use of
economic charges to control water use), tail-
water controls, and effluent standards for drain-
age systems  which incorporate both  quantity
and quality.

Potential Control Measures
  Area-wide  investigations will be required in
some  areas to define  the control measures that
are needed and feasible to improve downstream
water  quality. These  studies would   pinpoint
the sources  of salinity, nutrients, and other
pollutants,  and provide background information
to support  the most feasible approaches to con-
trol measures.  Once  the sources of pollutants
are defined, more detailed  studies will be re-
quired to specify how those sources may best be
controlled.  Such  broad  investigations will re-
quire  the cooperative efforts of several research
groups under the guidance of a central advisory
committee  representing local, state, and federal
interests. Local acceptance  of proposed control
measures  will  require demonstration  projects
and extensive  educational programs to demon-
strate  local,  regional, and interstate  benefits
to be gained.

-------
16     MANAGING IRRIGATED AGRICULTURE
          Challenge For The Future
  The completion  of research  studies, recon-
naissance  investigations,   and  technological
developments will not constitute the final solu-
tions to water quality problems.  The knowledge
and  technology gained  from such endeavors
must  then be applied to specific problems  to
recommend, initiate  and implement the most
feasible  irrigation  return  flow water quality
control measures. Of the major problem areas in
the western river  basins,  only the  Colorado
River  Basin  (10,  11) has been studied  on a
basin-wide scale for the  purpose of developing
potential  solutions and costs  for controlling
salinity within the  bas"in. Other similar recon-
naissance-type studies may be required in major
problem areas, such  as the Rio Grande River
Basin, San Joaquin Valley, and other smaller
areas like Yakima  Valley  and the Santa  Ana
Basin.
  Once the major problem areas within a river
basin  have  been  delineated,   more   detailed
investigations  become  necessary,  particularly
for those  areas  responsible for  the  greatest
water quality degradation. An example of such
an area in the Upper Colorado River Basin is
Grand  Valley  where  detailed  investigations
have been accomplished and further studies are
scheduled for the  immediate future.  At this
point,  it becomes necessary to  investigate the
water delivery system, farm water management
practices, and the water removal system to de-
fine  the magnitude of control  possible for a
number  of water  management  alternatives.
Knowledge gained and control measures recom-
mended  in Grand  Valley should  prove useful
and applicable to other similar irrigated areas in
the West. This is where the potential control
measures and research needs cited previously
play an important role.
  The two  major  problem areas  confronting
implementation of  water quality control pro-
grams are considered  to be: (a) economic evalu-
ation of the effects of alternative control mea-
sures;  and (b)  changes of the  legal, political,
and institutional  constraints to  water  manage-
ment  reform. In addition to evaluating the eco-
nomic benefits accruing to local, regional, and
river basin areas, economic incentives  must be
provided for the  individual irrigator. The fact
 that irrigation farming is an economic endeavor
 must not be overlooked. In order to maintain
 the  ability to  compete  in  farm  production,
 higher  levels  of production  without  undue
 added costs must be reached. This must be ac-
 complished by improved management practices
 and maintaining productivity levels of the land.
 To illustrate, consider the irrigator who may be
 required to improve his water distribution sys-
 tem,  improve water-use efficiency,  use an ir-
 rigation  scheduling  service, control tailwater
 losses, and/or  other measures dictated  by a
 basin-wide water quality control program. If it
 can  be  shown  that  such  improvements will
 result  in  increased   crop  yields,  increased
 fertilizer-use   (decreased   nutrient   loss)  ef-
 ficiency,  lower labor and operating costs,  lower
 system  maintenance  costs, and maybe  other
 accrued  benefits, then  the economic incentive
 will have been provided for the  irrigator to co-
 operate.  Demonstration  projects  and educa-
 tional programs will provide the key to obtain-
 ing support and cooperation at the local  level.
 These activities will also encourage the support
 of other  sectors  of the local community.
  Constraints imposed  by legal, political, and
 institutional influences  may be  even  more dif-
 ficult to deal with. In  addition to the needed
 changes  in interpretation of water law and the
 incorporation of water quality in a water  right,
 the operations of many small water districts in
 a valley  area often lead to inefficiencies and in-
 equities  among  water users. This suggests that
 one  master  district  managing  the water  re-
 source for an entire valley might be much  more
 efficient  in its operational control of water dis-
 tribution and  drainage facilities. Inequities and
 competition would be reduced, with the possi-
 bility that  distribution costs could be lowered.
 Considerations such as these are essential  if the
 challenge of the  future is to be met.
  The concept of effluent standards is present-
ly being explored with regard  to certain  in-
dustrial wastes.  It is not beyond  the realm of
possibility  that   similar approaches may  be
considered   for   certain  agricultural  wastes.
Studies and voluntary action programs designed
to solve  a problem before  enforcement actions
become necessary would be most desirable. The
urgency for immediate action is apparent.

-------
                                                    IMPLEMENTING QUALITY CONTROL
                                         17
REFERENCES
   1. Utah   State   University   Foundation,
Characteristics  and Pollution  Problems  of
Irrigation Return Flow,  EPA (FWQA), Robert
S. Kerr Water Research  Center,  Ada, Okla-
homa. (1969).

   2. Anonymous,  "U.S.  Irrigated Acreage,"
World Irrigation,  (Aug.-Sept. 1970).
   3. Pavelis,  George   A.,  "Regional Irriga-
tional Trends and  Projective  Growth Func-
tions," Report  by  Water  Resources   Branch,
Natural Resource Economics Division,  Eco-
nomic Research Service, USD A, Washington,
D.C. (Dec.  1967).
   4. Bower, C. A., "Irrigation  Salinity  and
the World  Food  Problem,"  Presented before a
joint meeting of the Crop  Science Society of
America and Soil Science Society  of America,
August 22,  Stillwater, Oklahoma (1966).

   5. National  Academy of Sciences-National
Research  Council,  Committee  on Pollution,
Waste Management and  Control, Pub.  1400,
Report submitted to the  Federal  Council  for
Science  and Technology,  Washington,  D.C.
(1966).
   6.  Busch,  C. D.,  and  Kneebone,  W. R.,
"Subsurface Irrigation with Perforated Plastic
Pipe," Trans. ASAE, Vol. 9, pp. 100-101 (1966).
   7.  Goldberg,  D., and Shmueli, M.,  "Drip
Irrigation—A Method Used Under Arid Condi-
tions of High Water and Soil Salinity," Trans.
ASAE, Vol. 13, pp.  38-41 (1970).
   8.  Jensen,  M. E.,  Robb,  D. C. N., and
Franzoy, C. E., "Scheduling Irrigations Using
Climate-Crop-Soil  Data,"  Jour.  Irrig.  And
Drain. Div., ASCE,  Vol. 96, No.  IR1, pp. 25-
38 (1970).
   9.  Bondurant, J. A.,  "Quality  of Surface
Irrigation Runoff Water," American Society of
Agricultural Engineers, Paper No. 71-247, 8 p.
(1971).
   10.  Colorado  River  Board  of California,
Need for Controlling Salinity of the  Colorado
River,  Report submitted by  the  Staff of the
Colorado River Board of California to Members
of the  Board,  Sacramento, California (Aug.
1970).
   11.  Environmental    Protection    Agency,
"Summary Report, Appendices  A-D,"  The
Mineral  Quality Problem  in the  Colorado
River Basin, Regions VIII and  IX (1971).

-------
              Salinity Control  Needs  in  The
                      Colorado River  Basin

                                MYRON B. HOLBURT
                                    Chief Engineer
                          Colorado River Board of California
ABSTRACT
  The salinity of the Lower Colorado River is
presently high. In the absence of any salinity
control measures, it  is projected that the aver-
age salinity at Imperial Dam, the major Lower
Basin diversion point,  will increase from  870
ppm to over 1,300 ppm by the turn  of the cen-
tury. Studies have indicated that without salin-
ity  control measures  damages in  the Lower
Colorado  River Basin due  to the above pro-
jected salinity will continue to grow, reaching
in the order of$45-$60 million per year.
  The U.S. Bureau  of Reclamation has com-
menced studies on salinity control projects that
could remove about  1.8 million tons of salt an-
nually from the Colorado River. Because of fu-
ture development, approximately three million
tons of salt per year would have to be removed
by the end of this century in order to keep the
river's  salinity at  present  levels.  Thus,  the
USSR has indicated the need for weather modi-
fication, a large scale desalting of concentrated
flows, or additional  salinity control  projects in
order to meet the objective of keeping salinity at
or below present levels.
  Work has commenced on a very promising
major salinity control program that is supported
by the seven  Colorado River Basin  States and
federal agencies.
INTRODUCTION
  The Colorado River Basin comprises 242,000
square  miles in the seven states of Colorado,
Wyoming, Utah, New Mexico, Arizona, Ne-
vada, and California, and  in the Republic of
Mexico. The river's salinity is a basinwide prob-
lem with the major impact falling on the latter
three Lower  Basin states and on Mexico. High
salinity adversely affects more than eleven mil-
lion people and over 900,000 acres of highly de-
veloped irrigated land in these three states.
  The  current  average annual salinity of the
Colorado River ranges from about 50  parts of
salt per million parts of water (ppm) at its head-
waters in Colorado and Wyoming to 870 ppm at
Imperial Dam, the lowest diversion point in the
United  States. Twenty-seven miles below Impe-
rial Dam the average annual salinity of the river
water used by Mexico at the Northerly  Interna-
tional Boundary is about 1,160 ppm.  Figure 1
shows the Colorado River  Basin, key  features
and the salinity at selected stations for the 1963-
67 period.


            Sources of Salinity
  Salinity concentrations are affected by two
basic processes: (1) salt loading, i.e., the pickup
of mineral  salts from  natural and man-made
                                           19

-------
20    MANAGING IRRIGATED AGRICULTURE
Numbers ore weighted overage  salinity
for calendar years 1963-1967 In

parts per million.


   \

      \
                NEVADA
                \
                    \
                  \
                                                             NEW  fOMK MIVCII


                                                             W  V 1 (  M  I N  «
                                                               FLAMING GORGE DAM
                                                                                  —
                                                                       NAVAJO DAM
                                   HOOVCM MM

                                    701
                                    AKE MOJAVE
                                     MAVASU LAKE


                                     PARKER
                                                                                       DAMS
                     HOO/ER DAAteC*\/
                                   \maifj      «77
                     DAVIS  DA
                        IMPERIA^
                              'CALIFORNIA
                                                                      so  11 y   ao   "}°   'y

                                                                          SCALE IN  MILES
                   Figure 1: Salinity at Selected Stations within Colorado River Basin

-------
                                                                    SALINITY CONTROL
                                          21
sources, and (2) salt concentrating, the loss  of
water from the  system through evaporation,
plant transpiration, and out-of-basin export. Ir-
rigation  is  a  major  contributor to  salinity
through both of the above processes.  Water is
consumptively used by irrigation, thus  reducing
the amount of water in the river and concentrat-
ing the salinity.  In addition, the water that is ap-
plied to the  farms in excess of consumptive use
requirements not only returns to the  river the
salts contained in all of the diverted water, but it
also  picks up additional salts from the soils  to
which it is  applied and from  the  underlying
formations.
  In its natural state, the Colorado River's dis-
solved mineral load averaged about 5.1 million
tons per year at  a concentration of about 250
ppm at Lee Ferry, Arizona, which is located just
a few miles downstream of Glen Canyon Dam.
About 80 percent of the naturally occurring salts
come from diffuse sources and about 20 percent
are from natural  point sources, such as mineral
springs and exposed formations  of highly solu-
ble minerals.
  When  pioneering farmers  moved  into the
Colorado River Basin, their activities increased
the salinity of the Colorado River. The  main use
of the river water was and  still is for irrigation.
About two  million acre-feet  of water is  con-
sumed by  irrigation  in the  Upper Colorado
River Basin, the area  upstream of Lee Ferry,
Arizona. Studies show that about 3.5 million
tons of salts are added annually to the Colorado
River by return flows from  irrigated areas in the
Upper Basin. This amount will  increase in the
future as additional water is  used on projects
now being developed. It has been found that  ir-
rigated  areas  in  the  Upper Basin have salt
pickup rates that vary from as  little as 0.1 ton
per acre per year to as much as 8.5 tons per acre
per year.  Many  projects  where irrigation has
been practiced  continuously  for over  80  years
still have pickup rates in excess  of six tons per
acre  per year; and it appears that in the absence
of  any  corrective  measures,  irrigation  salt
pickup will continue at approximately  the same
levels.
  Municipal and industrial uses in  the Upper
Basin consumptively use approximately 50,000
acre-feet  annually  and   add  approximately
130,000 tons of dissolved salts to the river  each
year, a minor amount compared with contribu-
tions from irrigated agriculture. Other sources
of increased  salinity in  the  Upper  Basin  are:
(1) the export of approximately  500,000 acre-
feet a year of low salinity water from the head-
waters of the Colorado River System to adjoin-
ing basins in the states of Utah, Colorado, and
New Mexico, further depleting the flow  down-
stream but  diminishing  only slightly the  salt
load,  and (2) evaporation from storage  reser-
voirs  of approximately  300,000  acre-feet per
annum.  Both of these items are expected to in-
crease in the future.
   In recent years, the salinity at Lee Ferry has
averaged about 610 ppm, considerably  higher
than the 250 ppm that prevailed prior to  the ad-
vent of modern man.
   The effect of man's activities in  causing in-
creased  salinity  also takes place  in  the  Lower
Basin. Over 900,000 acre-feet of water per year
are evaporated from Lakes Mead,  Mohave, and
Havasu.  Water losses between Davis and Impe-
rial Dams from phreatophytes and the  river's
surface   average  about  600,000  acre-feet per
year.  Recent measurements  indicate that  irri-
gated agriculture in the Lower Basin, primarily
from  Parker and Palo Verde Valleys, adds  over
200,000 tons of salts per year to the river by the
salt pickup process.


          Future Increases in Salinity
   Several major water  projects are  now under
construction in the Upper Colorado River Basin,
and others have been authorized and are await-
ing funds for design and construction. In addi-
tion to  these authorized projects, there are  a
number  of additional projects for which feasibil-
ity reports are being made, and others that  have
been identified as potential projects. The Colo-
rado River Board of California (CRB), the Envi-
ronmental Protection Agency (EPA), the  U.S.
Bureau of Reclamation (USER), and the partici-
pants in the recently completed  Type I Com-
prehensive Framework Studies of the Water Re-
sources  Council all analyzed the impact of fu-
ture basin development.  Although  the  projec-
tions  differed, all investigators agreed that un-
less salinity  control measures are undertaken,
the river's salinity will substantially increase in
the future.

-------
22    MANAGING IRRIGATED AGRICULTURE
  The CRB has concluded that, without salinity
control measures, the average salinity at Parker
Dam  (the point on the river where The Metro-
politan Water  District of Southern  California
now diverts and where the authorized Central
Arizona  Project will  divert) will increase from
the present  average of 740 ppm to  over  1,100
ppm by year 2000. At Imperial Dam, where the
bulk of diversions  are made for agricultural use
in  California   and Arizona,  the  comparable
values are 870 ppm at present and  over  1,300
ppm by the turn of the century. Approximately
75 percent of the increases will be due to Upper
Basin developments and 25 percent will be due
to  Lower  Basin  developments. It  should  be
noted that the above values are annual averages
and that  higher seasonal values will occur.

        Deleterious Impact of Salinity
  The deleterious  impact of high salinity water
on  municipal and industrial water users is felt in
a number of  ways, including high  soap con-
sumption, formation of scale in heating vessels,
corrosive  attack   on  distribution   pipelines,
plumbing systems  and  appliances, and  added
water treatment and conditioning costs. A num-
ber of estimates have been made as  to the cost
in dollars to municipal, industrial and agricul-
tural users due to increases in salinity. However,
it is difficult to place precise dollar values on the
above detriments.  Our estimate is that salinity
will cause increasing damages and unless action
is taken  to  control salinity, the total economic
impact in the Lower Colorado River Basin in the
United  States  due to  projected  increases in
Colorado River salinity would be in the order of
$45 to $60  million per year by the turn of the
century,

      Possible Ways to Reduce Salinity
  The importation of large quantities of low sa-
linity water would  hlep to reduce the river's sa-
linity. In the past  decade considerable publicity
has been given to augmenting the Colorado
River Basin by importing water to the  Basin
from  the Columbia River Basin or  from other
northerly river basins. Other  proposals have
been  to  desalt seawater  or  geothermal  water
and transport the  water to the Colorado River.
Consideration  has also been given  to weather
modification as a  method of  increasing the
river's runoff.
  The  1968 Colorado River Basin Project Act
placed  a ten-year moratorium on the study of
any plan to import water  from the Columbia
River Basin. To date there has been no develop-
ment of large-scale desalting plants  and, in any
event, the cost would be extremely high.  Geo-
thermal water development and weather modi-
fication are in very preliminary stages of devel-
opment. There is no indication that  large quan-
tities of water  will be developed from any of the
above sources  in the near future.
  The most promising approach of reducing sa-
linity is through the construction of projects and
implementation of measures which  would pre-
vent large  quantities of salt from entering the
river system.


           Salinity Control Projects
  Studies recently completed by the EPA and
its predecessors, by  the  USBR and the  CRB
have pointed the direction for a basinwide  salin-
ity control program. The EPA has  identified a
number of specific projects and, together with
the USBR, has conducted limited  reconnais-
sance level studies on them. The most promising
of these are now under feasibility investigation
by the  USBR.
  Preliminary  information on  the  USBR's in-
vestigations has been released and is referrrd to
herein. The USBR, in its appraisal  of potential
salinity control measures,  has  set its estimates
of salt  removal at lower levels than had the EPA
in its  studies, primarily reflecting  therein the
presently  foreseeable  volume  of salt removal.
The salinity control projects are categorized as
point-source control, diffuse-source control, and
irrigation-source control projects.

Point-Source Control
  These projects are in areas of localized salt
contributions  such as mineral  springs or out-
crops of soluble formations adjacent to or under-
lying surface water sources. They are generally
more susceptible of control than other types of
salinity control projects.  Potential point-source
control projects  are  located  at  La  Verkin
Springs,   Utah;  Glenwood-Dotsero  Springs,
Crystal Geyser, and Paradox Valley, Colorado;

-------
                                                                   SALINITY CONTROL
                                         23
and  Littlefield  Springs and Blue Springs, Ari-
zona.
  In crossing the Paradox Valley, Colorado, the
Dolores  River  picks  up a  large mineral  load
from underlying highly soluble  ground forma-
tions. The proposed  project  would include a
dam and regulatory reservoir  above the valley,
and an impervious channel through the valley to
minimize the interchange between stream  flow
and  ground  water, thereby reducing the salt
load by about  200,000 tons annually. A rough
estimate  of the cost of this project  is between
$25 to $35 million.
  Four projects would control mineral springs:
one  at Glenwood  and Dotsero  Springs, Colo-
rado; one at La Verkin Springs, Utah; and one
at Littlefield Springs, Arizona; and the fourth at
Blue  Springs,  Arizona.  The  USER estimates
that  about 40 percent of the flow of Glenwood
and  Dotsero Springs,  near  the  town of Glen-
wood Springs,  could  be controlled  and  about
200,000 tons of salt annually could be removed
by means of a  desalting plant. Brines could be
disposed  of by  injection into  deep  wells. This
project is estimated to  have  a  construction cost
of $40 to $60 million.
  The largest natural point source in the Colo-
rado River Basin is Blue  Springs on the Little
Colorado  River. One  proposal to eliminate its
500,000-ton annual salt contribution would con-
sist of a low dam downstream of Blue Springs to
intercept  the springs' flow, pumpint plants and
conveyance facilities to transport the water to
Flagstaff,  and  a  desalting  plant  at Flagstaff
where the desalted water could  be sold. The cost
of this project has not been estimated.
  The earlier studies by the EPA indicated a po-
tential for removing about 1,200,000 tons of salt
annually  from  the Colorado  River Basin by
means of all of the above point-source control
projects' except  Littlefield Springs and Crystal
Geyser. The USBR's  preliminary estimate for
salt removal by point-source control is 745,000
tons  annually.

Diffuse-Source  Control
  Diffuse-source control projects are designed
to control salt contributions from a larger areal
extent than point-source control projects. There-
fore,  these projects have  not  been sufficiently
studied to  formulate  even  tentative plans or
rough cost estimates. The basic concept is to se-
lectively remove the saline (over 1,500 ppm) low
flows from  a stream and  to  bypass  the high
flows. The low flows would be desalted or evap-
orated.  The  irrigated areas  on these streams
would also be investigated to determine if water
system  improvement and management  pro-
grams and improved irrigation scheduling might
contribute toward reduction of the salt load suf-
ficiently to justify feasibility studies.
  The  potential  diffuse source control  proj-
ects  which appear  to provide most favorable
prospects for salinity control  include the Price,
San  Rafael,  and Dirty  Devil Rivers  in Utah;
McElmo Creek in Colorado; and Big Sandy
Creek in Wyoming. Potential tonnages of salt
removed by means of diffuse-source  control
projects have been estimated at 390,000 tons
annually by the USER. Earlier studies by the
EPA had identified about 550,000 tons annually
to be removed by "Irrigation Improvement" fea-
tures that could be classified as diffuse-source
control  projects.  These included  all  of the
above-mentioned areas  except McElmo Creek.
They also included Henry's Fork and Duchesne
Rivers in Utah.

Irrigation-Source Control Projects
  These projects would consist of (1) improve-
ments of irrigation scheduling and  management
and  (2) construction of  improvements to exist-
ing water conveyance systems.
  The  irrigation scheduling  and  management
activities would be aimed  at reducing the vol-
ume  of deep  percolation  to  ground water
through   saline  geologic  formations,   which
would reduce the salt load being introduced into
the  Colorado River under present conditions.
The  water saved under  this program would be-
come available for other uses. Also,  increased
net  returns  would result  to  the  irrigators
through greater yields,  improved  crop  quality
and  lower production costs. The porposed tech-
nique is to schedule times  and amounts of wa-
ter to be applied to crops by analyzing the actual
soil moisture deficiencies in each area, probably
through use of computers. By  developing an ac-
curate water budget and giving operational con-
siderations to the root zone reservoir, a  basis is
provided for attaining  high irrigation efficien-
cies. This might prove to be one of the least ex-
pensive  of the various methods for reducing sa-
linity levels.

-------
24    MANAGING IRRIGATED AGRICULTURE
  Another method for controlling salts from ir-
rigation  sources  is  construction  of improve-
ments to existing water conveyance systems to
reduce losses, increase operating efficiency, and
further curtail salt loading into the river. Studies
will be made of irrigation systems to identify the
structural measures needed.
  The  irrigation areas chosen for this program
are the Grand Valley and Lower Gunnison Ba-
sins in Colorado, the Uintah Basin in Utah, the
Colorado River Indian Reservation in Arizona,
and the Palo Verde Irrigation District lands in
California. Potential tonnages of salt  removal
from the Colorado River Basin by means of irri-
gation-source control  projects have been  esti-
mated at 680,000 tons annually by the Bureau of
Reclamation. The earlier  studies by EPA  esti-
mated  1,100,000 tons annually of salt could be
removed by "Irrigation Improvement" projects,
which amount does not include the 550,000 tons
previously  mentioned under diffuse-source con-
trol projects.

      Effect of Salinity Control Projects
  Preliminary  reconnaissance studies  by the
EPA have  indicated projects with a potential to
remove approximately 2.9 million tons annually
from the river.  The USBR has selected projects
that it has estimated would  remove approxi-
mately 1.8 million tons annually for further re-
connaissance study  and  feasibility investiga-
tion.
  Based  upon projected water development in
the basin,  it is  necessary to  remove approxi-
mately 3 million tons of salt  from the  river in
order to  keep salinity at or below present levels.
The USBR has proposed large scale desalting
and weather modification to supplement salin-
ity control projects.  If these do not prove to be
successful, it will be necessary to develop addi-
tional salinity control projects in order  to meet
the objective of preventing  future  increases in
salinity.

           Recent Significant Actions
  In  August 1970  the Colorado  River Board
issued its report entitled "Need for Controlling
Salinity of the Colorado River." The report con-
cluded that salinity is a basinwide problem and
recommended  that  the  key  policy objective
should  be to maintain  salinity  of the Lower
Colorado River at or near present levels. To ac-
complish  this  objective,  it  was  also  recom-
mended  that  construction of  salinity control
projects should be scheduled for completion co-
incident with  the completion of water projects
that  would  increase salinity in the Colorado
River Basin.
  The EPA has  also given strong support for
constructing  salinity control projects. Its  No-
vember  1971   report,  "The Mineral Quality
Problem in the  Colorado River  Basin," con-
tained two  recommendations  that were  very
similar  to  the  two recommendations,  stated
above, of the Colorado  River  Board's August
1970 Salinity Report. However,  another EPA re-
commendation called for the establishment of
specific  numerical  criteria   at   key   points
throughout the basin by January 1, 1973.
  During February  15-17, 1972, the EPA  held
the "Seventh  Session of the Conference in the
Matter of Pollution of the Interstate Waters of
the Colorado  River  and  Its Tributaries" in Las
Vegas, Nevada. The conferees  consist of repre-
sentatives  from  the water  pollution  control
agencies  in  each of the  seven  Colorado River
Basin  States,  of an EPA representative from
each of the two regions of the EPA that encom-
pass the  Colorado River Basin, and of an EPA
representative from Washington, D.C.
  Testimony  at  the conference was  presented
by federal, state, and local agencies, as well as
by individual citizens. Commissioner Armstrong
of the USBR testified as the representative of
the Secretary of the Interior  and presented  a
ten-year Colorado  River Water   Quality Im-
provement Program. The program includes data
collection,  feasibility studies, construction, and
implementation of  projects. The  projects in-
clude the ones mentioned in the section on Sa-
linity  Control Projects.  The USBR  estimated
that the  costs would be $300 million or more.
  At the conclusion of the conference, the  state
conferees unanimously agreed upon a resolution
containing the following major points:
                            (
  1.  A salinity policy be adopted that  would
     have as  its  objective the  maintenance of
     salinity concentrations at  or  below levels
     presently found in the lower main stem.
  2.  Salinity is a basinwide problem that needs
     to be solved while the Upper Basin contin-

-------
                                                                    SALINITY CONTROL
                                          25
     ues  to develop  its compact-apportioned
     water.
  3.  To guard against rise in salinity, the salin-
     ity control program should be accelerated.
     The USBR,  assisted by the EPA  and the
     Office of Saline Water,  should have the
     primary responsibility for implementing
     the program.
  4.  Adoption of numerical salinity criteria be
     deferred until the potential effectiveness
     of salinity  control  measures is  better
     known.
  EPA agreed with the resolution in principle,
but felt it needed to be strenghened by adding to
it  a  statement  that  specific  tonnages  of salts
would be removed from the Colorado River at
specific times  by  a  salinity  control program.
EPA requested the USBR to prepare a proposal
for an  accelerated salinity control program and
recessed the conference  to await the USBR re-
port. In order to  accelerate the salinity control
program, it has been proposed that  demonstra-
tion  projects be  constructed  for the different
types of salinity control projects.
CONCLUSIONS
  A key policy objective which has been ex-
pressed by  the seven  Colorado  River Basin
States and by concerned federal agencies is that
salinity of the Lower Colorado River should be
maintained at or near present levels. Salinity is
recognized as a basinwide problem, and a major
salinity control  program has  been initiated to
meet the above objective. Mexico is also vitally
interested since the  Colorado River  salinity
problem is one of the major controversial issues
between the two countries.

REFERENCES
  1. Colorado River Board of California, "Need
for  Controlling Salinity of the Colorado River,"
August 1970.
  2. Gindler, Burton J., and Holburt, Myron B.,
"Water Salinity Problems: Approaches to Legal
and Engineering Solutions," July 1969, Natural
Resources Journal, University of New Mexico,
School of Law.
  3. Holburt, Myron B., and Valantine, Vernon
E.,  "Present and Future Salinity of the Colorado
River,"  Journal  of the  Hydraulics Division,
American Society of Civil Engineers, Vol. 98,
No. HY 3, Proc. Paper 8769, March 1972.
  4. International Boundary and Water Com-
mission,  U.S. Section, Reports on Operations
for Solution of  the Colorado River  Salinity
Problem Under Minute No.  218 for 1967, 1968,
1969, 1970, and 1971.
  5. Irelan, Burdge, "Salinity of Surface Water
in  the  Lower  Colorado  River — Salton  Sea
Area,"  U.S.  Geological  Survey  Professional
Paper 486-E, 1971.
  6. U.S. Department of the Interior, Bureau of
Reclamation, "Report on Operation of the Main
Outlet Drain Extension, Minute No. 218  with
Mexico," and Supplements Nos. 1-6.
  7. U.S. Department of the Interior, "Quality
of Water, Colorado River Basin," Progress Re-
ports Nos.  1, 2, 3, 4, 5, and 6 (1963, 1965, 1967,
1969, 1970, and 1971.

-------
       Economic  Impact of Salinity  Control
               in  The Colorado  River  Basin
                               L. RUSSELL FREEMAN
                                  Pacific Islands Office
                           Environmental Protection Agency
ABSTRACT
  Studies of the economic effects of salinity in
Colorado River water by the Colorado River Ba-
sin Water Quality Control Project are discussed.
The studies describe the  effects of salinity on
municipal,  industrial,  and  agricultural water
users.  They also describe  the secondary effects
of pollution costs on  the regional  economy.
Methods of analysis are described and summary
cost data are presented.

INTRODUCTION
  The Colorado River is situated in the south-
western United States  and extends 1,400 miles
from  the  Continental  Divide in the Rocky
Mountains of north central Colorado to the Gulf
of Califprnia. Its river basin covers an area of
244,000  square  miles,  approximately  one-
twelfth of the continental United States. The
Colorado River  Basin includes parts of seven
states; Arizona,  California, Colorado, Nevada,
New Mexico,  Utah  and Wyoming. About one
percent of the Basin drains lands in Mexico.
  The River rises on the east slope of the Conti-
nental Divide above 13,000 feet, and flows gen-
erally southwestward, leaving the United States
at an elevation of about 100 feet above sea level.
The Basin  is composed of a complex of rugged
mountains, high plateaus, deep canyons, deserts
and plains.
  Climatic extremes  in the Basin  range from
hot and arid in the desert areas to cold and hu-
mid in the mountain ranges. Precipitation is
largely controlled  by elevation  and the  oro-
graphic effects of mountain ranges. At low ele-
vations or in the rain shadow of coastal moun-
tain ranges, one finds desert areas which may
receive as little as six inches of precipitation an-
nually, while high mountain areas  may receive
more than sixty inches.
  An  average of about 200 million acre-feet of
water  a year is  provided by precipitation in the
Colorado River Basin. All but about 18 million
acre-feet of this is returned to the  atmosphere
by evapotranspiration. Most of the stream flows
originate  in the high forest areas where heavy
snowpacks accumulate and where evapotranspi-
ration is low. Thus, approximately two-thirds of
the runoff is produced from about six percent of
the mountainous area of the Basin.
  Stream flows fluctuate  widely from year to
year and season to season, and many reservoirs
have been constructed to make water available
for local  needs, for  exports and for meeting
downstream obligations. The usable capacity of
the Basin reservoirs  is about 62 million acre-
feet.
                                          27

-------
28
MANAGING IRRIGATED AGRICULTURE
                                    QUALITY
                               DEGRADATION
                      Additional  Water
                                        Additional  Water
Alter  Cropping
                      Not Available
                No Action
                                                 Available
Can't  Use
                                                                 Can Use
  E
  l_
  o
  .*•-
  c
  D
      Pattern
       No n U n if o rm
                                No  Action
                                 O   O
                                             (Soil Problems)
 Modify  Soil
 Use
Water
                                             O
                                                          0
Do Not
Use Water
                      Cost Based
                      ^t	tem
                        D e c is io n
                                  G)
                      ALTERNATIVE  DIAGRAM
                1.  NO ACTION. LEADS  TO YIELD  REDUCTION.

               2.  INCREASE  PURCHASE  AND  USE  OF WATER.

               3.  MAINTAIN SAME TOTAL USE ON  FEWER ACRES

                   LEADS TO  MORE USE/ACRE.

                          a. Remove  Least Profitable  Crops ($/A-ft.)

                          b. Remove  Least Tolerant Crops  (Yield in S/ppm)

                          c. Remove  Crops in Proportion to Acreage

               4.  INCREASED  PURCHASE OF  SOIL  CONDITIONERS.

   Figure 1:  Irrigation Water User Alternatives for Offsetting the Effects of Mineral Quality Degradation
   The Colorado River system carries a large salt
 burden (dissolved solids) contributed by a vari-
 ety of natural and man-made sources. Depletion
 of stream flow by natural evapotranspiration
 and consumptive use of water for municipal, in-
 dustrial, and agricultural uses reduces the vol-
 ume of water available for dilution of this salt
 burden.  As a result,  salinity concentrations in
 the  lower  river system exceed desirable levels
                                         and are approaching critical levels for some wa-
                                         ter uses. Future water resource and economic
                                         developments will increase  stream flow deple-
                                         tions and add salt which in turn will result in
                                         higher salinity concentrations.

                                                     Methods of Analysis
                                           The economic analysis proceeded along the
                                         following logic sequence: 1. select a target year

-------
                                                                      ECONOMIC IMPACT
                                           29
and  develop "normative" descriptive data on
the economy for that year; 2. select a water use
location and evaluate the direct economic ef-
fects of increasing salinity on water users at that
location; 3.  repeat step 2 for all water use loca-
tions significantly affected by salinity; 4. select a
base "state" of  the system and an  incremental
"change of state"  for analysis;  5. using data
from steps 2 and 3, determine the cost  to each
water user associated with the salinity change at
his location, and form the total (or aggregate)
direct cost to all users. This  total  represents the
direct cost associated with the change of state
selected for analysis;  6. using the  inter-industry
tables, calculate the indirect  economic effect  of
the change in salinity; 7.  repeat steps 5 and 6 for
other changes in the state of  the system over the
probable range of future mineral  quality condi-
tions; 8. repeat steps  1-7 for other target years.
   Each of the foregoing steps will be discussed
briefly.  The focus will be on critical points of the
analysis. For a  detailed  discussion, one should
turn to  references listed in the bibliography.
Step 1  - Selection of target years and normative
conditions:
   The salinity  studies described  here were be-
gun in  the  early 1960's.  The most recent data
available  at that  time   were  from 1960  and
earlier.  A fairly standard 50-year time horizon
was chosen for the  studies; and intermediate
years were  selected at  10-year  intervals as a
matter  of convenience in projection. Since sev-
eral years elapsed during the course of salinity
studies, the  base year was updated to 1965 prior
to release of the final project report.
Step 2 - Selection of water use locations and de-
termination of direct user costs:
   One should distinguish between cause and ef-
fect  in  discussing salinity. Causes of increasing
salinity are found  throughout   the Colorado
River.  However, dramatic increases in  salinity
levels are not  expected in  mountainous areas
where most of the upper basin diversions occur.
Therefore, analysis of the economic impact  of
mineral pollution was focused on  the water ser-
vice area of the  lower Colorado River.
   The Project attempted to identify practical al-
ternatives available to each major type of water
use in the Basin. The cost of these alternatives
was compared and one was selected for the pur-
pose of analyzing basinwide  effects.  It should be
emphasized  that, even  though one alternative
was selected for use in the analysis, the Project
did not determine whether such  an alternative
would  likely be implemented in  practice. This
analysis was carried out for the purpose of mea-
suring   the   economic  effect  of   anticipated
changes in a physical system.
  Initial evaluations of possible salinity effects
on  Basin water uses  indicated  that  adverse
physical  effects would essentially be limited to
municipal, industrial, and agricultural uses. Ag-
ricultural use is selected here to demonstrate
the analytical  technique.  An  alternative dia-
gram is shown in Figure  1. This diagram indi-
cates that there are four major alternatives to be
analyzed, and that there are three variations to
one of the alternatives.
  One  alternative available to an irrigator when
the quality of his water supply degrades is to
take no  remedial action. The  direct economic
loss in this case is a loss in crop yield. The per-
cent of optimum yield is calculated, for base wa-
ter  quality,  and  for the incremental change in
quality. The economic value of the difference in
crop yield associated with the two water quality
levels represents the direct cost of pollution for
this method  of analysis. This  calculation is
based  on experimental crop salt tolerance data
from the  U.S. salinity Laboratory at Riverside,
California. These data  are shown  in summary
form in Figure 2.
  A value for soil saturation extract is  calcu-
lated from data on irrigation water quality, con-
sumptive  water use  and irrigation practices in
the water use area.  Where consumptive  water
use, amount of applied water, and irrigation wa-
ter quality are known, the amount and quality of
drainage water may be calculated from the U.S.
Salinity Laboratory "Handbook 60" formulas10.
  The  mean conductivity of the soil column is
taken as the irrigation and drainage water qual-
ity values. This value is divided by two to con-
vert it  to the equivalent "soil saturation extract"
value,  in  accordance with  the recommendation
of the  Salinity Laboratory.
  Gross crop values are used in this calculation
since it is assumed that there are  no changes in
farm management practices, so that pre-harvest
costs remain unchanged.
  As an alternative to the no-action condition,
the irrigator could  choose to maintain existing

-------
30
MANAGING IRRIGATED AGRICULTURE
                        RELATIVE  YIELD  IN  PERCENT
               CARROTS

                 CITRU
                ONIONS

         ELONS-GRARES

               LETTUCE
            SWEET CORN
           CORN SILAGE
             GRAIN CORN
                ALFALFA
              HAY 8 SO.
        3 GRAIN SORGHUM
                 SILAGE
           SUGAR  BEETS
         BARLEY a OATS
       {BERMUDA GRASS
    O>
                   Figure 2: Salt Tolerance of Major Crops Grown in Study Areas

-------
                                                                     ECONOMIC IMPACT
                                          31
yields as  the  quality of his water  supply de-
grades. This is shown as alternative No. 2 in fig-
ure 1. This could be accomplished by applying
more water in order to increase the leaching
fraction.  In  this case  the major problem lies in
determining how much additional water would
be required.  Hill and Scofield  considered  this
problem  and set forth the concept of "equiva-
lent service"30 as one  method of calculating the
amount of water  required.  Equivalent  service
requires reduction in the concentration of the
drainage water in order to offset the increase in
concentration of dissolved solids in the applied
water.  This  concept calls for a substantial in-
crease  in  the leaching fraction in order to im-
prove the drainage water quality. However, Ag-
riculture  Department personnel who collabo-
rated in the study suggested1 using a more con-
servative  assumption (e.g.,  one which showed
less  cost  associated  with increasing pollution
levels). This  assumption was that maintaining
the quality  of the water percolating from the
field would result  in maintaining constant crop
yields.
  To determine the penalty costs associated
with quality degradation, it  is necessary to ac-
count for the  increase in conveyance losses and
to determine the dollar value of this quantity of
water.  The costs added by the need for extra la-
bor, more fertilizer, and additional drainage as-
sociated with the application of more irrigation
water were also included. The latter costs were
quite substantial, sometimes equal to the value
of the water itself.
   Costs were determined from studies of forms
in the water use area conducted for EPA by the
Economic Research  Service. Some data  were
collected  directly  by project economists.  Two
values were compared to arrive at a cost for ad-
ditional  water.  One  was the "residual  value".
This is a measure of the average amount an irri-
gator could afford to pay for water based on its
utility. This value is computed by developing a
farm budget; accounting for all income and for
all expenses  except  water  cost;  and  then by

 'The staff of the Economic Research Service, USDA,
 collaborated in the investigations; results were also
 reviewed with Dr. Bernstein and the staff of the Sa-
 linity Laboratory, as well as Dr. Vaughn Hansen and
 Mr. Raymond Hill who served as consultants to the
 Project.
dividing the excess income  by the total water
use to arrive at a maximum  average-water-cost.
Theoretically,  an  irrigator  would  pay  more
than this amount for small incremental amounts
of water, since he is paying less for the majority
of his supply.
  Another method of estimating water  value
was to estimate the cost of developing a supple-
mental supply.  Developmental costs normally
exceeded the residual values, so that the  latter
were used for the purposes of analysis.
  Note that  detailed cost data are not  given
here since they varied from area to area. Cost
data developed for  this project  are available
from the files of EPA. When more irrigation wa-
ter is  applied,  additional  labor  costs  are in-
curred; additional amounts of fertilizer are lost;
and   additional  drainage  facilities may  be
needed. In the case of additional labor costs it
was assumed that irrigators would tend to de-
crease the interval between irrigations. In order
to maximize  the  interval,  an irrigator would
apply  the  maximum amount  of  water which
could be benefically used during each irrigation.
It follows that any substantial increased  water
requirement would necessitate more irrigations
per year. The cost of additional irrigations was
assessed at $2 per foot of required additional ir-
rigation water in excess of three inches. The ini-
tial three inches  of  additional water  was as-
sessed no labor cost. The foregoing values are
based on an application of six inches per irriga-
tion at a cost of approximately $1 per irrigation.
   Fertilizer losses were calculated according to
a first-order chemical solution reaction  equa-
tion.  For  convenience,  this equation  was ex-
pressed in the form:

           L  =L0 Pn;

where:     L0 = Quantity of fertilizer presently
                applied,
           L  = Quantity of  fertilizer  remain-
                ing,
           P  = Percent  of fertilizer  remain-
                ing   under   present   condi-
                tions, and
           n  = Ratio   of   the    volume   of
                drainage with degraded  wa-
                ter  supply  to present volume
                of drainage.

-------
32     MANAGING IRRIGATED AGRICULTURE
From this equation, the loss in nitrogen fertilizer
associated with increases in drainage water may
be  calculated.  This amount is multiplied  12
cents per pound  to establish the direct cost of
fertilizer loss.
  The third alternative shown on Figure 1 in-
volves taking a portion of the irrigator's crop
land  out of production, using water thereby
saved to increase the amount of leaching water
applied to the remaining crop acreage. For this
alternative,  the profit  that would  have been
made on the crops taken out of production is the
measure of direct cost. Three methods of reduc-
ing acreage  were investigated: (1) removal of a
portion of all crops in proportion to total acre-
age (uniform reduction), (2) removal of the least
profitable crops, and  (3) removal of  the least
salt-tolerant  crops.  These  alternatives  are
shown as "3c", "3a",  and "3b", respectively, in
figure 1.
  The first  step  in determining detriments by
the uniform acreage reduction  technique is to
calculate the leaching  water requirements as-
sociated with  a base quality and an adjusted
quality  for the  target year. The "constant qual-
ity of percolate" method is used to obtain these
leaching water quantities. The  next  step is to
calculate the number  of acres of land  which
must be removed from production to obtain this
volume of water.  It should be noted that acreage
removal is reduced to account for the fact that
no added water is needed on those acres retired
from service.
  The fourth alternative shown on figure 1 in-
volves costs for soil conditioners in addition to
those for one of the previous alternatives involv-
ing  increased  use of  water. Thus, it  is clearly
more costly.
  Costs associated with the alternatives just de-
scribed were calculated for each  major irriga-
tion water use area below Hoover Dam, and the
results  were compared. Figure 3 shows such a
comparison for a typical water use area. The se-
lective  acreage reduction  technique  was  re-
jected since agricultural economists argued that
fixed investments, farm practices, and other fac-
tors not included in the analysis would preclude
practical adoption of this alternative.
  Thus, yield  decrement was  selected as the
method of analysis giving the best estimate of
the cost to irrigators of increasing levels of salin-
ity in irrigation water supplies.
  Similar analyses were made  for  municipal
and industrial water uses. Although irrigation is
the major water use in the basin, municipal use
is  also  significant.  Hardness, which is  closely
correlated with dissolved solids content,  creates
undesirable  effects  in domestic  uses. Three al-
ternative methods  of evaluating the economic
impact  upon municipal  uses were examined:
(1) the   acceptance  of  undesirable   effects,
(2) home water  softening, and (3) central soft-
ening. The  costs associated with each of these
alternatives  were  calculated  for each  major
municipality in  the geographic  region studied.
Except for the Colorado  River Aqueduct service
area,  the alternative resulting in highest penalty
cost  was  home  softening followed by  central
softening and acceptance of increased pollution
in that order. This ranking undoubtedly  reflects
the fact that the  latter method, based upon
soap-wastage, does not account for all the costs
incurred.    Nevertheless,   the   soap-wastage
method was selected as the measure of munici-
pal water use penalty costs for all municipal en-
tities  except those that actually have  central
softening  plants. These  are the Metropolitan
Water District of Southern California and the
city of Calexico, California, in the Southern Cal-
ifornia water service area. For these municipali-
ties the central  softening method was used to
calculate pollution costs.
  Steps 4 and 5 of the logic sequence involved
aggregating  salinity costs over Colorado River
Basin. The  large number and variety of manu-
facturing industries in the major centers of wa-
ter use, especially in Southern California2, made
it impracticable to  attempt an evaluation of sa-
linity effects on process waters within the scope
of this study. In addition, process water use gen-
erally falls  into one of  two categories:  (1) use
that  is insensitive to small incremental changes
in mineral  concentraction, or (2) use that re-
quires a completely  demineralized  supply. In
either case the effect of changes in mineral qual-
ity over the  range of concentrations expected to
prevail is considered to be unmeasurable.

2Bureau of the Census, "Statistical Abstract of the
United States,"  1964,  lists 17,  665 manufacturing
plants in the Los Angeles-San Diego metropolitan
areas. (Reference No. 32).

-------
H
U
|
y
S
O
o
U
w
















Iff
...J
„-«
0
o
„,*
z,
o
r~
i
i
as
'-*


2.0-
1.8-

1.6-

1.4-
1,2-
1,0-
,8
.6-
.4-
.2-
0
"2 -
~4 -
         O » O
600
                 LEGEND

EQUIVALENT SERVICE
CONSTANT QUALITY OF PERCOLATE f LABOR AND FERTILIZER
UNIFORM ACREAGE  REDUCTION 4- LABOR AND  FERTILIZER
YIELD  DECREMENT
SELECTIVE ACREAGE REDUCTION -1- LABOR AND FERTILIZER
             800
      Q5JAUTY 0?

-------
34     MANAGING IRRIGATED AGRICULTURE
  In view of these considerations the Project fo-
cused  upon cooling and boiler  feed waters.
Thus, the industrial penalty costs derived by the
Project's  study  are  somewhat  understated.
There  is no doubt, however,  that the included
costs cover  a  major portion  of the fresh water
used in manufacturing. In the United States as
a whole over  74 percent  of all industrial fresh
water is used in cooling and boiler feed,33 and in
the State of  California  67 percent  is so  em-
ployed.34 A  survey of water use in the chemical
and   metallurgical  complex  at  Henderson,
Nevada, made in  August 1964, by the Nevada
Department of Public Health35 showed 80 per-
cent of the  water to be employed for cooling,
four percent for boiler feed, and the  remaining
16 percent for processing, sanitary, and miscel-
laneous purposes.
  There are two pertinent types of cooling  and
boiler  systems. High-pressure boilers  require a
demineralized  supply; thus, they are  not sensi-
tive to minor changes in plant  intake water qual-
ity. Similarly,  specially designed cooling towers
can  accept  brackish or  highly saline waters;
thus, they are  insensitive to water quality. Low-
pressure boilers and cooling towers on fresh wa-
ter systems,  however, can tolerate only a limited
concentration  of dissolved mineral constituents.
These  systems, therefore, are directly affected
by changes  in mineral quality. This analysis  is
based entirely on an evaluation of penalty costs
associated with  Colorado River water used in
sensitive systems.
  Material balance in these systems establishes
the quantity of discharge  water required for any
level of water use, intake quality, and  system
tolerance.  Increasing   concentrations  of  dis-
solved  mineral  constituents  in the  feed  water
necessitates  an  increase  in  the  discharge re-
quirement and thus  an increase in the water in-
take requirement, in order to prevent salt accu-
mulation within the  system. The increase in wa-
ter use, the 1960 cost of water, and  feed-water
treatment costs  were used in  the  assessment of
industrial penalty costs.
  The direct economic costs of mineral quality
degradation may  be summarized in two basic
forms, total direct costs and penalty costs. Total
direct  costs  incurred for a given salinity level re-
sult from increases in  salinity concentrations
above the threshold levels of water  uses. Pen-
alty costs are the differences between total di-
rect  costs for a  given salinity level and for a
specified base  level.  They  represent the mar-
ginal costs  of increases in salinity concentra-
tions above base  conditions.
  Detailed  economic studies  were  aimed  at
evaluating  penalty  costs in order to provide a
basis for assessing the economic impact of pre-
dicted  future increases in salinity.  Water qual-
ity, water use patterns, and economic conditions
existing in  1960  were selected as  base condi-
tions. Water use  and economic conditions pro-
jected  for the target years  1980  and 2010 and
predictions  of future  salinity  concentrations
were utilized to estimate total direct costs in the
future. Direct panelty costs were then computed
from differences  in total direct costs. These  di-
rect  penalty costs  are summarized by type of
water use and by study area in Table 1. The  in-
direct and  total penalty costs, also presented in
the table, are discussed below.

Indirect Economic Effects
  Because  of the interdependence of numerous
economic activities, there are indirect effects on
the regional economy stemming from the direct
economic impact of  salinity upon water users.
These  effects, termed indirect penalty costs, can
be  determined if the interdependency of eco-
nomic activities are known.
  The Project's  economic  base  study  investi-
gated the interdependence of various categories
of economic activity or sectors. These interde-
pendent relationships, in the form of  transac-
tions tables, were  quantified for  1960 condi-
tions,  and  were  projected for the target years
1980 and  2010.  A digital  computer program
known as  an "input-output model" was devel-
oped to follow changes  affecting any given  in-
dustry through a chain of transactions in order
to identify  secondary or indirect effects on  the
economy stemming  from  the direct economic
costs of salinity. Application of the model to
evaluate indirect penalty costs  is  discussed in
Appendix  B,  Chapter V of the Project report.
The indirect  penalty costs predicted  by  the
model are summarized in Table  1.

 Total Penalty Costs
  Total penalty  costs represent the total mar-
ginal costs of increases  in salinity concentra-
tions above base  conditions. They are the sum of

-------
                                                                       ECONOMIC IMPACT
                                                      35
                                            TABLE 1

                                    Summary of Penalty Costs
         Location and Water Use
           1980                         2010
Direct    Indirect    Total    Direct   Indirect    Total
Penalty    Penalty   Penalty   Penalty   Penalty   Penalty
 Cost      Cost     Cost      Cost      Cost     Cost
                                                  ($1,000 Annually)*
                                      ($1,000 Annually)*
Lower Main Stem Study Area
  Irrigation Agriculture
  Industrial
  Municipal

               Sub-Total

Southern California Study Area
  Irrigated Agriculture
  Industrial
  Municipal

               Sub-Total

Gila Study Area
  Irrigated Agriculture
  Industrial
  Municipal
 1,096
   107
   275

 1,478
 4,617
   56
 1,347

 6,020
  765
    4
   14

  783
2,447
    3
  305

2,755
1,861
  111
  289

2,261
7,064
  59
1,652
2,424
 410
 779
2,237
   15
   39
3,613     2,291
4,661
 425
 818

5,904
         6,195     16,267
             5        108
           507      2,746
8,775     12,414     6,707
                                246
                               125
                  19,121
                              371
Sub-Total
Total
—
7,498
— —
3,538 11,036
246
16,273
125
9,123
371
25,396
* - 1960 Dollars

direct penalty costs incurred by water users and
indirect penalty costs suffered  by the regional
economy. Total penalty costs are  also summa-
rized in Table 1.
  Several  conclusions  can  be   drawn  from
Table 1.
  1. The majority of the penalty costs (an aver-
     age of 82 percent) will result from water
     use  for irrigated agriculture. This fact may
     be attributed  to the heavy  utilization  of
     Colorado  River water for irrigation along
     the  Lower  Colorado  River  and in  the
     Southern California area.
  2. Over three-fourths of the penalty costs will
     be incurred in the Southern California wa-
     ter service area. These costs will result pri-
     marily from agricultural use  in the Impe-
     rial and Coachella Valleys, and municipal
     and  industrial uses in the coastal metro-
     politan  areas.
            3.  Penalty costs in the Gila study area will be
               minor and will not occur until  after 1980
               when water deliveries to the Central Ari-
               zona Project begin.  (It was assumed that
               all Central Arizona Project water would be
               utilized for agricultural purposes).
            It  should be noted that the penalty costs do
          not represent the total economic impact of salin-
          ity; only the incremental increases in costs  re-
          sulting from rising salinity levels. There were
          costs being incurred by water users in 1960 as a
          result of salinity levels existing in 1960.  These
          costs would continue in the future if salinity lev-
          els remained at the 1960 base conditions.
            Penalty costs cannot be used for evaluation of
          the economic impact of basinwide salinity con-
          trols, especially when reductions in salinity con-
          centrations below  1.960 base levels are consid-
          ered. For  this reason, estimates of total salinity
          detriments have been prepared utilizing the ba-

-------
36     MANAGING IRRIGATED AGRICULTURE
sic information developed for penalty cost eval-
uations. These estimates, in the form of empiri-
cal   relationships  between  salinity  levels  at
Hoover Dam and salinity detriments, are shown
graphically for various target years in Figure 4.
  Hoover Dam is a key point on the Colorado
River system. Water quality  at  most points of
use in the Lower Basin and Southern California
water service area may be directly related to sa-
linity levels at  Hoover  Dam. Modifications of
salt  loads contributed by  sources  located  up-
stream from Hoover Dam also directly affect sa-
linity levels at this location. Salinity concentra-
tions at Hoover Dam were, therefore, utilized as
a water  quality index  to  which all  economic
evaluations were keyed.
  Total salinity detriments are the sum of direct
costs to water  users (including direct  penalty
costs) and indirect penalty costs.  It should be
noted that the salinity detriments are expressed
in terms of  1970  dollars.  It was necessary to
modify the basic data utilized in evaluating pen-
alty costs (expressed in terms of 1960 dollars) in
order to make  the salinity detriments compati-
ble with current estimates of salinity  manage-
ment costs discussed in the next section.
  Using the projected salinity levels for Hoover
Dam shown  in Table 1  and the salinity detri-
ment functions of Figure 4, it is possible to com-
pare the total  econonic detriments of salinity
under various conditions of water use and re-
source development. Under 1960 conditions, the
annual  economic  impact  of salinity was  esti-
mated to total $9.5 million. It is estimated that
present salinity detriments have increased to an
annual total of $15.5 million. If water resources
development proceeds as proposed and no salin-
ity controls are implemented, it is estimated that
average annual economic detriments (1970 dol-
lars) would increase to $27.7 million in 1980 and
$50.5 million in 2010. If future water resources
development is limited to those projects  now
under construction, estimated annual economic
detriments would increase to $21 million in  1980
and  $29 million in 2010. It should be noted that
all data in this  analysis are based on the period
of record 1942-61 adjusted  to 1960 conditions of
water use.

CONCLUSION
  The purpose  of  project activities was to eval-
uate the magnitude of the  pollution problem in
     40 t	
      600     700      800      900     ,1000     1100
      TOTAL DISSOLVED SOLIDS CO SCEHTUTI3* HS/L AT HOOVEI DAM
          Figure 4: Salinity Detriments

the Colorado  River;  and to stimulate  positive
pollution  abatement  action.  From what  has
been reported here,  one might get the conclu-
sion that the project was mostly an academic ex-
ercise. To avoid leaving that impression, let me
emphasize that  concurrently project staff was
working with  other State and Federal agencies
to identify the sources and/or causes of pollu-
tion and the costs of pollution control. An exten-
sive demonstration effort has been underway
here in  Grand Valley aimed at determining the
feasibility of controlling  salinity  from  irrigated
areas. Much research on control techniques has
been sponsored at universities throughout the
Western states. The Bureau of Reclamation has
done  reconnaissance  level studies of the feasi-
bility of  controlling natural salinity   sources.
These efforts culminated in the most recent ses-
sion of the Colorado River Conference held a
few weeks ago  in Las Vegas, Nevada. At this
conference, an objective of maintaining salinity
at or below existing  levels in the critical lower
reaches of the River was adopted,  and a pro-
gram was announced by  the Bureau of Recla-
mation  which would achieve this objective. The
cooperative effort to document the  costs of in-
creasing pollution was a prime factor in mobiliz-
ing support for this pollution abatement effort.

-------
                                                                    ECONOMIC IMPACT
                                          37
  The  Project's  activities  included  sampling
programs and field investigations to determine
the location and magnitude of agricultural, mu-
nicipal, industrial and  natural sources of min-
eral pollution throughout the Basin. Engineer-
ing studies involved the  assessment of adverse
effects on water users caused by saline pollution
and evaluation of technical  possibilities for con-
trolling or minimizing mineral pollution from
agricultural, municipal, industrial  and natural
sources.
  In the economic studies, the level and pattern
of economic  activity in  the Basin were devel-
oped for 1965 and projected for 1980 and 2010.
These data were used to construct inter-industry
input-output  tables for each major subbasin of
the Colorado for each  of these  three  target
years. The input-output tables were then used to
determine secondary effects of water pollution
costs on the regional economy. In  all economic
and engineering projections for the years 1980
and 2010, full cognizance was taken of legal al-
location of Colorado River water and of pro-
posed development of water resource projects
by various construction agencies.
   This paper focuses upon presentation of some
of the unconventional  phases of the economic
and technical analyses; upon presentation  and
interpretation of the results; and upon the cur-
rent status of recommendations for salinity con-
trol in direct economic losses to water users and
indirect economic losses to the regional econ-
omy.  Unless  salinity controls are implemented,
future increases in salinity concentrations  will
seriously affect water use patterns and may re-
sult in large economic losses.
   In 1963, based upon recommendations of con-
ferees from the seven basin states and the Fed-
eral government, the Colorado River Basin Wa-
ter Quality  Control  Project  began  detailed
studies of the mineral quality problem  in the
Colorado River Basin.  Mineral quality, com-
monly known as salinity,  is a complex  Basin-
wide problem that is becoming increasingly im-
portant to users of Colorado River water. Due to
the nature, extent,  and  impact of the salinity
problem, the Project extended certain of its ac-
tivities over the entire Colorado River Basin and
the Southern California water service area.
   The more  significant findings and data from
the Project's  salinity studies and  related perti-
nent information are summarized in the report
entitled, "The Mineral Quality Problem in the
Colorado  River Basin."  Detailed  information
pertaining to the methodology and findings of
the Project's salinity studies are presented in
three  appendices to  that report—Appendix A,
"Natural and Man-Made conditions Affecting
Mineral Quality," Appendix B.  "Physical  and
Economic Impacts,"  and  Appendix C, "Salinity
Control and Management Aspects."

BIBLIOGRAPHY
   1.  U.S. Department  of  Health, Education
and Welfare, Public  Health Service, "Drinking
Water  Standards,   1962,"  U.S.  Government
Printing Office, Washington, D.C., 1962.
   2.  State  of California, "Water  Quality  Cri-
teria," Second  Edition,  State  Water Quality
Control Board, Sacramento, California, 1963.
   3.  Sawyer, Clair  N.,  "Chemistry for Sani-
tary Engineers," McGraw-Hill Book Company,
New York, 1960.
   4.  U.S. Department of Commerce, Business
and  Defense Services Administration, August
20, 1964. Colorado River Project file correspon-
dence, Region VIII, U.S. Public Health Service,
Denver, Colorado.
   5.  State of California, Department of Water
Resources, Sacramento, California, "Water Use
by Manufacturing  Industries   in  California,
1957-1959," Bulletin  No.  124, April 1964.
   6.  Nordel, Eskel, "Water Treatment for In-
dustrial and Other Uses," Second Edition, Rein-
hold Publishing  Corporation, New York, 1961.
   7.  Betz Laboratories, Inc., "Betz Handbook
of Industrial Water Conditioning," Betz Labora-
tories, Inc., Philadelphia, Pa., 1962.
   8.  American  Water Works Association, New
York, "Water Quality and Treatment," Second
Edition, 1951.
   9.  Anonymous, "Progress Report  of Com-
mittee on Quality Tolerances of Water for In-
dustrial Uses,"  Journal  New England Water
Works Association, Volume 54,  1940.
   10.  U.S. Department of Agriculture, U.S. Sa-
linity  Laboratory, "Saline and Alkali Soils," Ag-
riculture Handbook No. 60, 1954.
   11.  U.S. Department  of  Health, Education
and Welfare, Public  Health Service, "Proceed-
ings of the National Conference on Water Pollu-

-------
38    MANAGING IRRIGATED AGRICULTURE
tion, December 12-14, 1960," U.S. Government
Printing Office, Washington, B.C., 1961.
  12. Eaton,  F. M.,  "Deficiency, Toxicity and
Accumulation of Boron in Plants," Journal Ag-
ricultural  Research,  Volume  69,  Illustration
1944.
  13. Eaton,  F. M.,  "Significance of Carbon-
ates in Irrigation Waters," Soil Science, Volume
69, 1950.
  14. Federal Water  Pollution Control Admin-
istration, U.S. Department of the Interior, "Wa-
ter Quality Criteria," Report of the National
Technical Advisory Committee to the Secretary
of the Interior, April  1, 1968.
  15. Ellis, M. M., "Detection and Measure-
ment of Stream Pollution," U.S. Fish and Wild-
life Service Bulletin No. 22,  1936.
  16. State of California, "The Ecology of the
Salton Sea, California, in Relation to the Sport-
Fishery," Department of Fish and Game, Sacra-
mento, California, 1961.  *
  17. State of California, Santa Ana Regional
Water Pollution Control Board, Santa  Ana,
California, Resolutions 54-4, January 28,  1955,
and 57-7, May 24, 1957.
  18. State of California, "The California Wa-
ter Plan," Department of Water Resources, Sac-
ramento, California, Bulletin No. 3, May,  1957.
  19. State of California, "Investigation of Al-
ternative Aqueduct Systems to Serve Southern
California," Appendix B, Bulletin No. 78, De-
partment of Water Resources, Sacramento, Cal-
ifornia, January, 1959.
  20. U.S. Department  of the Interior, "Qual-
ity of Water, Colorado  River Basin," January,
1965.
  21. Bureau of Reclamation, U.S. Department
of the Interior,  "Pacific  Southwest  Water
Plan," August, 1963.
  22. Geological Survey, U.S. Department of
the  Interior,  "Surface Water  Records  of  Ari-
zona," 1961.
  23. U.S.  Department  of  the Interior, "Sup-
plemental Information  Report on the  Central
Arizona Project," January, 1964.
  24. Eldridge, Edward  F., "Return Irrigation
Water—Characteristics and Effects," U.S. De-
partment  of Health,  Education and Welfare,
Public Health  Service,  Region IX,  Portland,
Oregon, May 1, 1960.
  25. Garnsey,  et al,  "Past and Probable Fu-
ture Variations in Stream Flow in the Upper
Colorado  River—Part I," Bureau of Economic
Research,  University  of  Colorado,  Boulder,
Colorado, October, 1961.
  26. Tipton and Kalmbach, Inc., Upper Colo-
rado River Commission, "Water Supplies of the
Colorado River—Part I," July, 1965.
  27. lorns, W. V., Hembree, C. H., and  Oak-
land, G. L.,  "1965 Water Resources of the Up-
per Colorado River Basin—Technical Reports,"
U.S. Geological Survey Professional Papers 441
and 442.
  28. U.S. Department of the Interior, "Quality
of Water,  Colorado River Basin," Progress Re-
port No. 4, January, 1969.
  29. U.S. Department of the Interior, "Qual-
ity of Water, Colorado River Basin," Progress
Report No. 3, January, 1967.
  30. Hill, Raymond  A.,  "Leaching Require-
ments in Irrigation," Reprint from Journal of
the Irrigation and Drainage Division, American
Society of Civil Engineers,  March, 1961.
  31. Soil and Water Conservation  Research
Division, "Salt  Tolerance  of Plants," Agricul-
tural Research Service, U.S. Department of Ag-
riculture, December, 1964.
  32. Bureau of the Census, U.S. Department
of  Commerce,  "Statistical  Abstract of the
United States," 1964.
  33. Bureau of the Census, U.S. Department
of Commerce, "Industrial Water Use, 1958 Cen-
sus of Manufacturers," U.S. Government Print-
ing Office, 1961.
  34. State of California, "Water Use by Man-
ufacturing Industries in California,  1957-1959,"
Bulletin 124, Department of Water Resources,
Sacramento, California, April, 1964.
  35. State  of Nevada, Department of Health
and  Welfare, unpublished  report on survey of
industrial water uses, August,  1964.
  36. American Boiler and Affiliated Industries
Manufacturers'  Association, "Limits for Boiler
Water  Concentrations in Units with a Steam
Drum."

-------
                                                                  ECONOMIC IMPACT
                                         39
  37.  Howson,  L. R.,  "Economics  of Water
Softening," Journal  of the American Water
Works Association, February, 1962.
  38.  Leasure, J.  William, "Polulation Projec-
tions  for the Three  Lower  Subbasins of the
Colorado River Basin," San Diego State  Col-
lege, San Diego, California, June, 1964.
  39.  Anonymous, "Statistics on Population of
Households," Journal of the American Water
Works Association, 42:904, September, 1950.
  40.  U.S.  Bureau of the Census, "Census of
Manufacturers: 1963 Water Use in Manufactur-
ing," Subject Report  MC63(1)-10,  Washington,
D.C.,  U.S. Government Printing Office, 1966.
  41.  Select Committee on National Water Re-
sources, U.S. Senate, Water Supply and  De-
mand, Committee Print No. 32, 1960.
  42.  Study by the National Aluminate Corpo-
ration reported in the State  of California, De-
partment of Water Resources, Sacramento, Cal-
ifornia, "Investigation of  Alternative Aqueduct
Systems to  Serve Southern California,"  Ap-
pendix B, Part Five.
  43.  University of Arizona, "The Quality  of
Arizona Irrigation Water," Report  223,  Sep-
tember, 1964.

  44.  University of Arizona, "Quality of Ari-
zona Domestic Waters," Report 217, November,
1963.

  45.  Miernyk, William H., "The Element  of
Input-Output Analyses,"  New York Random
House, 1965, Library of Congress No. 65-23339.

  46.  University of Colorado, "An Interindus-
try Analysis of the Colorado River Basin in 1960
with Projections to 1980 and 2010," Edited by
Bernard Udis,  Associate Professor of Econom-
ics,  Boulder, Colorado, June, 1968.

  47.  Vincent, James R. and Russell, James D.,
"Alternatives for Salinity  Management in the
Colorado  River,"  Water  Resources Bulletin,
Volume 7, No. 4, August, 1971, pp. 856-866.

  48.  United States Environmental  Protection
Agency, "The  Mineral Quality Problem in the
Colorado River," GPO 790-010, 1971.

-------
Soil,  Water and  Cropping  Management  for

   Successful  Agriculture  in  Imperial Valley

                              ARNOLD J. MACKENZIE
                     Imperial Valley Conservation Research Center
                          USD A Agricultural Research Service
                                  Brawley, California
ABSTRACT
  Colorado River water, containing about 900
ppm dissolved salts, is the only source of water
for Imperial Valley agriculture and is creating
salinity problems in that area. Imperial Valley
farmers  are successfully overcoming these sa-
linity problems, however, by  using soil, water
and cropping practices that minimize or elimi-
nate the detrimental effects of salt on crop pro-
duction. Underground tile drains installed in the
area have enabled a favorable salt balance to be
attained for the Valley since 1949. Profile modi-
fication to improve water movement through the
stratified soils is practiced to increase leaching
of salts. Salt tolerant crops are grown to mini-
mize the detrimental effects of salinity on crop
growth and yields.
  Successful agriculture in the Imperial Valley
is shown by the continued yield increases in the
major crops grown. Farmable acreage also has
not changed, indicating no loss of farmland be-
cause of increasing river water salinity.
  The existence of agriculture in the Imperial
Valley of California is dependent entirely upon
Colorado River  irrigation  water.  The annual
rainfall in the Valley is only 2'/£ inches and agri-
culture would not be possible were it not for the
irrigation water delivered  by the All  Ameri-
can Canal from the Colorado River. Because of
this source  of irrigation water, the desert has
been transformed into one of the most produc-
tive intensively irrigated areas of the world. In
1971,  the gross  agricultural production of the
one-half million acre  area  was valued at over
300 million dollars—an all time  high for the
area's 70-year agricultural history.
  While Colorado River water is the lifeblood of
Imperial Valley  agriculture, it is also a poten-
tial nemesis because of its  salinity.  Presently,
Colorado River water  delivered to Imperial val-
ley contains more than  900 ppm of dissolved
salts and the salt content is gradually increasing.
River salinity is projected (2) to reach 1000 ppm
by the year 1980 and 1340 ppm by the year 2000
if no upriver salinity control measures are taken.
  The water  presently delivered to Imperial
Valley brings with it annually approximately 3!4
million tons of salts. To prevent an accumulation
of these salts in  Imperial Valley soils, it is nec-
essary to continually leach  them from the soils
so that the salts will not have a harmful effect
upon  crop production. Controlling or reducing
salinity levels in  the soil depends upon manage-
ment practices that involve amount, frequency,
and methods of applying water. Leveling, plow-
                                          41

-------
42
MANAGING IRRIGATED AGRICULTURE
ing, cultivation, drainage, and other farm opera-
tions that affect soil structure affect water man-
agement and salt balance. Imperial Valley farm-
ers are utilizing many practices to minimize the
salt problem in that area. This discussion will re-
view the various soil management,  water man-
agement and cropping practices utilized by them
to assure permanent and successful farming de-
spite an ever present and increasing irrigation
water salinity problem.

           Irrigation Water Quality
  The  chemical composition of the  Colorado
River water delivered currently to Imperial Val-
ley is shown in Tabte 1. The water may be eval-
uated according to salinity, sodium, boron and
bicarbonate hazards using the classifications of
irrigation waters recommended by the U.S.  Sa-
linity Laboratory Staff (4) and Wilcox (5). At the
present  time the Colorado River water has an
electrical conductivity (EC) of 1400 micromho/
cm and  is classified as high salinity water. Water
in this classification  should not be used on soil


                  TABLE 1

  Chemical analysis of Colorado River water
         delivered to Imperial Valley

Electrical Conductivity,
  EC*103@25°C                        1.4
Boron, ppm                             0.2
Sodium, % of Total Cations              43
pH                                    8.1

                        Equivalents   Parts
                        Per Million Per Million
Dissolved solids
(1.25 tons/ac. ft.)
Calcium
Magnesium
Sodium
Potassium
Carbonate
Bicarbonate
Sulfate
Chloride
Nitrate

5.19
2.88
6.22
0.14
0
3.06
7.41
4.03
0.02

915
104
35
143
5
0
187
356
143
1
 Samples taken at All American Canal, Drop 1-Aver-
 age of biweekly samples taken during 1971.
                                          with restricted drainage,  and  even with ade-
                                          quate drainage, special management for salin-
                                          ity control may be required. Plants with good
                                          salt tolerance  should be selected if grown with
                                          this water.
                                            In the Salinity  Laboratory water classifica-
                                          tion, waters are divided  into four  classes with
                                          respect to the sodium hazard: low, medium, high
                                          and very high, depending upon the values for so-
                                          dium adsorption ratio (SAR) and EC. Colorado
                                          River water with a SAR value of 3.2 falls into
                                          the low sodium water  class. This water can be
                                          used for irrigation on almost all soils with little
                                          danger  of  the development of a  sodium-soil
                                          problem. However, sodium-sensitive crops may
                                          accumulate injurious amounts of sodium in  the
                                          leaves.
                                            The occurrence of boron in toxic concentra-
                                          tions in certain irrigation waters makes it nec-
                                          essary to consider this constituent  when assess-
                                          ing the  quality of water. The boron concentra-
                                          tion  in Colorado River water is low (0.2 ppm)
                                          and offers no  problem for Imperial Valley agri-
                                          culture.
                                            Bicarbonate  in irrigation water  is important
                                          because of its  tendency to precipitate calcium,
                                          and to some extent magnesium, in the soil solu-
                                          tion as normal carbonates. As calcium and mag-
                                          nesium are lost from a water, the relative pro-
                                          portion  of sodium is increased with an attendant
                                          increase in the sodium hazard.  This hazard is
                                          evaluated in terms of the residual sodium car-
                                          bonate  as  proposed by Eaton (3).  Colorado
                                          River water contains an excess of calcium and
                                          magnesium ions in relation to carbonate and bi-
                                          carbonate ions and therefore does not have a re-
                                          sidual sodium carbonate problem.

                                                      Imperial Valley Soils
                                            The water quality characteristics of Colorado
                                          River water establishes that the major problem
                                          related to its use is its high and increasing salin-
                                          ity content. Use of such water requires soil man-
                                          agement practices designed for salinity control.
                                          The control practices required depend upon  the
                                          soils to which the water is applied. Imperial Val-
                                          ley soils need  salinity control practices because
                                          of their unique chemical and physical proper-
                                          ties.
                                            Imperial Valley soils  are alluvial and are com-
                                          posed of highly stratified  Colorado River  de-

-------
                                             SOIL, WATER AND CROPPING MANAGEMENT    43
posits, largely  from  mixed  sedimentary rock
parent material. The variations in the soils are
mostly due to textural differences caused by the
manner and sequence in which the alluvial ma-
terial was deposited.  In  general, there are no
gravel and sand waterbearing strata and the dif-
fuse  groundwater is highly saline.  Stratum  of
any one type of soil does not extend over large
areas,  but  occurs  as small lenses or pockets.
These  stratified  soils  are more than  a mile  in
depth in some areas.
  The Imperial and Holtville series  are by far
the most widespread cultivated soils  covering a
little more than 75% of the  area. Other series in-
clude the Meloland and Indio  series. The  soils
are calcareous  (pH of 8.0) and are inherently
low  in organic  matter and nitrogen.  Nitrogen
fertilizer is needed for maximum  production of
non-leguminous crops. Phosphorus is needed for
legumes and winter vegetable  crops. Other nu-
trients (K, Ca, Mg, S and the various trace ele-
ments) are available  in sufficient quantities in
the soil or irrigation water and are no problem at
the present time.
  The soils  have little structure except for the
alternating horizontal textural layers in the pro-
files. The clay fraction is predominately mont-
morillonitic  and the  soils  generally  have low
water  intake and  low  hydraulic  conductivity
characteristics.  Impervious  clay  layers in the
profiles further  inhibit water movement through
the profiles for salt control.
  Imperial  Valley soils have the characteristics
as defined for saline soils and are managed ac-
cordingly to prevent  salt accumulation and  to
minimize plant growth inhibition by salt. By def-
inition (4) the electrical conductivity of a satura-
tion extract (ECe) of a saline soil is greater than
4 mmho/cm and the exchangeable sodium per-
centage (ESP) is less than  15.  The salts in the
saline  Valley  soils  are  mainly  neutral  salts,
such as the chlorides and sulfates of sodium, cal-
cium and magnesium. Sodium seldom comprises
more than one-half of the  soluble cations. Re-
stricted plant growth is almost directly related to
the total salt concentration of  the soil solution
and  is largely independent  of the kind of salts
present. If drainage is adequate and the excess
salts are removed  by leaching, saline  soils can
approach  the   classification  of  normal  soils.
Combinations of soil, water and cropping prac-
tices have  been established in Imperial Valley
for this purpose.

           Soil and Water Practices
  During the early years of Imperial Valley de-
velopment, it became very apparent that natural
drainage was not adequate for salt control after
the productivity of many thousands of acres was
seriously affected by rising water tables and in-
creased salinity. In 1922, the Imperial Irrigation
District  started  a system  of  open  drainage
ditches that emptied drainage waters into  the
New and Alamo Rivers which in turn flowed in-
to the Salton  Sea.  Over  1400 miles of these
drains are in use today.
  Installation  of underground farm  tile drain
systems began in 1929 to augment drainage by
open  ditches.  Underground  tile drains were
needed to remove water and salt during leaching
operations, to maintain ground water at a safe
level during normal irrigation, and to prevent
water and  salt from returning to the surface
from below. There are now about 16,800 miles
of drain tile  installed  in  the Valley, serving
370,000 acres. Underground drain systems have
been successful in helping to maintain the favor-
able salt balance attained in 1949.
  Correct water management practices in com-
bination  with drain  tile are necessary for salt
control. Sufficient water must be applied for the
water  requirements  of each  crop as  well as
enough extra water for leaching salts down
through the soil and out of the root  area. The
fraction of the  irrigation  water that  must be
leached through the root zone to control salinity
has been defined  as the leaching requirement
(LR) and is expressed as the ratio of the EC of
the applied irrigation  water to the EC of  the
drainage  water required to  maintain the soil sa-
linity at a specified level (4). To maintain an av-
erage root zone EC of 8 in  Imperial Valley soils
to which Colorado River  water is applied  re-
quires about 17 percent excess  water for leach-
ing. At the  present time this amount of leaching
on  the average is being accomplished as indi-
cated  by the Imperial Irrigation District posi-
tive salt balance data.
  Equally  important to the  quantity  of water
necessary for crop production and salinity con-
trol is the uniformity of water application. Good
land leveling assures uniform surface  flooding

-------
44     MANAGING IRRIGATED AGRICULTURE
and the equal water distribution needed to pre-
vent salt accumulation on high spots. Land level-
ing practices vary depending upon soil type, but
generally,  land in Imperial Valley is  graded to
between a 1- to 2-foot slope per thousand feet of
distance.  Fields are  landplaned frequently to
assure good grade control and surface condition
for uniform water application.
  Where water movement through the soil is re-
stricted  by soil strata that prevent leaching of
salts,  deep tillage  operations are performed to
modify  profiles. Chiseling, slip plowing or deep
moldboard plowing operations to depths  of 4 to
5 feet are performed to break up impervious soil
layers and improve water penetration in the pro-
file.
  Water losses by  seepage from canals and farm
ditches cause  some salinity problems in Imperial
Valley.  To minimize this problem and  to  im-
prove  water  conveyance,  canals  and  farm
ditches  are being lined with concrete and seep-
age from the  main canals is being recovered for
re-use (6). A  program of concrete lining of lat-
erals initiated in 1954 by the Imperial Irrigation
District in cooperation with landowners  has re-
sulted in the concrete lining of over 500 miles of
laterals. Almost all farm lateral ditches are con-
crete lined at the present time.
  Furrow-irrigated crops are  cultivated using
various  planting and bed shaping techniques be-
cause of  high soil  salinity.  Usually as water
moves by capillarity up into  a bed during the
first irrigation, salts are dissolved from  the soil
and carried along with the water so that the top
of the bed becomes very salty. With single-row
flat-beds,  soil salinity as low as 4 mmho/cm in
the top  soil at the  time of bedding may result in
excessive salinity  in the seedbed without special
control. Double row beds  and sloping beds (1)
are used  to  prevent the accumulation  of salt
around  the seed. Good stands are obtained with
these types of beds even where  the  salinity in
the topsoil exceeds 8 mmho/cm. Alternate  row
irrigation  with single-row flat-beds  is also help-
ful in getting good  stands where  soil salinity
problems  exist.
  Sprinkler application of water has proven to
be very helpful for improving crop germination
and reducing problems related to salinity in
planting beds. Many acres of winter vegetable
stands are being established now by a single ini-
tial sprinkler irrigation which leaches salt from
the surface soil where the seeds germinate. Af-
ter emergence and for succeeding irrigations the
crops are irrigated by conventional furrow meth-
ods.
  More frequent  irrigations than normally nec-
essary  for evapotranspiration requirements are
applied to Imperial Valley crops to lessen the ef-
fects of soil salinity on crop growth. As water is
removed from the soil by plant use and evapora-
tion, most of the salt is left behind. The longer
the time between irrigations, the saltier the re-
maining  soil moisture becomes, and the more
plants  suffer  because of salinity and water short-
age. By irrigating more often,  the soil water is
kept from becoming too salty  and the harmful
effects of the salt on plant growth can be mini-
mized.

              Cropping Practices
   Crops vary in  salt  tolerance. For  some vari-
ties both crop growth  and  yields are reduced
considerably  by the presence of salts in the soil.
The salt tolerance of many crops  has been ap-
praised (4) and they have been grouped accord-
ing to their  relative salt tolerance under man-
agement practices that are  customarily em-
ployed when they are  grown  under  irrigation
agriculture.  Because of the  relatively high level
of soluble salts in the soils and irrigation water,
only moderately tolerant and tolerant crops are
grown by Imperial Valley farmers. Salt tolerant
and moderately salt tolerant crops such  as cot-
ton, sugar beets, barley, wheat, alfalfa, lettuce,
cantalopes, and sorghum accounted for almost
all of  the farmable acreage in Imperial Valley
for the year  1971.
   Cultivation of some crops such as alfalfa may
result  in soil salinity  increases  after several
years because of the water management prac-
tices required for the crop do not allow sufficient
excess water to be applied  for adequate leach-
ing. In such  cases as these the  areas are rotated
with other crops such as sugar beets or lettuce to
which  excess water may be applied for leaching
without causing harm to the crop.
   The success of  Imperial Valley  agriculture
in overcoming the hazards of saline  irrigation
water is shown by the continued yield increases
for the area's major crops (Figure 1).  The yield
of barley grain for Imperial Valley, for example,

-------
                                             SOIL, WATER AND CROPPING MANAGEMENT
                                           45
s  5
^  4
•v

|r  2

i  '
   0

ISQO
en 800
^ 7OO
       ALRVLFA HAT
                             BARLEY GRAIN
                           COLORADO RIVER WATER
                56  98
                       6O  62
                       YEAR
                               64  66  68  TO
 Figure  1: Trends in Alfalfa and Barley  Yields,
 and Colorado River Water Salinity for Imperial
 Valley During the Period 1951-1970.

has increased at the average rate  of 0.06 tons/
acre/year during the last 18 years.  For the same
period, alfalfa hay  yields  have increased  at the
average  rate   of  0.24   tons/acre/year.   This
achievement has  been accomplished despite an
increase of 200 ppm in the salinity of the Colo-
rado  River irrigation water  during  the same
period.
  Acreage of farmable land in the Valley also is
remaining constant showing that  no areas are
being lost  because  of salinity problems.  Total
area farmable in the Imperial Valley for the last
ten years has been approximately 473,000 acres.
If drainage  continues  to  be adequate  there
should be no problem in keeping Imperial Val-
ley agriculture successful.

REFERENCES
   1. Bernstein, L., A. J. MacKenzie  and B. A.
Krantz.  1955.  The interaction of  salinity  and
planting practice on the germination of irrigated
row crops.  Soil Sci.  Soc.  Amer. Proc. 19:240-
243.
   2. Colorado  River  Board of California. 1970.
Need for controlling salinity of the  Colorado
River. August.
   3. Eaton,  F. M. 1950.  Significance  of  car-
bonates  in  irrigation waters. Soil Science 69:
123-133.
   4. U.S. Department of Agriculture, U.S. Sa-
linity Laboratory.  1954. Diagnosis and improve-
ment  of saline and  alkali  soils.  Agricultural
Handbook No. 60. February.
   5. Wilcox, L. V. 1955. Classification  and use
of irrigation waters.  U.S.  Department  of agri-
culture Circular No. 969.
   6. Willardson,   L. S., A. J. Boles,  and  H.
Bouwer.  1971.  Interceptor  drain  recovery of
canal  seepage. ASAE  Trans.  14(4):738-741.
July-August.

-------
 Irrigation  Return Flows  in Southern  Idaho
                                  DAVID L. CARTER
                        Snake River Conservation Research Center
                          Agricultural Research Service,  USD A
ABSTRACT
  The quantity and quality of irrigation return
flows from a 203,000-acre  irrigation tract in
southern Idaho were measured and compared to
the quantity and quality of the irrigation water.
Return flows for a typical water year amounted
to 929,350 acre feet  representing 64%  of the
total water input to the tract. The total salt con-
centration  in the  subsurface drainage water
was more than twice that found in the irrigation
water. The mean electrical conductivity of the
subsurface drainage water was 1040 n mhos j cm
which is as low as in some irrigation waters. The
Na+  concentration increased more than four
times as water passed through the soil and be-
came subsurface drainage water.  Similar com-
parisons  were  made for  other  cations  and
onions. Surface runoff water did not differ from
irrigation  water in chemical quality. Surface
drainage water from a 3,000-acre subregion con-
tained up to 3.06 tons of sediment per acre foot
during the midpart of the irrigation season.

INTRODUCTION
  The quantity and quality of irrigation return
flows depend  upon the quantity and quality of
water diverted for irrigation, the proportions
of the return flow from surface runoff and sub-
surface drainage and other  factors. Generally,
water passing across the land surface picks up
little salt or fertilizer elements except those as-
sociated with sediment picked  up by erosion.
Water passing through the  soil reacts chemi-
cally  with soil materials and dissolves  soluble
salts where contact  is made. As soil water is
used  in evapotranspiration, salts are concen-
trated in the water that remains in the soil or
drains from it. Thus subsurface drainage water
usually differs markedly in quality from the irri-
gation water passing through the soil influences
the quality of the subsurface drainage water and
the outputs of total salts, specific ions and fertil-
izer elements.
  Information is needed on  the quality of irri-
gation return flows under various management
systems as a basis for improving the quality of
irrigation return flows. This  paper presents in-
formation on the quality and quantity of irriga-
tion return flows from a large irrigation tract in
southern Idaho.

           Methods and Materials
  The study area (Figure 1) was a 203,000-acre
tract developed  by the Twin Falls Canal Com-
pany  about  1905. Water is  diverted from the
Snake River and delivered to farmers at a con-
                                            47

-------
                                                                                   Milner Dam
                                                                                    Diversion
THOUSAND
     SPRINGS
    SNA
       KE  RIVER
23    -^^-s. CA
26V25-19   \  24  8x
    Mud Creek  \
        9       \Cedar Draw
     *'
                                   Rock Creek   17
       Low Line C^nal
                  High Line Canal

                  SOUTH  HILLS
        Figure 1: The 203,000-acre Twin Falls Can Company Irrigation T

-------
                                                           IRRIGATION RETURN FLOWS
                                          49
stant rate of 0.5 cubic feet per second (cfs) for
each 40 acres during the irrigation season when
requested. Water is  in the canal  system from
about April  1 to November 15 each year. Canal
flows in the  early spring and late fall are consid-
erably  lower than during  the peak  irrigation
season of June, July, and August because many
farmers have crops  that do  not require early
spring or late fall irrigation.
  Soils  over most of the study area  are mod-
erately deep, uniformly textured silt loams de-
rived from calcareous, wind deposited material,
varying from a few inches to  50  feet in depth.
These soils are  generally well drained. They are
underlain by fractured basalt  to depths of sev-
eral hundred feet.  Most crops are irrigated  by
small furrows,  and infiltration rates  are fairly
high.  The mean, annual  precipitation  for the
area is 8.5 inches.
  The most  important crops grown on the tract
are alfalfa, dry beans, sugarbeets, small grain,
corn and pasture.1  Row crops  are  normally
seeded in April and May, and  harvest is usually
completed by late October.
  Soon after irrigation was initiated, high water
tables appeared in localized  areas throughout
the tract.  To alleviate this problem, the Canal
Company used two drainage methods. For the
larger areas, horizontal tunnels 4  feet wide  by
7 feet high were excavated where  test wells  in-
dicated  significant amounts of water in basalt
fractures.  These tunnels  effectively convey ex-
cess  water to natural surface  drains.  The 49th
and  final tunnel was completed  in 1948. The
other method combined shallow drainage wells
and  tile lines  in  networks  to drain  smaller
areas. The wells are connected by tile lines 31A
to 10 feet below the soil surface. The wells flow
from hydrostatic pressure, and the water is con-
veyed to natural surface drains by tile  lines. The
practice has proved effective  and is  still used
today.  All' surface and subsurface drainage  re-
turns to the Snake River which flows in a can-
yon about 500  feet deep, forming the northern
boundary of the irrigation project.
  Sampling  sites were selected throughout the
area (numbered on Figure 1) including the proj-
ect diversion  at  Milner  Dam on the Snake
River,  four  main  surface  drains, 15  drainage
tunnel outlets, five tile-relief  well network out-
lets,  and one small stream draining the South
Hills watershed. Initially, water  samples  were
collected from  each  site  at two-week intervals
and  analyzed for total soluble salts, Na+,  K+,
Ca++,  Mg++, Cl",  HCO;, S04-S, P04-P  and
NO3-N  concentrations.3 4S Water temperature
and pH were also measured at each site. Analy-
sis over a few months showed that the concen-
trations of some components were nearly con-
stant. Therefore, only PO4-P, NO3-N and total
salt concentrations and water temperature de-
terminations were continued at two-week inter-
vals. Analyses for other components were made
at four-week intervals.
  The flow rate from each tunnel and tile-relief
well network was measured. Weirs  were used
where  possible. Parshall flumes existed at  two
sites, and  the  remainder were  gaged  periodi-
cally by current metering. Water stage recording
stations were maintained on the main  surface
drains.  Existing  USGS  gaging  stations  were
utilized on Cedar Draw and Deep Creek. Flow
hydrographs were developed from the data and
the monthly flow volume computed for each
site. Hydrograph separation techniques2  were
applied to  the streamflow data to establish the
amounts of surface runoff and subsurface drain-
age from the area for a typical water year, Octo-
ber 1, 1968 through September 30, 1969.'
  A water balance for a typical water  year of
the Canal  Company, October 1, 1968  through
September 30, 1969, was computed.1 Using this
water balance along with the concentrations of
total  salts,  PO4-P   and  NO3-N,  input-output
balances for these components were calculated
and previously reported.1 Input-output balances
for specific ions and related information have
also been calculated (submitted for publication).
  The sediment concentration in the major sur-
face drain  waters was  determined by removing
the sediment from samples of  known  volume
collected at two-week intervals during the 1971
irrigation season. The sediment was removed by
allowing it to  settle and by centrifuging.  The
sediment was then dried and weighed.
            Results and Discussion
  The mean concentrations of all ionic compon-
ents except PO4-P were  higher in subsurface
drainage  water than  in  the irrigation  water
(Table 1). The concentration of each component
was nearly constant at each sampling site even
though the flow fluctuated with seasons at most

-------
50
MANAGING IRRIGATED AGRICULTURE

                                   TABLE 1
        Mean ionic concentrations at input and subsurface drainage sampling sites for the
                      water year October 1, 1968 to September 30, 1969
No.
1
2

3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

18
19
20
21
22
Name
Input Streams
Milner
Rock Creek (HL)
Drainage Tunnels
Claar
Fish Hatchery
Grossman
Nye
Tolbert
Walters
Mendini
Neyman
Galloway
Cox
Herman
Harvey
Peavy
Padget
Hankins
Tile-Relief Well Networks
Brown
Hutchinson
Kaes
Molander
Harvey
Na*
me/I
0.90
0.22

3.65
2.92
3.01
3.64
4.06
3.86
4.73
4.06
3.82
3.38
3.00
3.70
3.93
3.92
4.49

4.06
4.38
2.73
2.80
3.67
K+
me /I
0.12
0.12

0.10
0.13
0.10
0.11
0.12
0.14
0.21
0.23
0.12
0.12
0.12
0.13
0.13
0.13
0.18

0.16
0.21
0.19
0.21
0.14
Co"
me 11
2.54
1.09

5.22
3.55
3.99
3.81
4.95
4.47
3.60
5.42
3.88
3.81
5.71
3.64
3.57
3.46
4.27

4.36
4.15
5.07
4.82
3.59
Mg** Cl
me 11 me / 1
1.23 0.66
0.34 0.17

3.62 1.33
2.85
2.81
3.02
3.23
3.98
2.88
2.71
2.94
2.96
3.01
2.84
3.12
3.34
3.11
.30
.18
.54
.65
.52
.72
.55
.19
.24
.42
.37
.57
.62
.63

3.50 1.67
3.06 1.61
3.20 1.83
3.59 1.94
3.10 1.42
HCO]
me /I
3.38
1.79

6.66
5.57
5.97
5.91
6.45
6.67
7.22
7.52
7.12
6.74
7.08
6.93
6.47
6.24
6.62

6.78
7.65
6.25
6.12
6.27
N03-N
ppm
0.12
0.11

4.02
2.24
2.25
2.44
3.30
3.47
3.97
3.40
3.58
3.44
3.00
3.39
3.02
3.01
3.55

3.01
3.20
3.40
3.79
3.30
SO4-S
ppm
14
4

69
36
45
55
68
56
47
50
35
35
47
31
36
42
53

55
44
54
57
36
PO4-P
ppm
0.066
0.015

0.013
0.013
0.014
0.009
0.012
0.008
0.009
0.011
0.014
0.015
0.017
0.008
0.007
0.008
0.012

0.009
0.012
0.023
0.009
0.023
Mean, Subsurface
Drainage
3.69
0.15
4.27
3.14 1.52
6.61
3.24
48
0.012
sites.  The concentration variation among sam-
pling  sites for subsurface  drainage  water was
generally within 25% for any specific chemical
component and for total salts.
  The relative difference in  individual cation
concentrations between the  irrigation water
and the subsurface drainage water was  greatest
for Na+ and least for K+ (Table 2). For anions,
the relative increase was  greatest for  NO3-N
and least for Cl". The PO4-P concentration de-
creased so that the concentration in the drain-
age water was less than 1/5  of that in the irri-
gation water.
                                            The  water balance1 showed that 50% of the
                                          total input water (diverted water plus precipita-
                                          tion) became subsurface drainage water, 14%
                                          returned to the  Snake River as surface runoff
                                          and  evapotranspiration accounted for the re-
                                          maining 36% (Table 3). The net output of total
                                          salts amounted to approximately one  ton  per
                                          acre per year (Table 4). The portion of the water
                                          passing through the soil and the net salt output
                                          indicates that more leaching is taking place than
                                          is necessary to maintain a salt balance.6
                                            During a typical water year, about 1 acre-foot
                                          of surface runoff per acre returned to the River,

-------
                                                           IRRIGATION RETURN FLOWS
                                                               51
                                          TABLE 2

             Specific ion concentrations in irrigation and subsurface drainage waters
               and the relative concentration change that occurred as water passed
                                       through the soil
Component
Na+
K+
Ca++
Mg++
Cf
HCOJ
NO3-N
SO4-S
P04-P
C units
me/1
me/1
me/1
me/1
me/1
me/1
ppm
ppm
ppm
Cf
0.90
0.12
2.54
1.23
0.66
3.38
0.12
14.0
0.066
at
3.67
0.15
4.27
3.14
1.52
6.61
3.24
48.0
0.012
Csd
Ci
4.08
1.25
1.68
2.55
2.30
1.96
27.00
3.43
0.18
          Total Salt
   JLI mhos/cm
        460
               1,040
2.26
           Ci is the concentration in the irrigation water.
           Csd is the concentration in the subsurface drainage water.
                 TABLE 3

    The water balance for the 203,000-acre
  Twin Falls Tract for a typical water year1

                            Acre-feet      °/
Input
Diverted from Snake River
Runoff from South Hills
Precipitation
City of Twin Falls
TOTAL

1,290,100
32,000
130,000
900
1,453,000

89
2
9
—
100
Output
Evapotranspiration
Surface runoff
Subsurface drainage
            TOTAL
 523,650
 203,880
 725,470
1,453,000
 36
 14
 50
100
and subsurface drainage amounted to more than
3.5 acre-feet per acre. Thus, the total quantity of
return   flow   from   the  203,000-acre  tract
amounted to 929,350 acre-feet,  or 64%  of the
total input.
  There was a net output of all chemical com-
ponents  measured  excepting  PO4-P  and  K+
(Table 4). The net K+ input amounted  to ap-
proximately 13 pounds per acre per year which
is  significant from  the  standpoint of fertilizer
needs. More than 70% of the PO4-P entering the
tract  in the  irrigation water reacted with and
remained in the soil. The total output of PO4-P
in the drainage water was only 28 tons com-
pared to a total input of 116 tons in the irriga-
tion water and 2,580 tons applied to the land as
fertilizer. Results from this study show that ap-
plied  phosphorus fertilizers remain associated
with the soil. They are not dissolved into PO4-P
and leached away by drainage water.
  The chemical quality of the subsurface drain-
age water from the Twin  Falls tract is better
than that of irrigation water diverted at some
points from the lower reaches of the Colorado
and Rio Grande Rivers and that from  some

-------
52
MANAGING IRRIGATED AGRICULTURE
                                          TABLE 4

                    Input-output balances for measured chemical components
                     and total salts for a typical water year on 203,000 acres
Component

Na+
K+
Ca++
Mg^
Cf
HCOJ
NO3-N
S04-S
PO.-P
Total
Input
Tons
36,511
8,444
90,006
26,475
41,284
365,880
210
25,598
116
Total
Output
Tons
88,946
7,070
98,272
41,818
59,723
454,400
3,226
51,342
28
Net
Input
Tons
1,374
—
—
—
—
—
—
88
Net
Output
Tons
52,435
8,266
15,343
18,439
88,520
3,016
25,744
	
         Total salts
                         520,977
738,077
217,100
other sources. However, the total salt concen-
tration in this subsurface  drainage  water was
more than double that in the irrigation water.
The  quality of the irrigation water was  high,
and  even though  the  salt  concentration is in-
creased by irrigation, the quality of the drainage
water is still fairly  high.
  In some respects, the quality of the drainage
water  was  superior to that of  the irrigation
water. For example, the PO4-P load in the drain-
age water was less than in  the irrigation water,
and  the temperature of the drainage water was
lower than  that of the irrigation water during
midsummer when  irrigation water temperatures
exceeded 20° C.
  The NO3-N concentration was 27 times higher
in the subsurface  drainage water than in  the
irrigation water.  This concentration  increase
coupled with the water balance represents  a net
NO3-N output of 30 pounds per acre per  year.
However, the  mean  NO3-N concentration was
3.24 ppm which is well below the  10 ppm set by
the Public Health Service as a maximum for
drinking water.
  Surface runoff water quality did  not differ
from that of the irrigation water except that the
                                          runoff water had a much higher sediment con-
                                          centration during part of the irrigation season.
                                          The four main surface drains from the area con-
                                          tained both surface runoff and subsurface drain-
                                          age water. Therefore the quality of  water  in
                                          these  drains was between the  irrigation  water
                                          and subsurface drainage water qualities. For ex-
                                          ample, the electrical  conductivity of waters  in
                                          Rock Creek, Cedar Draw, Mud Creek and Deep
                                          Creek (sites 23, 24, 25  and  26 respectively)
                                          ranged from about 600 to 800ju mhos/cm during
                                          the time that water was being diverted into the
                                          canal system and from 1,000 to  1,100n mhos/ cm
                                          during the winter months when essentially all
                                          water  in these drains was from subsurface drain-
                                          age. Concentrations  of specific chemical com-
                                          ponents fluctuated similarly.
                                            The sediment load in surface  runoff  water
                                          varied through  the irrigation season. Our sedi-
                                          ment  data have not been summarized for all
                                          streams  at this date,  but preliminary results in-
                                          dicated  that the surface  runoff from a 3,000-
                                          acre subregion  contained  from about 0.25  to
                                          3.00 tons  of sediment per acre-foot during an
                                          irrigation season (Table 5). The amount of sedi-
                                          ment removed from  the 3,000 acres by erosion

-------
                                                          IRRIGATION RETURN FLOWS

                                         TABLE 5

                    Sediment load in surface runoff water from 3,000 acres
                                         53
Sampling
Date

May 25
June 2
June 15
June 29
July 13
July 26
August 10
August 24
September 8
Sediment
load
Tons/ acre- ft.
0.97
.54
.28
.97
3.06
2.88
1.92
1.11
.37
Accumulative
Sediment output
Tons
—
—
620
2,760
6,070
9,025
10,750
11,700
12,500
Accumulative
runoff
acre-ft.
—
—
1,000
2,065
3,179
4,411
5,541
6,805
8,791
during irrigation was over 12,000 tons or over 4
tons per acre per season.
  In conclusion,  irrigation return flows from a
203,000-acre irrigation tract in southern Idaho
represented nearly 2/3 of the input water. Most
of the return flow was subsurface drainage. The
total salt concentration  in subsurface drainage
water was more than twice that in the irrigation
water, but  the concentration was lower than
occurs in irrigation waters used  in some areas
of the  U.S. The  PO4-P load in the drainage
water was less than 30% of that in the irrigation
water. Surface runoff water did  not differ from
irrigation water in chemical quality, but high
sediment concentrations were measured  during
part of the irrigation season.

REFERENCES
  1. D.  L.  Carter, J. A. Bondurant and C. W.
Robbins, "Water-soluble NO3-Nitrogen, PO4-
Phosphorus and Total Salt Balances on a Large
Irrigation Tract," Soil Science Society of Amer-
ica, Proceedings 35:331-335, 1971.
  2. R. K.  Linsley, M. A. Kohler and J. L. H.
Paulhos, "Applied  Hydrology," McGraw-Hill
Book Co., New York, 1949.
  3. "Standard Methods for the Examination
of Water and Wastewater," 13 Edition, Ameri-
can Public Health Association, New York, 1970.
  4. U.S. Salinity Laboratory Staff, "Diagnosis
and  Improvement  of Saline and Alkali  Soils,"
USDA  Agricultural  Handbook  No.  60, L. A.
Richards, Editor, 1954.
  5. F. S. Watanabe and S. R. Olsen, "Test of
an Ascorbic Acid Method for Determining Phos-
phorus  in  Water and  NaHCQ  Extracts from
Soil,"  Soil Science  Society of  America, Pro-
ceedings, 29:677-678, 1965.
  6. L. V.  Wilcox, "Salt Balances and Leaching
Requirements in Irrigated Lands," USDA Tech-
nical Bulletin No. 1290, 1963.

-------
 Salinity Problems  in  the  Rio Grande  Basin
                                   JOHN W. CLARK
                            Water Resources Research Institute
                              New Mexico State University
ABSTRACT
  The Rio  Grande system carries a  large salt
burden contributed by a variety of natural and
man-made  sources.  Increased  efficiency  in
water management for agriculture  and con-
tinued economic  development will  result in
higher salinity concentrations.  Additional re-
search is necessary before adequate salinity
control and management programs can be con-
sidered.
INTRODUCTION
  The Rio Grande system carries a large salt
burden  (dissolved  solids)  contributed  by  a
variety of natural and man-made sources. De-
pletion of stream flow by consumptive use of
water for irrigation, municipal, and industrial
use and  by natural evapotranspiration reduces
the volume of water available for dilution of this
salt burden.  As a result, salinity concentrations
in portions of the system exceed critical levels
for certain water uses. Increased efficiency in
water management for agriculture and  con-
tinued economic development in the basin will
increase  stream  flow  depletions and add salt
which, in turn, will result in higher salinity con-
centrations.
  As salinity concentrations increase, additional
adverse physical effects are produced on  some
water uses. Unless adequate research informa-
tion is obtained to support appropriate salinity
management programs in the basin, future in-
creases in salinity concentrations will seriously
affect water use patterns and will result in large
economic  losses to water users and to the re-
gional economy.
  Because the Rio Grande Basin is primarily an
arid region, those few perennial streams within
the basin have  considerably  more  influence
upon  the  lives  and  livelihood  of the region's
inhabitants than any other element of the physi-
cal environment. Any alteration, modification,
or subtle change of this resource must, there-
fore, be carefully evaluated.
  The Rio Grande has its headwaters in south-
central Colorado and flows 1900 miles to the
Gulf of Mexico, see Figure 1. From Colorado it
flows  south  through New Mexico, and  then
turns  southeast at El Paso, Texas, on its way to
the Gulf.  For the 81 miles from  El Paso to Fort
Quit man, the river is the international boundary
between the Republic of Mexico and the United
States.'
  The river is generally considered to have three
regimens as far as water use is considered in the
United States: these are the river from its head
in Colorado to Fort Quitman, Texas, the Pecos
River, and the main stem from Fort Quitman to
the Gulf. The reach of the river above Fort Quit-
man is known as the Upper Basin and below
Fort Quitman as the Lower Basin.
                                           55

-------
56    MANAGING IRRIGATED AGRICULTURE
                m
                                     Figure 1: Rio Grande Basin

-------
                                                                   SALINITY PROBLEMS
                                           57
  Nearly all of the water produced in the Upper
Rio Grande is consumed  in that sub-basin.
Much  the same situation prevails in the drain-
age area of the Pecos River, the major tributary
of the Rio  Grande  in the United States. The
lower  Basin receives little flow from Colorado
or New Mexico. Because of this separation, the
three sub-basins will be discussed separately.

             Upper Rio Grande
  Physical  Description — The   Upper   Basin
drains  an area  of  about 32,000  square  miles
above  Fort  Quitman, see Figure 2.  Fort  Quit-
man, Texas, is approximately 650 miles down-
stream from the river's head  in Colorado and
about 80 miles downstream from the City of El
Paso, Texas. The Upper Basin includes parts of
Colorado and New Mexico, and very small parts
of Texas and the Republic of Mexico. More than
99 percent of  the water  comes in about equal
amounts from  Colorado and New Mexico.
  The  Upper  Basin  comprises three sub-areas
designated as the San Luis Section in Colorado,
the  Middle  Section  in New Mexico,  and the
Elephant  Butte  Project  Section  of Southern
New Mexico, including extreme West Texas and
the  adjacent river  valley in  the  Republic of
Mexico.
  San  Luis  Valley — The San Luis  Valley is a
large north-tending structural depression down-
faulted in the eastern border and surrounded
on the  west,  north and east by mountains. It is a
high flat mountain valley with an average alti-
tude of about 7,700 feet. Underlying the valley is
as much as 30,000 feet of alluvium, volcanic de-
bris, and interbedded volcanic flows and  tuffs.
Most of the  valley floor is bordered  by alluvial
fans deposited  by streams  originating in the
mountains: the most extensive of these is the Rio
Grande fan.
  The northern half of the San Luis Valley is in-
ternally drained and is referred to as  the "closed
basin." The  rest  of the valley is drained by the
Rio  Grande and  its tributaries. A  number of
small streams enter the closed  basin and  prac-
tically all water produced by these streams that
is not consumed in irrigation  is lost by evapo-
transpiration. In  addition, all water diverted
from the Rio Grande for a large irrigated acre-
age  in the closed basin and not  consumed in irri-
gation is also lost in the low portion  of the area
by nonbeneficial  evapotranspiration.  Mineral
concentrations in  the shallow groundwater  in
part of the closed basin range is nearly  14,000
mg/1.2
   Total  annual water supply to the San Luis
Valley is about 2,500,000 acre-feet. Approxi-
mately  1,500,000 acre-feet is stream flow from
snow melt in the  surrounding mountains, and
1,000,000 acre-feet is from precipitation  within
the valley. Discharges from the valley are esti-
mated at about 2,000,000 acre-feet per year by
evapotranspiration and about 500,000 acre-feet
as flow into New Mexico. The stream flow at the
state  line averages 445,000 acre-feet per year,
and the  groundwater underflow is estimated  at
55,000 acre-feet.
   Groundwater in the San Luis Valley is from
confined and unconfined aquifers which contain
at least 2 billion acre-feet of water in storage.2
The unconfined aquifer is of modern origin and
occurs almost everywhere in  the valley. Re-
charge to the unconfined aquifer  is mainly by
infiltration of applied irrigation water and leak-
age from the distribution system. Some water
percolates from the  many streams flanking the
valley and from local precipitation. Subirriga-
tion is  widely practiced in the  valley and the
phreatic  surface must be brought  very close to
the land  surface for at least part of the year.
  The valley  is arid, and a  successful agricul-
tural economy would not be possible without ir-
rigation. The main irrigated crops produced are
alfalfa, potatoes, barley, oats, hay, and pasture.
Irrigation development  began in the San Luis
Valley after 1850 and the oldest water right  in
the valley is dated 1852. Rapid and extensive
settlement occurred which utilized waters from
the Rio Grande after construction started  on the
Denver and Rio Grande Railroad through the
valley in 1879.
   The  Middle Section —The  Middle  Rio
Grande Section is a land of flat dry desert floors,
and rocky precipitous mountains in the north-
central part of New Mexico. It includes the Rio
Grande and tributary valleys from the Colorado
state line to San Marcial at the head of Elephant
Butte Reservoir, a river distance of about 270
miles.
  The Rio Grande flows into New Mexico from
a southerly extension of Colorado's San Luis
Valley through a deeply entrenched channel for

-------
58    MANAGING IRRIGATED AGRICULTURE
    UTAH•COLORADO
           NEW  MEXICO
ARIZONA
                                         COCHITI
                                          ..  9 Scta  Fe
                                      Albufauerque
                                   San  Marcial
                                    PLQB Cruces
                                    bx
                                    at
          TED    STATES	u^ LEI Paso
                                   Juairez
                                                           NEW MEXICO
                                                              TEXAS
                                                            SCALE
                                                              I I '
                                                         0 10 20 30 40 50 mi
            MEXICO
                                                     Ft, Quitman
                                                     v
                              Figure 2: Upper Rio Grande

-------
                                                                    SALINITY PROBLEMS
                                           59
some 75 miles downstream. At the mouth of this
canyon the river enters the several valleys that
make up the  Middle Section. These basins with-
in the Middle Section are aligned in a generally
southerly direction and consist of broad plains
flanked by mountains. The basins are partially
filled with  Quaternary and Tertiary sediments
known collectively as the Santa Fe Group. Irri-
gated lands lie in narrow strips from one to five
miles wide  on each side  of the river within the
basins.
  The  drainage  area between Lobatos Bridge,
Colorado, gauging station and the San Marcial
gauging station above Elephant Butte Reservoir
is approximately 20,000 square miles. This area
comprising the Middle Section contains about
175,000 acres subject to irrigation, about 38,000
of which are  on  Indian lands.1
  The  average  annual  water production  of
stream flow in the Middle Section is about 1.3
million acre-feet, and the outflow at San Marcial
averages  about 0.9 million acre-feet. The total
inflow from Colorado is about 0.5 million acre-
feet resulting in about 0.9 million acre-feet con-
sumed  in this reach.
  Groundwater is primarily from valley fill sedi-
ments that  have been deposited along tributary
streams and have filled much of the Rio Grande
trough.  These  fills  are  generally  stream-con-
nected  and are recharged mainly from stream
flow. The status of groundwater in the basin is
not well  known and  only generalized physical
properties and water bearing characteristics are
available.
  The Middle Rio Grande section is one of the
earliest areas in  the Western Hemisphere to be
occupied  by  man. Agriculture was first  prac-
ticed here by the Anasazi Indians who evolved
into the people of the Pueblo I and II eras, 700-
1050 A.D.  By 900 A.D.  irrigation was in wide
use, and 20 to 30 thousand people are estimated
to have lived in- 70 to 80 pueblos.3 A severe
drouth  between  1276 and  1300  resulted in a
large influx  of  agricultural  peoples  from the
Mesa Verde of Colorado on the Colorado River.
Subsequent irrigation developed  until  an esti-
mated 25,000 acres were in use  at the  time  of
arrival  of the first Spaniards. In  1598 Juan de
Onate built the first Spanish irrigation  ditch  in
the new world about 30 miles  north of Santa Fe.
ty 1800 A.D. over 100,000 acres were under cul-
tivation utilizing 70 diversion canals from  the
Rio Grande in this section. Rapid expansion of
irrigated agriculture occurred  during the late
1800's due to the railroads providing a transpor-
tation link to new markets, and resulted in about
230,000  acres  being under  irrigation in  the
Middle  Section and its tributaries by 1890.
  Elephant  Butte Project Section — The  Ele-
phant Butte  Project Section is  located in that
stretch  of the river from San Marcial, New
Mexico, to Fort Quitman, Texas, a distance of
about 258 river miles. The Rio Grande between
these  points contains  4,335  square  miles of
drainage area, 1,429 of which are in the Repub-
lic of Mexico. While this reach  of the river is in
two states of the United States  and in two na-
tions, it is one  economic and  hydrologic unit
separated from other populated centers of either
country by vast expanses of semi-arid deserts.
  Altitudes  range from 4,200 feet  at  Elephant
Butte Reservoir to 3,400 feet at Fort Quitman.
Like the  Middle  Section, the  Elephant  Butte
Project  Section  is a succession of valleys sepa-
rated by canyons and narrows. Of these valleys,
Rincon, Mesilla, and the El Paso Valley contain
the major irrigated areas.  The  El  Paso Valley
area southwest of the river is in  Mexico.
  The production of water in this  reach of the
river is negligible. Although there are numerous
ephemeral  tributaries,  some  of  which  have
rather large drainage areas,  there are no peren-
nial streams  tributary to the Rio Grande. The
Rio Grande  Joint  Investigation5 estimates  the
mean annual surface water production for the
years 1890-1935 to be 164,800 acre-feet for this
area,  but  this  is  subject  to   a  considerable
amount of  estimation. The flow  of the Rio
Grande  is normally depleted in this reach and
the river bed is usually  dry or has  only a small
flow of highly mineralized water at Fort Quit-
man except in times of local flood.
  The river alluvium in this reach constitutes a
major source of irrigation water, although the
quality of this water varies from place to place
and some areas do not have suitable irrigation
quality  groundwater.  Withdrawals  for the irri-
gated lands using river water are estimated to
have averaged in excess of one acre-foot per year
since 1951.
  Irrigation  in this section of  the Upper Rio
Grande  dates to the establishment of a mission

-------
 60     MANAGING IRRIGATED AGRICULTURE
church in 1659 in what is now Ciudad Juarez,
Mexico.  This mission served local Indian tribes
and provided a way station between the seats of
government in Mission City, Mexico, and Santa
Fe, New Mexico.
   Salinity — Salinity  concentrations  progres-
sively increase from the headwaters in Colorado
to Fort Quitman, Texas. This increase results
from salt concentration produced by stream flow
depletions that increase salinity by concentrating
the salt burden in a lesser volume of water, and
by salt loading due to the addition of mineral
salts from various man-made and natural sources
which increases salinity by increasing the total
salt burden carried by the river.
  Most of the water an the Rio Grande is used
for irrigation. This use, together with  that of
native  vegetation and surface  evaporation,  is
responsible for the largest  depletions.  Consump-
tive  use of water for municipal and industrial
purposes accounts for a much smaller depletion.
  Natural salt loading is contributed by some
tributaries of the Rio Grande in north-Central
New Mexico, such as the Rio Chama and the
Rio  Puerco,  that drain areas outside  the  Rio
Grande  depression  and   have   large   surface
areas covered with shales and clays containing
fairly soluble minerals. Man-made sources in-
clude municipal and industrial waste discharges
and return flows from land irrigated by ground-
water.  The relative effects of the various salt
concentrating and salt load  factors on salinity
concentrations  are  summarized  in Tables 1
and 2.
  There was a generally favorable salt  balance
on the Upper Rio Grande  to El Paso, Texas, un-
til the drouth  starting in 1946.  This  drouth
period, coupled  with  the heavy use  of a more
saline groundwater below Elephant Butte Res-
ervoir starting about 1953, created some serious
soil salinity  problems in this reach.  The prob-
lem has been further complicated due to the de-
crease  in  surface water delivered to Elephant
Butte Reservoir.  The 1970,  75-year  mean dis-
charge  for the Rio Grande  at San  Marcial is
924,000 acre-feet while   the ten-year   moving
average is 621,000  acre-feet. The twenty-year
moving average  would be less than this value
due to the drouth in the 50's. This situation has
caused  serious economic  loss to  New  Mexico,
Texas,  and Mexico portions  of the basin above
                  TABLE 1

     Salinity Concentrations on the Upper
   Rio Grande 1963 — Milligrams Per Liter
Headwaters
Otowi Bridge
Caballo Dam
El Paso
Fort Quitman
<100
235
440
870
4,000
                  TABLE 2

        Upper Rio Grande Salt Burden
             1963 — 1000 Tons
     Otowi Bridge
     Caballo Dam
     El Paso
     Fort Quitman
136.2
310.3
313.8
126.8
Fort Quitman. The Hudspeth District below El
Paso has practically ceased to exist as an eco-
nomic unit during this period.4
  Colorado River  water being delivered to the
Rio Grande through the San Juan-Chama Proj-
ect  is apparently of excellent quality, but the
consumptive use of this water  in  the  Middle
Section increases the dissolved  solids concen-
tration  below San  Marcial. Over  the  years,
studies have been made to develop  plans to re-
cover some of the  previously irrigated acreages
in the closed portion of the San Luis Section.
Recovery would involve a system of wells and
conveyance channels in the closed basin which
would deliver water to the Rio Grande, but such
a program would contribute an additional salt
load to  the basin and further increase the salin-
ity problems in the reach below San Marcial.
  It is  recommended that a  salinity policy be
adopted for the  Upper Rio Grande  that would
have as its objective the maintenance of salinity
concentrations below the levels presently found
in the lower portion of the  basin.  It is further
recommended that a comprehensive study be
made to provide detailed information on the
causes,  sources,  and effects of salinity  in the
basin.
                 Pecos River
  For its size, the basin of the Pecos River prob-
  ably  presents a greater aggregation of  prob-

-------
                                                                    SALINITY PROBLEMS
                                           61
  lems associated with land and water use than
  any other irrigated basin in the Western United
  States. These involve both quantity and quality
  of water supplies, the problem of salinity being
  particularly acute; erosion and silting of reser-
  voirs and channels; damage from floods; and
  interstate controversy over the use of the wa-
  ters. There  is an abundance of  good land  so
  that the limit of development is the availability
  of water of satisfactory quality. The use of the
  water of the river has been fully appropriated.
  — National Resources Planning  Board, 1942.6
  The principal tribuary of the  Rio Grande in
the  United  States,  the  Pecos River, rises in the
high altitudes of the  Sangre de Cristo Moun-
tains in north central  New  Mexico, and flows
southward  some   900  miles to  join  the  Rio
Grande  in Texas, see  Figure 3. In a distance of
160 miles, elevations in the  drainage area drop
from 13,000 feet in the headwaters to 4,300 feet
above sea level at Alamogordo Dam. The water
yield in the upper  mountain areas is a function
of snow pack.  The  average snowfall is deter-
mined by elevation, and  varies from 340 inches
above 12,000  feet to 85 inches at 8,000 feet.7
  Below Alamogordo Reservoir the narrow can-
yon-valley sections give way to an area of wide
plains bounded on the west by a  chain of moun-
tains and on the east by the escarpment of the
Southern High Plains. This is generally a semi-
arid area  and  practically all important tribu-
taries originate in the mountains to the west.
Approximately 100  miles  below Alamogordo
Reservoir, the Pecos Valley opens onto the  Ros-
well Basin.  This is  an area 6 to 20 miles in width
along the river for a distance of about 90 miles
and includes  McMillan Reservoir.  The  area
of the Carlsbad project operated by the Bureau
of Reclamation occupies the  river valley from
Avalon Reservoir about 20 miles to a point
where the river .again  enters a narrows section
near the New  Mexico-Texas state line above
Red Bluff Reservoir.
  Below Red Bluff Reservoir the river meanders
through a broad expanse of gently rolling plains.
The area  irrigated from the  river is confined
to the reach  between Red  Bluff Reservoir and
Girvin,  a  distance of  about  110  miles. Below
Girvin the  topography becomes rough and the
valley becomes narrow, developing into an  ever-
deepening canyon section which extends to the
river's confluence with the Rio Grande. At that
point the river canyon is some 300 feet deep.


Surface Water
  Average annual  precipitation  in the Pecos
River basin is more than 14 inches per year. The
highest  precipitation occurs in  the mountains
and  the lowest in the central river valleys, thus
making irrigation  essential  for successful agri-
culture.
  The National Resources Planning Board esti-
mated the total mean annual water production
from run-off available for irrigation in the Pecos
River basin as 1,095,000 acre-feet.6 This repre-
sents 4  percent of the total  precipitation in the
basin. Of this  total,  18  percent  occurs in that
reach of the river below Girvin, Texas, which is
below the major irrigated areas. Of the  water
production above  Girvin, 82 percent originates
in New  Mexico and 18 percent in Texas.
  Studies indicate that the area of highest flood
potential in the state is in the southwestern part
of the basin, and the Pecos  River has produced
many of the  major floods in New Mexico. The
flow of  the Pecos River is partially regulated by
Alamogordo, Two  Rivers, McMillan, and Red
Bluff Reservoirs  with many smaller reservoirs
in the basin area used to control sediment and
runoff.
  Some evidence of early man has been found
in the upper Pecos  basin. Projectile points from
Paleo-Indian  down to  modern  Pueblo  types
have been identified. Early  Pueblo Indians oc-
cupied  the  river  valley  and   its  tribuaries
throughout its  course, but they abandoned the
middle  and lower area  before 1540. Plains In-
dians were in this  portion of the Pecos Valley
when the  Spaniards arrived, and their presence
may account for the abandonment  of the area
by the Pueblo groups. When Coronado's expedi-
tion  passed through  the  upper  valley in 1540,
the  explorers found  Pueblo  Indians  near the
present  village of Pecos. They lived in a walled,
multistoried town and practiced irrigation farm-
ing  by  diverting  water through a system  of
ditches.
  When the Santa  Fe Trail  was opened in the
1820's, the Spanish were irrigating in the upper
valley:  these  settlements expanded downstream
as the Plains Indians were subdued. Settlement

-------
62     MANAGING IRRIGATED AGRICULTURE
                                     10 0  102030 mi.
                                        SCALE
                                       Figure 3: Pecos River

-------
in the middle valley did not take place until the
Civil War and the Homestead Act of 1862.  Rail
transportation was installed between 1878-1909
and  stimulated  the  rate of development. In
1891, following the discovery of flowing arte-
sian groundwater during the drilling of a domes-
tic well,  irrigation was substantially  increased
in the  Roswell  area.  Development  in  the
Middle Section was further  stimulated  in the
1920's by the discovery of oil and potash.
  Salinity — Quality of the  water is  a  serious
problem  throughout most of the Pecos River
basin. Salinity concentrations due to salt con-
centration and salt  loading  progressively in-
crease from  the headwaters to  Girvin, Texas.
Water high in dissolved solids is contributed to
the  Pecos River from  springs,  from  irrigation
return flow, and  from leakage  of old  oilfield
brine pits. Consumptive use of water  by the
heavy growth of phreatophytes, especially at
the  head  of  McMillan Reservoir, probably in-
creases the concentration of dissolved  solids.
  Before  1963, about 420 tons daily of dissolved
salts,  mostly sodium  chloride,  were  added to
the  river  by springs  and seeps  along a 3 mile
stretch about  17 miles below  Carlsbad, New
Mexico.  A  salvage  project  operated  by the
Bureau of Reclamation  has  reduced this  salt
load about 70 percent by pumping this spring
water away  from the  river to  an evaporation
disposal area.9
  There are  between 60 and 70 thousand acres
of salt cedars along  the river between Santa
Rosa, New  Mexico,  and Girvin, Texas.  The
growth was  first noted in 1915,  and by 1939 it
had  spread  over about 14,000  acres along the
river between  Alamogordo Dam and the state
line. The Bureau of Reclamation indicates  that
some 75,000 acres will be infested by the year
2010, and unless corrective action is taken, the
entire supply of the Pecos River could be de-
pleted by these plants within a few decades.10
  The relative effects of the various salt concen-
trating and salt load factors on  salinity concen-
trations  are summarized  in  Tables  3  arid 4.
Some of the  salt  concentration  has  probably
been due to  the diminution of good quality in-
flow from parts of the drainage  basin.  Part  is
attributed to the development and use of water
from major groundwater reservoirs in the basin
                   SALINITY PROBLEMS    63

                 TABLE 3

  Salinity Concentrations on  the Pecos River
       Ten Year Average —  1955-1964
             Milligrams Per Liter
     Headwaters
     Alamogordo Dam
     Carlsbad
     Red Bluff Dam
     Langtry
<100
1,420
2,320
6,970
1,460
                 TABLE 4

          Percos River Salt Burden
Ten Year Average — 1955-1964 — 1,000 Tons
    Alamogordo Dam
    Carlsbad
    Red Bluff Dam
    Langtry
247.4
154.3
550.6
432.0
as under natural conditions these groundwaters
discharge to the Pecos River.
  The fundamental difficulty is the lack of an
adequate and  reliable water supply to meet all
of the demands that are placed on the system.
The  controllable  part of the  supply  is vested
mainly  in  the individual  water-right owners.
This  imposes  legal  and  administrative  con-
straints on an already difficult physical system.
Programs for  solving most of the critical prob-
lems of the system have been undertaken or are
contemplated.

              Lower Rio Grande
  The Rio Grande leaves Fort Quitman through
a narrow, steep-walled canyon and  continues
through this  rugged  mountain  area,  flowing
occasionally through small isolated valleys as
it approaches  the Big Bend country, see Figure
4. This region is filled with geologic phenomena
such as tilted and folded strata, uplifts, and vol-
canic cones.  Some canyons in the area attain
depths of nearly 1,500 feet. Below the Big  Bend
country, the Rio Grande passes through a hilly
section with  broken  terrain,  narrow  valleys,
and  broad high tablelands. After crossing the
Balcones  Escarpment,  the  river  enters  the
Lower Rio Grande Valley.

-------
                                                           0  25   50  75  100 ml
                                                                                         : •
                                                                                         yo
                                                                                         g

                                                                                         5
                                                                                         >

                                                                                         w
                                                                                         a

                                                                                         3
                                                                                         C
                                                                                         -••:

                                                                                         n
Figure 4: Lower Rio Grande

-------
                                                                  SALINITY PROBLEMS
                                          65
  This valley extends about 100 miles upstream
from the Gulf of Mexico and contains approxi-
mately 2 million acres of alluvial soil divided
about equally  between the United  States and
Mexico. It  is a semitropical region which has
mild winters and hot humid summers  with  a
moderate rainfall of about 24 inches. Most  of
this rain falls in the early summer, making irri-
gation a necessity during the  remainder of the
year.
  Five  major  streams  join  the  Lower Rio
Grande. About 70 percent of the water  is con-
tributed, together with several smaller streams,
by three of them  which originate  in Mexico:
these are the Conchas, the Salado, and the San
Juan. The  American  tributaries,  primarily the
Pecos and Devils rivers, provide the rest.  Con-
sumptive use of  the waters of  the  Lower Rio
Grande  has increased substantially in  recent
years. Some of this increase is due to irrigation
development on the  Mexican  tributaries and
some  is  due to flood control, especially  Falcon
Dam. The flow of the Rio Grande to the Gulf of
Mexico below Brownsville prior to the construc-
tion of Falcon Dam averaged 2.6 million  acre-
feet: after the  construction this flow has  aver-
aged about 0.6 million acre-feet per year.
  Groundwater  supplies provide a  relatively
minor part of total irrigation water;  however,
groundwater is used extensively when surface-
water supplies are  below normal.  The  major
sources of groundwater in the region are the allu-
vial deposits in  the  river  delta. This alluvial
aquifer  contains sizeable amounts  of ground-
water within a belt five  to ten miles wide ex-
tending  along  the  river and is  recharged  from
surface  drainage and seepage from the Rio
Grande.
  The lower valley was important in  the  eigh-
teenth century as an outpost of Spanish coloniz-
ers.  Reynosa  and  other  settlements on the
Mexican side were settled as early as 1749. The
early  settlers'  attempts  at  growing  irrigated
crops were discouraged by intermittent drouths
and  river floods,  and consequently  ranching
became  the  important  activity.  During the
nineteenth  century,  the  development  of the
valley by Anglo-Americans was stimulated  by
military activity  and  the area's strategic  loca-
tion on the Mexican border. The first major
irrigation  systems  in The  Lower Rio Grande
were begun around 1905 by large land develop-
ment companies.
  Salinity — Salinity  concentrations   in   the
Lower Rio Grande are fairly constant above
Falcon Dam. There is a concentrating effect be-
low that point due to heavy consumptive use in
irrigation. Some soil salinity problems have de-
veloped because  of poor  drainage,  and this
problem  is expected  to increase as improved
irrigation  management techniques are applied
upstream.

SUMMARY
  The Rio Grande system carries a large  salt
burden contributed by a variety of natural and
man-made sources.  Future  developments will
increase  salinity   concentrations.  As  salinity
concentrations  increase,  adverse physical ef-
fects  are  produced which  result in economic
loss to the region. Additional research is neces-
sary before adequate salinity control and man-
agement programs can be considered.

REFERENCES
   1. E..V. Jetton  and J. W. Kirby, "A Study
of Precipitation, Stream Flow and Water Usage
on the Upper Rio Grande," Report Number 25,
College of Engineering,  University  of Texas,
Austin, Texas, June, 1970.
   2. P. A.   Emery,   A. J.  Boettcher,  R. J.
Snipes,  and  H. J.  Mclntyre,  Jr., "Hydrology
of the San Luis  Valley South-Central  Colo-
rado," Hydrologic Investigations Atlas HA-381,
U.S. Geological Survey, 1971.
   3. R. A. Wortman, "Environmental Implica-
tions  of Surface Water Resource Development
in the Middle Rio Grande Drainage, New Mexi-
co," M.S. Thesis, The University of New Mexi-
co, Albuquerque, New Mexico,  1971.
   4. Bureau of Reclamation, "Rio Grande Re-
gional Environmental  Project,"  Information
Report, Department  of the  Interior,  October
1971.
   5. Natural Resources Committee, "Regional
Planning  Part  VI — Upper Rio  Grande," U.S.
Government, February, 1938.
   6. National Resources  Planning  Board.  Re-
gional Planning, Part X, The Pecos River Joint
Investigation in the Pecos River Basin in New

-------
 66     MANAGING IRRIGATED AGRICULTURE
 Mexico and Texas — Summary, Analysis, and
 Findings, U.S. Government, 1942.
   7. J. W. Hernandez, "Management Alterna-
 tives in the Use of the Water Resources of the
 Pecos River Basin in New Mexico," Report No.
 12, New Mexico Water Resources Research
 Institute,  New Mexico  State  University,  Las
 Cruces, New Mexico,  December, 1971.
   8.  W. L.  Heckler,  "Surface  Water  Avail-
ability and Quality Characteristics in the Pecos
River  Basin  in  New Mexico,"  Proceedings,
Tenth Annual New Mexico Water Conference,
New Mexico State University, Las Cruces, New
Mexico, 1965.

   9. W. E. Hale, L. S. Hughes, and E. R. Cox,
"Possible Improvement of Quality of Water of
the  Pecos  River  by  Diversion of Brines at
Malaga Bend, Eddy  County,  New Mexico,"
Pecos  River  Commission,  New  Mexico  and
Texas, Carlsbad, New Mexico, 1954.

  10. New  Mexico  State  Engineer  Office,
"Water Resources of New Mexico," State Plan-
ning Office, Santa Fe, 1967.

-------
  Hydrologic  Modeling for Salinity Control
            Evaluation  in  the  Grand  Valley
                                WYNN R. WALKER
                                        and
                            GAYLORD V. SKOGERBOE

                         Agricultural Engineering Department
                              Colorado State University
                                Fort Collins, Colorado
ABSTRACT
  The format and supporting  information for
a salinity control evaluation  model for the
Grand Valley is presented.  The scope  of the
model, while developed specifically for applica-
tion in the Grand Valley incorporates most of
the general characteristics prevalent in irrigated
agricultural areas. The objective of the model is
a quantitative evaluation of a specific salinity
management measure, and the comprehensive
description of the ionic interchanges occurring
in the soil profile has not been included.
  In  order to provide an  additional degree of
confidence  in the results,  the model has been
designed to compare  the quantities of ground
water return flows through the subsurface aqui-
fers computed by an analysis of the ground wa-
ter characteristics to one using a mass balance
analysis.  Since both of these methods are not
entirely  independent,  the  adjustment  of the
model until both are predicting the same flows
results in the measurable surface flows and the
unmeasurable subsurface flows being comple-
mentary.
INTRODUCTION
  The salinity problems associated with agricul-
tural water use in the Grand Valley are being fo-
cused upon at the local, state, and federal level
in an effort to reduce the downstream impacts.
Although the need to implement effective salin-
ity management programs in the valley is rec-
ognized, the most practical alternatives for ac-
complishing this requirement have not been en-
tirely formulated. A major step in delineating
the  promise  of available alternatives has re-
cently been  taken in  the completion of  the
Grand Valley Salinity  Control Demonstration
Project. The purpose of the project was to exam-
ine canal and lateral linings as possible salinity
control methods.
  The evaluation of the effects of canal and lat-
eral linings on the hydro-salinity flow system in
the  Grand Valley cannot be made indepen-
dently of the various other fact6rs which influ-
ence the flow system. Consequently, in order to
substantiate the impacts of controlling one seg-
ment, all other phases of the hydro-salinity net-
work must also be delineated and quantitatively
                                         67

-------
68     MANAGING IRRIGATED AGRICULTURE
defined. This process of modeling the complete
water and salt flow system involves the prepara-
tion of water and salt budgets.
   A difficulty often encountered while  prepar-
ing water and salt budgets is the variability in
the accuracy and reliability with which  the  hy-
drologic and salinity parameters are measured.
Usually the measurement precision varies with
the scope  of the research and the size of the lo-
cale. Because of these factors, it is helpful to the
understanding of the results of an investigation
if the  techniques employed in  computing  the
budgets and the  simpliflying assumptions which
are made are examined.
   since the  hydrologic system  is difficult  to
monitor and predict, it js impractical to expect
their models to operate without some  experi-
enced adjustments. In short, the budgeting pro-
cedure  is  usually the adjustment of the  seg-
ments in the water and salt flows according to a
weighing of the  most reliable data until all  pa-
rameters represent the close  approximation of
the area. The vast and lengthy computation pro-
cedure of  computing budgets is facilitated by a
mathematical model programmed for a digital
computer. For  the purposes  of  this  paper,  the
more important aspects will be extracted for dis-
cussion. A  schematic  diagram  of  a  general
hydro-salinity model is shown in Figure  1.
  The hydro-salinity system of the Grand Val-
ley can be divided into four general areas:
  (1) Inflows representing  the total  water po-
     tentially   available   for    use   within
     the area and  the  dissolved minerals car-
     ried  by the water. Included in  this group
     are river, tributary, and ground water in-
     flows,  importations, and precipitation.
  (2) Cropland  diversions  form the available
     supply of water diverted into the agricul-
     tural segment  of the valley.  From  the
     main supply canals, small turnout struc-
     tures are  used  to  divert the water into
     small lateral ditches  which lead  to  the
     fields.
  (3) Ground water results from the seepage
     from canals and  laterals and  the water
     which  deep percolates  :hrough the root
     zone into the water table region.
  (4) Outflows are  the  flows  which return to
     the river system or are lost to the  atmos-
     phere  through  evapotranspiration. Nu-

r~
L

r




Ev


Flows
operation
t

f
Cono
Seepage

.J

J
Latera
Seepage



f
Deep
Percolation
u
Gnx
Wot
StO


er
age
r
Ground Water
Outflows


r
Lateral
Diversions





r~
Canal
Diversions



i
i
~i
Spillage
i
i
i
.j
t
Root Zone
Supply




i

Consumptive
Use



Kiver
Inflows
i
	 1
U-
r-


1
Field
Talwater



1
..


1
Soil Moisture
Storage
! I— ® 	 1


1
	 1 J
Drainage
Return Rows

i



r
1


1
— I
-i
—



f
»
•
^•— —
m — —


River
Outflows

Tributary
Inflows
— i i
i

Ground Water
Inflows
— i !
/
Imparts
— i !

Preci potation
	 i
i
Rreotophyte
Consumption

—|t
Exports

— » i
Municipal
Uses

 Figure  1: Schematic Diagram of  Generalized
 Hydro-salinity Model

      merous routes are taken by the water to
      reach  the river, including drainage and
      ground water return flows.
In the following sections, the aspects of these di-
visions are explored in greater detail  to give a
degree of insight into the real system which is to
be modeled.

                   Inflows

 River Inflows
   The exclusive sources of irrigation  water in
 the Grand Valley are the Colorado and Gunni-
 son Rivers. Together, the two rivers represent
 an average combined flow  of 6500 cfs  from
 which large check type structures are used to di-
 vert the  water into the principal supply canals.
 Owing to the increased development of the wa-
 ter resources in the Upper Colorado River Ba-
 sin, the  discharges in the rivers  in the Grand
 Valley region are highly affected by transbasin
 diversions, reservoirs,- power development, and
 irrigation. The Colorado River at the east en-
 trance to the valley drains  an area of about
 8,050 square miles resulting in an average dis-
 charge of about 4000 cfs containing between

-------
                                              HYDROLOGIC MODELING — GRAND VALLEY    69
300 to 400 ppm of dissolved minerals. The river
at this point also carries large sediment loads,
especially during the high flow periods in the
spring, which aids irrigation in the valley by re-
ducing intake rates during the early growing
season when smaller irrigations suffice. The ef-
fect of development upstream of Grand Valley
is clearly noticeable in examining historical rec-
ords. A large part of these conditions are related
to the approximately 190,000 acres which de-
plete about 190,000 acre-feet of water annually
from the river system (2). The Gunnison River
Basin, while  about the same size, uses about
350,000  acre-feet  annually on about  270,000
acres. The resulting flow in the Gunnison River
as it enters the Grand Valley is about 2500 cfs
containing between 500 and 600 ppm of salts.

Tributary Inflows
  Tributary inflows which are the ungaged wa-
ter resulting from precipitation on the adjacent
area in  the valley, actually account for only a
minor portion of the water passing through the
Grand Valley. Aside from the estimated 60,000
acre-feet added by Plateau Creek, the estimated
yield from the surrounding watershed  is prob-
ably on the order of 55,000 acre-feet (2), (4),
(5). The precipitation averages  about 8 to 9
inches per year, and the intensity is usually low
enough  to allow the soils to simply absorb the
water. The area is marked by natural washes,
evidence that some periods of tributary inflows
occur, but for the majority of the time, the flows
in these washes are field tailwater and drainage
return flows.

Imports
  Importation of water from nearby mountain
watersheds currently  supplies the bulk of do-
mestic  demands  in  the  valley although  new
treatment facilities have been built to use water
from the Colorado River. In addition, several
deep wells are used for domestic and commer-
cial use. However, none of the water is used for
irrigation as a rule because of the abundance of
cheap river water.

 Ground Water Inflows
   Ground water inflows to the valley are essen-
tially impossible to measure but should be ac-
counted for in the water and salt budgets. Hyatt
(4), using an electronic analog computer system
model of the Grand Valley, indicated little or no
ground water inflows to the region. His conclu-
sion seemed  well justified  as  both the rivers
enter the valley  through  rocky mountainous
channels.
             Cropland Diversions

Canal Diversions
  The source of the irrigation supply is the di-
version into large  canals by  means  of  large
check-type diversion dams. Distribution of the
canal flows occurs  in four ways: (1) diversions
into  the  farm  lateral  system,  (2) seepage,
(3)  spillage  into  wasteways,  and  (4) evapora-
tion.
  By using the natural  washes as wasteways,
the individual canal companies maintain regula-
tion points along the  system where control of
downstream flows  can be made. This practice
has the advantage of being able to readily com-
pensate for drastic events such as irrigation cut-
backs due to foul weather or increased demands
by periods of warmer weather. Even though this
practice is not a desirable water management al-
ternative, the long  length between canal  head-
works and  the end of the distribution system,
along with the abundance of  cheap water,  make
spillage the  most employed  regulation tool in
the  Grand Valley. One situation  does  exist
where spillage  is used to supply  another seg-
ment of the system. The Grand Valley Canal
spills water into what  is known as Lewis  Wash
which  is then diverted  by  the Mesa  County
Ditch. A noticeable consequence of this particu-
lar operation is the poorer quality of irrigation
water being used  by  irrigators served by the
Mesa County Ditch resulting from  mixing the
spillage with the drainage waters in the washes.
The  salinity being  added to the  spilled  water
throughout the valley is difficult to determine,
but in the course back  to the river system evapo-
ration and phreatophyte consumption probably
further concentrate existing salt concentrations.
   Seepage from the conveyance channels enters
the ground water  basin directly  and in the
Grand  Valley these flows complicate an already
serious drainage problem. Most estimates  of the

-------
70     MANAGING IRRIGATED AGRICULTURE
 magnitude  of the  seepage losses range in the
 neighborhood of 20%, but considerable variabil-
 ity has been noted.

 Lateral Diversions
  The term lateral  as used in this text refers to
 those  small conveyance  channels  that  deliver
 water  from the company  operated canals to the
 farmers fields. These small conveyance chan-
 nels usually carry under 5 cfs, and range in size
 up  to  4 or 5 feet of wetted perimeter.  Since the
 lateral system in the Grand Valley is  so im-
 mense, seepage from these small channels is
 probably much higher than from the canal sys-
 tem.
  In  a manner  similar to the distribution  of
 canal  discharges, lateral  diversions may be di-
 vided  into four classes: (1) diversions reaching
 the crop root zone, (2) flows that are allowed to
 run off the ends of the fields during irrigation,
 otherwise known as field tailwater, (3) seepage
 losses,  and  (4) evaporation from the water sur-
 faces.  The nature of irrigation in the  region is
 not conducive to efficient handling of water in
 the laterals as their lengths  are so vast. Other
 factors influencing  water  management methods
 include the slope of the land, which is about  50
 feet per mile  in some places, and the inexpen-
 sive water  supply.  Numerous  instances  exist
 where laterals flow continuously with the water
 not  being  used  in  irrigation  being  simply
 dumped into the drainage  system. A preliminary
 sampling of these discharges  indicates some salt
 pickup from the soil surfaces.

 Root Zone Diversions
  The purpose of irrigation is to supply the root
 zone of crops with  sufficient water to  meet the
evapotranspiration  demands. During the proc-
ess of crop water use, the dissolved minerals  in
the water  are isolated resulting in accumula-
tions  occurring in the root zone which necessi-
tates a leaching of  these  salts by an additional
quantity of irrigation. There are often difficul-
ties in irrigating farm lands because of the  rela-
tive effects of the water on the plants. For ex-
ample,  when  insufficient  moisture is  provided
during critical growth periods, the reduction  in
production  may  be enormous. On the other
hand, in most situations a  small excess produces
very little damage.  As a result, the tendency is
 usually to over-irrigate. The consequences of
 over-irrigation,  while  not  as  severe  to the
 farmer, are unnecessary fertilizer leaching, high
 water tables and drainage  requirements,  and
 large salt additions to the river systems.
   The movement of water within the root zone
 is either lost to the atmosphere through evapo-
 transpiration, or lost through deep percolation
 into the ground water basin below the root zone,
 or it may simply be stored within the root zone.
 The delineation of these flows by measurement
 is extremely difficult and is impractical  on a
 large scale. Consequently, the procedure most
 often used  is empirical computational methods.
   Numerous methods of estimating evapotran-
 spiration have been  developed for agricultural
 lands. Probably the  two most adaptable meth-
 ods  for the  semi-arid western regions are the
 Blaney-Criddle   method (4) and the  Jensen-
 Haise method (6). The  Blaney-Criddle method
 involves  the  determination of consumptive use
 as a function of mean monthly temperature and
 the percentage of daylight hours occurring dur-
 ing the month. The general equation can  be ex-
 pressed as,
     U = (t-p/100)-(Kc-Kt)-(A)/12  .. (1)
 Where U is the  water use in acre-feet, t is the
 mean monthly temperature  in degrees Fahren-
 heit, p is the yearly daylight hour percentage oc-
 curring during a month, Kt  is a climatic coeffi-
 cient expressed as,
    Kt = 0.0173- t-0.314	 (2)
 Kc is the crop growth  stage coefficient  deter-
 mined experimentally, and A is the acreage of a
 particular crop or phreatophyte.
  The Jensen-Haise method  was  formulated
from  the evaluation of about 3,000 published
and unpublished reports on  short period  meas-
urements of evapotranspiration using soil sam-
pling procedures during a 35-year interval in the
western USA. The resulting equation is,
    ET=KcEtp  	  (3)

in which ET  is  the evapotranspiration, Kc is a
crop coefficient  much like the Blaney-Criddle
Kc value, and Etp is the potential evapotranspi-
ration in a well watered soil in a semi-arid area.
The value of  Etp is computed by a relationship
between air temperature and  solar radiation,
    Etp = (0.014t - 0.37) • Rs 	  (4)

-------
                                              HYDROLOGIC MODELING — GRAND VALLEY
                                           71
where t is  the temperature in degrees Fahren-
heit and Rs is the solar radiation in Langleys.
  In order  to determine the magnitude of deep
percolation losses and root zone storage, some
simplifying assumptions  must be made.  Since
vegetation  is  only capable of transpiring at its
potential rate when the soil moisture storage is
adequate, an adjustment  based  on the storage
and irrigation supply whenever insufficient wa-
ter is available must be  made to the potential
value. The  measured values of canal and lateral
diversions do not reflect the application on each
field or each crop and so the assumption is made
that  the  irrigation is made uniformly over the
cropland. Thus, the water used  from the root
zone would only involve crop requirements and
phreatophyte use would be from the moisture
below the water tables. Evaluation of reported
crop and soil characteristics in the area can be
made to determine  root  zone  depths and soil
moisture storage  capacity as they change with
time (1), (3), (6), (8), (9).  Once  these parame-
ters have been established for an area, a budget
of root zone  water can be made.  The calculated
total potential consumptive use  is  first  com-
pared with the total water added to the root
zone from irrigation and precipitation. Three al-
ternatives are assumed:
  (1) If the supply  to the root zone is less than
      the potential demand, but  ample water is
      stored  within  the root zone to meet  the
      deficit, then the use would  be equal to the
      potential demand.  Since  the  usual bud-
      geting  procedure  is carried   out   on a
      monthly time  interval, the  next period of
      study would  have  an unused  root zone
      storage that  would be filled  or supple-
      mented with  that period's supply.  It  has
      been assumed that whenever  a deficit in
      root  storage  exists, no water is  lost to
      deep percolation.
   (2) If the sum of the supply and the available
      storage is less than the potential demand,
      the actual use is assumed  to be the total
      quantity  available.  A term called con-
      sumptive use deficit is defined as the dif-
      ference between  potential demand and
      the actual use. Again, there would be no
      deep percolation.
   (3) If the supply  to the root zone is sufficient
      to meet the potential demand, the actual
      use would equal this demand. If the  ex-
      cess is sufficient to refill the soil moisture
      storage, the deep  percolation would be
      the water above which is necessary to re-
      fill the soil moisture storage to  field  ca-
      pacity.
An illustrative flow  chart  of this routine is
shown in Figure 2.
  Although the salt flow system is generally de-
pendent  upon the nature and magnitude of the
water flows, the behavior of specific ions in the
root zone and ground water basin are very com-
plex. When the irrigation water is applied to the
soils, the processes of ionic exchange,  adsorp-
tion, and precipitation occur. The mathematical
ability to describe these reactions has developed
a long way,  but in terms of evaluating the ef-
fects of a salinity control program, the expense
necessary to collect sufficient  data to operate
such a comprehensive model is  presently pro-
hibitive.  As  a result, the modeling which  has
been  done for evaluation in the Grand Valley
has  considered only  the quantity of total  dis-
solved solids  as a parameter. In addition,  the
                        PCU- Potentiol Con«ut*>lM Use
                        PREC= Precipitation
                        RZC* Root Zone Capacity
                        RZSC- Root Zone Storage
                            Consumption Use Deficit
                            Additions to Smod Water
CUD-
AGW*
  Figure 2:  Illustrative  Flow Chart of Root Zone
  Budgeting Procedure

-------
72
MANAGING IRRIGATED AGRICULTURE
root zone model has not focused upon the spe-
cific behavior of the various components of sa-
linity.
   These assumptions seem well justified in light
of present technology. In the evaluation of a sa-
linity  control  alternative,   the  results  of the
model should evaluate a net effect and predict
in a  quantitative  manner the  results of an ex-
panded program.


                Ground Water
   The discussion of ground water flows in this
section is limited to those flows below the water
table as the root zone discussion  of the proceed-
ing section involved the-region above the water
table.  The ground water flows  in the agricul-
tural region  are comprised of  canal and lateral
seepage  as well as  deep percolation. The hy-
draulic gradient resulting from  the water  in-
flows causes  the movement of the water towards
and into the  river system.
   The  ground  water discharges  involve  two
phases: (1) drainage  interception,  and  (2) sub-
surface outflows. Since the water table is often
intersected by the drainage system, these flows
are easily  measured by flow measuring devices
in these drains. The subsurface outflows cannot
be measured, but  with water table elevation
data throughout the area, along  with hydraulic
conductivity  measurements in  the various sub-
surface strata,  these flows  can  be reasonably
computed. Even though  considerable effort can
be made to  monitor the pertinent subsurface
variables, the data usually obtained do not war-
rant a non-steady state analysis unless a ground
water study is the specific objective of a project.
For the purposes of this study,  Darcy's steady
state equation has been used, (7).
    Q = AK(dh/dx)  	  (5)
in which Q is the discharge, A is the cross-sec-
tional area of the ground water flow, K is the hy-
draulic conductivity, and dh/dx is the hydraulic
gradient in the  direction of flow. During the
course of the investigation, it  was  felt  that the
weakest link  in the data collection was in deter-
mining  the values  of hydraulic conductivity;
however, it seemed  reasonable to  assume that
the relative magnitude of conductivity between
one  strata and  another could be determined
with  reasonable accuracy.  With this  type  of
                                           data, it is possible to formulate two independent
                                           methods of calculating the ground water flows
                                           and thus increase the hydrologic budgeting pro-
                                           cedure  accuracy by  forcing an alignment be-
                                           tween the two methods.
                                             The ground water  analysis illustrated in Fig-
                                           ure 3 involves the comparison of the ground wa-
                                           ter outflow based on a mass balance arrived at
                                            i   -Refers to Its Strata
                                           Grad  = Hydraulic Gradient
                                           Q    -Computed Ground Water Outflow
                                           PHC  *Reld  Values of Hydraulic Conductivity
                                           AHC  ^Adjusted  Values of Hydraulic Conductivity
                                           TCWOF= Total Ground  Water Outflow from Mass Balance  Analysis
                                           AQ   -TCWOF
                                           Figure 3: Illustrative Flow Chart of Ground Wa-
                                           ter Modeling Procedure

                                          through a general  budgeting procedure (inflow
                                          equals outflow  minus  storage  changes)  and
                                          computations of outflows based upon measured
                                          gradient  and  conductivity  data.  Because  the
                                          model only uses  relative magnitude of strata hy-
                                          draulic conductivity, the values are adjusted on
                                          their  relative proportion until  each value of
                                          monthly ground water outflow is consistent be-
                                          tween analyses. Then the variability of the com-
                                          puted values of conductivity is examined. If they
                                          become homogenous, the model represents the
                                          "best fit" between all monitored data. The com-
                                          putation  of ground water outflow based on
                                          Darcy's equation for a number of strata  can be
                                          written:

-------
                                              HYDROLOGIC MODELING — GRAND VALLEY
                                                               73
               dh
dh.
                               dh_
                               TT	(6)
where An is the area of the nth strata, K'n is the
hydraulic conductivity of the nth  strata,  and
dhn/dxn is the hydraulic gradient acting on the
nth  strata. Occasionally  when no confining lay-
ers exist, the  hydraulic gradients  would be the
same  for each  strata; however, in the Grand
Valley an underlying cobble aquifer is partially
confined and  this condition does not exist. The
value computed from Eq. 6 will generally not be
comparable in magnitude to the expected values
determined  from the mass balance analysis be-
cause  no  need  exists  to  keep  the measured
values of conductivity in proper units. Usually,
conductivity is  determined in in/hr, while the
ground  water outflows will be  in acre-feet/
month.  To  avoid confusion, these inconsisten-
cies are  compensated for in the  adjustments. In
addition, it is often difficult to evaluate the re-
spective  strata areas accurately. Consequently,
a representative thickness  can  be  determined
and then a convenient unit width selected for
the  computations.  If this  practice is used the
flows will be altered, but the adjustment can be
absorbed in the adjustments to hydraulic con-
ductivity. The adjustments to the field hydraulic
conductivity measurements can  be made as in-
dicated:
    Kj + (TGWOF • KJ)/value of discharge ob-
      tained from Eq. 6  	 (7)
where K, is the adjusted  hydraulic conductivity,
TGWOF is the ground water outflow estimate
from the mass balance analyses, and Kj  is the
field measurement of hydraulic conductivity.
  In summary, the computational procedure for
evaluating ground water flows is as follows:
  (1) From field data collected  on hydraulic
     gradients  and conductivities,  along with
     physical dimensions  of  the   system,  a
     value of ground water discharge can be
     computed using Darcy's equation.
  (2) Comparison  of the values obtained in
     Step (1) with the estimates of ground wa-
     ter outflows based  on a mass  balance
     analysis is used to adjust  the field values
     of conductivity according  to their relative
      magnitude until both methods of comput-
      ing discharge agree.
   (3) Compare the monthly values of adjusted
      hydraulic  conductivity for homogeneity.
      If the values are found to be in error, all
      parameters in the  hydro-salinity  model
      are not aligned and further adjustment of
      various budget factors is necessary.

                 Outflows
   The water leaving the Grand Valley,  aside
from evapotranspiration, includes the river out-
flows and ground water flows occurring beneath
the gaging station. Another common form of re-
gional outflow is exportation, but in the Grand
Valley none of these flows exists.

River Outflows
   The Colorado River exits from the Grand Val-
ley in the western end  with a mean annual dis-
charge  of about  6,500 cfs  and a salt load of
about  900  parts per million of total dissolved
solids. The resulting salt contribution from the
Grand Valley during most normal water years
varies between 0.7 and 1.0 million tons.

Ground Water Outflows
   The estimated  discharge under the exit gag-
ing station operated by the U.S. Geological Sur-
vey is  shown to be small in comparison  to the
river flows, (4). However, the reliability of this
type of an  assessment  should be questioned in
light of the difficulty of monitoring the exact ef-
fect of the Grand Valley within acceptable accu-
racy.

Hydro-Salinity Model
  The  vast computational requirements  requi-
site to formulating water and salt budgets, often
called  hydro-salinity modeling, is  best  facili-
tated by digital computers or, as in the case of
Hyatt (4), by electric analog systems. The model
formulated for evaluating the salinity  control ef-
fects of canal and lateral linings in the Grand
Valley is also somewhat general, so it is helpful
to  examine  the  computer framework  under
which it was formulated.

Model Operation
  The mathematical  model derived for this
study attempted to simulate the hydrologic con-

-------
74
MANAGING IRRIGATED AGRICULTURE
ditions of the agricultural system in the Grand
Valley, but the concepts are general and can be
extended with modification to other areas which
are similar in nature. The program was written
in  individual  but  interconnected subroutines
which give the program a measure of flexibility
during operations by separating the calculation
phase from either input or output. By structur-
ing the program in this manner, several of the
subroutines become optional if their functions
can be replaced by input data, or if certain out-
puts are not desired. This general nature of the
program  is  illustrated in  the  schematic flow
chart shown in Figure 4 with name and func-
tions defined in Table 1.
  The main portion of the program is used to
read  necessary  input data and  to control the
order  of budget calculations. There are certain
advantages in separating the input, output and
computational stages of a program,  including:
  (1) Input order is not important as the data
      are completely available at all  stages of
      computation.
  (2) Variable sets of data can be utilized in the
      model when  several budgets are desired,
      or when some form of integration is de-
                                                           TABLE 1

                                                 Hydro-salinity model subroutine
                                                          descriptions
                                           Subroutine
Description
 Figure 4: Schematic Flow Chart of Hydro-salinity
 Model
                                          WATER    Computation of monthly and annual
                                                      values of the water budget. Relatively
                                                      little independent data is generated by
                                                      WATER directly,  since it functions
                                                      primarily as a summary.

                                          BUDO      Outputs data generated from WATER
                                                      as the water budget.

                                          PCUS       Computation of monthly and annual
                                                      value of potential consumptive use for
                                                      irrigated crops, dryland crops, munici-
                                                      pal uses, industrial uses, open  water
                                                      surfaces, and  phreatophytes.

                                          PCUO       Outputs data generated in subroutine
                                                      PCUS.

                                          CGSC       Outputs values  of  crop growth stage
                                                      coefficients.

                                          ACUS       Computation of the estimated actual
                                                      consumptive use and the various root
                                                      zone parameters.

                                          GWMOD    Computation of the discharges through
                                                      the ground water model and adjusted
                                                      strata hydraulic conductivities. When
                                                      these  values of conductivity approach
                                                      equality,  the  ground water outflows
                                                      are correct.

                                          GWMOP    Output of ground water model compu-
                                                      tations.

                                          SALT       Computation of the salt budget for the
                                                      area.

                                          SABU       Output of salt budgets.
                                               sired. This is useful especially when  an
                                               area can be broken down into smaller de-
                                               pendent areas.
                                           (3) The functions of the subroutines are  in-
                                               dependent of input  making  each a unit
                                               that can be  implemented  in other pro-
                                               grams.

-------
                                              HYDROLOGIC MODELING — GRAND VALLEY     75
  (4)  Corrections and adjustments  are  easily
      made  without  detailed consideration to
      other segments of the program.
  In controlling the computational order of the
program, the main program separates the calcu-
lation of  the  water  and salt  budgets.  Conse-
quently, the modeling  procedure  involves only
the water phase of the flow system.  This has
been possible in this study because of the detail
in which data have been collected. Once the wa-
ter flow system has been simulated, the individ-
ual flows  are  multiplied by measured salinity
concentration and converted to units of tons per
month.  At this point in the formation of the
budgets, careful attention must be given  to the
salt flow system since irregularities may be  pre-
sent and further model adjustment is necessary.
Thus, when the final budgets have been gener-
ated, the salt system, ground water system,  and
surface flow system must be reasonably coordin-
ated and additional reliability is assured.
SUMMARY
  The  evaluation of specific salinity manage-
ment schemes in the  Grand Valley involves the
development of models which detect and pre-
dict the  quantitative  effects  of the program.
Such a modeling effort encompasses an evalua-
tion of the hydro-salinity system and thus indi-
rectly evaluates the potential impacts of other
possible measures.
  The  current  hydro-salinity model  for the
Grand  Valley is only the first step in salinity
control evaluation. While its  purpose has been
to simply establish effects, future models should
incorporate greater  detail of  the behavior and
specific interactions of the salinity flow system
in order for more comprehensive predictions of
capabilities.  The  next  development in Grand
Valley  modeling must  not only  be capable of
evaluating   the impact  of a salinity control
measure, but also to indicate its influence in a
correlated program involving a broad variety of
alternative salinity management alternatives.
REFERENCES
  1.  Blaney,  H. F., and W. D.  Griddle. 1950.
Determining  water requirements  in  irrigated
areas from climatological and irrigation data.
U.S. Department of Agriculture,  Soil Conserva-
tion  Service. SCS-TP 96, 44 p.
  2.  Colorado  Water Conservation Board and
U.S. Department of Agriculture.  1965. Water
and related land resources Colorado River Basin
in Colorado. Denver, Colorado.  May.  183 p.
  3.  Hagan, Robert M.,  Howard R. Haise, and
Talcott W. Edminster,  Editors. 1967. Irrigation
of agricultural lands.  No.  11  in the series,
Agronomy.  American  Society  of Agronomy.
Madison, Wisconsin. 1180 p.
  4.  Hyatt,  M. Leon.  1970. Analog computer
model of the hydrologic and salinity flow of sys-
tems within the Upper Colorado  River Basin.
Ph.D. dissertation,  Department  of Civil Engi-
neering,  College  of Engineering, Utah State
University, Logan, Utah. July.
  5.  lorn,  W. V.,  C. H. Hembree, and G. L.
Oakland.  1965. Water resources of the Upper
Colorado  River Basin - Technical Report. Geo-
logical Survey Professional  Paper 441.  U.S.
Government  Printing Office, Washington,  396
P-
  6.  Jensen, M. E., and H. R.  Haise, 1963. Esti-
mating evapotranspiration from solar radiation.
American Society of Civil Engineers, Irrigation
and Drainage Division, Proc. 89.
  7.  Luthin,  James  N.  1966. Drainage Engi-
neering. John Wiley & Sons, Inc. New York.
250 p.
  8.  Pair, Claude  H.,  Walter W.  Hinz, Craw-
ford Reid, and Kenneth R. Frost, Editors. 1969.
Sprinkler irrigation. Third Edition. Sprinkler Ir-
rigation Association, Washington, D.C. 444 p.
  9.  U.S.  Department of Agriculture, Soil Con-
servation  Service,  and Colorado  Agricultural
Experiment Station. 1955.  Soil  Survey, Grand
Junction Area, Colorado. Series 1940, No.  19.
118 p. November.

-------
         Sediment Control in  Yakima  Valley
                                     B. L. CAROLE
                             Agriculture Experiment Station
                               Washington State University
 ABSTRACT
  A study of irrigation return flows in the Roza
 Irrigation District in the Yakima River Basin in
 central  Washington  was initiated during the
 1971 irrigation season to evaluate methodology
for sediment removal from drainage waters and
provide some assessment of the costs and bene-
fits of sediment control.
  Water management in furrow irrigation re-
gimes has a significant  effect on sediment con-
centration,  nutrient content and overall water
quality of return flows from such systems. Sedi-
ment concentrations in return flows under the
best managed furrow irrigation systems did not
meet the water quality  standards for turbidity
for streams of the region. Some treatment for
sediment control will therefore be necessary.
  Sedimentation basins alone leave much to be
desired in  treating return flows for subsequent
 discharge to streams since only a fraction of the
sediment is removed by such treatment and the
discharge waters still contain turbidities in ex-
 cess of water quality criteria. Other alternatives
alone or in conjunction  with sediment  basins
 will be needed to meet the water quality criteria.

 INTRODUCTION
  Sediment in streams and reservoirs is neither
 a new problem nor one which is peculiar to any
one location in the country. However, it is be-
coming  an increasingly important problem in
many areas of the Northwest from the  stand-
point of potential impact on recreation and fish
life, on clogging of channels, on eventual  reduc-
tion of reservoir storage capacity and on the cost
of treating water for subsequent use.
  Sediment is sometime referred to as inert sol-
ids carried in suspension by water. Considerable
evidence is available to indicate that most sedi-
ment in agricultural runoff is certainly not inert
material but surface active particles which serve
as adsorbing and transport media for toxic and
nutrient chemicals, some  radioactive materials
and some pathogens.
  Agriculture,  in  addition  to  its role as the
largest single  consumptive user of water, is a
major source of sediment pollution. Therefore it
has been reasoned that the ideal approach to
control of sediment pollution would be through
use of proper land management practices. Soil
and water conservation methods can and do re-
duce the quality  of sediment reaching  water-
ways, but the realization of widespread effective
implementation of land management techniques
for the purpose of pollution control is unlikely
until sufficient legal .and/or economic motiva-
tion for such practices is provided. However, the
technological groundwork should be laid now
for eventual curtailment of sediment pollution,
                                            77

-------
78
MANAGING IRRIGATED AGRICULTURE
especially in view of the efforts already under-
way to abate pollution from municipal and in-
dustrial sources.

            Nature of the Problem
  Control at the  source seems  to be the most
feasible  approach to  curtailment of sediment
pollution. Prevention by means of better water
and land management practices would undoubt-
edly be the optimum method of combating sedi-
ment pollution arising from agricultural and ur-
ban regions.  Until such practices can  be pro-
moted,  interim   measures  are  required   to
alleviate the existing situations in some prob-
lem areas.
  A case in point is the situation which exists in
the Yakima River Basin in southcentral Wash-
ington. This region is devoted almost entirely to
agriculture and is one of the more extensively ir-
rigated regions in  the western United States. Ir-
rigation constitutes such a major water use that
irrigation return flow comprise nearly the entire
summer  flow in the lower  eighty miles of  the
river. A study by Sylvester and Seabloom1 indi-
cate that  irrigation return  flows are the major
factor influencing overall water quality of  the
Yakima River.
  Sediment pollution has become an ever wors-
ening problem in  downstream diversion canals
and in the Yakima River itself. Canal water re-
ceiving irrigation return flows often carries such
a heavy  sediment  load that untreated water is
unsuitable  for irrigation. Some farmers have
constructed  silting basins  in an attempt to  re-
move enough of the particulate matter to render
the water fit for use. However, only a portion of
the sediment can be removed and the remainder
still causes serious problems in pumps, sprinkler
heads and pipes. It is ironic that, because of sed-
iment  pollution and  the  associated problems,
many fanners are reluctant to convert from fur-
row to sprinkler irrigation. At the same time, to-
tal conversion to sprinkler irrigation would be a
major step toward alleviating sediment pollu-
tion.
  Economic consequences of sediment pollution
are  now  being  felt   by  cities  and counties
throughout the Yakima Valley. Benton County,
located in  the  lower  region  of the  Valley,
spends upward to $50,000 per  year to  remove
sediments from road  ditches and bridge ap-
                                          proaches. City residents are faced with the task
                                          of replacing sprinkler heads and flushing pipe-
                                          lines day after day when using irrigation water
                                          to  keep lawns  green  and  gardens  blooming.
                                          Pumps have  to  be torn down and cleaned and
                                          city officials are  faced with the never-ending
                                          task of hauling silt from flush points in the city
                                          irrigation system. Coupled  with impaired recre-
                                          ational and  aesthetic  values,  public  reaction
                                          against sediment pollution in the Yakima Valley
                                          has reached the point where interim measures
                                          are needed to combat the existing problem until
                                          improved water  and land management practices
                                          can be implemented.

                                                       Sediment Studies
                                           A preliminary study of irrigation return flow
                                          characteristics in  the Yakima  Valley  was  ini-
                                          tiated  in 1971 to evaluate the feasibility of sedi-
                                          ment removal from drainage waters and to pro-
                                          vide an assessment of the costs and  benefits of
                                          sediment control.
                                           The study was aimed at determining the kinds
                                          and amounts of sediment entering drains from
                                          selected agricultural watersheds and  evaluating
                                          methodology for removing  these sediments from
                                          drainage waters prior to their entry into streams.
                                           The study was conducted in the Roza Irriga-
                                          tion District of the lower Yakima  Valley  on
                                          three separate drainage basins, each represent-
                                          ing a different concept of water management in
                                          a  similar geographical  region. Each drainage
                                          unit was relatively small, representing from one
                                          to three land owners with  a total area of from
                                          200 to 400 acres. Duration of the study was from
                                          the beginning of one irrigation season in spring
                                          until surface  flow  ceased following the final ir-
                                          rigation in late summer.
                                           The study included obtaining information on
                                          water  intake and return flow; on area devoted to
                                          various classes of crops; and on the use of fertil-
                                          izers for each crop. It  involved field measure-
                                          ments of intake  and  drainage water temper-
                                          ature,  and collection of water samples on a bi-
                                          weekly basis for determination of suspended sol-
                                          ids, turbidity, COD, ammonium and nitrate ni-
                                          trogen, total and  soluble  phosphate,  chloride,
                                          electrical  conductivity and  an occasional analy-
                                          sis for coliform bacteria. The water sample to be
                                          analyzed  was immediately cooled  and main-
                                          tained at a  few  degrees  above freezing  until

-------
                                                                    SEDIMENT CONTROL
                                          79
analysis was completed. Nutrient and microbial
analysis were completed within 2 days of sample
collection.
  Quantities  of suspended sediment from each
drainage outlet were characterized as to particle
size, mineral  composition, nutrient content and
adsorption, light scattering ability (turbidity po-
tential) and settling rates.  Flocculation and sedi-
mentation studies were conducted to measure
the effectiveness  of chemical  flocculation  and
coagulation with neutral  salts  and/or synthetic
polyelectrolytes as a means of sediment removal
in agricultural waters.

          Description of  Study Area
  Eight sampling stations on  farms within the
three drainage units were selected as represent-
ing the inflow and return flow water  quality
characteristics for the study areas.
  Stations A-l, A-2 and  A-3 are located within
the unit designated as Drainage A occupying an
area of approximately 300 acres with 80  percent
being  classified  in the  Irrigation Capability
Class II representing medium  textured soils on
slopes of less than  2 percent gradient2.  The
sampling  stations were selected to allow mea-
surement of water quality of the irrigation water
supplied to the basin; of the return flow leaving
the basin; and of the return flow from one indi-
vidual field  of sugar beets. This  basin repre-
sented a  unit where  excellent water manage-
ment practices were carried out on land ideally
suited for surface irrigation.
   Stations B-l and B-2 represents inflow and re-
turn flow characteristics of a drainage unit simi-
lar in size to  drainage A but representing signifi-
cantly different  water management  practices.
Only 20 to 30 percent of the area within Drain-
age B falls in the Irrigation Capability Class II,
the remaining being classified as Class  III and
IV, representative of medium  textured soils on
slopes of greater than 2 percent. This unit repre-
sents water management  practices  where opera-
tors apply significantly  more  water to  similar
crops on land less than ideally suited for surface
irrigation.
   Drainage unit C is a 200 acre unit similar in
soils and topography to unit B but where all the
surface drainage water from the unit is treated
in  a large sediment basin having a  detention
time of approximately 2 hours  prior to discharge
to adjacent canal.  Station C-l represents inflow
water to the unit while stations C-2 and C-3
were located at the inlet and outlet of the sedi-
ment basin to assess the effect of such basins on
the overall water quality of return flows.
  Irrigation in  all units was by furrow applica-
tion with crop rotations consisting of field crops
(sugar beets,  mint and  potatoes) five years out
of six and cereal grains to complete the rotation.
  Intake  of  water  (diversion from  the  Roza
Canal) for use on the  irrigated units averaged
about five and one-half feet depth during the six
months from April to September. An additional
three  to four feet of water were applied to units
B and C from a drainage source adjacent to the
units. An equivalent of about two feet depth was
discharged from unit A and five to 6 feet depth
discharged from units B and C as return flow in
the drains.
  There is no  way to  estimate influent or ef-
fluent  seepage and there is some inherent error
in the flow measurements, so that  the inflow-
return flow balance may be of question although
it was considered sufficient for the purpose of
this study.

          Return Flow Characteristics
  Average values for the water quality param-
eters  measured throughout  the irrigation sea-
son are presented in Table  1. Inflow quality
characteristics  were  sufficiently uniform  such
that the three  stations  A-l, B-l and C-l  were
compiled together  and reported  as applied
water.
   From the values shown, it is evident that wa-
ter and land management factors do have a sig-
nificant influence  on quality parameters of irri-
gation return flows. However, at this point the
relationship is  not clear due to the  large varia-
tions  in  many of the measured  parameters.
Much of the variation noted in Table 1 is related
to  the range in suspended solids concentration
and the related parameters  of  turbidity, COD
and total phosphorus from each drainage unit.
This reflects the inherent difficulty of sampling
watersheds for  sediment discharge  since many
factors affect the suspension, transport and de-
position of sediment from  the point of origin to
the place of sampling.
   It is evident that furrow application of irriga-
tion water does have a significant influence on

-------
80
MANAGING IRRIGATED AGRICULTURE
                                           TABLE 1
             Irrigation and Return Flow Water Quality Average of Biweekly Sampling
                                      for Irrigation Season

Station


Applied
Water
A-2
A-3
B-2
C-2
C-3

Temp
°c


19 A
25.5
20.0
22.2
20.0
19.3

NO3-N
t
\

0.2
0.4
0.4
2.5
e.6
0.4
Soluble
PO4-P



0.1
0.5
0.3
0.3
0.3
0.3
Total
PO4-P

mg/ 1

1.1
34.2
22.6
16.1
34.8
23.2

COD



8.8
162.2
64.2
24.0
75.8
26.8
Total
Salts



70
70
74
79
79
83
Suspended
Solids
X


91
7850
2800
751
3350
770
r
Turbidity
ITU


29
2500*
900*
250*
1300*
400*
 'approximate values due to high concentration of suspended solids.
the water quality of surface  drains. Water re-
turned from furrow  application is greatly en-
riched compared to  its quality incoming.  The
principal degradation relates to increases in sus-
pended solids and the related  parameters of tur-
bidity, total phosphorus and COD.
  Treatment of return flows in sedimentation
basins affords  considerable  removal  of  sus-
pended solids   and  related  contaminants  as
noted from sampling stations C-2 and C-3. From
a  water  quality  standpoint, such treatment
leaves much to  be desired in  treating water for
subsequent discharge  to streams since water dis-
charged from the basin still contained turbidity
values of over 400 JTU units and total phospho-
rus concentrations of over 20 mg/1. Such values
are far in excess  of what is recommended for dis-
charge to Class B waters of the region3.

         Sediment Control Measures
  Much of the sediment carried in irrigation re-
turn flows in the Yakima  Valley never reach the
river or major canals  but  are deposited adjacent
to the field  of origin or in the  small drainage-
ways  serving individual or small groups of oper-
ators.  This  does not  negate the significance of
the problem, however, since sediment in  trans-
port depletes the land  resource from which it
comes and  sediment  deposition  creates  un-
sightly and troublesome accumulations of mate-
rial in nearby drains and  structures and may be
quite  costly  to  the downstream operator. The
fraction of fine  material not deposited in adja-
                                          cent structures may still be of such nature to
                                          create  substantial water degradation in the ma-
                                          jor rivers and canals.
                                             With improved  soil and  water  management,
                                          sediment yields in the Yakima Valley can be re-
                                          duced but it is virtually impossible to reduce ero-
                                          sion in furrow irrigation systems  to  the extent
                                          that sediment and related contaminants in re-
                                          turn flows will meet water quality standards  rec-
                                          ommended for the region. To irrigate most effi-
                                          ciently, water must be applied uniformly to all
                                          parts of the field. With surface irrigation it is im-
                                          possible to  irrigate every spot uniformly,  but
                                          present techniques will permit  operators to ap-
                                          proach  this ideal. In practice, the irrigator finds
                                          that the rate  of irrigation for proper application
                                          varies from irrigation to irrigation and from year
                                          to  year. Soil characteristics, cultivation, mois-
                                          ture, crop, climatic and  many  other conditions
                                          all influence  the water intake  rate of the soil.
                                          This variability probably causes the irrigator the
                                          greatest difficulty and requires  an  excess of wa-
                                          ter to be applied to offset intake rate variances
                                          within  the field. With such water application
                                          where the soil is necessarily  a medium for trans-
                                          porting water to other parts of the field, suspen-
                                          sion of particles is an inevitable result of such a
                                          system.
                                             As a practice which greatly  reduces the load
                                          of sediment in drainage waters, sprinkler irriga-
                                          tion is  finding increased favor in the  Yakima
                                          Valley.  One  of the greatest advantages of the
                                          sprinkler irrigation method  is that it overcomes

-------
the difficulty of varied intake rate since the wa-
ter  is applied at a design rate somewhat lower
than the intake rate of the soil. The water man-
agement problem is greatly simplified and little
or no surface return flow occurs. One deterrent
to conversion to sprinkler irrigation in the Ya-
kima Valley, in addition to the cost involved, is
the problem of sediments in the  supply water
currently available. Plugging of sprinkler heads,
abrasion of pipe and wearing of pumps will be a
major problem  where surface and  sprinkler sys-
tems are being operated simultaneously.
  One alternative to conversion to sprinkler sys-
tems would  be the installation of basins or ponds
to intercept  and collect drain waters for return to
the  head  ditch with no external surface dis-
charge. The same problems of pump  and pipe
abrasion,  channel clogging and basin  sedimen-
tation would still be inherent to this system but
would have  the effect of requiring each operator
to treat the  problem which his individual opera-
tion creates  and would probably make many op-
erators take a  close look at the efficiency and
management of his irrigation system.
  It appears to be most impractical to construct
sediment basins with sufficient detention time
for natural  settling to produce waters  meeting
recommended quality criteria.  Detention times
in the order  of two to three days would be neces-
sary to produce waters of less  than 50 units of
turbidity for drainage waters of the lower Yak-
ima Valley.  Considerable reduction  in basin
capacity can be achieved by the use of chemical
additives in  conjunction with sediment basin to
create forced settling of suspended solids.
  Some preliminary studies  have  been  con-
ducted with  a great number of synthetic organic
polyelectrolytes and several inorganic  salts  to
determine  the effectiveness of each in facilitat-
ing sediment flocculation and removal. The re-
sults of the  best  of  each type  of chemical are
shown in Figure  1.  The results are reported as
percent turbidity removed versus dose rate after
a one hour  settling  period following  chemical
addition. The sediment size fraction used in this
study was the less than 0.05 mm diameter parti-
cles which account for approximately 70 percent
of the total  suspended load in return flow wa-
ters. Some of the cationic and non-ionic poly-
mers at dose rates less than 1 mg/1 appear to be
as effective as the standard flocculants at dose
                   SEDIMENT CONTROL    81

                                 One hour settling time




                                        f—°
                    Dose Rate mg/1
 Figure 1: Percent Turbidity Removed by Chem-
 ical Flocculants
rates up to  100 mg/1. The cost of chemical ap-
plication for flocculation and sedimentation ap-
pear to be prohibitive at this stage but studies
will continue to evaluate the cost of application
as well as the total effects on water quality.

SUMMARY
  Sediment pollution has become an ever wors-
ening problem in downstream diversion canals
and in the Yakima  River itself. Canal water re-
ceiving  irrigation return flows often carry such
a heavy sediment load that untreated water is
unsuitable for sprinkler irrigation. Coupled with
impaired recreational and aesthetic values, pub-
lic reaction against sediment pollution  in  the
Valley  has  reached  the  point where  interim
measures are needed to combat the existing
problem until improved water and land manage-
ment practices can be implemented.
  Treatment of drainage waters  in  sedimenta-
tion basins does afford considerable removal of
suspended solids, total phosphorus and  COD
from such waters. From a water  quality stand-
point, such  treatment leaves much to be desired
for discharge to streams since waters discharged
from the basins still contained turbidity values
in excess of 400 JTU units and total phosphorus
concentrations of over 20 mg/1.
  To  produce return  flow  waters meeting  the
recommended quality  criteria,  established by
the Washington State Department  of Ecology,
other alternatives to natural settling must be im-
plemented.  These  alternatives appear to  be:

-------
82     MANAGING IRRIGATED AGRICULTURE
1) installing practices  which greatly reduce the
load of solids in drainage waters (sprinkler irri-
gation for  example); 2) use  of basins as inter-
ception ponds to collect drain waters for return
to the head ditch with no external discharge; or
3) use of chemical additives in conjunction with
settling basins to create forced settling of solids
and  reduce detention  time  requirements. Each
or some combination of these or other manage-
ment programs  may be needed, but implemen-
tation should be initiated soon in view of the ef-
forts already underway to abate pollution from
non-agricultural sources.
REFERENCES
  1. R. O.  Sylvester  and  R. W.  Seabloom.
"A  Study of the Character and Significance of
Irrigation Return Flows in the Yakima River Ba-
sin".  USPHS  Research  Report  May  1963.
  2. J. L. Rasmussen, et al. "Soil Survey of
Benton County Area, Washington". U.S. Dept.
Agr., Soil Cons. Service. 1971.

  3. "Implementation  and Enforcement  Plan
for  Water  Quality  Regulations"  Washington
State Department of Ecology. Sept. 1970.

-------
     Treatment of  Irrigation  Return Flows
                 in  the San  Joaquin  Valley
                                   LOUIS A. BECK
                           California Regional Water Quality
                                     Control Board
                                   Fresno, California
ABSTRACT
  Agricultural tile drainage in portions of Cali-
fornia's San Joaquin Valley must be disposed of
because of their brackish nature. It is planned to
collect this tile drainage and dispose of it in the
San Francisco Bay System. The only constitu-
ent of the tile drainage that might create prob-
lems in the receiving waters is nitrogen.  The ni-
trogen is essentially all in the nitrate form.
  Two  methods  of nitrogen  removal  were
studied and found to be feasible. These are bac-
terial denitrification  and algal production and
harvesting  (algae  stripping). Operating condi-
tions, design criteria and costs  were developed
for both  methods. Denitrification was  studied
both in deep ponds and packed columns.
  Denitrification reduced the nitrate concentra-
tion from 20 mg/1 to less than 2 mg/1 through-
out the year.  It was necessary  to add organic
carbon to  the process to achieve this efficiency.
Algae stripping reduced the nitrate concentra-
tion from 20 mg/1 to 2 to 4 mg/1 depending on
the time of the year.
  Studies are  now being conducted of a symbi-
otic method of nitrogen removal.  This method
combines algae or plant growth and bacterial
denitrification in one pond. Preliminary results
indicate that cost of the symbiotic method may
be one-quarter to one-third the  cost of either
bacterial denitrification or algae stripping.
  Studies are also now being conducted on the
use of reverse osmosis techniques for desalina-
tion  of the  tile drainage. If the reverse osmosis
studies are successful, the complete reclamation
process will be studied in pilot scale. This will
include boron removal and brine disposal.

INTRODUCTION
  In recent years an ever increasing emphasis
has  been  placed  on nitrogen removal  tech-
niques, both for reasons of public  health and the
control of  eutrophication.  The subject studies
were instituted because  of the possible eutro-
phying effect of the discharge of large quantities
of agricultural tile drainage into the San Fran-
cisco Bay System. The tile  drainage to be dis-
posed of results from the irrigation of large por-
tions of California's San Joaquin Valley that
have been classified as drainage problem areas

This  paper was presented  at  the National Confer-
ence  on Managing Irrigated Agriculture to Improve
Water Quality in Grand Junction, Colorado, May 16-
18, 1972.
                                           83

-------
 84      MANAGING IRRIGATED AGRICULTURE
                                                           SCALE OF MILES
                                                     I0    0    10    20   30   4C
Figure I:  Existing and Potential Agricultural Waste Water Disposal Problem Areas. San Joaqui i Valley-Cali-
fornia

-------
                                                         TREATMENT OF RETURN FLOWS
                                          85
(Figure 1). Of the over 7 million acres (2.8 mil-
lion ha) of irrigable land in the Valley, approxi-
mately  1.7 million acres (0.7 million  ha) have
been classified as existing or potential drainage
problem areas — lands where brackish high wa-
ter tables have or will become a problem1.
   Most of the  saline agricultural wastewater
that is presently  being generated within the Val-
ley is discharged to the San Joaquin  River. As
the quantity of this waste increases, due to the
increased availability of water for irrigation, the
waste load  on the river would reach a point
where downstream users would be deprived of a
usable water supply.
   For this reason, a system is now under con-
struction to collect much of the saline drainage
and transport it from the Valley. The system will
serve 300,000 acres (120,000 ha) of the drainage
problem area. It is being built by the U.S. Bu-
reau of Reclamation (USER)  to handle primar-
ily the  wastewater from the  Federal  San Luis
Service Area. California's Department of Water
Resources (DWR) is also aware of the drainage
problem and  is  making  plans to serve those
areas outside  of the Federal  San  Luis Service
Area. The predicted quality of the waste to be
carried by the San Luis Drain is  presented in
Table  I2.  It is estimated that by the year 2000
there will be 500,000 acre-feet (6.2 x 108-cu m)
of tile drainage to  be disposed of annually. (This
is only about 4 percent of  the water  to be ap-
plied to the Valley in the same  period.) The
peak flows, during the summer irrigation sea-
son, are expected  to reach 700 mgd (2.65 x 106 -
cu m/day).
   Prior to the study reported herein, the DWR
and  USER had  investigated disposal  of tile
drainage  for several years. Of all the alterna-
tives studied, they concluded that the most eco-
nomical and safest method bf disposal would be
a gravity  concrete-lined canal, or drain, that
would discharge into the upper reaches of San
Francisco Bay3.  They further concluded that the
only  major constituent of  the drainage that
could create problems in the receiving water is
nitrogen.   A  1967  Environmental Protection
Agency (EPA),  then  Federal Water  Pollution
Control Administration (FWPCA), study of wa-
ter quality problems which could result from the
drainage   water   concurred  with the  earlier
study4.  The FWPCA  concluded that  problems
would  not occur if the nitrogen content of the
subject waste  could be reduced to 2 mg/1 or
less.
  The  DWR,  USBR,  and FWPCA joined to-
gether  to determine if a method of nitrogen re-
moval for tile drainage could be developed or if
the costly alternative of ocean disposal would
have to be adopted. In order to evaluate possible
methods  of nitrogen removal under field condi-
tions, the Interagency  Agricultural Wastewater
Treatment Center was established. It is located
near Firebaugh,  California (40 miles west of
Fresno) in one of the existing drainage problem
areas. This paper presents a comparison of the
various treatment methods studied at the Cen-
ter.

            Treatment Techniques
  Ninety-five percent,  or  more, of the nitrogen
in tile drainage is in the nitrate form. Therefore,
the problem of reducing the nitrogen content of
tile drainage becomes  one of removing nitrate-
nitrogen. Two basic methods of nitrate removal
were investigated: bacterial  denitrification, and
algal biomass  production and harvesting (algae
stripping).  Bacterial denitrification was studied
in both anaerobic filters and deep ponds. A sym-
biotic  method  (combination of photosynthetic
and bacteriological in one pond) of nitrogen re-
moval is now being studied. In addition to the
nitrate removal  studies,  desalination is being
studied at the  Treatment Center.

Bacterial Denitrification
   Bacterial denitrification is a process in which
bacteria  utilize  highly oxidized anions for the
oxidation of organic matter5. The overall pro-
cess can  be expressed stoichiometrically, as  a
two-step or three-step reaction. Nitrate (No3) is
reduced  to nitrite (NO2~) and then from nitrite
to molecular nitrogen (N2) with or without the
intermediate formation  of  hyponitrite (N2O2).
The process can be shown to be thermodynami-
cally sound (capable of producing energy for or-
ganisms  utilizing it)6. Several commonly occur-
ring facultative  bacteria are denitrifiers,  for ex-
ample, Bacillus brevis1. The occurrence of bac-

 'One of  the three bacteria of the  Bacillus genus
which are cultured commerically to produce polype-
tide-type  antibiotics.

-------
       MANAGING IRRIGATED AGRICULTURE
terial denitrification  in existing  sewage  treat-
ment plants is well documented and several
pilot-scale  units  have  been built  and success-
fully operated for denitrification7 8 9.10)
  Unless  biodegradable   organic  material  is
present  in adequate quantity, effective denitrifi-
cation will not  take place.  Most agricultural tile
drainage has a low organic content.  Therefore,
methanol was added to the waste prior to entry
into the anaerobic  filters and deep ponds oper-
ated at the Center".
  The main factors, with regard to bacterial de-
nitrification, requiring  study  were: the general
feasibility of the  process, detention time, meth-
anol  requirement,  and operating efficiencies
under field conditions.


Anaerobic  Filters
               GAS RELEASED TO ATMOSPHERE
                t 1J I ! t! I I
  In  the  anaerobic filter the waste  is  passed
through a packed column-type reactor (a vessel
containing an inactive medium). The filter me-
dium provides a  solid support for the denitrify-
ing bacteria.  The advantage of the filter is that
the solids retention time of the unit is longer
than  the hydraulic  detention time and, there-
fore,  there is no need for the recycling of sol-
ids12.  Anaerobic filters can be operated with up-
flow or downflow configurations, as long as the
unit is kept anaerobic (or nearly so). All of the
filters operated at the Center have been  upflow
units (Figure 2).
  Four different sizes of anaerobic filters have
been used: 4-, 18-, and 36-inch (10-, 46-, and 91-
cm) diameter13  and  10-feet by  10-feet  (3-m  x
3-m)  square. The 10-feet by 10-feet filter has a
false  bottom (of the type used  in water treat-
ment rapid sand filters) with an 8-inch (20-cm)
plenum and a 6-foot (1.8-m) bed depth. The pri-
mary purpose for building this  filter was to in-
vestigate the effect of scale-up  on process effi-
ciency due to changes in the hydraulic  regime.
   WASTE INFLUENT
                   LEGEND
                 I OPTIONAL SAMPLE TAPS
 Figure 2:  Schematic Diagram. Upflow Anaerobic
 Filter Denitrification Process

Deep Ponds
  The feasibility of bacterial  denitrification in
deep  ponds  under field  conditions  was deter-
mined in  3-foot (0.9-m)  diameter  simulated
deep ponds14. Once the process was shown to be
feasible,  two large ponds were constructed as
shown schematically  in Figure 3. The larger of
the two ponds constructed at the Center is 200 x
50-feet (61-x 15-m) and has a floating cover. The
smaller pond is 50-x  50-feet (15-x 15-m) and  is
not covered. Both ponds are approximately 14-
feet  deep (4.3-m). Through the parallel opera-
tion of these two units, it was  possible to evalu-
ate the significance of wind  mixing, and algal
growth on process efficiency.  The solids reten-
tion time and hydraulic detention time are the
same,  therefore, the  controlling parameter in
pond operation is the regeneration time of the
bacterial population.  To  increase the solids re-
tention time, recycle is  necessary.  If an  ade-
quate  residence time  is not provided, effective
denitrification is not maintained.
                                            - OPTIONAL POND COVER
 BIODEGRADABLE ORGANIC
 CARBON INJECTION   ~-~

                                                                                    TREATED EFFLUENT
       WASTE INFLUENT

                             RECYCLE =25% OF INFLUENT

                     Figure 3:  Schematic Diagram. Deep Pond Denitrification Process

-------
                                                         TREATMENT OF RETURN FLOWS
                                                            87
Algae Stripping

  Research on the treatment of domestic  and
industrial wastes in photosynthetic systems has
been studied for a number of years15. As a result
of the findings of such studies, it has become ap-
parent that nutrients are assimilated by algal
cells in significant amounts and, hence, are re-
moved as part of the operation.
  Figure 4 is a schematic diagram of algae strip-
ping.  It is immediately evident, upon observing
any ponded tile  drainage, that algae definitely
grow in this type of waste. The main problem in
the process is not the mere growing of algae, but
the optimization  of growth factors to maximize
their  nitrate uptake rate. Some of the  factors
that   must  be  investigated   before  the algal
growth system could be evaluated were: growth
pond  depth, hydraulic detention time, recircula-
tion, mixing, pH, and nutrient addition. All of
these factors must be  optimized before a photo-
synthetic  system can be designed.  Once  the
algae  have been grown; the problem of separat-
ing them from the waste becomes paramount.
Flocculation,   sedimentation,   flotation,   and
microscreening are all processes that were eval-
uated to  determine the least costly technique.
  Three  types  of growth vessels  were used at
the Center: a light  box,  small-scale growth
ponds (miniponds), and a  1/4-acre  (0.10-ha)
large-scale rapid  growth pond16. The light  box
was used for the preliminary screening  of  fac-
tors such as pH and nutrient  addition. These re-
sults were then verified in the miniponds. In ad-
dition  to varifying light box  results, miniponds
CHEMICAL ADDITION -
      WASTE INFLUENT
 ALGAE
GROWTH
 POND
                  MARKETABLE
                  BY-PRODUCT
                 were used to study detention times, mixing, wa-
                 ter depth, and climatic conditions. Final testing
                 was conducted in the large-scale rapid growth
                 pond. The pilot-scale growth  pond also served
                 as the primary source of algal-laden waste for
                 the separation studies.
                   The  harvesting  studies  were  divided  into
                 three  categories — separation, dewatering, and
                 drying.  The majority of the work that was done
                 on separation has consisted of jar tests to eval-
                 uate various flocculants.  Pilot-scale separation
                 techniques studied included: centrifugation, mi-
                 croscreening,  filtration,  and  flocculation and
                 sedimentation.  The  following dewatering and
                 drying  equipment  were investigated:  cylindri-
                 cal  and basket-type centrifuges,  a vacuum fil-
                 ter, sand drying beds, and flash drying.

                 Symbiotic Method
                   During the study of the  algae stripping and
                 denitrification methods,  there were indications
                 that   under  certain  conditions  photosynthesis
                 and  bacteriological denitrification  could  occur
                 in the same pond.  This phenomenon occurred
                 both  when algae and emergent vegetation were
                 the plant material. The addition of methanol
                 was not required for the symbiotic reaction. It
                 has been theorized that the decomposition of
                 the plant material provided the  organic carbon
                 necessary  for  the  bacteriological  denitrifica-
                 tion.  A two-year study was started in July of
                 1971  to define the mechanisms   and nitrogen
                 pathways in this process  and to determine the
                 efficiency,  operational procedures, and costs of
                 the process.
                                                        ALGAE
                                                        LADEN
                                                        WASTE
SEPARATION
EQUIPMENT
TREATED
EFFLUENT
                                                     ALGAL PASTE
                        Figure 4: Schematic Diagram. Algae Stripping Process

-------
88
MANAGING IRRIGATED AGRICULTURE
Desalination
   In  July 1971, reverse  osmosis studies  were
started  on  the agricultural wastewater.  It is
planned to study different configurations of re-
verse osmosis.  If the reverse  osmosis  studies
demonstrate  that the process will operate effi-
ciently and economically, complete reclamation
studies will be instituted as an alternative to ni-
trogen removal and disposal to San Francisco
Bay. The  reclamation studies, if needed, will in-
clude pilot-scale operation of reverse osmosis
and boron removal facilities and demonstration
of brine disposal techniques.

            Analytical Techniques
  The routine analyses used to monitor the op-
eration of the various units at the  Center are
summarized as follows:
                                          temperatures were monitored daily with maxi-
                                          mum-minimum  thermometers and periodically
                                          with  8-day recording thermographs.  The in-
                                          fluent pressure required  for an anaerobic filter
                                          to maintain a constant hydraulic detention time
                                          was monitored with varying frequency through-
                                          out the study. Flows were calibrated volumetri-
                                          cally. In addition tracer  studies using  the chlo-
                                          ride  ion as  a tracer were run to  determine the
                                          actual hydraulic regime  of the different units.
                                          Pre-and postinjection  density corrections were
                                          made using sodium  sulfate.  The tracer studies
                                          were analyzed using the  volume apportionment
                                          technique18'19.

                                                         Influent Quality
                                            The tile drainage used for the investigation
                                          reported in this paper came from a drainage sys-
             ROUTINE ANALYSES PERFORMED AT THE INTERAGENCY
               AGRICULTURAL WASTEWATER TREATMENT CENTER
                    Analysis
                                                  Technique
        Nitrate-Nitrogen
        Nitrite-Nitrogen
        Total Kjeldahl Nitrogen
            Ammonia Nitrogen
            Organic Nitrogen
        Orthophosphate
        pH
        Alkalinity
        Dissolved Oxygen
        Suspended Solids
        Volatile Suspended Solids
        Optical Density
        Electrical Conductivity
        Algal Cell Counts and Identifiers
        Methanol
                                Brucine Method and/or Selective Ion Electrode
                                Standard Methods, 12th Edition17
                                Kjeldahl Method
                                Distillation Method
                                Kjeldahl Method
                                Stannous Chloride Modification
                                Glass Electrode
                                Standard Methods, 12th Edition
                                Winkler Method
                                0.45/u Glass Paper, 103° C
                                0.45^ Glass Paper, 600° C
                                450^, 5 cm Cell
                                Galvanic Cell
                                Sedgewick-Rafter Cell
                                Gas Chromatograph, Carbowas Column, Flame loni-
                                  zation Detector
Most samples were normally collected between
8:00 and 9:00 AM and  analyzed immediately.
Some samples were taken in the afternoon to
gather information on changes that may occur
during the  peak  photosynthetic period. Diurnal
studies were also conducted, as the need  arose.
  In  addition  to  chemical analyses,   several
physical  parameters  were  monitored.  Water
                                          tem that was in existence when the Center was
                                          constructed. This system, which serves approxi-
                                          mately 400 acres (160-ha), was chosen because
                                          of its proximity to land available for construc-
                                          tion of the Center and the quality of the drain-
                                          age, which was considered to be quite "typical".
                                          The range of constituent concentrations in this
                                          drainage are as follows:

-------
                                                         TREATMENT OF RETURN FLOWS
                                                                     89
 CHARACTERISTICS OF TILE DRAINAGE
  USED AT THE INTERAGENCY WASTE-
      WATER TREATMENT CENTER
      Constituents
Range of Concentrations
         mg/1
Total Dissolved Solids
Salts
Sulfate
Sodium
Chloride
Calcium
Magnesium
Bicarbonate
Boron
Potassium
Nutrients
Nitrogen
Phosphate
Pesticides
Others
5-Day BOD
COD
DO
2500-7600

1500-3900
620-2050
310-640
160-390
70-230
280-330
4-15
4-11

5-25
0.13-0.33
0.001

1-3
10-20
7-9
The variations are seasonal and due primarily to
irrigation applications and rainfall. With the ex-
ception of the variation in nitrogen concentra-
tion, it was decided that all other constituents
should be allowed to vary naturally. Sodium ni-
trate was used to maintain an average influent
nitrate-nitrogen concentration of approximately
20 mg/1,  the predicted average  nitrogen con-
centration of the waste in the San Luis Drain.

                 Discussion
  Figure 5  illustrates the  predicted  seasonal
variations of flow and nitrogen concentration.
The  distribution   of  flow  was  determined
through a  monitoring program and computer
simulation3. The nitrate-time relationship was
developed from data collected by the DWR in a
special three-year water quality monitoring pro-
gram20. Figure 6 is  a summary of westside San
Joaquin Valley air temperature data from three
stations in the general area in which a treatment
plant might be constructed (between Los Banos
and Tracy, California)21. These figures are pres-
ented to illustrate the fact that the majority of
the waste load will occur during the warmer
        JAN    FE8     MAR     APR     MAY     JUN     JUL     AUG     SEPT     OCT     NOV
                                                                                    DEC
Figure 5: Predicted Seasonal Variation of Tile Drainage Flow 8 Nitrogen Concentration from San Joaquin
Valley, California

-------
90     MANAGING IRRIGATED AGRICULTURE
   120
   no
   9O
 0  TO
   "
   40
    IO
                                                    _„-' N
                                                                     JULY 1962 THROUGH SEPT 1967
                                 -MAXIMUM DAY OF RECORD
                                                                                        30
        JAN
               FEB    MAR     APR     MAY
                                          JUN     JUL

                                            MONTH
                                                       AU6
                                                              SEPT
                                                                     OCT
   Figure 6: Annual Air Temperature Variation on the West Side of the San Joaquin Valley, California
summer months. Seventy percent of the annual
nitrogen load,  in pounds per day (grams/day),
will arrive at the plant between April 1 and Sep-
tember 30.  The significance of  this lies in the
fact that  all of the nitrogen removal  systems
being evaluated at the Center are biological sys-
tems, and are  temperature and/or light  sensi-
tive.
  Operational  data will be presented for each
system to illustrate individual operating criteria,
etc. The individual systems  will then be com-
pared.

Anaerobic Filter
  The data from experiments designed to inves-
tigate the significant  media  size,  texture, and
sorptive quality are summarized below13. Based
on  the results of an initial feasibility investiga-
tion,  all filters  operated to  generate this  data
had hydraulic  detention times (based on  void
volumes)  of 2  hours. Medium surface  texture,
size, and sorptive  quality have no apparent af-
fect on removal efficiencies.  After an extended
period of  continuous operation the bacterial
mass  within filters containing media with dia-
meters of less than 1-inch (2.54-cm) built  up to
the point where the required influent pressures
(as high as 60-psig) rendered them uneconomi-
cal.

  NITROGEN REMOVAL EFFICIENCIES
 FOR FILTER DENITRIFICATION UNITS
     CONTAINING VARIOUS MEDIA
        Medium
 Nitrogen Removal
 Efficiency, Percent

Mm.   Max. Average
Activated Carbon
Washed Sand
5/16" Coal
5/16" Volcanic Cinders
3/8" Aggregate
5/8" Volcanic Cinders
1-Coal
1" Volcanic Cinders
1" Aggregate
89
84
80
85
82
87
81
89
89
99
97
98
98
97
97
98
97
98
96
93
93
94
94
91
93
96
94
  The data gathered from the long-term opera-
tion of three filters containing 1-inch diameter
media are summarized in the following table.

-------
                                                        TREATMENT OF RETURN FLOWS
                                                                91
               SUMMARY OF LONG-TERM PERFORMANCE OF FILTERS
                    CONTAINING 1-INCH DIAMETER AGGREGATE
Detention Days
Time
Mrs.
0.5
1.0
2.0
of
Operation
275
240
244
Percent Nitrogen

Min.
40
64
79
Removal
Max.
91
97
97
Required Influent
Pressure, psig
Average
68
88
91
Min.
3.5
3.2
3.5
Max.
11.8
9.6
9.7
Average
7.2
5.4
6.2
For about the first 150 days of operation, the re-
quired influent pressure fluctuated within a nar-
row range. There did not appear to be any need
for backwashing, etc.  After that time, an  up-
ward  trend appeared. Investigations were con-
ducted to determine the best technique to re-
duce the pressure. Backwashing with a combi-
nation of air and water appears to be the best
method  of cleaning the filter.
  The temperature of the influent to the units
varied from a high of 22° to a low of 10° Centi-
grade. There were some indications of  a rela-
tionship  between temperature and nitrogen re-
moval efficiency;  however, the relationship was
far from obvious. By grouping the nitrate and
nitrite removal data for samples taken  at  dif-
ferent depths within a filter, the reduction in ef-
ficiency  becomes more  apparent as illustrated in
the next table.

PROFILE DATA GROUPED BY PERCENT
   OF FILTER REQUIRED  FOR 80% OR
      GREATER NO3 + NO2 REDUCTION

        (1-inch diameter aggregate media,
        2-hour hydraulic detention time)
    Percent of Filter
         Used '
Average Influent
Temperature ° C
      25% or less
      25% to 50%
      50% to 75%
      14.5
      12.8
      11.9
As the temperature of the influent dropped, the
percent of the filter bed required for the same
degree of treatment increased. The 6-foot bed
depths used for this investigation and the range
of temperatures encountered were such that the
effect was not reflected  in the overall  treatment
                      efficiencies of the units. This information is only
                      qualitative,  since steady-state conditions  were
                      not reached at any given temperature.
                        The following equation  was developed  by
                      McCarty to express the concentration of metha-
                      nol required for denitrification11.
                      Cm
                      Cm

                      cr

                      Cr
                      No


                      Nl
         (1.90 N0+ 1.77 N, +0.67 D0)'
         required   methanol   concentration,
         mg/1
         consumptive ratio   (determined  ex-
         perimentally)
         total quantity  of  organic  carbon
         source  consumed during  denitrifica-
         tion divided by the stoichiometric re-
         quirement for denitrification  and de-
         oxygenation
         initial nitrate-nitrogen concentration,
         mg/1
         initial  nitrite-nitrogen concentration,
         mg/1
         initial   dissolved  oxygen  concentra-
         tion, mg/1
  The average  consumptive  ratio  calculated
from the data gathered at the Center equals 1.47
mg/1  with a standard deviation of + 0.367. The
fairly  large standard deviation claculated for the
consumptive ratio is more likely due to inherent
difficulties in field studies than fluctuations in
system requirements. With the development of
an  automatic  methanol control  system, which
would respond to changes in influent and efflu-
ent quality, this  variation should be drastically
reduced. The methanol requirement for a typi-
cal influent containing 20 mg/1 of nitrate-nitro-
gen and 8 mg/1  of dissolved oxygen calculated
by equation1 equals 64 mg/1.

-------
92     MANAGING IRRIGATED AGRICULTURE
Deep Ponds
  The initial field studies of bacterial denitrifi-
cation in deep ponds were conducted in the Fall
of 196714. In these studies, units were operated
at various detention times with and  without
pond  covers.  Adequate denitrification  took
place in ponds with detention times as low as 5
days.
  Data gathered from the large ponds,  which
started operation in  the  Fall of 1968, are sum-
marized in Table 2. Covered ponds have consis-
tently  outperformed  uncovered ponds.  There
are two  primary reasons for this. Large algal
populations develop in uncovered ponds and the
resultant high  dissolved  oxygen concentration
inhibits denitrification.  The second reason is
that more  of the influent  is  short-circuited
through the unit because of wind mixing and
temperature variation in an uncovered  pond.
The high nitrogen removal efficiencies recorded
for the uncovered pond  when operated with a
10-day detention time was probably caused by
the formation  of a natural partial cover. The
cover was composed of decayed algal and bacte-
rial cells floating on  the surface of the  pond.
Some type of covering is  necessary for sufficient
denitrification to take place.
  There were  numerous  mechanical  equipment
breakdowns associated with these units;  there-
fore, it is impossible to  accurately predict the
minimum detention time possible. However, a
ten-day detention time, with 25% recycle, and a
pond depth of 14-feet (4.27-m) have been shown
to be  effective for covered pond  summertime
operation (Table 2) and is used  in this paper.
  The methanol requirement for pond denitrifi-
cation follows the same equation developed for
filter  denitrification.  The average consumptive
ratio for deep  ponds  has  been found  to be stati-
stically equal to that  found for anaerobic filters,
within the accuracy possible under field condi-
tions.
  Based  on  the  feasibility and  operational
studies conducted at the Center, the cost of both
methods  of bacteriological denitrification (an-
aerobic filters and deep ponds) is estimated to
be about 90 dollars per million gallons.

Algae  Stripping
  The results of the  algae stripping studies are
divided into three subsections; growth, harvest-
ing, and algae utilization.
  Growth. Light box studies conducted early in
the investigation indicated a serious phosphorus
limitation in tile drainage. The addition of ap-
proximately 2 mg/1 of phosphorus was found
necessary. This was further confirmed in studies
in the miniponds.
  Once the required phosphorus addition was
determined, work began on optimizing nitrate
assimilation. Of all the nutrient additions inves-
tigated, iron and carbon were the only two that
have significant effects on assimilation. The op-
timum amount of iron that should be used has
been found to be  approximately 2-3 mg/1. A
seasonal variability  in the iron requirement was
indicated and  2-3 mg/1 appears to be a maxi-
mum requirement. A mixture of carbon dioxide
(5%) and air improved nitrate assimilation. Car-
bon dioxide studies in the miniponds indicated
that carbon additions  enhanced growth  only
during the summer months.
  Physical parameters which  have been studied
that affect the  process are: water depth, mixing,
and hydraulic  detention time. Water depths of
from  8- to  16-inches  (20.3-  to 40.6-cm) have
been studied in combination with various deten-
tion times and mixing schedules. The following
data summarizes nitrogen removal data for sev-
eral miniponds  operated  to  study the signifi-
cance of these  parameters.
  This table is intended to show the effects of
physical parameters only.  The low efficiencies
resulted because the  other factors that  affect
growth were   not optimized.  For summertime
operation a design depth of 12-inches (30.5-cm)
and a detention time of 5-days has been found
to be  adequate. Because  of the difference be-
tween  summer and winter temperatures and
light a 13- to  15-day detention  time is required
in the winter. Mixing was found not required for
optimum growth when carbon was added to the
system.
  Harvesting.  Oswald,  et.  al.,  determined that
flocculation  and sedimentation was the most
economical and efficient method of algae sepa-
ration at the time of their study22. Consequently,
the first work  done on separation at the Center
involved jar   tests.  The  following flocculants
were investigated:  alum,  lime, ferric chloride,
ferric sulfate, and various polyelectrolytes. Bet-
ter  than ninety percent removal  of algae cells
from the liquid has been achieved with  alum,
lime, ferric sulfate,  and several  polyelectrolytes.

-------
                                                        TREATMENT OF RETURN FLOWS
                                          93
         NITRATE REMOVAL EFFICIENCY OF ALGAE STRIPPING GROWTH
                  UNITS AS RELATED TO PHYSICAL PARAMETERS
                  Detention Time

           Days       Nitrate Removal in %

                    Max.     Min.    Average   Inches
       Depth

           Nitrate Removal in %

         Max.     Min.    Average
2.5
5.0
8.0
10.0
75
80
82
90
43
59
74
80
59
75
78
85
8
12
16

98
87
87

71
74
61

83
79
75

The most economical flocculant used was ferric
sulfate.
  A  shallow-depth sedimentation unit with in-
clined tubes in the sedimentation chamber con-
sistently  removed 95  to 97 percent of the  sus-
pended solids. The algal slurry from this  unit
contained 1 to 2 percent suspended  solids.  A
second concentrating device, a Sanborn rapid
sand filter also removed about 95 percent of the
suspended solids and produced an  algal slurry
containing 1 to 3 percent solids. Other units (mi-
croscreen,  upflow  clarifier,  and   centrifuges)
were tested  as concentrating devices but were
not found to  be as effective or reliable as the
sedimentation  and rapid sand filter systems.
  A  vacuum filter using a multifilament nylon
belt produced  an algal cake containing about 20
percent solids from an influent with about 0.3 to
3 percent solids. A self-cleaning centrifuge  pro-
duced an algal product with about 10 to 12 per-
cent solids and removed up to 95 percent of the
influent algae  (influent concentrations ranging
from 500 to 30,000 mg/1). This centrifuge was
found to be more effective than either a solid
bowl or nozzle type of unit.
  The algae were normally dried in the open at-
mosphere to about 85 to 95 percent solids, a con-
centration which allowed for safe storage of the
material. One algae sample containing about 15
percent  solids was spray dried. No  problems
were experienced in attaining the  desired  level
of moisture in the product.
  Algae  Utilization. At peak flows more  than
one million  pounds (4.5 x 105-kg) of dry algae
will be produced each day. The use of algae as a
protein supplement  in the feed formulations
for sheep and hogs has been shown to be feasi-
ble22. A study investigating the use of algae for
poultry feed determined that in addition to the
value of the protein content, a benefit is  derived
from the presence  of significant quantities of
Xanthophyll23.  (A carotenoid  which is  used to
increase the color of egg yolks and bird  flesh to
enhance market value.) Other possible uses for
algae include soil conditioning or fertilization.
Depending on  use,  algae have  been found to
have a value of between 80 and  160 dollars per
ton (912 kg).
  Based on the studies conducted at the  Center,
the estimated cost of nitrogen  removal by algae
stripping is about 100 dollars per million  gal-
lons. This includes all phases of algae stripping:
growth, separation, dewatering and drying.  The
cost estimate also includes income from  the  sale
of algae as a by-product at 60  dollars per ton.

Symbiotic Nitrogen Removal
  The  USBR operated a ten-acre test in a rice
paddy  configuration  with  water grass  as  the
emergent vegetation.  This test plot had many
operational  problems,  but  the effluent  consis-
tently contained less than 2-3 mg/1 of nitrogen.
At  the Treatment Center  two miniponds were
operated in a symbiotic mode  with algae as the
plant material.  These miniponds nearly  always
achieved the effluent object of 2 mg/1. Prelimi-
nary cost estimates based  on  these two opera-
tions were  25 to 35 dollars per million gallons.
These costs are much lower than bacterial deni-
trification and  algae  stripping and led  to  the

-------
94
MANAGING IRRIGATED AGRICULTURE
present symbiotic studies at the Treatment Cen-
ter. Units are being operated  to  determine  if
deeper ponds can be utilized and what detention
times are required. The  units under study are
still in the shake-down period and the results are
still erratic.

Desalination
  A tube-type reserve  osmosis unit is being
tested at  the Center. The results to  date indicate
that the  total dissolved solids  can be reduced
from 4,000 mg/1 to 600 mg/1 at 600 psi with a
flux of 20 gallons per square  foot per day. This
datum is based  on the first  set of membranes
which  covered a  range of curing conditions.  A
second set of membranes  have been installed
based  on the first series of tests and long term
operation will now be tested.

           Comparison of Methods
  The pilot-scale studies at Firebaugh have de-
monstrated  that three  methods  can  remove
ninety percent of the nitrogen (mostly nitrate)  in
agricultural wastewaters. Differences in the bio-
logical systems and physical configurations  of
the processes result in  many differences be-
tween the methods.
  Anaerobic filters operate efficiently with the
shortest  detention times of the three methods
studied, one hour, because" of the concentrating
                                          and containing of bacteria within the units. Al-
                                          gae are not as readily washed out of ponds as
                                          bacteria. Therefore, algae growth ponds operate
                                          at  a five-day  detention time compared to ten-
                                          days for deep  pond denitirfication. For a design
                                          flow of 700 million gallons per day, the relative
                                          surface areas  would be 37-acres (15-ha) for fil-
                                          ters, 1,400-acres  (570-ha)  for denitrification
                                          ponds  and 10,700-acres (4,300-ha)  for  algal
                                          growth ponds. All three treatment methods re-
                                          quire some chemical additions. The detrification
                                          systems require methanol. The algal system will
                                          require phosphorus, iron, and carbon for growth
                                          and ferric sulphate for separation. Algae strip-
                                          ping requires  more mechanical equipment than
                                          bacterial denitrification. Chemical feed pumps
                                          for  methanol  addition, influent pumps for an-
                                          aerobic  filters,  and  recycle  pumps  for deep
                                          ponds are the  major pieces of mechanical equip-
                                          ment required for bacterial denitrification.  Be-
                                          cause of the  low dissolved oxygen  content of
                                          wastes  that have been treated by  these  pro-
                                          cesses, it may be necessary to reaerate the ef-
                                          fluent prior to discharge. Algal stripping will re-
                                          quire phosphorus, iron, and carbon dioxide feed
                                          systems;  separation  equipment;  dewatering
                                          equipment;  drying equipment; and  equipment
                                          to  handle to algae to prepare it for marketing.
                                          The foregoing information is summarized  be-
                                          low.
          COMPARISON OF NITRATE REMOVAL METHODS FOR ONE MOD OF
               AGRICULTURAL WASTE CONTAINING 20 MG/L NITRATE
                                        NITROGEN
                                    Bacterial Denitrification

                             Filters           Deep Ponds
                                                               Algae
                                                              Stripping
          Detention          1 hour

          Chemical Addition   Methanol
         By-product Produced None
                                        10 days

                                        Methanol
         Surface Area of
         Treatment Unit

         Equipment
         Requirements
                     2,300 ft.

                     Chemical
                     feed; influent
                     pumps; aeration
                                        None
2.0 acre

Chemical feed;
recycle pump
  aeration
                   5 days

                   Phosphorus
                   Iron
                   Carbon Dioxide
                   Ferric Sulfate

                   1 ton
15.3 acre

Chemical feed;
Separation;
dewatering; drying;
by-product handling

-------
  Each  of the methods has special operational
requirements. The solids level in denitrification
ponds and  algal growth ponds must be main-
tained at a level high enough to achieve the nec-
essary nitrogen removal efficiency. The solids
level in  the filters  must  be maintained  low
enough  to prevent clogging and excessive head
loss through the units. The chemical feed rates
have to be carefully monitored to provide the re-
quired additive concentrations for the biological
cultures, as economically as possible. The three
methods are temperature sensitive, and as tem-
peratures vary throughout the year, detention
times will have to be adjusted. The separation
procedures may vary throughout the year as the
nature of the algal culture  changes.
  Algae stripping is  the only method that pro-
duces a by-product.  Approximately one ton of
dry algae will be produced for each million gal-
lons per day of waste treated. The value of the
algae produced will be between 80 and 160 dol-
lars per ton.
  Estimated costs for bacterial denitrification
and algae stripping are between 90 and 100 dol-
lars  per million gallons   for each treatment
method

SUMMARY
  1. Bacteriological  denitrification  is the best
method  of removing  nitrogen from  agricultural
wastewater in the San Joaquin Valley at  this
time. It meets the effluent objective of 2 mg/1
nitrogen and involves less  operation than algae
stripping.
  2. Bacteriological denitrification (both in an-
aerobic  filters and deep ponds) and algal strip-
ping have  several advantages and disadvan-
tages.
  3. The high- cost (90 to  100 dollars per mil-
lion gallons) for denitrification and algae strip-
ping have  instigated  studies of  a symbiotic
method  of nitrogen removal and reclamation of
the wastewater.
       TREATMENT OF RETURN FLOWS     95

                TABLE 1

 ESTIMATED CONCENTRATIONS OF
CHEMICAL SUBSTANCES IN SAN LUIS
            DRAIN WATERS


                       1970        2020
                     Conditions  Conditions
Substance
Total Dissolved Solids
Salts
Sulfate
Sodium
Chloride
Calcium
Magnesium
Bicarbonate
Potassium
Boron
Nutrients
Total Nitrogen
Total Phosphate
Pesticides
Others
5 Day B.O.D.
C.O.D.
mg/1
7,000

3,000
1,500
1,200
300
200
200
20
10

20
0.15
0.001

1-3
10-20
mg/1
3,000

700
700
900
100
50
100
10
3

20
0.15
0.001

1-3
10-20

-------
96
MANAGING IRRIGATED AGRICULTURE
                                         TABLE 2

         NITROGEN REMOVAL-DETENTION RELATIONSHIP FOR DEEP POND
                                    DENITRIFICATION
        Number of Days of Operation
                                                Nitrogen Removal In Percent
Detention Time Uncovered Covered Uncovered Pond Covered Pond
Days Pond Pond Min. Max. Avg. Min. Max. Avg.
7.5
10
15
20
34 ... 67 92 80
61 31 60 93 77 75 98 88
30 35 30 95 56 88 94 91
31 40 54 84 71 84 94 90
REFERENCES
   1. Stetson, C. L., "A Drainage System for
the San Joaquin Valley," 4th International Wa-
ter Quality Symposium, San Francisco, Califor-
nia (August 14, 1968).
   2. Price,  E. P., "The San Luis Drain," 4th
International  Water  Quality Symposium, San
Francisco, California (August 14, 1968).
   3. Lindholm,  R.  R., "San Joaquin Valley
Drainage  Investigation-San   Joaquin  Master
Drain," California Department of Water Re-
sources, Bulletin No. 127, Preliminary Edition
(January 1965).
   4. "Effects of the San Joaquin Master Drain
on Water Quality of the San Francisco Bay and
Delta," U.S. Department of the Interior, Federal
Water Pollution Control Administration, South-
west  Region,  San Francisco, California (Janu-
ary 1967).
   5. Hutchinson, G. E., A  Treatise on Lim-
nology-Volume 1, John  Wiley and Sons,  Inc.,
New York, New York (1957).
   6. McCarty,  P. L.,  Beck,  L. A.,  and  St.
Amant, P. P., "Biological Denitrification  of
Waste Waters by Addition of Organic Materi-
als,"  24th Industrial Waste Conference, Purdue
University, Lafayette, Indiana (May 1969).
   7. Christenson, C. W., et  al., "Reduction of
Nitrate-Nitrogen   by   Modified   Activated
Sludge,"  USAEC, TID,  7517,  Pages 264-267
(1956).
                                            8. Ludzack,  F. J.,  and  Ettinger,  M. B.,
                                        "Controlling  Operation  to Minimize  Activated
                                        Sludge Effluent Nitrogen," Journal Water Pol-
                                        lution Control Federation, Vol. 34, No. 9, Pages
                                        920-931 (1962).
                                            9. Bishop, D. F., "Ammonia and Nitrogen
                                        Removal," Internal Report,  Advanced Treat-
                                        ment Research, FWPCA, R. A. Taft Water Re-
                                        search Center, Cincinnati, Ohio.
                                           10. Finsen, P. O., and Sampson, D., "Deni-
                                        trification  of  Sewage  Effluents,"  Water  and
                                        Waste Treatment Journal, Vol. 7, Pages 298-300
                                        (1959).
                                           11. McCarty, P. L., "Feasibility of the Deni-
                                        trification Process for Removal of Nitrate-Nitro-
                                        gen from Agricultural Drainage Waters,"  Ap-
                                        pendix  California Department of Water  Re-
                                        sources Bulletin 174-3 (May 1969).
                                           12. Young, J. C., and McCarty, P. L., "The
                                        Anaerobic  Filter for Waste  Treatment," Stan-
                                        ford University, Technical Report No. 87, Stan-
                                        ford, California (March 1968).
                                           13. Tamblyn, T. A., and Sword, B. R., "The
                                        Anaerobic Filter for the Denitrification of Agri-
                                        cultural Subsurface Drainage," 24th Industrial
                                        Waste  Conference,  Purdue  University,   La-
                                        fayette, Indiana (May 1969).
                                           14. Brown, R. L., "Field Evaluation of An-
                                        aerobic   Denitrification  in  Simulated  Deep
                                        Ponds," California  Department of Water  Re-
                                        sources Bulletin No. 174-3 (May 1969).

-------
                                                       TREATMENT OF RETURN FLOWS
                                         97
  15. Golueke, C. G., Oswald, W. J., and Gee,
H. K., "Increasing High-Rate Pond Loading by
Phase  Isolation-Final  Report,"  Sanitary  En-
gineering Research  Laboratory,  University of
California, Berkeley, California (April 1962).
  16. Beck,  L. A., Oswald, W. J.,  and Gold-
man, J. C.,  "Nitrate  Removal from  Agricul-
tural Tile  Drainage  by  Photosynthetic Sys-
tems," Second  National Symposium,  Sanitary
Engineering  Research  Development and  De-
sign,  American Society  of Civil  Engineers,
Cornell  University,  Ithaca, New  York (July
1969).
  17. "Standard Methods for the Examination
of Water and  Wastewater", American Public
Health Association, Inc., New York, New York,
12th Edition (1965).
  18. Cholette, A., Blanchet, J., and Cloutier,
L., "Performance of Flow Reactors at  Various
Levels of Mixing," Canadian Journal of Chemi-
cal Engineering, Vol. 38, (1960).
  19. Milbury, W. F., "A  Development  and
Evaluation of a Theoretical Model Describing
the Effects of Hydraulic Regime in Continuous
Microbial Systems,"  dissertation  presented to
Northwestern University at Evanston, Illinois,
in 1964, in partial fulfillment of the requirement
for the degree of Doctor of Philosophy.
  20. California  Department  of  Water  Re-
sources, San Joaquin District, Fresno, Califor-
nia, unpublished data.
  21. California  Department  of  Water  Re-
sources, "Hydrologic Data,  Volume  IV:  San
Joaquin  Valley, Appendix A,  Climatological
Data," Bulletin  130, (September  1968).

  22. Golueke,   G. G.,  and  Oswald, W. J.,
"Harvesting  and Processing of  Sewage-Grown
Plantonic Algae," Journal Water Pollution Con-
trol Federation, Vol. 37, No. 4, Pages 471-498
(1965).
  23. Anonymous, "A Study of the Use of Bio-
mass Systems  in  Water  Renovation,"  North
American Aviation Inc., (May 1, 1967).
  24. Oswald,  W. J., Crosby, D.  G., Golueke,
C. G.,  "Removal  of  Pesticides  and  Algal
Growth Potential from San Joaquin  Valley
Drainage  Waters (A  Feasibility Study,), Cali-
fornia Department of Water Resources unpub-
lished report (1964).

-------
   Examination  of  Agricultural  Practices for
     Nitrate Control  in  Subsurface  Effluents

                    DONALD R.  NIELSEN and JOHN C. COREY
                                University of California
                                   Davis, California
ABSTRACT
  This paper outlines the results of a total nitro-
gen balance study on the Santa Ana Basin in
Southern California and shows the flux of nitro-
gen between the atmosphere, land surface, soil,
and ground water for this  356,000 acre area.
These fluxes are then discussed with respect
to the effect of agricultural production and ferti-
lizer use on the amount of nitrogen available for
leaching. Recommended fertilizer  and water
application rates in this irrigated basin are pro-
posed based on the  current level of fertilizer
technology and production practices.

INTRODUCTION
  Although there is  a wide range  of opinion
about the effect of irrigated agriculture on our
environment, particularly as it  relates to  the
quality of irrigation return flows, it must gener-
ally be agreed that our research and engineering
efforts have to be augmented if we are to mea-
sure and manage our water resources as cur-
rently demanded by  society. Agricultural pro-
duction  techniques have traditionally involved
practices made to guarantee or achieve maxi-
mum biological production per acre. Except for
the establishment and maintenance of a favor-
able salt balance within an  irrigated basin, we
have seldom anticipated the results of our ac-
tions on the total  environment as they slowly
accrue over decades of years, nor have we made
very many  attempts at measuring and under-
standing the impact and consequences of these
production  practices in terms  other  than crop
yields.
  One consequence of considering crop yields
as the only goal of a production oriented re-
search program coupled with an ever-increasing
population  educated to  emphasize animal in-
stead of plant protein in its diet has been an in-
creased requirement for nitrogen fertilizers from
chemical sources. Past applications of farm-site
nitrogen fertilizers and the detection  of nitrate-
nitrogen in ground water has focused attention
on  the deep percolation of soluble  forms of
nitrogen. Viets1 has summarized and evaluated
some of the evidence pro and con regarding
water quality in relation to farm use of fertilizer.
His review and an earlier effort2 attest to the
assertion that measurements of and information
about the exact fate of applied nitrogen ferti-
lizers are not available.
  The scope of the nitrogen problem can be best
placed in context by briefly examining the total
nitrogen picture before narrowing our field of
vision. According to Bartholomew and Clark3, a
global overview of the distribution of nitrogen
                                          99

-------
100     MANAGING IRRIGATED AGRICULTURE
 reveals that 98.03% of the total supply is  held
 essentially immobile in the lithosphere  of the
 earth while 1.96% is found in the atmosphere.
 Hence, the amount of nitrogen available in ter-
 restrial and aquatic life forms  together with soil
 organic  matter  capable of being  managed by
 man's  activity is only about 0.01% of the global
 supply. This seemingly small  percentage figure
 represents 1.8 x 1011 English  tons of elemental
 nitrogen and is that a mount on which all life de-
 pends.
   To assess the importance of  this nitrogen pool
 to irrigated agriculture, we must first examine
 its influence  on a specific basin. Moreover to
 be able to ascertain the effects of irrigation and
 related agricultural  practices  on the retention
 and  redistribution of nitrogen not only  in soil
 but also in drainage and return flow waters, it is
 essential  to  recognize the magnitudes  of all
 sources and sinks of ntrogen  and  the potential
 rates of transfer from  one location to another.
 In other words, an understanding of the various
 processes involved  in the so-called nitrogen
 cycle treated comprehensively by Bartholomew
 and  Clark3 must be quantified and coupled with
 processes of water management.
   The  principal purpose  of this paper is to re-
 view a recent effort4 made to ascertain the mag-
 nitude  of agricultural  contributions  of nitrate-
 nitrogen in ground water  supplies in the upper
 Santa  Ana Basin in California. In addition to
 the  guidelines  established from  that  effort,
 recommendations  will be given for potential
 irrigation practices and future research  activi-
 ties  based  upon  analyses of  field-measured
 water and solute behavior.

           Santa Ana Basin Study
  Only a limited number  of areas are suitable
for a detailed analysis of the magnitude of agri-
cultural  contributions of nitrate-nitrogen in
ground water supplies. Such studies require ac-
curate  and  regular measurements  over long
periods of time of such information  as ground
water quality, land use patterns, rates of nitro-
gen  fertilizer, waste disposal from humans,
cattle and poultry, water application rates,  and
water use. The upper Santa Ana Basin located
in Southern California  was chosen as the site of
this type of an investigation because necessary
information was available. The basin comprises
356,000  acres  overlying 900-foot thick water
bearing sediments. The basin is surrounded by
mountains and is drained  by the Santa Ana
River. This  river is  impounded at  the egress
from the basin by Prado Dam.
   In an attempt to assess the impact of agricul-
ture  on  the  nitrate-nitrogen  in  ground water
supplies  in the Santa Ana Basin the total nitro-
gen pools and fluxes were first determined. The
pools (sources and sinks) and fluxes (rates of
transfer from sources to sinks) of nitrogen, are
presented in 4 zones (Figure 1), those in the at-
mosphere, land surface, surface six feet of soil,
and in the substrata. The magnitude of each flux
is given  in thousands of tons of nitrogen per
year next to each arrow representing the various
transport pathways.   These  fluxes were deter-
mined by evaluating the various mechanisms
or processes in the nitrogen  cycle, nitrogen con-
tents of natural and man-made substances, land
use, and  hydrologic data.
  Specifically for the Santa Ana Basin, the at-
mosphere loses about 4,600 tons of nitrogen per
year, of  which 2,100 tons returns to the land
surface nitrogen  pool in rain, dry fallout, and
sorption  of ammonia (Precip.).  The remaining
2,500 tons is returned by symbiotic and nonsym-
biotic fixation. The  atmosphere gains 9,000
tons, 6,400 tons of gaseous nitrogen from vola-
tilization from fertilizer applications and  com-
bustion  products emitted  by motor vehicles,
boilers, etc., (Combustion) and 2,600 tons from
denitrification in the  soil zone (Gas Losses). Be-
cause  the total atmospheric gains above the
basin exceed  the losses, there is 4,400 tons avail-
able for export by the atmosphere to other areas
of the world.
  In addition to bulk precipitation and soprtion,
the land surface  receives about  5,200  tons of
nitrogen  per year in the water supply (200 tons
from streamflow and  imported water and 5,000
tons from pumped ground  water. The  return-
flow  waters from urban and agricultural irri-
gations amount to 700  tons. Water analyses at
Prado Dam,  the only ground water drainage
pathway  from the basin, indicate 800 tons left
the basin in the surface water.
  A net total of 17,400 tons of nitrogen per year
are distributed on the land surface from munici-
pal sewage and solid  wastes  (3,900 tons), indus-
trial sources (100 tons), manures from  feedlot

-------
                      4.6
CI
UJ

CL.
CO
O
S
r—
•s'
         Ci,
         E
        CO
        E
        n
               C3
                    Losses
                        AU1  N  POOL

                        1.25 x 109  tons
                                                             NITRATE CONTROL
                                                                          101
                             Interbasiii Transfers
   I.'-
  CO

  C3
  s:
        CJ
        Co

        0.2
               CM
Water
5.0
  UJ
  2
  O
  IS!
  CO
      €0
      *_*^
  CO
  CO
      es
      10

            yVecin. 2.1
                      5.2
                    Supply
           Waste,N Pert.
                17.4
                                  LAND
                                SURFACE
                                N  POOL
                               29X103  tons
                                   -'u.l1
                                            6.4
                                          Combustion
 Gas  Losses

Return Flow
                                                    0.7
                 19.9
                              •>
                                           Plant Uptake
                                  SOIL
                                 N  POOL

                               2.8xlo6 tons
                                  8.4
           SUBSTRATA
             ji  t: -J -u L
                 6 .
            40K10  tons
                                 3.0
                               m-ZZZX:*
                                       CO
                                       CO
                                      C9
                                                             ,
SURFACE
 WATER
 \\  POOL
5.9 x 103 tons
  Gas  Losses

         JJ	j

                                                                         A
                                                                               ra
                                                                        va
                                                                            CO
                                                                        0.8
                                                                           o
                                                    Infiltration
                                                 Subsi'rfacs  Flow
                                                 V/atsr Table
                              V
                              GRDUHDWAT
                                  K  POOL
                             To
                        t-^t
                        i ft i
                                            Groundwater Flow
                                                                       CO
                                                                       cj:
                                                                       LaJ
                  5 0
Figure 1:  Nitrogen Pools and Fluxes for the Santa Ana Basin Based on the Level of Development in 1960. The
Basin is 356,000 acres in Extent and Assumed to Have 6 feet of Soil  and 900 feet of Substratum. The Mass of
Nitrogen in Pools is in Tons. The Flux of Nitrogen between Pools (numbers for arrows) is in Thousands of Tons
of Nitrogen per Year.

-------
102     MANAGING IRRIGATED AGRICULTURE
and  dairy  cattle (3,200  tons),  precipitation
(2,100  tons), and chemical forms of nitrogen
fertilizer (8,100 tons).
  The  total flux  of nitrogen transferred to the
soil nitrogen pool is 25,100 tons per year.  This
amount  accrues  from  fixation  from  the at-
mosphere (2,500 tons), nitrogen from the water
supply (5,200 tons) and nitrogen transferred to
the soil from the land surface, including precipi-
tation and sorption (17,400 tons).
  The soil, in turn, loses about 2,600 tons of ni-
trogen per year to the atmosphere by denitrifi-
cation and about 14,100 tons to plant uptake.
This calculation  leaves  8,400 Jons potentially
available for leaching to the substrata, accumu-
lating  in the soil, and/or  plant uptake in suc-
ceeding years.
   If we assume that all 8,400  tons of nitrogen
per  year potentially available for leaching is
completely nitrified to nitrate and leached by
about  300,000 acre-feet of water percolating
beyond the soil zone, the resultant nitrate con-
centration in the percolating  water  would be
about  80-100 ppm in the unsaturated zone. This
estimate  is  unrealistic,  however, for  several
reasons:  1) nitrogen in the soil does not  exist
wholly in the  nitrate form; 2) nitrogen waste
loads  contain  organic  nitrogen  compounds,
which may take years to degrade completely to
nitrate;  3) nitrate can  be  assimilated by mi-
crobes if the carbon-nitrogen ratio of the waste
load or the humus in the  soil is high. A more
realistic concentration level of nitrate in the re-
charge water would probably be 40 to 50 ppm
basin-wide, with perhaps  100 ppm or  more in
localized disposal sites.
   Following this overview of the nitrogen cycle
for the  basin, an effort was made to identify
locations of high nitrate concentrations in the
ground  water. After identification of  problem
areas was complete, a program to determine the
cause  of these nitrate accumulations was begun.
Finally guidelines were established for control-
ling nitrate accumulations  stemming from agri-
cultural practices.
   At locations of high nitrate concentrations (45
to 90  ppm nitrate and higher) measured in well
waters today,  early water  analyses (1919-1937)
confirmed that nitrate concentrations were low,
ranging from only 2 to 22 ppm nitrate. Time
sequence  studies of nitrate concentrations in
well waters revealed that they increased during
the last 40 years and were in general agreement
with the increasing amounts of nitrogen (con-
centration times volumetric flow rates) observed
in the Santa Ana River passing through Prado
Dam between  1950 and 1970.
  To determine the cause of these nitrate in-
creases, the major nitrogen inputs into the basin
were considered — agricultural  fertilizers  and
wastes from dairy cows, poultry and  humans.
Other  recognized  sources  of  nitrogen  were
native  or fossil nitrogen accumulated under na-
tural conditions  (before agricultural  develop-
ment) and nitrogen fixed from the atmosphere
by  leguminous plants,  nonsymbiotic  microor-
ganisms and electrical storms, and internal com-
bustion engines.  These nitrogen sources  were
related  to early and present day land use  pat-
terns and to the  movement of nitrogen on the
basis of examination and  analysis of nitrogen
transformations,  soil  permeability,  irrigation
water use, and surface and ground water hydrol-
ogy.
  An examination of changes in  fertilizer use
and vegetation patterns with  regard to citrus,
other fruit crops,  nut crops, field crops, and turf
grass reveals the  impact of land use and  tech-
nology on agricultural practices.
  Early studies (before  1950) on nitrogen ferti-
lizer use by citrus  in California indicated  that
maximum returns required applications of 100
to 350  pounds of nitrogen per acre annually.
Since  long-term  fertilizer  experiments at the
University  of California  Citrus  Experiment
Station indicated annual needs of about 300 ,
pounds of nitrogen per acre for maximum re-
turns, many growers adopted this  as a minimal
rate. Analyses of  nitrogen fertilizer use on citrus
in Riverside and San Bernardino County  por-
tions of the Santa Ana Basin since 1930 (Figure
2) indicate a steady utilization of fertilizer until
1950 and then a  rapid decrease  that begins
around 1950  and stabilizes approximately ten
years later.  This precipitous  decrease is  due
to  both a changing land  use  pattern and a
changing fertilizer use. During this period there
was a reduction in citrus acreage due to popula-
tion growth in the area and a reduction in ferti-
lizer use because of the development of leaf
analysis  as a  guide to citrus  fertilization. Ex-
periments (circa  1950) showed  that  100 to 150

-------
                                                                      NITRATE CONTROL



                                                                San Bernardino County Portion

                                                                of the Santa Ana Watershed
                                                                  103
r-l
t
£
Riverside County Portion

of the Santa Ana Watershed
   I  2000
     1000
          1930               1914O                 1950                I960                 1970
Figure 2: Tons of Nitrogen Applied as Fertilizer Yearly to Citrus in the Santa Ana Basin Between 1930 and
1970
pounds of  nitrogen  per acre  from  chemical
sources gave yields equal to those from higher
rates  using  leaf analysis. Presently 85-90% of
the citrus  acreage in  the  Upper Santa Ana
Watershed  is fertilized  according to leaf analy-
sis programs. The consequences of this chang-
ing fertilizer use on the tons of nitrogen ferti-
lizer subject  to  leaching is  shown in  Figure 3.
These graphs were prepared using the assump-
tions: a) citrus acreage data is  valid;  b) 15,000
pounds of fruit were  produced  per  acre an-
nually;  c) 300 pounds  of nitrogen applied per
acre annually prior to  1950, and  75 pounds of
nitrogen applied per acre annually in San Ber-
nardino County and 150 pounds  in  Riverside
County in  subsequent  years; d) 2.5 pounds of
nitrogen are removed per ton of fresh  fruit; and
e) 20% of applied  nitrogen is lost  by volatiliza-
tion or denitrification.
  Similar studies to the citrus utilization  of ni-
trogen fertilizer were  conducted  on the  15 or
                        more principal noncitrus fruit and nut crops in
                        the  basin. This acreage is dominated by grapes
                        with annual rates of nitrogen fertilization rang-
                        ing  from 60 pounds per acre for grapes to 300
                        pounds  per  acre  for  strawberries.  Over  the
                        period from 1950 to 1970, fertilizer nitrogen ap-
                        plied to these crops has decreased from 1800 to
                        600 tons per year,  owing primarily to acreage
                        reductions  lost  to  urbanization. Field crops,
                        noncitrus fruit  crops, and turf grass  were con-
                        sidered to not contribute appreciably to nitrogen
                        increases in  the ground waters  of the basin.
                        Vegetable  acreages  combined  with  nitrogen
                        fertilization  studies  revealed that in  1930,  12
                        pounds  of  nitrogen  per acre was available  for
                        deep percolation below the root zone, and by
                        1969, that  value had increased nearly tenfold
                        to 107 pounds per acre.
                          In addition to fertilizers, agriculture can in-
                        troduce  nitrogen from organic wastes produced
                        by people and animals. The amount of nitrogen

-------
104
MANAGING IRRIGATED AGRICULTURE
j=
x:
 u
      sooo
      5000
      4000
     3000
     2000 .
     1000  .
                                                     Sen Bsmardino County Portion
                                                     Of Tte Safitc Ana Watershed
 0

I
                         Riverside County Portion
                         Of The Santa Ana Watershed
         1930
                     1940
1950
I960
1970
Figure 3: Annual Amount of Nitrogen from Fertilizer Available for Leaching in the Santa Ana Basin between
1930 and 1970
produced in this manner in the basin increased
from 4 million  pounds  in  1930 to 30 million
pounds in 1970.
  Inasmuch as any  significant contribution of
nitrogen reaching the ground water in the basin
from agriculture is likely to be associated with
crop  fertilization,  water use,  and manure  dis-
posal, recommendations for reducing the poten-
tial of nitrate pollution relative to fertilization
and  water use focus on  rates of fertilization,
timing of fertilizer applications, type of ferti-
lizer,  evapotranspiration   requirements,   and
leaching fraction.
  All crops require nitrogen for growth. In the
Santa Ana Basin, naturally occurring soil nitro-
gen reserves are  too low to produce yields high
enough  for  substained  commercial   agricul-
ture  production.  Except with legume crops,
which utilize atmospheric nitrogen through con-
version by legume bacteria living on their roots,
                                         supplemental fertilizer nitrogen must be applied
                                         to all crops  if agriculture is to  continue in the
                                         basin.
                                           The rates of nitrogen application required for
                                         good yields and good quality vary from crop to
                                         crop. Citrus and vegetables are two crops to
                                         which  nitrogen is applied at rather high rates
                                         and which may, under inefficient management,
                                         result  in  appreciable  excessive nitrogen addi-
                                         tions to  the underground water supply. The
                                         recommendation for citrus is to use leaf analysis
                                         as a guide to nitrogen fertilization and to apply
                                         the fertilizer in one application in  the  winter.
                                         Leaf analysis  integrates the effects  of many
                                         factors  into  one figure — the nitrogen concen-
                                         tration  in a specific type and age of leaf. These
                                         guides  were  developed  to  permit maximum
                                         yield at lowest cost,  so the great reduction in
                                         pollution  potential is an extra benefit.  A further
                                         reduction  in the suggested  optimum nitrogen

-------
                                                                   NITRATt CONTROL
                                         105
level in leaves might improve the fruit quality
of oranges and further reduce the pollution po-
tential. Such a program is likely to reduce yields
slightly, but that could be counterbalanced by
improved  fruit quality. Analyses of field experi-
ments  for vegetables  indicate that  maximum
yields  will require  120 to 300  pounds N/ac,
the rate varying with crop, and applied in two or
more  applications.  Grower  practices  on the
average  result  in  greater  applications  than
necessary.  For annual crops (vegetables  and
field crops) split applications of  fertilizer are
to be  encouraged — one-fourth  to one-half of
the nitrogen needs of the crop applied at or
near planting time and  the  remainder at the
main period of most rapid growth.
   Fertilizing of  tree  crops  or spring-planted
annual crops  in  late-summer and fall is inef-
ficient, and excessive deep percolation losses
of nitrogen to underground water supplies may
result from rainfall and normal fall-winter irri-
gations or irrigations  made in preparation  for
planting. The most critical period for tree crops
is generally bloom  time. Recommended to en-
sure maximum yields is a single nitrogen appli-
cation timed sufficiently ahead of bloom to as-
sure high nitrogen at bloom.  High nitrogen at
bloom will usually be enough to supply crop
needs for  the rest of the season.
  The use of a relatively new group, slow-re-
lease fertilizers, offers some interesting possi-
bilities of reducing nitrogen  losses below crop
root  zones by  releasing  nitrogen  rates that
match crop needs more closely. Although some
promising  materials have  been developed  re-
cently, their use is still primarily on an experi-
mental basis.  Desired goals with slow-release
fertilizers  must await improvements in  formula-
tions and  more extensive field testing.
  Animal manures can be thought of as a form
of slow-release  nitrogen. About  one-half the
nitrogen in manures is released during a crop-
ping period of two to three months. Recommen-
dations for use of manures include the follow-
ing:  1) Apply well ahead  of planting. 2) Disc or
otherwise  incorporate  the manure at leat  six
inches deep into the soil. 3) Irrigate adequately
or let  winter rainfall wet to leach detrimental
salts away from germination zone of the crop to
be planted. 4) A light supplemental  nitrogen
application  may  be  needed near planting time
if  crops are planted in the cool season, since
nitrogen release from manure is slower in cool
weather. 5) Do not apply bulk manure to grow-
ing crops,  because  of plant  disease problems,
fly problems,  and other generally  undesirable
effects. 6) To  reduce deep percolation  losses,
disposal of dairy wash water or waste effluent
by irrigation  (either surface ditch,  pipe,  or
sprinkler)  should be timed to apply quantities
consistent with crop or soil requirements for
water and nitrogen.
   Soil analysis and/or plant tissue analysis can
be helpful in  timing fertilizer applications  by
evaluating nitrogen needs or crop  response to
nitrogen  application at  a  given  time.  Soil
samples (or soil solutions  extracted by  soil-
solution probes) taken below crops  can be ana-
lyzed to evaluate losses of nitrogen resulting
from  various timings of fertilizer applications.
   The actual  amount  of  nitrogen reaching
underground waters is important, but the  pri-
mary concern  is generally the concentration of
nitrate. The concentration in the soil water can
be varied by  changing either the  quantity of
nitrogen or the volume of irrigation water ap-
plied. Control  of nitrogen pollution may require
not only reduced rates of fertilization but appli-
cation of water in excess of crop needs (evapo-
transpiration)  sufficient that waters percolating
downward will dilute the nitrogen to an accep-
table  concentration.  Evapotranspiration can be
calculated  from  various  empirical formulae,
measured  indirectly,  or  measured  directly  by
weighing  lysimeters.  These  values  of  evapo-
transpiration may be used as a guide to crop
needs and as a mesaure of irrigation efficiency.
Another way  of estimating  the quantities of
water moving  downward is based on the leach-
ing fraction. The leaching fraction is the volume
of water that moves past the root zone, ex-
pressed  as a  fraction of the irrigation  water
entering the soil at the surface. A leaching frac-
tion5 of 0.20 to 0.50 is common (20 to 50% of the
water entering the soil during irrigation perco-
lates  beyond the reach of roots and becomes
drainage water). The leaching fraction for well-
drained soils  may  also  be  estimated  more
directly by comparing the concentration  of a
constituent such  as chloride  in the irrigation
water with the chloride in the soil solution be-
low the root zone. For example, if  chlorides in

-------
106
MANAGING IRRIGATED AGRICULTURE
the soil solution below the root zone are 5 me/1,
and the chlorides in the water are 2 me/1, the
leaching fraction is 2/5 = 0.40 or 40%.
            Future Research Effort
  Accurate appraisals of the major known facts
pertaining to nitrogen balance in a basin, field,
plot or greenhouse pot have eluded soil scien-
tists and  engineers owing to  insufficient devel-
opment of technology and the previously-held
limited objective of determining the percentage
nitrogen  recovery from  fertilizer additions in
the crop  only.  In the field,  both gaseous losses
of  nitrogen to the  atmosphere and  leaching
losses below the root zone too often have  been
lumped together without a direct measure of
either. Without the use of tracers, an accurate
measure of gaseous losses is precluded. On the
other hand, although they remain more or less
untested,  field  techniques are available for ex-
amining the rates of water moving below the
root zone. In addition to those mentioned in the
previous paragraph, soil  water flow may be as-
certained  by in situ  measurements of the un-
saturated  hydraulic conductivity coupled  with
those of the hydraulic gradient.6 Assuming that
water  soluble  nitrogen forms  move  with the
water, measurements of their  concentrations,
accompanying  the measured soil  water flow,
yield estimates of rates of nitrogen appearance
in irrigation return flows. Although such tech-
niques are advantageous in that they do not de-
pend upon knowing the physical, chemical and
biological processes occurring in the root zone,
they do involve  temporal  and  spatial integra-
tions plagued  with  several  degrees of uncer-
tainty7.  Research  effort  resolving these uncer-
tainties will undoubtedly increase  in  the  near
future.
  Knowledge  concerning the forms,  amounts,
transformations and importance of organic and
inorganic  soil nitrogen compounds has been ac-
cumulating for more than a century. Yet, for a
laboratory soil  column, much less for a field, re-
search has not been sufficient to allow a quanti-
tative description  or prediction of biochemical
reactions  of nitrogen  and the resulting concen-
tration  distributions  within  soil  profiles.  In
nature,  and especially under irrigated agricul-
ture, nitrogen  and other beneficial and poten-
tially harmful  solutes are subjected to a down-
                                          ward translocation at the same time that they
                                          may be absorbed by plants  or  undergo micro-
                                          biological transformations. If we are to identify
                                          agricultural practices which  affect the  amount
                                          of surface-applied fertilizers and  salts  and the
                                          time  required  for them  to  reach the ground
                                          water, we must begin to examine the biochemi-
                                          cal course of events within soil profiles both as a
                                          function of time of flow and depth in  the soil.
                                          Field techniques,  like those  described by Mar-
                                          cura and Malek8  and by McLaren9 for labora-
                                          tory samples,  must be developed to  ascertain
                                          values  of physical,  chemical,  and  biological
                                          parameters   essential  for  microbial   activity.
                                          These  techniques  together with qualitative in-
                                          formation  already at hand regarding soil tex-
                                          ture,   depth  of  profile,  crop  requirements,
                                          growth characteristics, and others will provide
                                          better guidelines for fertilizer management.
                                            Soil scientists as well as plant  scientists and
                                          irrigation engineers  are  relatively unfamiliar
                                          with  management techniques  for identifying
                                          and controlling the  concentration  of  the  soil
                                          solution once it has moved beyond the root zone.
                                          This is especially true for deep,  well-drained
                                          alluvial soils having water tables at consider-
                                          able depths. It is often assumed that below the
                                          root zone in the absence of microbial  activity,
                                          nitrate  concentrations are not  diminished ex-
                                          cept by dilution with water containing little or
                                          no nitrate.  How  does dilution actually  take
                                          place if a  pulse of nitrate-laden water is dis-
                                          placed  by an  irrigation  with water relatively
                                          free of  nitrates?  This  question  is not easily
                                          resolved,  and  it  is especially  complex  when
                                          other less desirable salts must be leached pref-
                                          erentially.10  The   characterization of solute
                                          mixing  and displacement coupled with the un-
                                          saturated hydraulic properties  of the vadose
                                          zone remains  obscure, yet essential,  to long-
                                          term, rational irrigation basin management.
                                            When the concepts expressed  in the above
                                          paragraphs have been integrated and connected
                                          to  theoretical  developments  of ground water
                                          management by hydrologjsts,  the  technology
                                          of  managing nitrate  in  potentially saline irri-
                                          gated basins will  be well advanced and more
                                          mature than it  is today. Until that time, agricul-
                                          tural practices  for nitrate control  in subsurface
                                          effluents will depend largely  on guidelines es-
                                          tablished by studies similar to those in the Santa
                                          Ana Basin.

-------
                                                                  NITRATE CONTROL
                                        107
REFERENCES
   1. Viets, F. G., Jr., "Water Quality in Rela-
tion  to Farm Use of Fertilizers," BioScience 21,
460 (1971).
   2. Viets, F. G., Jr., "Soil Use  and Water
Quality — A Look into the Future," Journal of
Agriculture and Food Chemistry 18, 789 (1970).
   3. Bartholomew,  W.  F.  and  F. E.  Clark,
"Soil Nitrogen,"  American Society of Agron-
omy, Madison, Wisconsin (1965).
   4. Ayers, R. S. and R. L. Branson, editors,
"Nitrates in the Upper Santa Ana Basin in Re-
lation to Groundwater Pollution," unpublished
report,  Kearney  Foundation of  Soil Science,
University of California, Davis, California.
   5. Richards, L. A., editor, "Saline and Alkali
Soils," Agriculture Handbook  No. 60, United
States Department of Agriculture (1954),
   6. LaRue, M. E., D. R. Nielsen and R. M.
Hagan,  "Soil  Water  Flux below  a Ryegrass
Root Zone," Agronomy Journal 60:625 (1968).
   7. Nielsen,  D. R., J. W. Biggar and J. C.
Corey, "Application of Flow Theory  to Field
Situations," Soil Science (in press).
   8. Macura,  J. and I,  Malek, "Continuous
Flow Method for the Study of Microbiological
Processes  in Soil Samples," Nature 182,  1796
(1958).
   9. McLaren, A. D., "Steady-State Studies of
Nitrification in Soil:  Theoretical  Considera-
tions," Soil  Science Society of America  Pro-
ceedings 33, 273 (1969).
  10. Biggar, J. W.  and D. R. Nielsen, "Mis-
cible Displacement and Leaching,"  in Irrigation
of Agricultural Lands, American Society of
Agronomy Monograph 11, 254 (1967).

-------
               Grand  Valley  Salinity Control
                      Demonstration  Project
                                  T. JOHN BAER, JR.
                                Colorado State Legislator
                                Grand Junction, Colorado
ABSTRACT
   The Grand  Valley Salinity Control Demon-
stration Project was formulated to evaluate the
salinity control effectiveness of canal and lateral
linings. Introduction of water into the highly sa-
line environment of the soils and aquifer in the
Grand Valley results in excessive salinity contri-
butions to the Colorado River. Consequently, by
reducing these ground water  inflows  by a pro-
gram of canal  and lateral linings, significant re-
ductions in the Grand Valley's salt contribution
could be achieved.
  Ponding tests were made at the beginning of
the project in order  to establish the initial con-
ditions regarding seepage rates, as well as final-
izing the construction recommendations. While
the results of seepage measurements  in the ca-
nals  do not indicate potentially high  losses, re-
sults of some  preliminary  lateral seepage  rate
measurements  show very high rates. Approxi-
mately 8 miles of canals and 5 miles of laterals
were lined.
INTRODUCTION
  Because the salinity problem in the Colorado
River Basin  is rapidly becoming acute, meas-
ures  must be taken concerning various salinity
control alternatives if future development is to
occur. But most measures which could be con-
sidered  may be found to be very  poorly con-
ceived unless a thorough understanding of the
physical,  economic, and social conditions in
each area is gained. The attitudes of local peo-
ple are important  management considerations
and for the most part these attitudes are easily
traced to the events surrounding the  develop-
ment  of the area. In addition, it is  necessary to
be aware of the natural characteristics of an
area so that restrictions are not imposed which
would destroy an area as a functional commu-
nity. And finally, the  institutional  structure of
an area should be known in order to effectively
implement control measures.
  Probably the most significant salt source re-
sulting from irrigated  agriculture in the upper
basin  is the Grand Valley area  (Fig. 1) in west
central Colorado. Consequently, this valley was
selected for study in order to evaluate irrigation
water conveyance lining as  a possible salinity
control practice.
  The primary source of salinity is from the ex-
tremely saline aquifers  overlying the marine de-
posited Mancos Shale formation. The shale is
characterized by lenses of salt in the formation
which are dissolved by the water from excessive
irrigation and conveyance seepage  losses when
it comes  in contact with  the Mancos shale for-
                                          109

-------
110
MANAGING IRRIGATED AGRICULTURE
                                                                 Grand Valley Salinity
                                                                 Control Demonstration
                                                                       Project
                                                                /"x	f-
                                                           ~-v>N Gunnison
                                                                   River
                              Figure 1:  The Grand Valley, Colorado
mation. The introduction of water through these
surface sources deep percolates into the shallow
ground water reservoir where the hydraulic gra-
dients it produced displace some water into the
river. This displaced water has usually had suf-
ficient time to reach chemical equilibrium with
the salt concentrations of the soils and shale.
  Aside  from  the  numerous  studies  in  the
Grand Valley to  evaluate local  conditions, this
effort, the Grand Valley Salinity Control Dem-
onstration  Project,  is the  first study conducted
in the area to determine the effect of  a salinity
management  practice on  the conditions in the
basin. The project  was funded  on a  matching
basis (10% Federal  and 30% local) by  the Envi-
ronmental  Protection  Agency  in conjunction
with the Grand Valley Water Purification Proj-
ect,  Inc., which consists of representatives from
the Grand Valley Water Users Association, Pal-
isade Irrigation District, Mesa County Irrigation
Company,  Grand Valley Irrigation Company,
Redlands Water  and Power Company, and the
Grand Junction Drainage District. The project
was undertaken in order to  further the develop-
ment of pollution control technology in the ba-
sin. The objectives of the project include:
  (1) Demonstration of the feasibility  of reduc-
      ing salt loading in the  Colorado River sys-
                                              tem by lining conveyance channels to re-
                                              duce unnecessary ground water additions.
                                           (2) Extension of the  results of this study to
                                              the problem in the upper basin.
                                           The project is comprised of three study areas
                                         selected for  their different characteristics com-
                                         monly found in the valley. Area I, shown in Fig.
                                         2,  was  chosen as an intensive  study  area in
                                         which the bulk  of the investigation was  to be
                                         made and also includes most of the construction
                                         effort. This area was established to study in de-
                                         tail the  results of conveyance linings on the  wa-
                                         ter  and salt flow systems in an  irrigated area.
                                         The intensive study area was selected for its ac-
                                         cessibility in isolating most  of  the  important
                                         hydrologic parameters,  but had  the  important
                                         advantage that it allowed five irrigation com-
                                         panies to participate in one unit. Area II was se-
                                         lected because it  represented  a  different  land
                                         form several  miles west of Area I along a short
                                         section  of the Grand Valley Canal where high
                                         seepage losses had resulted in a severe drainage
                                         problem. Area III is  located along a section of
                                         the Redlands First Lift Canal, which is supplied
                                         from the  Gunnison  River and was  selected to
                                         evaluate the effect of different drainage  and soil
                                         types. Both Areas II and III were to be studied
                                         with sufficient data  to  confirm the results in

-------
                   Canals
                                  Government
                                  Highline
                                  Canal
                                              Stub Ditch
                                                              GRAND VALLEY PROJECT     111
  	Drains
       20
          Price Ditch
Boundary
Section Number

 Grand  Valley  Canal
                 ..-••A     1
       Mesa  County
       Ditch
                                               Colorado  River
                   Figure 2: Intensive study area, Area I, of the Grand Valley Project
Area I and involved a minimum number of ob-
servations.
  The procedure outlined for execution of the
project consisted of a two-phase study.  Phase 1
was planned to evaluate the conditions in the
study areas prior to the construction of the lin-
ing. These results have been reported (6). Phase
2 consisted of re-evaluating  the conditions after
the lining had been completed, which will be re-
ported in the following paper. In both cases, the
study was conducted to collect and analyze suf-
ficient data  to define in detail both the water
and salt flow systems.

       Description of the Grand Valley

Exploration and Settlement
  Although  numerous hieroglyphics and aban-
doned ruins testify to the occupation of the Col-
                             orado River Basin long before settlement began,
                             the first people encountered in the Grand Valley
                             were the Ute  Indians. The  first  contact these
                             peoples  had with white  men was recorded in
                             1776 when  an expedition led by Fathers Dom-
                             iniquez and Escalante passed north of what was
                             later to be Grand Junction and across the Grand
                             Mesa (3). The region was subsequently visited
                             by  fur trappers,  traders and explorers.  In 1839
                             one such trader named Joseph Roubdeau built
                             a trading post just  upstream from the present
                             site of Grand Junction.
                               In 1853,  Captain John W. Gunnison led an
                             exploration party into the Grand Valley from up
                             the Gunnison  River Valley in search  of a feasi-
                             ble transcontinental railroad route (1). As Cap-
                             tain Gunnison and his party traversed the con-
                             fluence of the Colorado and Gunnison Rivers,
                             an error was made by the expedition recorder as

-------
112
MANAGING IRRIGATED AGRICULTURE
to the proper naming of the rivers. Beckwith re-
ferred to the Gunnison River as the Grand River
and the Colorado River as  the Blue River, or
"nah-un-Kah-rea" as it was known to the Indi-
ans.  The mistake was later corrected, however,
since  the  Colorado River was known as  the
Grand River as early as 1842 (2).  Field surveys
conducted  by  Hayden  (4) in 1875  and  1876
found only  the Ute Indians in the valley,  and
skirmishes  with some  of the  hostile Utes  cut
short  the  1875 expedition. As  a  result of the
Meeker Massacre of 1879, the Utes were forced
to accept a treaty moving them out of Colorado
and onto reservations in eastern Utah. After the
completion of the Utes'~exit in September 1881,
the valley was immediately opened up for settle-
ment with the first ranch staked out on Septem-
ber 7, 1881 near Roubdeau's trading post. Later
that year on September 26, George A. Crawford
founded  Grand Junction as  a  townsite  and
formed  the  Grand Junction Town  Company,
October 10,  1881. On November  21,  1882, the
Denver  and Rio Grande Railroad narrow-gage
line was completed to  Grand Junction via the
Gunnison  River Valley and thus assured  the
success of the settlement.

Water Resource Development
  Early exploration did  not conclude  that the
Grand Valley had much potential for agriculture
since  the  terrain  appeared very  desolate. A
great deal of appreciation for this  judgment can
be acquired just passing through the area  and
noting the  landscape outside the  agricultural
boundaries.  In 1853,  Beckwith described  the
valley as, "The valley, twenty miles in diamater,
   UNCOMPAH5RE UPLIFT
                                        enclosed by these mountains, is quite  level and
                                        very barren except scattered fields of grease-
                                        wood and sage varieties of artemisia—the mar-
                                        gins of the Grand (Gunnison) and Blue (Colo-
                                        rado) Rivers affording but a meager supply of
                                        grass, cottonwood, and willow." Soon after the
                                        settlement began, it  was realized that the cli-
                                        mate could not support a non-irrigated agricul-
                                        ture. As a result,  irrigation companies were or-
                                        ganized to divert water from  the river for irriga-
                                        tion.

                                        Geology
                                          The plateaus and mountains in the Colorado
                                        River Basin are the product of a series of up-
                                        lifted  land  masses deeply  eroded  by wind and
                                        water. However,  long before the earth move-
                                        ments which created  the uplifted land masses,
                                        the region was the scene of alternate encroach-
                                        ment and retreat of great inland seas.  The  sedi-
                                        mentary rock  formations underlying large por-
                                        tions of the basin are the result of material ac-
                                        cumulated at the bottom of  these seas. In areas
                                        similar to  the Grand Valley area, the upper
                                        parts contain  a large number of intertonguing
                                        and overlapping formations of continental sand-
                                        stone and  marine shales, as shown in Fig. 3.(8).
                                        The  lower parts are  mostly  marine  Mancos
                                        shale and the  Mesa Verde group of related for-
                                        mations. This  particular geology is exhibited in
                                        about 23 percent  of the basin in such common
                                        locations as the Book Cliffs, Wasatch, Aquarius,
                                        and  Kaiparowits  Plateaus,  the  cliffs around
                                        Black Mesa, and large areas in  the San Juan
                                        and Rocky Mountains. Many of the interior val-
                                        leys attracted  settlers  such as the valleys in the

                                                                   GRAND MES«
  v>'-/.,v- V
   -T- r'" ~.~vr!  •
  '"J- tWUHTf ,"&MCi*S — V
  - NAMPM«OL'T£ , CTC .. /
  ^T-  !U»*COt«FOm*iTY). """ •'
  ''•'^^:<^A
        %.,
                         -^K^S^ft^^
                         -s. , - •-. -. x . * ; S I - -^ ' -' s - ^N. ^ / /-  -' ^ / \. i  \. » - - •  -• - v — l * - \ s . v   - J \ - v • p«orE«ijD^:
                         ,,/ I f\ , /-. I v/ N'rs < * ' -\ ' \-' V -GRANGE , GN£,S5 ftMPM 9 LlfE EC" , J«0«f (WT f V / -,9.»
                            ^ 'V^^^^vT^i^r^^T-^^^^^}^:-^^ l-"««
                          Figure 3: General geology of the Grand Valley

-------
                                                              GRAND VALLEY PROJECT
                                          113
 vicinity  of Price,  Vernal, and  Green  River,
 Utah; Rock  Springs,  Wyoming; and  Grand
 Junction, Delta, and Montrose, Colorado (5).
   The geology of an  area has a profound in-
 fluence on the occurrence, behavior, and chemi-
 cal quality of the water resources. In the moun-
 tainous origins of most water supplies, a contin-
 uous interaction of surface water and ground
 water occurs  when precipitation in the form of
 rain and melting snow enter  ground water res-
 ervoirs. Eventually,  these  quantities  of ground
 water return  to  the  surface flows  through
 springs,  seeps,  and  adjacent  soil  in regions
 downstream.  A further consequence of such a
 flow  system  is  the  addition  of water  from
 streams to the ground water storage during pe-
 riods of high flows and the  subsequent  return
 flows  during  low flow  periods.  The resulting
 continuous  interaction of surface  water  and
 ground water allows contact with rocks and soils
 of the region  which cause their chemical char-
 acteristics to be imparted to the water.
  The interior valleys  "of the  basin, the Grand
 Valley being a good example, do not receive
 large enough amounts of precipitation to signifi-
 cantly recharge  the  ground  water  storage.
 Usually, the water  bearing aquifers are buried
 deep below the valley  floor and are fed in and
 along the high precipitation areas of the moun-
 tains. Shallow ground  water supplies  are pre-
 dominately the result of irrigation. Although the
 water in the consolidated rock formations of the
 valley  region does not contribute to the stream
 flows as is  the case in  higher  elevations, it does
 have a pronounced effect on water quality. High
 intensity thunderstorms bring surface runoff in
 contact with the rocks  and soils, which  then ac-
 quire certain chemical  ions from the native ma-
 terials. The. long erosion by rivers and streams
 has deposited alluvium along certain lengths
 and  thus serves to produce an interchange of
 water in these areas.


Soils
  The physical features describing the test area
are similar  to  those of the  whole of Grand Val-
ley. The soils  in the area were classified by the
 Soil  Conservation Service  in  cooperation with
the Colorado  Agricultural Experiment Station
(8) in 1940.
  The desert climate of the area has restricted
 the growth of natural vegetation, thereby caus-
 ing the soils to be very low in nitrogen content
 because of the absence of organic  matter. The
 natural inorganic content is high  in lime car-
 bonate, gypsum, and sodium, potassium, mag-
 nesium, and calcium salts. With the addition of
 irrigation,  some locations have experience high
 salt concentrations with a resulting decrease in
 crop  productivity. Although natural phosphate
 exists in  the soils, it  becomes  available  too
 slowly for use by cultivated crops and a fertilizer
 application greatly aids yields. Other minor ele-
 ments such as iron are available except in those
 areas where drainage is inadequate. The soils in
 the test area are of relatively  recent origin as
 they contain no definite concentration of lime or
 clay  in the  subsoil as  could  be  expected in
 weathered  soils. Some areas in the valley have
 limited  farming use because of poor internal
 drainage which results in water logging and salt
 accumulations.
  Lying on top  of the Mancos shale and below
 the alluvial soils is a large cobble aquifer ex-
 tending north from the river to about midway up
 the test  area. The importance of  this aquifer
 with  respect  to the drainage problems  of the
 area  has been  demonstrated by  a cooperative
 study in 1951 between the Colorado Experiment
 Station in conjunction with the Agricultural Re-
 search Service  (ARS) (7), which  evaluated the
 feasibility of pump drainage from the aquifer.

 Land Use of Study Area
  Evaporation  and  transpiration  from  crops,
 phreatophytes, and other land uses results in a
 loss of salt-free  water to the atmosphere and a
 deposition  of salt in the soil profile. The magni-
 tude of these losses depends on the acreage of
 each water use.  As a part of a valley wide evalu-
 ation, the various acreages of land uses in Area I
 were mapped and the acreages for each land use
for Area I  are shown in Table 1,  while the irri-
gated  acreage for each  irrigation  company is
 listed in Table 2 (10).  One of the most quoted
 statements  in  the  literature  concerning  the
 Grand Valley is that approximately 30% of the
farmable area is unproductive because of the in-
 effectiveness of the drainage in these areas. Ex-
amination of the results presented in Table 1 in-
dicates that 70% of the study area can be classi-

-------
 114     MANAGING IRRIGATED AGRICULTURE

                  TABLE 1

         Land use in 1969 for Area I
 Classification

 Corn
 Sugar Beets
 Potatoes
 Barley
 Oats
 Wheat
 Alfalfa
 Cultivated Grass Hay
 Pasture
 Native Pasture
 Orchards
 Idle
 Other
 Farmsteads
 Residential
 Urban
 Stock Yards
 Water Surfaces
Cottonwoods
Salt Cedar
Willows
Rushes
Greasewoods
Precipitation Only
Acreage

  487
    1
    8
  255
   14
    9
  545
  141
  476
  387
  349
  559
    6
                                      3237
  258
   61
   85
    8
   70
    6
  268
   70
   10
  333
  225
                                       412
                                        70
                                       687
                                       225

                             TOTAL  4631

fied as irrigable land, however only 52% can be
considered productive. It should be noted that
the use of the term productive relates to those
areas producing cash crops such as corn, beets,
grains, orchards, alfalfa, etc.

Climate
  The effects of the mountain ranges in the Up-
per Colorado River Basin have much more in-
fluence  on the climate than does the latitude.
The movement of air masses is disturbed by the
mountain ranges to the extent that the high ele-
 vations are wet and cool, whereas the low pla-
 teaus and valleys and dryer and subject to wide
 temperature changes. A common characteristic
 of the climate in the lower altitudes is hot and
 dry summers and cold winters. Moist Pacific air
 masses can move across the basin, but dry polar
 air and moist tropical air rarely continue all the
 way across the basin.  Movement of both types
 of air masses is obstructed and deflected by the
 mountains so that their effects within the basin
 are weaker and more erratic than in most areas
 of the country.
   Most of the precipitation to the  basin is pro-
 vided  from the Pacific Ocean and the Gulf of
 Mexico whose shores  are 600 and 1000 miles
 from the  center of the basin, respectively. The
 air masses  are forced  up to high altitudes and
 lose much of their precipitation before entering
 the basin.  During the  period from October to
 April, Pacific moisture is predominant, but the
 late spring and summer  months receive  mois-
 ture from the Gulf of Mexico.
   The monthly distribution of precipitation and
 temperature for Grand Junction is shown in Fig.
 4  (9).  The climate  in the  area is marked by  a
 wide  seasonal  range,  but  sudden or  severe
 weather changes are infrequent due mainly to
 the high ring of mountains  around the valley.
 This  protective topography results in a rela-
 tively low annual precipitation of approximately
 eight  inches. The usual  occurrence of precipita-
 tion during the growing season is in the form of
 light  showers  from thunderstorms which  de-
 velop over the western mountains. The nature of
 the valley  location with typical  valley breezes
 provides some spring and fall frost protection
 resulting in an average growing  season of 190
 days from April to October. Although tempera-
 ture have ranged to as high as  105°F, the usual
 summer temperatures range  in the middle and
 low 90's in the daytime  to the low 60's at night.
 Relative humidity  is usually  low during the
 growing season, which is common in all of the
 semi-arid Colorado River Basin.

Irrigation
  The system of irrigation most common to the
area is surface flooding either by borders or fur-
rows.  The study area itself is located in the nar-
row eastern part of the valley which has a relief
of about 50 feet per mile sloping south towards

-------
                                                                GRAND VALLEY PROJECT
                                                               115
                                            TABLE 2

             Summary of 1969 land use under each irrigation company in Grand Valley
 Canal System

 Grand Valley
   Irrigation Company

 Grand Valley Water
   Users Association

 Orchard Mesa Irrigation
   District

 Palisade Irrigation
   District

 Mesa County Irrigation
   District

 Redlands Water & Power
   Company
 Cropland Domestic  Industrial   Water   Phreato-     Prec
 Acreage   Acreage  Acreage   Surfaces   phytes     Only

29,722       5,540     682       798      6,340
25,169


 7,694


 4,306


  608


 3,046
1,629


1,363


 547


  32


 885
               GRAND JUNCTION, COLO.
                 Alt. 4843 ft.
                                8 29" annual
                 ,   190  frost-free  days
                Apr 16
                                   Oct 23
   80'F-
        525°F annual
    60-
 a

 i
        JFMAMJJASONO
                       MONTH
 Figure 4: Normal precipitation  and temperature
 at Grand Junction, Colorado

the river. As a result, care is  taken to prevent
erosion in  most  cases by  irrigation  with small
streams. Most farms in the area  are small and
have short  run lengths. However, the small irri-
gation stream allows adequate application. The
635
135
 37
 31
 63
          3,591


6,554     10,429
  850
  177
  51
1,202
 964
 337
  51
1,235
 Total

46,678


44,416


11,006


 5,404


  773


 6,431
                    quantity of  water delivered  to the farmer is
                    plentiful so the usual practice is to allow self-
                    regulated diversions.

                    Conveyance System
                      The delivery system  in the valley is  divided
                    into the canal and ditch subsystem and  the lat-
                    eral subsystem. The distinction is made  primar-
                    ily on the basis of responsibility. The canal com-
                    panies  and   irrigation  districts   divert  the
                    appropriated water right from the river  and de-
                    liver it to the canal turnout, but they assume no
                    responsibility  for  the  water below this point.
                    The lateral network, which  originates at the ca-
                    nal turnout, is managed by cooperative  actions
                    of the individual users served by the turnout.

                    Canal System
                      The canals  and ditches in the Grand Valley
                    are shown in Fig. 5. Discharge capacities range
                    from above 700 cfs in the Government Highline
                    Canal to 30 cfs in the Stub Ditch, and these fig-
                    ures diminish through the length of each canal
                    or ditch. The lengths of the  respective canal sys-
                    tems are approximately  55 miles for the Govern-
                    ment  Highline  Canal, 12 miles for Price, Stub,
                    and Redlands Ditches,  110 miles for the Grand
                    Valley system,  and 36 miles for  Orchard Mesa

-------
 116    MANAGING IRRIGATED AGRICULTURE
                                                                      X OrM V.U.,
                          Figure 5: Grand Valley canal distribution system
Canals. In the following parts of this subsection,
individual  aspects of the canal system  will be
examined.
   The management of the canals and ditches in
the area varies between canals and also changes
with  the water supply. For example, it can be
noted that  during  periods  when river flows
become  small,  restrictions are placed  on the
diversion from the Government Highline Canal.
This  is possible because the flows are measured
and recorded at each individual turnout in that
system, and it is a necessity as their water rights
are junior to others. On the other hand, in most
instances along the other canals,  no measure-
ments are taken because little shortage is exper-
ienced. Another practice used  extensively in
the  region  is  regulating canal  discharges  by
varying the amounts of spillage into the natural
wasteways and washes. Neither  of these prac-
tices,  poor  measurement  and   spillage,  are
conducive to salinity control.
  The dilemmas  being faced by the irrigation
officials are numerous but can be traced to  one
factor. When the demand for irrigation was real-
ized  and the canal  alignments located, the ex-
pected demand for water was based on the total
area of land under the canal. However, when
the acreages of roads, homes, phreatophytes,
 etc., are deducted, the water available for each
 acre is significantly increased. For example, un-
 der the Grand Valley  Canal are 44,774 acres of
 which only 28,407 are irrigable. Consequently,
 instead of having a total before system losses of
 5.76 acre-feet  per  acre,  there is  more  than 9
 acre-feet per acre. The result is a two-fold prob-
 lem:

   (1) With a considerable excess  of water as
      appropriated stating that the water is to
      be  used  exclusively for irrigation,  it is
      very economical to  be wasteful.  Thus,
      how does  an  irrigation official  justify
      stringent management control to the peo-
      ple he serves?
   (2) The  history of development  in  the west-
      ern  United States has  always shown wa-
      ter to be a valuable  commodity to an area
      and as such,  the rights one has  are to be
      protected. Under the common water law,
      rights not  historically diverted  are  lost.
      Consequently, the Grand Valley must di-
      vert its rights for fear of losing them.

In short, it is not the practice of agriculture to be
wasteful, but the  laws regulating the use of wa-
ter dictate that a  user either be wasteful or give
up a valuable right.

-------
                                                              GRAND VALLEY PROJECT
                                          117
Canal Seepage
  The initial phase of the study involved the de-
termination of the seepage rates from the canals
and laterals in the test areas, Area I, II and III.
As noted previously, the ponding technique was
employed to assure the reliability of the results.
The first tests were conducted on all canals in
Area  I  except the Grand  Valley  Canal. The
lengths evaluated included a 2.6 mile section of
the Stub Ditch, 2  miles  of the Government
Highline Canal, 1.9 miles of the Price Ditch, and
2.2 miles of the Mesa County Ditch. In addition,
the tests  were made along the '/£  mile length of
the Redlands First Lift Canal in Area  III. The
0.15  mile length of the Grand Valley Canal in
Area II was not evaluated because of the high
seepage losses being evident, and the construc-
tion costs for dikes and tests in relation to the
costs of  the lining would have resulted in the
testing being more expensive than the lining. A
summary of the test results is shown in Table 3,
indicating only moderate seepage rates  in Area
I and a relatively high rate in the Redlands First
Lift Canal.  The  average seepage  rates were ap-
proximately 0.15 ft3/ft2/day (cfd) in the  Stub,
Price, and  Mesa County ditches; 0.25 cfd in the
Government Highline Canal; and an  average
rate of 0.40 cfd in the Redlands First Lift Canal.
  Although size and location of the water table
may  be  significant factors  influencing  seepage
rates, the results seem to contradict those of pre-
vious investigations. The water table depths in
the area  of the Stub Ditch, Government  High-
line Canal  and the Price Ditch are similar, but
the large canal has a  noticeably large  seepage
rate,  yet  it is much more affected by the  shale.
Nonetheless, it seems justified to conclude that
seepage rates in the Grand  Valley canals  prob-
ably  vary between 0.15 and 0.40 cfd. Extending
these results to the entire valley, the estimated
conveyance losses  attributable to seepage  prob-
ably  range between 25,000 and 65,000 acre-feet
annually, which would mean that  between 4%
and 10% of the annual diversion is lost  by seep-
age from canals.
  Because  the  influence of the  water  table
would  serve to reduce canal seepage, the loca-
tion of linings in the low lying areas may not be
as effective as in the upper elevations in the val-
ley. In the  principal study area, the water table
is higher than the bottom of the Grand Valley
 Canal in many places, and consequently these
 reaches probably pick up rather than lose water.

Linings
  Based upon  the results of the ponding tests,
two modifications to the original plan were rec-
ommended and incorporated in the project. The
first was the lining of the Government Highline
Canal in Area I should be made on the downhill
bank  of the last mile in the study area. The ob-
jective of this change from  a complete perimeter
lining as  originally proposed  was to  make an
evaluation of the  effectiveness of the  downhill
lining which should prove to be useful for future
projects. The second modification involved re-
ducing the construction scheduled for the drain-
age system to two small surface problems in the
area  and the  remainder of these funds being
spent on lateral linings. The linings which were
made in the canal system are  illustrated by the
darkened lengths in Fig. 6.  The cost estimates of
the final plans are tabulated in Table 4.
  The construction along the Stub Ditch, which
consisted  of the standard trapezoidal  slip-form
concrete lining, is shown in Fig. 7. The location
of  the Stub Ditch, which  runs to just east  of
Grand Junction, is along the  extreme northern
edge of the irrigated area of the valley, charac-
terized  by rapid and numerous undulations  in
the topography, thereby causing the canal align-
ment  to assume a very winding path. Along the
path  of the Stub Ditch,  there is often close
proximity to the Government Highline Canal as
indicated in Fig.  7. The  linings in the Price
Ditch system in Area I and the Redlands First
Lift Canal in Area III are of the same  nature as
the Stub Ditch linings.
  The other commonly used type of lining in the
area is  the  gunnite material used  for the Mesa
County Ditch,  as shown in Fig. 8. It is of some
interest to note the white  covered field in the
background; a  good illustration of the alkalinity
resulting from poor drainage. This type of lining
has several advantages  to the local irrigation
companies since they are already equipped to do
the work and the limited preparation for the lin-
ings does not tie up too many people. This same
procedure was  also used along the short section
of  the Grand Valley Canal in Area II and the
downhill bank lining of the Government High-
line Canal in Area I.

-------
118     MANAGING IRRIGATED AGRICULTURE

                                            TABLE 3
                        Results of ponding tests on canals in Area I and II.
                   Canal
          Mesa County
            Ditch
           Government
           Highline Canal
           Price Ditch
           Stub Ditch

Pond

1
2
3
4
5
6
7
8
9

Length
(ft)
1300
1075
1367
1227
1025
1316
1250
1449
1184

Mid-
station
6+50
18+37
30+57
43+55
54+82
66+52
77+85
92+85
106+02
Avg.
Wetted Pe-
rimeter (ft)
11.81
12.12
12.14
11.20
10.40
12.22
12.42
12.34
12.59
Seepage
Rate
(cfd)
0.12
0.13
0.12
0.15
0.15
0.17
0.18
0.13
0.13
e*
Acre-ft
Season
8.46
7.78
9.12
9.40
7.13
12.40
12.80
10.62
8.59
           Redlands First
             Lift Canal
1
2
1
2
3
4
5
1
2
3
4
5
6
 1
 2
4925
5278
1930
1960
2000
2035
1865
2175
2150
2600
2300
2275
2375
1213
1318
6+50
18+37
30+57
43+55
54+82
66+52
77+85
92+85
106+02
28+37
79+39
10+67
30+60
50+00
70+17
89+67
11+62
33+25
52+06
81+50
104+37
127+62
6+06
19+02
11.81
12.12
12.14
11.20
10.40
12.22
12.42
12.34
12.59
54.92
59.18
15.91
16.41
10.01
12.22
14.12
10.53
12.34
10.84
13.50
11.69
10.68
15.30
14.38
Total

0.25
0.25

Total

0.12
0.13
0.11
0.11
0.16

Total

0.15
0.13
0.15
0.12
0.13
0.22

Total

0.45
0.35

Total
  86

310.70
358.30

 669

 16.67
 19.00
 19.20
 12.40
 19.20

  86

 15.60
 15.60
 19.40
 17.10
 15.80
 25.60

109

 38.20
 31.80

 70
           * Based on 200-day season.

-------
                                                                GRAND VALLEY PROJECT
                                                                         119
                                           TABLE 4

                              Recommended construction program
   Map
  Desig-
  nation
Company Name
  Canal Name
  Area I
    A     Grand Valley Irrigation Co.
            Mesa County Canal

    B     Palisades Irrigation Dist.
            Price Ditch

    C     Grand Valley Water Users
            Assn. Govt Highline Canal

    D     Mesa County Irrigation Co.
            Stub Ditch

    E     Grand Junction Drainage Co.
            Open Drains
            Closed Drains
            Laterals

 Area II
    F     Grand Valley Irrigation Co.
            Grand Valley Canal

 Area III
    G     Redlands Water and Power
            First Lift Canal
                Perim-          Unit  Misc.   Total
Type of  Length   eter    Area   Cost  Costs*   Cost
Lining    (mi)    (ft)    (yd*)  ($/ytf)   ($)     ($)
                         Gunnite    2.2


                        Slip Form  1.9


                         Gunnite   1.0


                        Slip Form  2.5
                        Slip Form
                           Tile
                        Slip Form
                        Slip Form  0.5
                 14    17,500   3.25    2,100   58,975


                 15    16,720   3.25    1,900   56,240


                 15+    8,800   3.50    5,800   36,600


                 10    14,700   3.25    3,500   51,275
                                              4,000
                                             16,000
                                            110,815
                         Gunnite   0.15     15+    1,320  3.50   4,000    8,620
                 12     3,500   3.25    1,600   11,475
                                TOTAL     354,000
"Costs of pre-construction and post-construction ponding tests above amounts in CSU contract, plus costs of
installing headgates, etc.
+Downhill bank lining, only.
   The  construction  of the small drainage im-
 provements in the project is illustrated in Fig. 9.
 The figure indicates  the utility of such a design
 for a fanning area in that it only temporarily dis-
 rupts  farming  operations in  the vicinity and
 avoids  the land area occupation encountered in
 locations using an open drainage system.
   An often overlooked aspect of the lining pro-
 gram is the installation of appurtenances to the
 land  and to rehabilitate the system in general.
 Two  examples are  shown in Fig. 10,  showing
 first a  box-type  culvert arrangement to convey
 surface runoff from adjacent lands over the Stub
                              Ditch to avoid destruction of the lining. The sec-
                              ond (lower) picture shows a typical circular slide
                              gate turnout along the Price Ditch. These addi-
                              tions to the  canal  allow and promote better
                              management of diversions being made into the
                              lateral system.
                                In addition to the canal lining, a modification
                              was made to the original proposal to allow par-
                              ticipation in the project by individual farm own-
                              ers by setting some funding aside  for lining of
                              laterals.  A summary of lateral lining which  has
                              been constructed, all of which is in Area I, is
                              listed in  Table 5, as follows:

-------
120
MANAGING IRRIGATED AGRICULTURE
TABLE 5
Size and length of lined
Description
*14" trapezoidal
12" trapezoidal
10" trapezoidal
6"X10" rectangular
12"* 10" rectangular
1 2" buried pipe
8" buried pipe
6" buried pipe
Total


laterals
Length
5,941'
11,435'
624'
1,478'
1,987'
978'
2,111'
950'
25,504'
4.83 miles
                                                      Figure 7:  Slip Form concrete lined section of the
                                                      Stub Ditch
*Note - The first dimension listed  in the description
refers to the bottom width of the lateral, except where
pipe was used.
                                                     Stub Ditch
                                     Government
                                     Highline
                                     Canal
                                Price  Ditch
                        Grand Valley  Canal
           Mesa  County
            Ditch
                                                  Colorado  River
                   Figure 6:  Location within Area I of the linings that were constructed

-------
                                                              GRAND VALLEY PROJECT
                                                                                          121
  Figure 8: Gunnite  lined  section  of the  Mesa
  County Ditch

REFERENCES
   1. Beckwith, E. G.  1854. Report of explora-
tions for a route for the Pacific Railroad: U.S.
Pacific R.R. Explor. V. 2. 128 p.
   2. Fremont,  J. C.  1845. Report of the ex-
ploring expedition to the Rocky Mountains in
the year 1842 and to Oregon and North Califor-
nia in the years 1843-44: Washington, Gales and
Seaton, U.S. Senate. 693 p.
    3. Hafen, L. R. 1927. Coming of the White
 Men; Exploration and acquisition, in History of
 Colorado:  Denver Linderman Co.,  Inc.,  State
 Hist. Nat. History Soc. Colo., V.  1. 428 p.
    4. Hayden, V. F.  1877. Report of  progress
 for the year 1875: U.S. Geol. and Geog. Survey
 Terr., embracing Colorado and parts of adjacent
 territories. 827 p.
    5. lorn,  W. V., C. H. Hembree, and G. L.
 Oakland. 1965. Water Resources of the Upper
 Colorado River Basin - Technical Report. Geo-
 logical  Survey Professional  Paper  441.  U.S.
 Government  Printing Office, Washington, 369
 P-
    6. Skogerboe,  G.  V.,  and W. R.  Walker.
 1971. Preconstruction Evaluation of the Grand
 Valley Salinity Control Demonstration Project.
Agricultural Engineering  Department,  College
of Engineering, Colorado State University, Fort
Collins,  Colorado. June. 140 p.
    7. U.S. Department of Agriculture and Col-
orado Agricultural Experiment Station. 1957.
   Figure 9: Tile drain line in the study area
 Figure 10: Special
 of the lining work
                                                                        structure built as part

-------
122
MANAGING IRRIGATED AGRICULTURE
Annual Research Report, Soil, Water, and Crop
Management Studies  in the  Upper Colorado
River Basin. Colorado State University, Fort
Collins, Colorado. March. 80 p.
   8. U.S.   Department of  Agriculture, Soil
Conservation Service, and  Colorado  Agricul-
tural  Experiment Station. 1955. Soil Survey,
Grand Junction Area, Colorado. Series 1940,
No. 19. 118 p. November.
                                          9. U.S. Department of Commerce, Environ-
                                       mental  Science  Services  Administration,  En-
                                       vironmental Data Service. 1968. Local Climato-
                                       logical Data, Grand Junction, Colorado.
                                          10. Walker,  W. R.   1970.  Hydro-salinity
                                       model of the Grand Valley. M.S. Thesis CET-
                                       71WRW8.  Civil Engineering Department, Col-
                                       lege  of Engineering, Colorado State University,
                                       Fort Collins, Colorado. August.

-------
           Salinity  Control Measures in  The

                              Grand Valley

                GAYLORD V. SKOGERBOE and WYNN R. WALKER

                          Agricultural Engineering Department
                                Colorado State  University
                                 Fort Collins, Colorado
ABSTRACT
  Introduction of seepage and deep percolation
losses to the saline soils and aquifers, and the
eventual return of these flows to the river sys-
tem with their large salt loads, make the Grand
Valley in  Colorado one of the more significant
salinity sources in the  Upper Colorado  River
Basin.
  The Grand Valley  Salinity Control Demon-
stration Project was formulated to evaluate the
salinity control effectiveness of canal and lateral
linings for reduction  of seepage losses into the
ground water.
  The results indicate  that  the linings, even
though seepage rates  are relatively low, are fea-
sible based solely  upon salinity damages down-
stream. However,  it is concluded that the major
impact of  conveyance channel linings and sys-
tem rehabilitation  is the improved water control
necessary  for achieving more efficient  water
management practices on the farm, which is the
key to salinity control.

INTRODUCTION
  Salinity  is  the most pressing problem facing
the  future  development  of water resources in
the Colorado River Basin (Figure 1). Because of
the progressive deterioration in mineral quality
towards the lower  reaches, the detrimental ef-
      Figure 1: The Colorado River Basin
                                         123

-------
124     MANAGING IRRIGATED AGRICULTURE
fects  of  using  an increasingly degraded  water
are first  seen in the lower basin. As a result of
the continual development in the upper  basin,
most  of which will be diversions out of the basin
to meet  large  municipal  and industrial  needs,
water ordinarily available to dilute the salt flows
will be deleted  from the system, causing signifi-
cant   increases   in   salinity  concentrations
throughout the basin. The economic penalty re-
sulting from a use of lower quality water will be
primarily incurred  by those uses in the  lower
system.
  In order to alleviate  this problem, the  upper
basin must concern itself  with the aspect  of salt
loading in the  river system from municipal, in-
dustrial,  agricultural and  natural sources.  The
other aspect, which is the  salt concentrating ef-
fects, is  related to consumptive use,  evapora-
tion,  and transbasin diversions. Although  sev-
eral  methods   of  controlling salinity,  such as
phreatophyte eradication  and evaporation sup-
pression  on reservoirs  are desirable, the most
feasible solutions are in reducing inflows from
mineralized  springs and  more  efficient  irriga-
tion practices. In any case, the salinity manage-
ment objectives in  the upper basin must  neces-
sarily be concerned with a reduction in the total
salt load being carried to the lower basin in or-
der that the detrimental  salinity effects antici-
pated from further development can be limited.
  The major emphasis  on reducing  the salt in-
puts from the upper basin must be placed on ef-
fective utilization  of  the water  presently di-
verted for  irrigation  by  comprehensive pro-
grams of conveyance channel  lining, increased
irrigation efficiency on  the farm, improved  irri-
gation  company  management  practices,  and
more effective  coordination  of local objectives
between  the various institutions in the problem
areas. Salinity  is no longer a local problem  and
should  be considered  regionally  by the local
people.  In  irrigated areas,  it is necessary to
maintain an acceptable salt balance  in the crop
root zone which requires  some water for leach-
ing.  However,  when irrigation efficiency  is  low
and conveyance seepage losses are high, the ad-
ditional  deep  percolation  losses are subject to
the highly saline aquifers  and  soils common in
the basin and  result in large quantities  of salt
being  picked up and carried back to the river
system.
  A need exists to quantify the high input areas
and  examine  the   management  alternatives
available to establish the most effective salinity
control scheme. A first important step in this re-
gard  has  been taken in  the Grand Valley with
the Grand Valley Salinity Control Demonstra-
tion Project. The purpose of this paper is to pre-
sent a summary of the important findings of the
study.
        Summary of Previous Research
  Although the first  engineering investigations
in the region were made to determine the feasi-
bility of developing more irrigated land, almost
every study since has been related to salinity in
some way. The early settlers were able to dig
shallow fresh water wells, but as irrigation con-
tinued,  the  groundwater became  too saline for
any use. Not until low lying lands began to show
signs of high  water tables and older apple or-
chards began to fail was the  necessity for salin-
ity investigations realized. With the saline soil
conditions and natural  composition of the soil
producing a very low internal drainage capacity,
the lapse between first irrigation and the follow-
ing salinity  crisis was about  twenty years. This
seemingly protracted period would have  been
greatly reduced if the expansion of the irrigated
acrease to the higher regions had been accom-
plished  earlier.


Drainage
  Although many early settlers attested to the
seriousness of the drainage problems in the low
lying lands of the  valley, the conditions  were
alarming by about  1914. In 1908,  D. G. Miller,
Senior  Drainage Engineer,  Bureau  of Public
Roads  and  Rural  Engineering,  initiated  an
eight-year investigation  of the problem covering
the irrigated acreage  now served by the Grand
Valley  Irrigation Company (7). The objective of
the  study was  to  point out the contributary
causes  of the  high  water table condition in the
valley,  and to emphasize the necessity of drain-
age  as a  remedy.  Observations  were  made
throughout  the region to determine water qual-
ity of both soil abstracts and groundwater, water
table elevations, and the effects  of high water
table conditions.

-------
                                                        SALINITY CONTROL MEASURES
                                          125
  The early manifestations of injury due to alka-
linity in the Grand Valley were  first  realized
when regions in the more mature apple orchards
began to fail. Almost invariably, when the older
trees died younger trees would still grow  owing
to their  shallower  rooting system. Eventually,
not even field crops could be produced. Soon
after the causes of crop failure had been  some-
what determined, several farms were selected in
the valley  where the influence of the  Mancos
Shale could be studied. In addition, water sam-
ples were taken from wells that had been drilled
to the shale. Samples were also collected from
1,000 feet of drainage tile connecting six wells
drilled into a water bearing strata in the Mancos
Shale.
  The  conclusions  formulated from  this  re-
search had much to do with the organization of
the Drainage District in  1924. Some of the per-
tinent results of the study include:
  1.  Drainage in the Grand  Valley, because of
     the  magnitude of the problem, must be a
     regional  undertaking in  three  types  of
     drainage. These are  (1) relief of the pres-
     sure condition existing in the cobble aqui-
     fer,  (2) interception of  canal  and  lat-
     eral seepage and excessive irrigation from
     irrigated  land  in the higher parts of the
     valley, and (3) lateral drainage of the  wa-
     ter logged soils in locations where natural
     drainage is entirely deficient.
  2. The positive hydraulic gradients in the aq-
     uifer  are significant in  numerous locales,
     but rarely   sufficient to bring  the  water
     within three to six feet of the surface. Con-
     sequently, high water tables were probably
     the result of  the excessive  irrigation  and
     seepage.
  3. The analysis of soil samples and ground-
     water samples showed  a definite and di-
     rect relation between the quantities of ni-
     trate  in the samples and the total soluble
     salts.  The percentage of nitrate increases
     as samples get closer to the shale. As a re-
     sult, the solution of the alkaline problem,
     irrespective  of the  source  of the  nitrates,
     will  eliminate the  problem completely.
     Further, those salts  that do  remain in the
     soils can be easily ammended by  standard
     methods.
  In early  1946, an investigation was launched
by the Soil Conservation Service on the "Will-
sea" farm  west  of Grand Junction (3).  Piezo-
meter readings indicated the presence of a ver-
tical gradient of 0.48 ft/ft. Since an open drain
on the north side of the farm had little effect on
the problem, it was decided to drill an 8" well in-
to the cobble to test pumping as a drainage re-
lief.  On April 15, 1947 a pump  test was con-
ducted by  the SCS personnel. Pumping at the
rate of 100 gpm had only  very local  effects,
thereby indicating the need for a larger pump
and  certain well  improvements. The possibility
of  pump  drainage  was clearly  evident  and
aroused support  for a more detailed study. In
1948, an agreement with the Drainage  District
and the Soil  Conservation Service  was  for-
malized which initiated a test program in the
area. The  conclusions reached from this study
include:
   1. Analysis of the piezometer data indicated
     that the  pump produced a  decline in the
     water table  near the well of approximately
     0.10 inch per day, which decreased to 0.03
     inch per day at a radius of about  Vi mile.
   2. It was observed that  the pumping pro-
     duced greater and  more rapid changes in
     the aquifer  pressure.  The  fact  that  the
     dense clay  layer overlying the cobble re-
     stricted flow into the cobble from the soil
     was stated  as the  reason for the relative
     slow effect on the water table.
  3. When the pump test was conducted, it was
     noted that a "hole" in the clay layer ex-
     isted  in the radius of the influence of the
     pump. It was through this hole that the
     water above the  cobble entered the aqui-
     fer. Consequently, a drainage well in any
     part of the  lower valley can be  successful
     only if openings exist in the upper confin-
     ing stratum which  will allow the  ground-
     water above to enter the aquifer.

Canal and Lateral Seepage
   A complementary investigation to drainage is
evaluation of the sources of the excess  ground-
water.  Several   experiments  have  been  con-
ducted  in  the valley starting in 1954. The first
was a cooperative enterprise between the  Colo-
rado Agricultural Experiment Station and Agri-

-------
126     MANAGING IRRIGATED AGRICULTURE
cultural Research Service, USD A (1). An eight-
mile length of the Grand Valley Main Line Ca-
nal was isolated. Both inflow-outflow and seep-
age  meter  measurements  were   taken. The
results indicate an average loss rate in the sec-
tion of about 2 acre-feet per day per mile of ca-
nal (0.30  cfd).  Later,  on July 12,  1955, A. R.
Robinson (8) conducted a similar seepage loss
investigation on the Government Highline Ca-
nal,  but little is known of the results. In 1958,
Salomonson  and  Frasier investigated  lateral
seepage on several 200-foot sections (4). These
tests utilized the  ponding method of analysis
and  indicated a  loss  rate of 0.40 acre-inch/day/
mile.
  Summarizing the results of these studies:
  1. Seepage rates  in  the canal  were  signifi-
     cantly  higher where the  canals  were lo-
     cated in the alluvium material. In the shale
     cuts, the  seepage  rates  were noticeably
     lower,  probably  owing  to quantities  of
     groundwater interception.
  2. Results show that an average of about 5
     acre-feet  per month per mile  of canal
     through the test area would alone exceed
     the internal drainage capacity  of the soils.

Farm Efficiency
  It  has been concluded by most investigators
in the area that excess irrigation is  the primary
cause of the drainage problem in the lower val-
ley. To quantify the  magnitude of these  contri-
butions, efficiency studies were  conducted  in
1954 and 1955 by the Colorado Agricultural Ex-
periment Station (1), and by the U.S. Bureau of
Reclamation (9). The earlier studies determined
that  an average of 7.4 acre-inches excess was
being applied seasonally, but the farms were lo-
cated in a small  area. In order to provide more
reliable data and to provide a basis for extend-
ing the results to the valley, the 1955 study in-
volved six new farms between Grand  Junction
and Fruita which were irrigated by the Govern-
ment Highline Canal and the Grand Valley Ca-
nal. The results  of this study indicate  a  signifi-
cant amount being applied in addition to con-
sumptive  use  and leaching  requirements.  A
result showed  that an  excess as high as 23.7
acre-inches per acre was being applied annu-
ally. The Bureau of Reclamation study was con-
ducted from 1965 through 1968 on three locally
operated farms about 10 miles west of Grand
Junction. Careful consideration was given to the
parameters affecting the farm efficiencies being
attained, as well as the efficiencies that could be
attained. The fields ranged in slope from 0.7 to
1.5 percent, and in irrigation runs from 250 to
1252 feet. In addition, all three  operators prac-
ticed different farming methods  to some extent,
which  gave the investigators an opportunity to
observe the effect of farming methods on irriga-
tion efficiency. The results indicate the  follow-
ing general conclusions:
   1. The nature of irrigation in the area in the
     early part of the season is not conducive to
     efficient  irrigation as the process of "wet-
     ting across," or getting  the seed bed prop-
     erly moistened between furrows requires
     larger furrow  streams  and long run peri-
     ods.
   2. Careful observation of  efficient water use
     not only significantly improved yields, but
     also required less fertilizer.
   3. Control of the Western Cutworm in sugar
     beet fields by keeping  a moist soil is seen
     to decrease efficiency. Since drainage was
     not a problem in this  area, this practice
     was common and economical.
   4. Consideration of such factors as leaching
     and salt  loads during critical growth peri-
     ods, as well as soil tilth, not only improved
     yields, but  also  affected  irrigation  effi-
     ciency.

Regional Hydrologic and Salinity Studies
  The  increasing salinity problem in the Colo-
rado River Basin has necessitated the collection
and analysis of data on water and salt flows in
order to evaluate the contributions from various
sources.  Although  several  interested  govern-
mental  agencies  have  conducted  short term
studies in the  basin, the primary source of data
is  the U.S.  Geological Survey's stream monitor-
ing system. One of the most comprehensive ef-
forts to summarize and analyze this data was
made by lorns, Hembree, and Oakland (6) for
the period between 1914 and 1957 and adjusted
to the  1957 conditions. The  study was inclusive
of the  entire  Upper Colorado River Basin,  but
for the purposes of this report only the  section
dealing with the Grand Valley area has been ex-
tracted. Because the location of the exiting gag-

-------
ing station is below the confluence with the De-
lores River, some of the data is not uniquely rep-
resentative of the Grand Valley.
  Some  of the results of this study provide  an
interesting overview  of the  basin wide implica-
tions caused  by water use in the Grand Valley.
An extraction of part of the data is shown sche-
matically in  Figure  2, showing  the fraction  of
flows for water and  salt that flow in the Grand
Valley related to the total flows at Lee's Ferry,
Arizona. The effect  caused by  water resource
development  in the basin  above the Grand Val-
ley was shown to be  an increase from 178 to 592
ppm in the Gunnison River Basin and from 272-
382  ppm above Grand Valley along the main
stream.  The  net effect  of  man's activities
through  the Grand Valley were  determined  to
be a salinity increase of 256  to 547 ppm. The to-
tal salt loading to the Colorado River from the
Grand Valley averaged about  750,000 tons dur-
ing the period, although  when adjusted to the
1957 conditions the value  was  set at 440,600
tons. In terms of tonnage  contribution per acre,
the salt pickup would  be  between 5 or 6 and 8
tons per  acre  depending on the  time  period
used.
  The  wide fluctuation in salt pickup through
the Grand  Valley is shown in  the USGS report
(6) to  be related to the  river discharge.  This
poses  an interesting source of salinity control
speculation.  For example,  during  the  period
from about  1930 to  1942,  the  average  was
860,000 tons  pickup per year, in  1943 to  1951
the average pickup  was  745,000  tons, and  in
1951-1957 period, the addition was only 490,000
tons of salt pickup annually. The significance of
this is  that during each of these  periods the an-
nual  river discharge  was  decreasing,  which
would  somewhat  indicate  that  during water
short years the farm  efficiency increased, there-
by resulting in dramatic salinity reductions.
  The  1963-1967 water years  were selected for
study by  the Colorado River Board of California
(2) and other governmental  agencies in order to
appraise  the salinity sources in the basin, as well
as evaluate the future impact  of water resource
developments on mineral water quality. The re-
sults pertaining to the Grand Valley, in particu-
lar, indicates  the salt pickup to be about 8 tons
per acre per year, which is  the result Hyatt
(5) established for the 1963-1964 years (Figure
         SALINITY CONTROL MEASURES     127

3).  Both of these  references are  useful data
sources for examination of the Upper Colorado
River flow system.

            Experimental Design
  The evaluation of canal and lateral linings as
feasible salinity control measures depends to a
large  extent  on  the  success of  isolating and
measuring the various segments comprising the
water and salt flow  systems discussed in the pre-
vious section.  The  primary emphasis  of the
Grand Valley Salinity Control Demonstration
Project took place in Area I, the intensive study
area shown in Figure 4. Since the purpose of the
project was to demonstrate canal and lateral lin-
ings for salinity control, the main advantage of
this site is that it involved a large number of the
irrigation companies in the valley. Areas  II and
III  included one additional company, but were
selected for their  different land  conditions. The
principal test  area also represented an area in
which hydrologic conditions could be studied in
reasonable detail  and in which  the local  effects
of poor water management were significant.
  The  instrumentation in the study area indi-
cated by Figure 4 provides valuable data con-
cerning  many of the  important water and salt
movements.  While some  of  the parameters can
be  measured  directly such  as drainage dis-
charges,  lateral  diversions,  water quality, and
precipitation, others cannot and must be inves-
tigated indirectly. These budget parameters re-
late mostly to groundwater movement and are
monitored for changes using such techniques as
piezometers, wells, and soil sample analyses.

Instrumentation
  The area I instrumentation has provided valu-
able data for analyzing  the hydrology of the
area.  The data collected from these locations
were used to delineate the canal and lateral di-
versions, drainage outflows, water table and aq-
uifer pressure fluctuations, and  water quality at
the  various locations. In order to gain a clearer
picture  of the type  and  limitations  of data
gathered from each type of measurement, it is
useful to examine them individually.
  The  hydrostatic  pressures and gradients in
the  region of alluvium between the cobble aqui-
fer  and  the water table were studied using nu-

-------
128
MANAGING IRRIGATED AGRICULTURE
                                                           Percentage of  Combined Flow
                                                           of Colorado River at
                                                           Lee's  Ferry   Arizona

                                                           Percentage of Combined Dissolved
                                                           Solids pischarge of Colorado River
                                                           at Lee's  Ferry,  Arizona
                                                             Weighted  Average Dissolved
                                                             Solids Conentration in ppm
                                          592
                                                                                         547
               387
                                                           Grand  Valley

                                                           Grand  Junction, Colorado
                             Figure 2: Water Quality in Grand Valley

-------
                                                        SALINITY CONTROL MEASURES
                                               129
                             \
      River Outflows
      4,444,OOOac-ft
      GroundWater Outflows
      105,000 ac-ft
                                       Phreatophyta  Use
                                         75,000 ac-ft
Agricultural  Use
 185,000  ac-ft
                         River Inflows
                         4,690,000 ac-ft
                                                                   Tributary Inflow
                                                                   40,000 ac-fc
Precipitation
                                                                    75,000 ac-ft
       River Outflows
       3,907,000  tons
       Ground Water Outflows
         295,000 tons
                                    Natural  Sources
                                      700,000 tons
                                   Agriculture
                                   265,000 tons
                           River Inflows
                           3,180,000 tons
                          Tributary  Inflow
                           57,000 tons
      Figure 3: Schematic Water and Salt Budgets for the Grand Valley During 1963-64 Water Years
merous clusters  of  % inch steel pipe piezome-
ters. In each cluster,  which contained three  to
seven  piezometers,  the  depths  that  each
extended were varied so the vertical hydraulic
gradient could be evaluated. The small pipe sec-
tions were placed using a jetting technique and
the same apparatus could also be used to period-
ically flush the pipes to insure reliable readings.
The small pipes  were also used to evaluate hy-
draulic conductivity of the subsurface strata.
  Due  to ' the  limitation  of  the piezometer
depths, the examination of the movement of wa-
ter and salt in the cobble aquifer was conducted
by drilling two-inch steel pipes into it. The func-
tion of these instrument points has been essen-
tially the same as that of the smaller piezome-
ters. The larger diameter better facilitates the
collection  of water  samples and are somewhat
easier to read.
  The drillers log from each installation located
the topography of the subsurface strata to and
including the Mancos Shale. In addition, a few
      of the pipes were initially continued a small dis-
      tance into  the  shale  formation indicating the
      presence of small aquifers within the shale. Be-
      cause the number and  distribution  of these
      larger piezometers is  small, all pipes were in-
      stalled in the overlying cobble strata and the in-
      fluence of water inside the shale has been as-
      sumed small.
        The small Pars hall and Cutthroat flumes used
      in this study have not only been used for moni-
      toring drainage  discharges as shown in Figure 5,
      but have also been used  extensively as part of
      the  special studies on farm  efficiency, lateral
      seepage, and lateral diversions. At the most im-
      portant locations  in the area, flumes with con-
      tinuous stage recorders have been installed in
      order to minimize the error in evaluating these
      discharges.  Since  both of these  type flumes are
      most accurate when critical depth occurs in the
      flume, attention to these conditions was made
      upon installation  and continually  monitored
      throughout the  study.

-------
 130
MANAGING IRRIGATED AGRICULTURE
       • Piezometers
       «2"  Wells
       »Conol  Rating Section
       0 Drainage Measurement
	Drains
	Area Boundary
                            Stub Ditch
            Mesa
            County
            Ditch
Government
hkghline
Canal
                                                                          Price Ditch
                                                                             rand Valley Canal


j
«> /
? 1
X
                                                  do River
                           Figure 4: Location of Instrumentation in Area I
   An essential  item  in the analysis was the de-
 termination of  the magnitude of the quantities
 of water diverted from the canals into the lateral
 system for irrigation. This was accomplished in
 two ways:
   1.  The  smaller capacity  canals  such  as the
      Stub Ditch, Price Ditch, and Mesa County
      Ditch were rated at both the inlet and exit
      sections in the test area.
   2.  The  discharges in the Government  High-
      line Canal and the Grand Valley Canal are
      so large that the error in the inlet and exit
      ratings would  over-shadow the diversions.
      To remedy this situation the turnouts from
      each canal were individually rated.

 Land Use Inventory
   The quantity of water transpired from vegeta-
 tion  surfaces or evaporated from soil add water
 surfaces can only be measured  using expensive
 equipment.  As  an  alternative, computational
 methods can be used to estimate evapotranspi-
 ration for each  type of vegetation. In order to
 meet  the data requirements of these methods, it
                                         is necessary to  determine  the type and area  of
                                         each land use. This analysis was performed for
                                         the  entire Grand Valley (12) with the data sub-
                                         divided by sections within townships and also
                                         by section within townships served by individual
                                         canals. The procedure involved  carrying  aerial
                                         photos (1 inch  =  1000 feet) into the field and
                                         marking  the various  land uses according to a
                                         prescribed index. Then the data was  transferred
                                         to inked  base maps where the acreage  of each
                                         land use was evaluated.

                                         Seepage Investigations
                                           The seepage from conveyance channels in the
                                         three project areas of the  study were examined
                                         in two parts: (1) lateral seepage in Area  I and
                                         (2) canal  seepage in Areas I and  III. This type
                                         of study was conducted before and after the lin-
                                         ings had  been constructed along the lengths  of
                                         the  Stub  Ditch, Price  Ditch, Government  High-
                                         line  Canal, and Mesa County Ditch. In addition,
                                         a before  and  after  measurement  was  made
                                         along the Redlands First Lift Canal, which is lo-
                                         cated in Area III.

-------
                                                         SALINITY CONTROL MEASURES
                                          131
  Figure 5: One-Foot Cutthroat Flume Located in
  an Open Drain in the  Test Area

   The ponding method for determining seepage
loss  rates was selected  to  assure accurate mea-
surements, especially in the  two large  canals.
The  degree of error in using  inflow and outflow
measurements would have tended to  mask the
magnitude  of  seepage  losses.  The ponding
method, illustrated in Figure 6, consists of plac-
ing an impervious barrier  at two or more loca-
tions along the canal, filling each resulting iso-
lated section with water, and then taking peri-
odic measurements on  falling water elevations,
which can then be used to evaluate the seepage
rates.

Farm Efficiency Studies
  The efficient use of water on the farm  would
be achieved when the  total quantity of water en-
tering the field was just sufficient to meet the
demands of the crops, soil surface evaporation,
and the quantity of water necessary to leach the
residual salts of the evapotranspiration process
from the root zone. Factors which decrease the
farm  water use  efficiency  in  the Grand  Valley
are numerous and result  in excessive deep per-
colation losses and large field tailwater flows.
Since the waste flows result in additional salin-
ity problems, irrigation efficiency is a good in-
direct  examination of the areal  magnitudes of
deep  percolation,  drainage  interception,  con-
sumptive use, and field tailwater. As noted ear-
lier, without expensive  and  complex  instru-
ments, farm efficiency studies become the most
feasible method of estimating these variables.
   The principal  parameters  related to evalua-
ting on-farm water use include:
   1. Field tailwater which can be  measured in
     the Grand  Valley with small  Parshall or
     Cutthroat flumes.
   2. Deep  percolation and leaching  require-
     ments  derived from budgeting  the  root
     zone flows.
   3. Root zone  storage  changes measured  by
     soil sampling and correlated neutron probe
     techniques.
   4. Evapotranspiration which is computed.
   5. Uniformity  of applications evaluated from
     soil moisture samples, along with  advance
     and recession analyses.
   6. Precipitation.


    Results of the Canal and Lateral Linings

Basin-Wide Benefits of Salinity Control in the
Grand Valley
   If the economics of salinity control are ex-
amined closely, almost any  control method is
feasible on  the basis of the traditional benefit-
cost ratio. The  extension of this concept  to  a
basin-wide scale  yields some interesting points.
According to the estimated damages being ex-
perienced in the lower basin (10), a conservative
value of $100,000  per  ppm  per year at Lee's
Ferry,  Arizona is not an unreasonable datum.
In the period between  1931  and 1960  and ad-
justed  to 1960 conditions in the  basin, the an-
nual flow  rate  at  Lee's Ferry was 10,880,000
acre-feet carrying 8,570,000  tons of total dis-
solved  solids (5). Thus, a ton of  salt represents
0.676 x 10~4 ppm and results in seven dollars of
annual damage.   .
   If it  is assumed that a repayment period of 50
years is taken at an interest rate of 7%, the max-
imum present cost that should be incurred to re-

-------
 132     MANAGING IRRIGATED AGRICULTURE
                   Figure 6: Canal Seepage Measurement Using the Ponding Method
duce one ton of salt from the system is $97. It
should be noted that once the money has been
spent to control the ton of salt, it is assumed to
be a permanent improvement so that the money
is only needed at an initial point of time. The to-
tal  cost of  the  lining program was $420,000,
with $350,000 being spent for actual construc-
tion and $70,000 for administration, design, and
construction inspection. In terms of the Grand
Valley Salinity Control Demonstration Project,
a salt reduction of 4300 tons should be made in
order to economically justify the project, based
upon downstream damages  alone.  If for  com-
parison,  a  repayment period  of 30  years  is
taken, the acceptable present cost of each ton of
salt reduction is $87 and the break  even point
for this project would be a  reduction of 4800
tons.

Local Benefits of the Canal and Lateral Linings
  An adequate evaluation of the operation and
maintenance benefits attained from the linings
is difficult. Correspondence  with the officials of
the Grand  Valley  Water Purification Project,
Inc., most of whom are also serving as local irri-
gation and  drainage officials,  has  delineated
certain benefits  resulting from the lining pro-
gram.  The moderate gradient channels, when
running near capacity from April  1 to October
31, experience comparatively few problems with
bank vegetation, mossing, and sedimentation.
Records  and  comments from irrigation com-
panies indicate  an  average  maintenance  cost
per mile of between $250 and $370 per year, de-
pending of course on the canal size. In the inten-
sive  study area, the construction will probably
result in  a total savings of $2500 annually. Al-
though periodic maintenance is always  neces-
sary, there are linings ten or more years old that
have as yet required almost no attention. When
the canals are lined, the improvements to the
delivery system for new turnout structures and
measuring devices greatly  aid  handling,  distri-
bution, and charging for  water and provide a
stimulus  to  irrigators for more efficient water
management.
  The local benefits from  the linings in many
parts of the Grand Valley are  primarily indirect

-------
                                                           SALINITY CONTROL MEASURES
                                                      133
and include factors such as improvements to ad-
jacent lands. In the test area and in a large por-
tion of the valley, the value of land is primarily
determined  by the expanding urban areas and
as  such do not greatly  depend  on agricultural
production. Nonetheless, the increased utility of
well drained soils is worth far more than the cost
of the linings.

Seepage Reduction Due to the Linings
  The results of the seepage rate measurement
before and  after the canal linings (summarized
in Table  1) indicate that the linings are  not es-
pecially effective when the  seepage rates are al-
ready low. However,  based upon the modeling
results, the  reduction in  seepage rates attribut-
able to the canal linings was placed at about 600
acre-feet  per  year. While  this figure may  not
seem very large at this point, the effect on the
salinity loadings to the river is fairly substantial
and will be noted in a later section.
  Although  no  effort was made  to evaluate
           seepage  rates  in  the laterals after  they were
           lined in Area I, it is not unrealistic to assume the
           same  values  as found for the canal linings.  A
           typical loss rate of 0.1 cfs per mile would repre-
           sent a rate in a usual lateral of about 0.5 cfd, or
           about as large  as earlier studies indicated. Con-
           sequently, the lining of most laterals would re-
           sult in about a 90%  seepage rate reduction.
             The effect of the lateral linings constructed as
           part of this project indicates linings probably re-
           sult in a seepage  reduction of on the order of
           200-300  acre-feet  annually.  This  reduction  is
           from lining only five of the estimated 90 miles of
           laterals  in Area I. However, most of the linings
           were constructed  where water tables were low
           and thus represent high seepage rate areas. On
           a valley-wide scale, the lining of canals could be
           expected to reduce  seepage  losses from  5% to
           about 1% of the total canal diversions. The seep-
           age losses from the lateral system could  be ex-
           pected to be much greater, probably from 10%
           to about 2% of the total agricultural diversions.
                                           TABLE 1

                          Comparison of before and after lining results
                                        by ponding tests
                    Canal
Seepage rate
before lining
    (rfd)
Seepage rate
 after lining
    (cfd)
Reduction
          Stub Ditch
          (concrete slip-form
          lining)

          Price Ditch
          (concrete slip-form
          lining)

          Gov't Highline Canal
          (gunnite lining on down-
          hill bank)

          Mesa County Ditch
          (gunnite lining)

          Redlands First Lift Canal
          (concrete slip-form
          lining)
    0.15
    0.15
    0.25
    0.15
    0.40
    0.07



    0.07



    0.13



    0.03


    0.06
   53
    53
    48
    80
    85

-------
134    MANAGING IRRIGATED AGRICULTURE
Salt Reductions Due to the Linings
  In order to control the salt pickup from the
Grand  Valley,   the  irrigation  return  flows
through  the  cobble  aquifer and the overlying
soils must be reduced. It has been argued that
even reducing groundwater flows would not re-
duce the salt loadings since the reduced water
would simply become more concentrated  and
thus carry the same  load. The results of this
study indicate that groundwater is retained  in
the soils  and aquifer much longer than is neces-
sary to reach chemical equilibrium with ambient
salinity concentrations.  However, whether  a 50
percent   reduction  in   moisture   movement
through the soil would result in a 50 percent re-
duction in salt pickup is not known at the pres-
ent time. Nevertheless,  since the quantity of
water reaching the groundwater due to convey-
ance seepage is small, this assumption can be
made with confidence.
  The exact region of salt pickup through the
soil and aquifer profiles has not been defined in
this  study. It was originally though that when
water came in contact with the Mancos Shale
the most significant load of salts was picked up,
but later data analysis showed this assumption
to be false,  so  the specific nature  of the salt
loading   in  the area is unclear.  Nonetheless,
these unanswered questions do not significantly
affect the reliability of the model which was de-
veloped to evaluate on  a quantitative scale the
magnitudes   of  the  various segments  of the
hydro-salinity system.
  Although the complete water and salt budgets
derived during the course of the study will not
be presented  herein, it is interesting to examine
some of the annual figures. During the 1971 wa-
ter  year, the total salt input to the intensive
study area was approximately 23,000 tons and
the total outflows amounted  to about 79,000
tons. Consequently, a net gain of about 56,000
tons occurred, which amounts to almost 12 tons
per irrigated acre. Even though this figure  is
above the valley average (8 tons per acre), it is
reasonably certain that salt loading varies from
region to region within the valley and the condi-
tions encountered in the test area are probably
the most conducive  to  high salt loading any-
where in  the valley.
  Of the total salt input in Area I, 14,810  tons
(0.86 tons/acre-foot) is computed to be the  area
contribution to the groundwater  flow system,
with 12,520 tons resulting from over irrigation
(deep percolation), 170 tons from canal seepage,
and  2120 tons from lateral seepage.  Thus, for
each ton of salt added to the groundwater basin
(assuming that here is where all additional salt
is added),  3.8 tons  can be expected  to  result
when the irrigation return flows reach the river.
The  effects  of the canal and lateral linings can
now be seen. The canal and lateral linings which
reduced seepage into the groundwater basin by
about 930 acre-feet during the year reduced the
salt  contribution from  the area by 4700 tons.
The  canal  and lateral  linings  which were con-
structed as part of the project, while undertaken
for the purpose of investigation and demonstra-
tion, are thus economically feasible from a salin-
ity control standpoint with a benefit-cost ratio of
1.09. If the damage figure for each part per mil-
lion  at Lee's  Ferry, Arizona is  increased to
$150,000, the resulting minimum cost per ton of
salinity reduction is $138 and the  lining in this
project  would have a benefit-cost  ratio of 1.54
based solely on salinity control (50 year repay-
ment period at 7% interest).

Indirect Impact of Canal and Lateral Linings
  Although the formal results of canal and lat-
eral lining  will establish that the impact on sa-
linity reduction due solely to  reducing seepage
is small in comparison to the total problem, the
importance   of  canal   system  improvement
should not be underestimated. The greatest im-
pact  on the salinity problem from  canal and
lateral linings  is better system  management. Al-
most  every arid agricultural area  depending
upon a year to year fluctuation in the water sup-
ply can  produce evidence to substantiate the
fact that during periods of diminished supply,
farm production is often higher as a  result of
better water management on the farm. The ex-
planation of this observation does  not lie in the
use by the farmer alone, but  also involves the
distribution of an equitable water supply to the
irrigator. Thus, the irrigation  company or dis-
trict  which is faced with water distribution
among demands which totally exceed the supply
are the  primary controllers. This  indicates that
any presently  proposed  salinity control alterna-
tive to be implemented on a  valley-wide scale
must involve efficient canal management. In or-

-------
                                                         SALINITY CONTROL MEASURES
                                          135
der to improve canal system management, reha-
bilitation by programs of linings and installation
of adequate diversion,  measurement, and con-
trol structures is inherent.

              Study Conclusions
  The basis for reducing salt loadings from the
Grand Valley is to minimize percolation into the
groundwater basin. At the present level of salin-
ity control technology,  it is difficult at best to
determine what effect a 50% reduction in these
flows  valley-wide would have.  However,  the
bulk of the deep percolation losses are the result
of excessive irrigation applications and, as such,
the objective  of  any salinity control program
should include measures to improve the effi-
ciency of on-farm water management.
  The first and most important consideration in
improving farm water use  is sound canal com-
pany  management practices.  Implied in this re-
alization is the  requirement of  sound water
measurement at the farm turnout and again at
critical division points among farmers below the
turnout.  This  would necessitate a considerable
rehabilitation of both the canal and lateral sys-
tem, and the implementation of a "call period"
to allow canal  operators more time for flexible
water handling. In addition,  it is  an important
requirement that  the canal  companies extend
their control of the water below the canal turn-
out structure to include key division points with-
in the lateral system in order to insure equitable
allocation of water among users. Other neces-
sary aspects of efficient  on-the-farm water man-
agement  are  the incorporation  of irrigation
scheduling  programs and an effective drainage
system.
  An important conclusion of this study is that
no single salinity control  measure will  effec-
tively alleviate  the Grand Valley salinity  prob-
lem.  It is thus adamant that  an integrated pro-
gram  involving a  planned combination of water
delivery rehabilitation, on-the-farm water man-
agement,  and  effective  drainage  be  imple-
mented for efficient salinity control.
  Probably the most binding constraint for re-
ducing salinity  from the valley is not in the rel-
ative  feasibility of the  technological alterna-
tives,  but in the institutional structure of west-
ern water laws. In order for water management
in the area to  improve without devastation of
the already burdened agricultural system,  new
mechanisms for economic incentives for improv-
ing the efficiency of water use must be applied.
Present water laws are actually a deterrent to
better water management because of the result-
ing loss of a portion of a water right, which is
considered very valuable to an irrigated  agricul-
ture.
  The results of this study indicate that the lin-
ing in the test  area reduced salt inflows to the
Colorado River by up to 4700 tons annually.
The bulk of this reduction is attributable to the
canal linings, but clearly indicated is the greater
importance  of lateral  linings. The  economic
benefits to the lower basin user alone exceed the
costs of this project. Consequently, it seems jus-
tifiable to  conclude that conveyance lining in
areas such as the Grand Valley where salt load-
ings reach 8 tons per acre, or more, are feasible.
The local benefits accrued from reduced main-
tenance, improved land value, and other factors
add to the feasibility of conveyance linings  as a
salinity control alternative.

REFERENCES
    1. Colorado Agricultural Experiment  Sta-
tion.  1955.  Irrigation  water  application  and
drainage of irrigated land in the Upper Colo-
rado River Basin. Progress Report 1954,  Project
W-28, Department of Civil  Engineering, Colo-
rado A&M College, Fort  Collins,  Colorado.
March. 38 p.
    2. Colorado River Board of California. 1970.
Need for  controlling salinity of the  Colorado
River. The Resources Agency, State of Califor-
nia. August. 89 p.
    3. Decker,  R. S. 1951.  Progress report on
drainage project  (1945  to  Jan.,  1951),  Grand
Junction,  Colorado. Lower  Grand Valley  Soil
Conservation District, Mesa County, Colorado.
January. 37 p.
    4. Frasier,  G., and  V.  Solomonson. 1958.
Seepage study  on lateral ditches.  Unpublished
file data.
    5. Hyatt,  M. Leon.  1970. Analog  computer
model of the hydrologic and salinity flow of sys-
tems within the Upper Colorado River Basin.
Ph.D. dissertation, Department of Civil Engi-
neering,  College  of Engineering, Utah State
University, Logan, Utah. July.

-------
136     MANAGING IRRIGATED AGRICULTURE
   6. lorns, W. V., C. H. Hembree,  and G. L.
Oakland.  1965. Water resources of the Upper
Colorado  River Basin - Technical Report. Geo-
logical Survey Professional  Paper  441. U.S.
Government Printing Office, Washington,  369
P-
   7. Miller, D. G. 1916. The seepage and al-
kali  problems in  the  Grand Valley,  Colorado.
Office of Public Roads and Rural Engineering,
Drainage Investigations. March. 43 p.
   8. Robinson,  A. R.  Personal  Correspon-
dence with  W. J. Chiesman, Superintendent,
Lower Grand  Valley Water Users Association,
Grand Junction, Colorado. August. 1955.
   9. U.S. Department of the Interior, Bureau
of Reclamation.  1968. Use of water  on federal
irrigation  projects: Grand Valley Project, Col-
orado. Final Report 1965-1968. Volume, 1, Sum-
mary and Efficiencies. Region 4, Salt Lake City,
Utah.

  10. U.S.  Department of the Interior, Federal
Water  Pollution Control Administration (now
the Environmental  Protection  Agency).  1970.
The mineral quality problem in the Colorado Ri-
ver Basin, Appendix B. January. 166 p.

  11. Walker,  W. R.  1970.  Hydro-salinity
model  of the Grand Valley. M.S. Thesis CET-
71WRW8.  Civil Engineering Department,  Col-
lege of Engineering, Colorado State University,
Fort Collins, Colorado. August.

  12. Walker,  W. R., and G. V. Skogerboe.
1971. Agricultural land use  in the Grand Valley.
Agricultural Engineering Department,  College
of Engineering, Colorado State University,  Fort
Collins, Colorado. July. 75  p.

-------
        Management of  Irrigation Water  in
     Relation  to Degradation  of Streams  by
                              Return  Flow
                                   H. R. RAISE
                   Soil and Water Conservation Research Division,
                           U.S. Department of Agriculture

                               and F. G. VIETS, JR.
                               Research Soil Scientist
                           U.S. Department of Agriculture
ABSTRACT
  Alternatives for improving the quality of re-
turn flow to avoid the concentration of soluble
salts in  the return and seepage flows are  dis-
cussed. The process of concentrating salts is an
inevitable consequence of evaporation and tran-
spiration. Possibilities for using automated  wa-
ter management systems with a high degree of
water control to deposit salts either in the solu-
ble or insoluble form below the root zone or in
the aeration zone above the water table are  dis-
cussed.  The dilemma of keeping excess salts
out  of return flow and streams or continuing to
periodically leach them out of the root zone to
maintain a neutral or favorable salt balance can-
not  now be resolved.  Eventual solution of the
problem must take into  consideration the crop,
depth to water table, amount of rainfall, quality
of irrigation water,  and  water  management
practice.
INTRODUCTION
  The specter of rich, western irrigated valleys
turning into salty deserts or marshes like the
once productive plains of the Tigris and Euphra-
tes Valleys of ancient Mesopotamia has induced
us to  think that every irrigation project and
every field should have a favorable salt balance.
Favorable salt balance means that total salt in
drainage return flows should be at least equal to
or greater than that of the irrigation water ap-
plied. Since the water evaporated from soils or
transpired by plants is essentially distilled wa-
ter, the  inescapable consequence of this con-
sumptive use is that the salt concentration in the
drainage is necessarily higher than that of the ir-
rigation water. Degradation of water quality by
increase in salinity is inevitable if there is drain-
age.
  These traditional  ideas on the  need for a fa-
vorable salt balance have been entirely consis-
                                        137

-------
138     MANAGING IRRIGATED AGRICULTURE
tent with the very practical need to keep salt in
the root zone at permissible levels so that yields
will not be reduced. Salts inhibit plant growth
largely by reducing water absorption by  the
roots, although  there are some specific ion ef-
fects such as the limited tolerance of some spe-
cies to chloride.
  Salts  originate from the weathering of pri-
mary minerals and of  some sedimentary rocks,
such as marine shales that may be high in salts.
Other sources of salt are the cyclic salts from the
sea that come down in rain or as particles, espe-
cially along coasts. Salt rarely accumulates in
soils from the weathering of rocks to levels of
significance  to  plants,  but  becomes concen-
trated in soils by the evaporation of salt carrying
waters.
  Our traditional approach to improving saline
soils or to controlling  the accumulation of salts
in  nonaffected  soils   is to  periodically leach
them. In sophisticated  terms,  this means the cal-
culation  of  the  leaching requirement, which is
the amount  of water that must be applied in ex-
cess of consumptive use to keep the soil solution
at the bottom of the root zone at a permissible
salt concentration, depending on the kinds of
crops to be grown. The salt concentration of the
input and drainage waters, and the consumptive
use are the  parameters needed in making such
calculations. Precipitation of  insoluble com-
pounds such as  lime and gypsum  or their solu-
tion is ignored in such calculations (U.S. Salin-
ity Lab Staff20).
   Obviously, the kind  of irrigation system and
its  management is just as important as the kind
of  cropping and cultural system and irrigation
water  quality in  management  of excess  salts
that no one wants, either in  the drainage or in
the soil.


    Alternatives for Improving the Quality
                of Return Flow

   If irrigation is to continue, then at this time,
we appear to be caught on  the horns of a di-
lemma. One horn is the need  to keep salts low in
the root zone, and thereby degrade the  quality
of the return flow to  rivers by increasing the
concentration of salt in the drainage or the per-
colate to ground water which eventually reaches
rivers or is pumped for reuse. The other horn is
the need to maintain good quality return flow by
having neither leaching  to  ground  water nor
runoff. The salts accumulating from consump-
tive use would have to be stored either in the
root zone or in the aerated zone between the
bottom of the root zone and the capillary fringe
of the water table. The amount of the  latter
space depends on  substrate porosity  and the
depth of the water table and  its fluctuations.
Possibilities  for  salt  storage  include  storing
them in part of the surface soil not used by roots
(as in trickle irrigation) or the precipitation of
lime or gypsum so that they no longer enter into
the salt balance. Another possibility is to grow
more salt-tolerant crops. Sections of the world
that are very arid and have no water for leach-
ing are further  along than  we in the United
States in knowing how to live with salts in the
root zone.
   Since we are caught on the horns of the di-
lemma,  it  seems  that  some  compromise will
have to be worked out between degradation of
quality of  return flow and degradation of soil.
The compromise will have  to be  made on a
river-by-river basis or regional basis, with all
benefits and detriments known or estimated. On
one river some degradation would have little ef-
fect; on others, it would add to an already acute
problem.
   Little is  known now about storage of salts in
the soil profile in insoluble forms so that they do
not enter the leaching requirement equations to
maintain   a  favorable  salt  balance.   Eaton5
pointed out that irrigation  waters high  in re-
sidual bicarbonate  should accumulate  lime and
sparked  a whole new set of investigations on
special  treatment of high carbonate waters in
assessing   the   sodium  hazard.  Thorne and
Thome19 examined 14 paired profiles that had
 12 different water sources in Utah as to changes
in chemical  compounds. Lime content changes
varied from an  increase of 2'/$ percent to a de-
crease of 7l/2 percent in the surface 6  inches of
soil.  Lime content decreases were associated
with  the waters highest in salts as measured by
electrical   conductivity,  calcium  plus  magne-
sium, chloride,  and sulfate. This behavior was
explained  by  increased  solubility  of calcium
 carbonate  in salt solutions. By statistical meth-
 ods  they  showed  that lime content  did  not
 change when the water had a conductance of 2
 mmhos per cm. Such water has high or very
 high salinity hazard.

-------
                                          DEGRADATION OF STREAMS BY RETURN FLOW    139
   Several studies (Bower and Wilcox2, Bower et
 al.3, and Bower et al.1) have dealt with the ques-
 tion of precipitation and dissolution of CaCO3
 when  soils  are  irrigated with high bicarbonate
 water. However, most irrigation waters do not
 contain residual carbonate as defined by Eaton5,
 and so the extent of precipitation of lime, gyp-
 sum, and other salts of low solubility in soils ir-
 rigated with more common irrigation waters are
 now unknown.  Extensive studies  of the phe-
 nomena  are now underway at the U.S. Salinity
 Laboratory.
   In the management of irrigation water for the
 final disposition of salt, the amount applied and
 time of application in relation to consumptive
 use and  the method of  irrigation are very im-
 portant management variables.

          Controlled Application of
          Irrigation Water to Achieve
         High Application Efficiencies
   Based  upon arguments presented  above, it
 appears that a salt  control achieved either by
 controlled leaching, by chemical precipitation of
 salts in the soil profile, or by storage  of salts in
 or below the surface soil is essential if a perma-
 nent irrigated agriculture is to be maintained. In
 any alternative,  the need for efficient  applica-
 tion and controlled distribution of irrigation wa-
 ter is  implied.  Recent technological  develop-
 ments to automate existing water control struc-
 tures on  surface irrigation systems  and the de-
 velopment of automated solid-set and traveling
 types of  sprinkler irrigation systems have stim-
 ulated  capital investment on  large or corporate
 farms to  achieve a high degree of water control
 and at the same time reduce irrigation labor re-
 quirements.  More recently, widespread interest
 in the solid-set trickle- or drip-irrigation method,
 although  limited in application, offers further
 promise  of achieving an even greater degree
 of water  and nutrient control  under certain
 soil, water and cropping conditions. Regardless
 of the  irrigation method used, application effi-
 ciencies and uniformity  coefficients > 90 per-
 cent are possible if the necessary capital invest-
 ments are made and the irrigation  systems are
 properly  designed and operated to  maximize
their potential capabilities.  In the  discussion
that follows, the viewpoint is that it is desirable
and essential to attain a  high  application effi-
ciency to avoid excessive deep percolation and
runoff losses, thereby using the deep subsoil as
a "salt-sink" to reduce the buildup of salt in the
return flow and/ or ground-water reservoirs.

Surface Irrigation Systems
  Dead level irrigation, where applicable,  pro-
vides an excellent system of water control and is
readily adapted to automation. This irrigation
practice  is  being used principally on  alluvial
soils in the Desert Southwest to reduce irriga-
tion labor requirements and to achieve a higher
degree of water control for efficient operation.
Water is applied to dead-level irrigated blocks
of land ranging from 5 to 10 acres in size. Large
streams of water, usually 10 to 20 cfs, are re-
leased through  standard jackgates installed in
open  channels   or from a  buried  pipeline
equipped with alfalfa valves and risers.
  Where jackgates are used, the irrigator need
only open one gate and close another to change
irrigation sets. This can be done in minutes. Ap-
plication  efficiencies approaching  95  percent
are possible by rapidly flooding the basin to irri-
gate a solid crop like alfalfa or applying water to
dead-level furrows  formed within the irrigated
block. No applied water or rainfall runs off. Wa-
ter  applied  to "fast" furrows is turned  around
at the far end to irrigate "slow" furrows. Thus,
the farmer  is able  to apply  nearly the exact
amount of water required on a timed basis to
satisfy crop  needs  and leaching requirements,
provided the stream size is matched to soil tex-
ture and/or intake rate, crop and the size of area
irrigated.  In the  Lower Rio  Grande Valley of
Texas, additional benefits accrue through max-
imum utilization of rainfall (about 20 inches an-
nually) for leaching and/or supplemental water
for  crop growth in an area where the average
annual irrigation allotment is about 1 acre-foot
per acre.
  An example of how a dead-level irrigation
system  has  been  automated  on  the  Bruce
Church Ranch, Inc., near Poston, Arizona, is il-
lustrated in Figures 1 through 6". The USDA
pneumatic diaphragm10, shown in Figure 1, is
used to control the remote release of water from
standard alfalfa valves  and risers attached to
buried pipelines.
  Although the  captions for each figure   de-
scribe the function of the various components

-------
140     MANAGING IRRIGATED AGRICULTURE
     Figure   1: Pneumatic   Diaphragm  in-
     stalled on  12-inch alfalfa valve and riser
     to automatically control discharge. When
     inflated, as shown, flow of water ceases;
     when deflated, as shown in Figure 2, wa-
     ter  is  released  (Bruce  Church,   Inc.,
     Poston, Arizona, February 1971)
Figure  3: Automatic  irrigation   system
shown in  operation  on 20 acres of wheat,
Bruce Church Ranch, Inc., Poston, Arizona,
February  1971. Zig-zag border permits dis-
tribution  of water in both directions from
buried  distribution  pipeline.  Each 5-acre
field is dead-level and 4, 12-inch valves dis-
charge  a  combined  flow of about  10  cfs.
Field at right with valves in "zigs"  has re-
ceived a 4-inch depth of application and at
left, 4 valves in "zags"  have just opened to
water opposite 5-acre field
     Figure 2: Same as Figure 1  with pneumatic
     diaphragm  deflated to permit discharge of
     water
 Figure 4: Border dikes surround each 5-acre
 field.  Water flowing  in dead-furrow com-
 pletely surrounds irrigated area permitting
 rapid  flooding from all directions

-------
                                       DEGRADATION OF STREAMS BY RETURN FLOW
                                           141
Figure  5: Controls  required  to  operate
automatic surface irrigation system consist
of pressure regulator (far left),  pilot valves
with 1/2-inch polyethylene control lines at-
tached,  a commercially available electro-
mechanical timer used to irrigate turfgrass,
and air compressor (far right) with 1/2 H.P.
electric motor
Figure 6: Nearest pilot valve is opened to
permit rapid discharge to atmosphere of air
in pneumatic closures for release of water
from riser and alfalfa valve in set 1. Holes in
wall of piston of extended air cylinder allows
rapid  discharge of air from control line con-
nected to pneumatic closures.  When in re-
tracted position, air  pressure is fed to pneu-
matic closures through 5/16-inch tubing to
close  alfalfa valve and stop the flow of wa-
ter
and collectively explain how the system works, a
few additional remarks are offered to clarify op-
eration procedures. When  manually  operated,
the irrigator opens 3 or 4 12-inch valves in the
first irrigation set.  He then alternately opens
and closes the required number of valves per set
until  the  40-acre  unit  has  been  irrigated.  The
combined flow from the 3 or 4 valves (usually 10
to 15 cfs) floods a 5- to 10-acre area surrounded
by a dike. From 1.5 to 2.0 hours are required to
irrigate each set. During the peak water-use pe-
riod,  as many as 14 irrigators are needed to  si-
multaneously apply water  to designated fields
on this 6,000-acre ranch. With automation, it is
anticipated that the labor requirement could  be
reduced by one-half.
  The dead-level system of irrigation described
here provides a high degree  of water control and
leaching capabilities from an engineering point
of view. Whether a favorable salt balance can be
maintained and at the same time promote the
precipitation and the deposition of salts at some
point within the soil profile is a matter of specu-
lation. Certainly the degree of water control this
system provides is adequate to accomplish  such
an objective if soil profile  conditions, depth to
water table, and  other chemical and physical
factors are favorable.
  Level or  low gradient border strips or checks
offer the same automated  capabilities and de-
gree of water control as the dead-level irrigation
system. But certain limitations  of slope, topog-
raphy, soil texture,  soil depth and development
must be recognized.
  In  1959 and 1960 a  low-gradient border  strip
irrigation system was constructed at the Newell
Field Station, Newell,  South Dakota, on Pierre
clay with slopes ranging from 0.5 to 5 percent4.
Based  upon the  high application  efficiencies
achieved  with low-gradient  irrigation systems
elsewhere and coupled with the possibilities for
automation, it was believed that maximum  utili-
zation could be  made of rainfall during years
when irrigation allotments to the Station  were
12  inches  or less. A full irrigation allotment of
24 inches was received in 10 of 34 years (1930 to
1964) and,  during  the same period,  10 allot-
ments were less than 12 inches. One further ad-
vantage anticipated was the possibility of adapt-
ing  automatic controls  to alfalfa valves and
risers in each border strip or check.

-------
142     MANAGING IRRIGATED AGRICULTURE
  Problems that were difficult or impossible to
solve soon appeared. Most  of the surface soil
that was moved downslope to create the level or
nearly level border strips was deposited in the
escarpment  or  berms  between each leveled
strip. After 12 years,  the soil productivity of the
"cut" areas on the original  steeper slopes has
not  been  fully restored. Where the  depth  of
"cuts" exceeded 12 inches, high salts in some of
the exposed subsoil  reduced productivity fur-
ther. Since it is difficult to maintain a perfectly
level surface, low spots with  standing water fol-
lowing irrigation or rainfall have interfered with
trafficability and timeliness  of cultural opera-
tions. Where grass was~ planted on  the escarp-
ments between border strips, large  soil cracks
appeared in the shrinking-swelling clays as they
dried, resulting in leakage from one border strip
to another. Rodent burrows further complicated
the problem. A 6-mm. plastic sheet placed in the
escarpment with a  "berm  splitter"  prevented
leakage but considerably increased the cost  of
the  installation.  Surface drainage  following
periods of heavy rainfall was inadequate on the
nearly level border strips, and border dikes at
the lower end often had to be opened to remove
excess water.
  Based upon these experiences,  the advan-
tages of low-gradient irrigation  systems  origi-
nally envisioned are far outweighed by the prob-
lems created. Where  site conditions are similar
to those described above, low-gradient bench-
leveled  border-strip  irrigation systems are not
recommended.
  In contrast, level or low-gradient border strips
or checks constructed on Tripp fine sandy loam
on  the Scottsbluff  Experiment Station  near
Mitchell, Nebraska, have been highly successful
on slopes up to 5.0 percent. Topographic condi-
tions are comparable to those of the Newell in-
stallation.   "Cut"  and  "fill"  problems  were
avoided by stockpiling the surface soil for reap-
plication to the cropped area14. Subsurface soil
was used to form the escarpments or berms be-
tween the border strips.
  If properly designed and operated, the border
irrigation  system gives a high degree of water
control. Jensen and  Howe16 17 working on the
Mitchell  site   obtained  application  efficien-
cies of 80 to 95 percent where alfalfa, corn, field
beans, and sugar beets were  grown in rotation.
Sugar beets restricted surface flow of water late
in the season, resulting in one or two irrigations
in which water was poorly distributed. They in-
dicated, however,  that distribution  could have
been  improved  either by  applying a greater
depth of water or  by using a larger stream size
with  only a small reduction in irrigation effi-
ciency.
   Howe and Heermann13 extended the Mitchell
investigations  of Jensen  and Howe16 17 by es-
tablishing low-gradient border  strips  on two
fine-textured soil  sites near Grand Junction,
Colorado.  Both stream  size  and  slope  were
varied in developing design criteria. They found
that for practical purposes  the uniformity coef-
ficient is independent of input  and  stream size
within the range of 0.03 to 0.12 cfs  per foot of
border width and is independent of slope in the
low-gradient range and for steeper  slopes with
close-growing  crops. Both the medium- and
fine-textured soils  had high initial intake, up to
3 or 4 inches in the first 2 hours. Proper opera-
tion of  the  irrigation system, and  not design,
was the key factor for efficient  irrigation appli-
cations. The need  for automatic control  where
frequent changes of irrigation sets are required
to achieve efficient irrigation was recognized by
Howe and Heermann as an essential ingredient
for widespread adaptation of the border irriga-
tion  practice.   Automation  can  be   easily
achieved by  adapting the pneumatic diaphragm,
Figure 10,  to  alfalfa valve risers from  buried
pipelines or from open channels with automated
check gates15.

Automated Furrow Irrigation Systems
  Automated  furrow  irrigation systems  with
tail-water reuse facilities also are being used to
attain the same degree of water control as pre-
viously  discussed.  Haise  and Fischbach9 sum-
marized the  progress made in research and de-
velopment of hardware for the  automation of
single- and double-pipe irrigation systems.
  As  implied, the double-pipe automated irriga-
tion system  requires two  pipelines: One that is
buried for conveyance of water and the other on
the surface to distribute water. The surface pipe
has adjustable gated outlets to direct the flow of
water into furrows  and, where automated, is at-
tached to a hydrant or valve body equipped with
a pneumatic diaphragm shown in Figure 1. The

-------
                                           DEGRADATION OF STREAMS BY RETURN FLOW     143
essential features of this  system are shown  in
Figures 7 and 8. Remote controls used to change
irrigation sets are the same as  those  previously
described to automate the level basin irrigation
(see Figures 5 and 6).
  Fischbach and Somerhalder6  report that reuse
systems are an essential part of any  automatic

    Figure 7: Auto-surface valve (1970 model)
    with double outlet. A 1/2-inch polyetheylene
    pipe  connected  between  controller  and
    pneumatic valve controls discharge by regu-
    lating the flow of air to inflate or deflate the
    pneumatic valve — developed by Fischbach9
   Figure 8: Automatic irrigation system (3200
   feet buried pipeline) being used to irrigate
   field  beans  near  Wiggins,  Colorado,  in
   1966.  Inflatable pneumatic valves shown in
   Figure 1 are placed inside of standard  hy-
   drant  attached to alfalfa valve  riser. Risers
   are spaced 200  feet  apart  and sufficient
   gated  pipe is available to irrigate 4 sets in 24
   hours. Control wire  for telemetry system
   that actuates 3-way pilot valve  was placed
   in same air line used to pressurize pneuma-
   tic valve for remote control operation. Bar-
   rell section was  used to cover valve when
   system is moved to the next point of control
irrigation system if high application efficiencies
and uniformity coefficients are to be attained.
Measured  uniformity coefficients ranged from
86.5  to  95.4  and  application efficiencies from
84.4 to 96.8 percent. The same automated irriga-
tion system operated  without the reuse or cut-
back  features resulted  in  an  application  ef-
ficiency  of 64.8 percent with 27.1 percent of ap-
plied water lost as runoff. Where reuse systems
are incorporated with automatic controls, mini-
mum water application depths ranging from  1.5
to 2 inches were possible.
  Automated double-pipe irrigation systems are
now commercially  available. With the increased
cost of labor  and the trend toward larger farm
operation,  farmer acceptance of automated fur-
row irrigation systems probably will be acceler-
ated. One word of caution, however: Automated
surface irrigation  systems are  best adapted to
soils with a medium to low intake rate but are
not recommended  where slopes exceed  1.5 per-
cent.  At present,  the  estimated  cost  of auto-
mated furrow irrigation with reuse features is
comparable to the self-propelled  center-pivot
and less  than  automated solid-set sprinkler irri-
gation systems.
  Automatic  pipe-gates  installed in a  single-
pipe irrigation system have the advantage of uti-
lizing  a  single pipe to transmit and  distribute
water  to furrow-irrigated  crops. The  experi-
mental pipe-gate, shown in  Figure 9, is a  nor-
mally open valve that  utilizes a small penumatic
     Figure 9: Pneumatic pipe gage in opera-
     tion.  Note that pipe outlets  have  been
     added to direct discharge into each of two
     furrow

-------
144
MANAGING IRRIGATED AGRICULTURE
diaphragm to provide the closing force. The ex-
ploded view of the pipe gate  (Figure  10) illus-
trates its simplicity. The  plastic disc,  a unique
   Figure 10: Exploded view of components used
   to develop normally open pneumatic pipe gate
   showing, from left to right: pipe locking mech-
   anism, rubber gasket,  valve body (white), seal-
   ing disc, pneumatic diaphragm and threaded
   diaphragm support

feature  of the pipe-gate, placed between the dia-
phragm and  valve  seat,  provides an effective
water-tight closure. The valve embodies a built-
in fail-safe feature whereby loss of control pres-
sure causes all valves  of the pipeline to open.
When this happens, water is dissipated by  reap-
plying it to all furrows irrigated by the pipeline
until the malfunction is located and repaired.
   A one-quarter-mile automated single pipeline
that incorporates the pipe-gate shown  in Figure
7 and a tailwater reuse system will be operated
near Waverly, Colorado, in 1972. Application
efficiencies comparable to those  systems previ-
ously described will be possible.  Measurement
of deep percolation and nitrate losses under two
water management practices at this site should
provide information pertaining to possibilities of
utilizing controlled water  applications  to reduce
percolation losses and  salts in return flow.
   Humpherys, et  al.,15   describes the  use  of
automated check gates in open conveyance dis-
tribution channels where  outlets are installed in
spreader ditches or distribution bays to water in-
dividual furrows or corrugations. A stream cut-
back  feature incorporated in  the design of this
system  provides for improved water control on
sloping or graded fields.

Sprinkler Irrigation  Systems
   Automatic  center-pivot, and to a lesser ex-
tent,  solid-set sprinkler irrigation systems are
                                         becoming a familiar sight in eastern Colorado
                                         (187,000 acres), in the sand hills of Nebraska, in
                                         the Columbia River  Basin of Washington, in
                                         Idaho  and  elsewhere  in the  Western  United
                                         States.  For controlling  irrigation  applications
                                         and the eventual movement of water, salts and/
                                         or nitrates within  the soil profile, few irrigation
                                         methods can equal the performance of a prop-
                                         erly designed and operated center-pivot or solid-
                                         set system.  Heermann and Hein12  report mea-
                                         sured uniformity coefficients of  87 and 90 for
                                         two center-pivots in eastern Colorado.
                                           Some operators of center-pivots tend to over-
                                         irrigate by continuous operation  of the system.
                                         Others apply less water than the crop needs, es-
                                         pecially  where  soil  intake  rates  restrict  the
                                         amounts of water that can be applied per irriga-
                                         tion. Preliminary data obtained by Heermann*
                                         on two center-pivots used to irrigate corn near
                                         Crook, Colorado,  in 1971 indicate that the total
                                         water applied was 26 inches including irrigation
                                         and rainfall on a coarse-textured site where only
                                         21 inches were needed based upon  estimates of
                                         ET using the  modified Penman equation18  and
                                         measured soil water  depletion. Irrigation plus
                                         rainfall on another field in the same area totaled
                                         21 inches, but here a slowly permeable layer at
                                         the  6-foot  depth reduced  deep   percolation
                                         losses. In the latter instance,  application effi-
                                         ciency exceeded  95 percent. In spite of  hail
                                         damage, both  fields produced about 125 bushels
                                         of shelled corn per acre.
                                           Center-pivot and/or solid-set  irrigation  sys-
                                         tems have a high  initial  cost (about $250/acre)
                                         but provide a  high degree of water  control with
                                         a low labor requirement. For example, one op-
                                         erator and  his  elderly  father irrigated  about
                                         1,000 acres  of corn with  a center-pivot and per-
                                         formed all  cultural operations except planting
                                         and harvest.

                                         Trickle Irrigation
                                           Perhaps the ultimate  in  water and  nutrient
                                         control has been achieved in Israel with the de-
                                         velopment of the trickle  irrigation system. This
                                         system is designed to drop water  into the soil at
                                         designated points along  the crop  or tree  row
                                         using nozzles or orifices to control discharge (2-
                                         10 liters per hour). A typical installation for a

                                         •Personal communication.

-------
                                         DEGRADATION OF STREAMS BY RETURN FLOW     145
deciduous orchard consists of one or two trickle
lines per tree row with nozzles spaced one meter
apart on each line. Crops irrigated include cit-
rus,  pears,  peaches,  apples, grapes,  bananas,
and some  high-valve row crops, including toma-
toes and melons.
  Advantages of the trickle  system  include
(a) an  exceptionally high degree of water and
nutrient control, (b) maintenance of low ampli-
tudes in the fluctuation of matrix and solute suc-
tion  by frequent irrigations especially where
highly  saline  irrigation  waters  are  used,
(c) maintenance  of dry  areas  between trickle
lines to facilitate timeliness of  cultural opera-
tions, (d)  possible savings in water under cer-
tain  soil,   crop  and  climatic conditions, and
(e) marked increases in yields.
  Water moves downward and outward into the
soil  from  the point of  application forming a
leached, onion-shaped zone beneath each noz-
zle. Goldbert and  Shmueli8  delineate  3  saline
zones: An upper zone  where  the salinity in-
creases as the distance from nozzle and the soil
surface decreases;  a  wide  intermediate  zone
where  salinity values are low; and a lower zone
where  the salinity level increases with  depth
and with  distance from the nozzle. Plant roots
are generally concentrated in the intermediate
zone where most  effective leaching occurs. Con-
centration of accumulated salts depends  upon
the salinity of the irrigation water applied, the
rate  of application, evapotranspiration, hydrau-
lic conductivity, and back diffusion of salts into
the leached zone during the offirrigation  cycle.
Goldberg, et al.7,  later reported on the use  of the
trickle  system for irrigation  of  grapes  using
three frequencies  of application.
  The question  of how long trickle  irrigation
systems can be operated without leaching is per-
tinent.  Most researchers  believe that  some
leaching in arid regions will be required to sus-
tain  high  yields even though the crop may be
planted in the  same wetted  strip previously
cropped. Even so, when considered in terms  of
immobilizing salts by  chemical  precipitation
and/or deposition, the trickle irrigation method,
more than any  other irrigation system,  offers
possibilities for achieving such an objective. The
dry  area   between the  trickle  irrigated rows
might serve as a salt sink. With controlled peri-
odic leaching, these salts might then  be moved
downward  into a zone  below  that  normally
wetted by the drip-nozzle, providing that leach-
ing by rainfall  is not excessive and water table
depths do not fluctuate throughout the layer of
accumulated salt.

                  Outlook
  In spite of the desirability of improving water
quality in rivers and aquifers by avoiding return
and seepage flows that are enhanced in soluble
salts,  we do not know if a permanent irrigated
agriculture can be achieved without having neu-
tral or favorable salt balances in river basins and
on individual fields. The  process of concentrat-
ing salts is an inevitable consequence of evapo-
ration and  transpiration.  Whether the answer to
this concentration will be the growth of crops
more tolerant of salinity, or water management
systems with  sufficient control that the salts can
be deposited either  in a soluble  or  insoluble
form  in part  of the root zone or in the aeration
zone  above the  water table, cannot  be given
now.  The alternative is to keep salts out of the
root  zone  by periodic leaching, with  some of
them reaching ground and surface water. This is
what is now being done.
  The problem will have to be solved differently
in different regions depending on kinds of crops,
depth to water table,  amount  of rainfall,  and
quality of irrigation water.
  Methods of irrigation that can be closely con-
trolled and fully automated, even though expen-
sive, are now available so that the consumptive
use requirements can  be almost  exactly sup-
plied. These systems can also be used to supply
the exact amounts of water for leaching   re-
quirements when leaching is needed. Such sys-
tems  include automated basin,  sloping border
and furrow systems with reuse  of runoff water,
and center-pivot and solid-set sprinkler systems.
Among the latest developments is trickle irriga-
tion,  which may be  most useful when water is
very scarce and there is little alternative to stor-
ing the accumulated salt  in part of the soil pro-
file out of reach of roots.
  Proper scheduling of irrigations and applying
the correct amount  of water on automated or
non-automated irrigation  systems remains  an
age-old  problem. Irrigation  scheduling proce-
dures developed by Jensen, et al.,18 offers prom-
ise of predicting when a farmer should irrigate

-------
146     MANAGING IRRIGATED AGRICULTURE
and how much water he should apply.  But ap-
plying the correct amount of water particularly
on sloping lands by surface-irrigation methods
in the absence of tail-water reuse facilities still
constitutes a major hurdle. Fanners generally
irrigate by experience. Seldom do they measure
the input to our runoff from individual fields. As
a consequence, irrigation systems  that can pro-
vide a high degree of water control become inef-
ficient if improperly operated.

REFERENCES
   1.  Bower, C. A.,  Ogata, G.,  and  Tucker,
J. M. 1966. Sodium hazard of irrigation waters
as influenced by leaching fraction and  by pre-
cipitation or solution of calcium carbonate. Soil
Sci. 106:29-34.
   2.  Bower, C. A., and Wilcox, L. V.  1965.
Precipitation and solution of calcium carbonate
by irrigation operations. Soil  Sci. Soc. Amer.
Proc.  29:93-94.
   3.  Bower,  C. A.,  Wilcox,  L. V.,  Akin,
G. W., and Keyes, M. G. 1965. An index of the
tendency of CaCCh to precipitate from irriga-
tion waters. Soil Sci. Soc. Amer. Proc. 29:91-92.
   4.  Dimick, N. A. 1965. Level and low-gradi-
ent irrigation at Newell. South Dakota Farm &
Home Research. XVI(l):32-37.
   5.  Eaton, F. M.  1950. Significance of car-
bonates in irrigation waters. Soil Sci.  69:123-
134.
   6.  Fischbach, P. E., and Somerhalder, B. R.
1969.  Efficiencies  of an automated surface irri-
gation system with  and without a runoff reuse
system. Paper No. 69-716, presented at ASAE
Winter Meeting in Chicago,  Illinois.
   7.  Goldberg, S. D., Rinot, M., and Karu, N.
1971.  Effect  of trickle  irrigation intervals on
distribution and utilization of soil  moisture in a
vineyard. Soil Sci. Soc. Amer. Proc.  35:127-130.
   8.  Goldberg, D.,  and Shmueli,  M.  1970.
Drip  irrigation—a method used under arid and
desert conditions of high water and soil  salinity.
Trans. ASAE
   9. Haise,  H. R., and Fischbach, P. E. 1970.
Auto-mechanization  of pipe distribution sys-
tems. Nat'l. Irrig. Sym. Papers, pp. M-l - M-15.
   10. Haise,  H. R., Kruse, E. G., and Dimick,
N. A. 1965. Pneumatic valves for automation of
irrigation systems. ARS 41-104. 21 pp.
   11. Haise,  H. R., Kruse, E. G., Payne, M. L.,
and Erie, L. J.  1970. Automated pipe-valves for
surface  irrigation. Paper  No.  70-745.  Winter
Meeting ASAE, Chicago, Illinois. 16 pp.
   12. Heermann, D. F., and Hein P. R. 1968.
Performance  characteristics  of  self-propelled
center-pivot sprinkler irrigation system.  Trans.
ASAE 11(1):11-15.
   13. Howe,  O. W., and Heermann, D. F. 1970.
Efficient border irrigation design and operation.
Trans. ASAE 13(1): 126-130.
   14. Howe,  O. W., and Swanson, N. P. 1957.
Stockpiling will keep surface soil on top of your
field. Nebraska Expt. Sta. Quart. 5(l):8-9.
   15. Humpherys, A.  S.,  Garton,  J. E.,  and
Kruse, E. G.  1970. Automechanization of open
channel distribution systems.  Proc. Nat'l. Irrig.
Sym. pp. L-l -  L-20.
   16. Jensen, M. E., and  Howe, O. W. 1961.
Operational characteristics of border checks on
a  sandy soil.  Proc.  Int's.  Comm. Irrig.  and
Drain. Div., Ann. B. pp. 5-10.
   17. Jensen, M. E., and  Howe, O. W. 1965.
Performance  and design of border checks on a
sandy soil. Trans. ASAE 8(1): 141-145.
   18. Jensen, M. E., Wright, J. L., and Pratt,
B. J.  1971. Estimating soil moisture depletion
from climate, crop and soil data. Trans.  ASAE
14(5):954-959.
  19.  Thorne, D. W., and Thome,  J. P. 1954.
Changes in composition of irrigated soils as re-
lated to the quality of irrigation waters. Soil Sci.
Soc.  Amer. Proc. 18:92-98.
  20.  United States Salinity Laboratory Staff.
1954. Diagnosis and improvement of saline and
alkali soils. U.S. Dept. Agric. Handbook 60.160
pp.

-------
Water  Quality  Aspects  of Sprinkler Irrigation
                                             by
                       J. KELLER, J. F. ALFARO, and L. G. KING
                                    Utah State  University
                                        Logan, Utah
 ABSTRACT
    Sprinkler irrigation affords a unique means
 for managing irrigated agriculture to improve
 water quality. In order to obtain maximum flexi-
 bility for managing water quality  the sprinkler
 system must be designed to produce a UC in the
 neighborhood of 84 percent, have  a capacity in
 excess of the maximum evapotranspiration de-
 mand, and be automated to allow maximum op-
 erational flexibility.  Both automatic solid-sets
 and center pivot sprinkler systems provide the
 flexibility for optimizing the management of ir-
 rigating with  saline and nutrient rich  waters.
 With good scheduling practices and salt man-
 agement within the  soil profile, the normal wa-
 ter quality hazards associated with  irrigation
 can be greatly reduced. Furthermore, nutrient
 rich waters'can be utilized to augment fertility
 requirements, thus providing an extra value, in
 addition to being  cleaned up for discharge  into
 receiving water supplies. Sprinkler  irrigation af-
 fords a means for  saline  land reclamation as
 well as an auxiliary system for leaching the salt
 build-up which may be associated  with other
 forms of irrigation.
    Sprinkler irrigation systems can be designed
 to afford a high degree of uniformity and opera-
 tional  flexibility.  Under  proper management,
 such systems have the potential of being oper-
 ated to minimize the salt build-up and pollution
problems normally associated with irrigated ag-
riculture, utilize waters of relatively low quality,
and even improve the quality of waters having
high organic and nutrient loadings through the
"living soil filter" concept.
  Sprinkler  systems  can be designed  to  effi-
ciently and uniformly apply water to almost any
type of soil and topographic condition. Further-
more,  areas can be  irrigated  in their natural
state requiring a minimum of disturbance to the
soil surface.  This is especially important where
topsoils are shallow and land leveling or exten-
sive surface preparation operations would dam-
age delicate soil profiles.

        System Design Requirements
  General information and instructions for de-
signing  sprinkler systems can  be found in a
number of books and publications of which
some of the best known are12-21-27. However, the
following special design considerations are  nec-
essary  where  irrigation systems are  to be used
for water quality control management purposes.


System Capacity
  Where it is desirable to manage the sprinkler
system to control the storage and release of salts
from the unsaturated  soil profile, the system ca-
pacity should be 10 to  50 percent greater than
                                             147

-------
148     MANAGING IRRIGATED AGRICULTURE
that necessary to meet the maximum evapotran-
spiration demand.
  It may be advisable to have system capacities
over double that which would normally be re-
quired for irrigation purposes when: a) high sa-
linity  waters are apt to cause leaf burn and it
may be necessary to stop sprinkling during the
hottest and windiest parts of the day, or b) the
living soil filter concept  is employed to reduce
BOD  and nutrient loadings of sewage  effluent
prior to entrance into the groundwater system.

Application Uniformity
  In order to obtain, a high degree of control
over the deep percolation or effluent from the
unsaturated profile underlying a sprinkler sys-
tem high application uniformities are essential.
The most widely used and  best  known of the
sprinkler  application uniformity  coefficients
was presented by Christiansen7 as,
        UC = 100 (1 - -|1)  	  (1)
where d is the deviation of individual observa-
tions from the mean value m; and n is the num-
ber of observations.
  The UC value is a single parameter which has
proven quite useful for predicting the entire dis-
tribution  of water  throughout  the  sprinkled
area. Tables giving the distribution for various
UC values are available.14 19 With UC = 84 per-
cent the wettest 5 percent of the area  will re-
ceive over one-third more water than the aver-
age application depth (conversely  the driest 5
percent of the area will receive less than  two-
thirds of the average application).  This should
be a sufficiently high uniformity for most sprin-
kler systems which are designed for water qual-
ity control purposes.

Timing and Application Depth
  To exercise maximum control over the quality
of deep percolating waters  and  best  utilize
sprinkler systems for waste disposal, maximum
system  operational  flexibility is  essential.  It
must be possible to commence irrigation at will
and have full control over the depth applied dur-
ing each irrigation cycle.

              Types of Systems
  To have full control over the depth and timing
of applications without excessive labor require-
ments automatic sprinkler systems must be em-
ployed. Sprinkler systems which are not auto-
matic are useful for waste disposal and the ap-
plication of less desirable waters on agricultural
lands. However, they do not provide the needed
flexibility for the management of the quality of
effluent from the unsaturated  soil profile under-
lying the sprinkler field.

Manual Systems
  Hand move, wheel move, tow move, boom
sprinklers,  giant sprinklers,  mobile solid-set,
giant traveling sprinklers, and manual valved
solid-set systems  all require sufficient labor in-
put to reduce their flexibility in terms of timing
and  depth of water applied.  It  is  beyond the
scope of this paper to describe each system in
detail, however, sufficient descriptions and pho-
tographs are presented  elsewhere.12 21 27 Most
any of these manual systems can be designed to
provide a very efficient level of irrigation with a
UC  =  84 percent or higher.  Of course, wind
speeds and directions should be taken into  con-
sideration and  large water droplets or high ap-
plication rates should be avoided  where the soil
infiltration rate is low.

Automatic Systems
  Solid-set  sprinkler systems with automatic
valves and  center pivot  systems are the  only
fully automatic sprinkler systems with proven
reliability. The  automatic solid-set system is the
most versatile, however, it may be several times
more expensive per unit area covered than a
center pivot  system. Although it  is possible to
obtain an  ordinary  automatic solid-set system
for less than $500 per acre, a high capacity,  high
uniformity system could  very well cost in the
neighborhood of $1,000 per acre.  Typical center
pivot systems will cost between $150 and $250
per acre, depending upon the  mechanical qual-
ity and versatility desired. (The above cost fig-
ures  do not include storage reservoirs, pumping
plants and the general water distribution and
drainage systems which will be required.)
  A  paper covering the general criteria for the
design, use, and management  of  solid-set agri-
cultural sprinkler systems  is available.19 While
solid-set systems can be designed  for any appli-
cation rate desired, center pivot systems  nor-
mally produce  very high application rates  near

-------
                                                                 SPRINKLER IRRIGATION
                                          149
 the outer end of the moving pipeline. For many
 soils and topographic conditions, the infiltration
 capacity of the soil is exceeded creating runoff
 problems. In order to solve these problems light
 depths of application at frequent intervals and/
 or small circles are required. Under such condi-
 tions, the versatility of the center pivot in terms
 of water quality management is greatly reduced.

                 Soil Moisture
   Figure 1 shows the moisture content variation
 in the unsaturated soil profile or root zone. This
 PAIM AND IP-MGATIOM
                    CVAPOTPAWPI P-ATION
                                 Fltu>  •SATURATION
                                CAPACITY
  Figure 1: Moisture Content Variation in the Un-
  saturated Soil Profile-Root Zone

would be reasonably uniform throughout a ho-
mogeneous  field which  is irrigated by a high
uniformity sprinkler system.

Moisture Flow
  Under sprinkler irrigation, water is uniformily
distributed over the soil  surface. Therefore, the
moisture movement can be depicted as a one-
dimensional flow system  operating in the down-
ward vertical direction. Under furrow irrigation
there is lateral flow between furrows as well as
downward flow. With trickle irrigation  the flow
is three-dimensional radiating from each source
point.  In high water table subsurface irrigation
the flow is one-dimensional but in the upward
direction.
 Moisture Content
   Border or flood irrigation provides a source
 for  one-dimensional  downward flow. However,
 since the surface is inundated, water enters the
 soil at a maximum rate and the surface layer of
 soil becomes essentially saturated. Under sprin-
 kler irrigation, the rate at which water enters
 the  soil  is controlled by the rate at which it is
 applied.
   It has been found18 that the soil moisture con-
 tent in a deep soil profile after long periods of
 sprinkler irrigation could be depicted by:
 B = J(A)J  +00	 (2)
 in  which  6 is  the  soil moisture  content  by
 weight, A is the application rate, and J, j, 00 are
 soil  parameters which  are positive  in sign and
 differ for each soil but are unaffected by consoli-
 dation or compaction. From Equation 2 it is ap-
 parent that as  the application rate increases the
 soil moisture content increases.
   It was found18 that with very low application
 rates the soil profile  could be watered without
 increasing  the  moisture  content  appreciably
 above the so called "field capacity".  It is expen-
 sive  and difficult to achieve very low application
 rates and high  uniformity under  windy condi-
 tions. With automatic  solid-set systems, how-
 ever, it is possible to obtain similar results by
 frequent cycling of higher application rate sys-
 tems.

 Salt  Flow
  Since  salts  move  with  the water, the  salt
 movement or leaching operation under sprinkler
 irrigation is vertically downward  in the unsat-
 urated  soil profile. There are no salt concentra-
 tion  areas left  behind  and the salt movement
 within  the unsaturated  profile can be carefully
 controlled by varying the depth and timing of
 water applications. The  depth to which the pro-
 file is leached depends upon the quantity of wa-
 ter applied.  Studies have been conducted which
 demonstrated that the leaching efficiency is im-
 proved by decreasing  the application rate.20
  In many  instances reduction  of  soil water
evaporation and, therefore, upward movement
 of water will result in a more efficient leaching
 of the top portion of the soil profile. By combin-
ing surface mulches, to  reduce evaporation and
 upward water movement, with periodic sprin-

-------
150
MANAGING IRRIGATED AGRICULTURE
kling, more effective leaching can be attained as
compared to the conventional leaching practice
of flooding the ground surface.6
  The changes  in the effluent  salt concentra-
tions resulting from leaching a uniform soil pro-
file may be similar to those changes represented
by  Figure 2.1 The maximum concentration of
   16 r
 s
 z
 x
 o
 o
                              16
                                    20
                                          24
           VOLUME OF EFFLUENT, Odw (cm5/cm2)
 Figure 2: Leachate  Concentration  Curves for
 Uniform  Sandy Loam Soil Profiles of Various
 Depths, Ds1

the drainage water will  be equal to the maxi-
mum salt  concentration within the soil profile at
the beginning of the salt removal process. It will
gradually  decrease as the volume of the drain-
age  water increases. A  stratified soil  profile
where two  different salinity strata  exist  will
yield a leaching curve similar to any of those
shown in  Figure 3.' The peak value of salt con-
8
ti
S
   OS            IO          15
          VOLUME OF EFFLUENT, D*> (Cm3/en?)
 Figure 3: Leachate  Concentration Curves  Re-
 sulting from Sprinkling of Stratified Soil Profiles
 of Various Depths, Ds, of Salinized Sandy Loam
 Soil over a Non-Salinized Sand1

centration will depend upon the maximum con-
centration in the profile  before water applica-
tion, its location with respect to the soil depth,
and  initial  water  content.  This indicates that
                                          higher peak values may be obtained for both
                                          uniform and stratified soil profiles at lower wa-
                                          ter contents.
                                             Under sprinkler irrigation the entire surface
                                          layer of soil is leached by each irrigation.  If just
                                          enough water were applied by each irrigation to
                                          supply the moisture deficit, salts would build-up
                                          in the lower portion of the root zone. With care-
                                          ful management, the salinity and moisture con-
                                          tent at various levels in the root zone can be con-
                                          trolled and additional leaching water applied
                                          when it becomes desirable to discharge the salts
                                          into  the water table. Details of the management
                                          schemes necessary for such an operation will be
                                          discussed later.

                                                      Irrigation Scheduling
                                            Scheduling of irrigation can be based on ob-
                                          servations of the soil  moisture status, estimates
                                          of moisture  used, or a combination of the  two.27
                                          These methods of scheduling may not necessar-
                                          ily deal with the quality of the water, however,
                                          water quality aspects  can  be built  in to the
                                          scheduling program.
                                            A  recent innovation for irrigation scheduling
                                          using climate-crop-soil  data,  computers  to fa-
                                          cilitate tedious computations, and field observa-
                                          tions by  experienced personnel17 is being  met
                                          with widespread popularity.15  16 Scheduling ser-
                                          vices of this type have the  potential of increas-
                                          ing the skills available for managing the  water
                                          quality aspects under  sprinkler irrigation. The
                                          general framework and the  use of  computer
                                          scheduling in Idaho and Arizona  will be  dis-
                                          cussed in  this conference.  The  necessary com-
                                          puter program and operational details are avail-
                                          able from the USDA Agricultural Research Ser-
                                          vice.

                                                    Managing Saline Waters
                                            Special sprinkler irrigation management tech-
                                          niques must be applied where salinity is of a ma-
                                          jor concern. Since all water contains some salt
                                          which is  left behind by the evapotranspiration
                                          process, the salinity  of percolating  waters is
                                          greater than the salinity of the irrigation water.
                                          Although this is a problem inherent in any irri-
                                          gation system,  the  discharge and  quality of
                                          these saline  waters from the unsaturated profile
                                          can be controlled by careful management  of the
                                          irrigation systems.

-------
                                                               SPRINKLER IRRIGATION
                                         151
  When high salinity irrigation waters must be
utilized, special attention is required for select-
ing the crops to be grown and the proper soil
management practices. Various crops have dif-
ferent tolerances or abilities to produce within a
saline environment.  It is beyond the scope of
this paper to discuss the salt  tolerance levels of
various  crops,  however, a brief review of the
subject is available which cites much of the lit-
erature  where  specific crop-soil salinity toler-
ance information may be obtained.23

Leaf Burn
  Salinity,  or presence of soluble  salts in con-
centrations above  the recommended  levels for
plant growth may result in absorption of salts
by plant roots and subsequent accumulation in
the plant  leaves causing leaf burn. Moreover,
some plants can absorb nutrients and other ions
directly through their leaves. If the foliage of sa-
line sensitive crops is wetted by sprinkler irriga-
tion, it may result in harmful accumulation of
toxic ions  and characteristic  leaf  burn. Symp-
toms  of toxicity in leaves of sensitive plants are
similar  when  the  toxic ions are  absorbed by
either roots or foliage3 but the rate at which they
accumulate may differ greatly. The difference in
rates of root and foliar absorption is further em-
phasized by  the fact that during the growing
season root absorption is continuous while foliar
absorption during sprinkling occurs only about
one percent of the total time.10
  In some cases low saline  waters that would
not have harmful effects  if applied directly to
the soil may affect plant growth when applied
by sprinkler irrigation. The rate at which foliage
injury develops varies  with  species, salt  type
and concentration, and time and mode of water
application. •
  Direct foliage injury as a result of sprinkling
with saline waters  is not likely to occur in vege-
tables and forage crops if the crops can tolerate
similar  salt concentration in the  root media.
However, gradual accumulation of salts which
are not removed by an excess of water may in
time injure the leaves.10
  Citrus is quite sensitive  to damage by  foliar
absorption of sodium and  chloride. Water con-
taining as little as 70 to 100 ppm sodium or chlo-
ride may be harmful if applied by sprinkler irri-
gation.  In  orchards  sprinkled with waters con-
taining less than 40 ppm of solium or chloride
no leaf damage was observed.13 However, these
ions may accumulate in the leaves more rapidly
by direct foliar absorption than by root uptake
even from soil solutions 10 times more concen-
trated.2 Therefore, some fruit crops absorb salt
through the leaves 100 times faster than through
the roots. Apricot and almond trees accumulate
these ions most rapidly, orange and plum are in-
termediate, while avocado, a sensitive fruit  to
foliar concentrations of sodium and chloride,
has a remarkably low rate of absorption through
its leaves.10
   Strawberries are very sensitive to soil salinity
 and furrow irrigation may cause salt accumula-
 tion as a result of unequal salt concentrations in
 the root zone. However, indications are that fo-
 liar absorption under sprinkler irrigation should
 be relatively unimportant since much more chlo-
 ride would be absorbed  from the root  media
 containing equivalent  levels  of chloride than
 from the foliage.9
   Sugar cane plants were  not adversely affected
 by sprinkler irrigation using water containing up
 to 15 milliequivalent per liter, or about 500 ppm
 of chloride salts.4
   Cotton sprinkled with saline water containing
 an average salt content over 3,000 ppm, at a rate
 of 14 inch per hour resulted in leaf  burn when
 the plants were less than  12 inches tall. Higher
 sodium  content  of washed cotton  leaves oc-
 curred when plants were sprinkled during day
 time than a night. With the same quality of wa-
 ter, night sprinkled and furrow irrigated cotton
 produced comparable yields. The day sprinkled
 cotton, however, produced only 68  percent as
 much of the furrow irrigated short staple cotton
 and only 43 percent as much of the furrow irri-
 gated long staple cotton.5
   Earlier foliar injury may be expected in sensi-
 tive crops when sprinkled with waters contain-
 ing CaCl2 or NaCl than when sprinkled with
 waters containing NaSO*. Sodium may be more
 rapidly  accumulated  from  NaCl  than from
 NaSo4 solutions because of the difference in sol-
 ubility.  Chloride accumulations  may  be  ex-
 pected to be equal from waters containing either
 NaCl or CaCl2.
   Saline  waters  applied to sensitive crops by
 sprinkler irrigation during the hottest part of the
 day should be avoided since the foliar uptake of

-------
152     MANAGING IRRIGATED AGRICULTURE
salts may be twice the amount absorbed during
evening irrigations.10 In general more severe in-
jury may be expected by sprinkling with saline
waters in hot, dry climates than in cool, humid
regions.
  More  sodium and  chloride may accumulate
under intermittent conditions than when sprin-
kled continuously. Intermittent sprinkling, such
as for crop cooling, allows for evaporation with
a subsequent increase in concentration of salt
from the water film left  on the leaves. On warm
days with low  humidity rapid evaporation  re-
sults.  The first salts to precipitate  and be de-
posited on the  leaves  are- the relatively insouble
salts such as CaCOa, and CaSCX. It is theo-
rized13 that sodium and chloride ions remain in
solutions until most  of the  water  has evapo-
rated. Thus, relatively  high  concentrations of
these ions in solution  will result, accounting for
the fact that only sodium and chloride appear to
be absorbed by the plant foliage.
  To  minimize foliar absorption of salts and,
therefore, leaf burn, water should be applied by
sprinkler  irrigation at low angles below the fo-
liage of tall crops such as trees. Cool and cloudy
day or night irrigations should be preferred to
maintain  the foliage  continuously  wet during
sprinkling, thus minimizing leaf burn hazards.

Soil Profile
  Managing sprinkler irrigation systems to con-
trol the movement of salt through the soil profile
is a practical possibility. Salinity control in the
soil profile is an important part  of controlling
the quality  of  irrigation return flow,  i.e., that
part of the  water  diverted for  irrigation which
eventually returns  to the stream or groundwater
body. Details of models for describing the water
and salt  movement in the root zone are being
presented at this conference.22  These models
provide the design and operation  criteria for the
sprinkler system.  These proposed  models for
managing irrigation are  based on the concept of
temporarily storing salts in the soil profile and
leaching  only when necessary to prevent the salt
from  becoming too concentrated in  the root
zone. This  concept requires a  fairly accurate
knowledge and control  of the timing of irriga-
tions  and depth of water applied  during  any
given irrigation. In any given irrigation, the per-
centage of applied water  required for leaching
 may be as high as 50 percent. This extra water
 can be obtained by increasing either the  rate of
 application, the duration of irrigation, or a com-
 bination of both.

 Profile Effluent
   The models of the movement of water and
 salt through the soil profile predict the quantity
 and quality of water leaving the bottom of the
 root zone (deep percolation) as a  function of
 time during the irrigation season.22  In addition,
 through  the mass balance the amount of water
 and salt  existing in the root zone at any  time is
 also known.
   With all other conditions the same, varying
 the interval  between  irrigations changes  the
 time at which the peak salt  load in the profile ef-
 fluent occurs. Thus, particularly in areas with an
 extensive man-made  drainage system wherein
 the residence  time  for  the effluent  is short, a
 wide range of possibilities for quality control ex-
 ists. For control in entire valleys or irrigation
 districts,  possible alternative  modes  of  opera-
 tion might be:  (a)  schedule  the irrigations in-
 volving leaching of salts from the profile  on dif-
 ferent farms at different times so as to smooth
.the peaks of salt in the drainage effluent and
 provide a more nearly constant quality of efflu-
 ent during the season;  or  (b) attempt to peak
 the salt  removal from all  farms at nearly the
 same time and divert this high salinity water to
 an evaporation basin or other suitable salt sink.
   Under  certain  conditions,  however,  deep
 leaching  may not be needed every year. Again,
 it should be  emphasized  that controlling  the
 depth and timing of irrigations is necessary for a
 successful profile effluent  quality control  pro-
 gram.

           Managing Waste Waters
   Irrigation in general and sprinkler irrigation
 in particular provides an effective  method for
 utilizing  and/or improving the quality of waste
 waters. Through the "living soil filter" the bio-
 logical oxygen demand (BOD), as  well as the
 concentrations of nitrogen, potassium and phos-
 phorous  can be reduced and the suspended sol-
 ids and bacteria removed from the waste water.
 Excess irrigation will result  in recharge to the
 groundwater of  a  portion of the  effluent. If
 properly managed,  the water leaving the "living

-------
                                                                SPRINKLER IRRIGATION
                                          153
 soil filter" will be  equivalent to  or  higher  in
 quality than that produced by a third stage  or
 tertiary treatment facility.
   It has been estimated that over 400 cities  in
 the United States are currently using crop irri-
 gation as a means of disposing of domestic and
 industrial  waste waters.11 The proceedings  of
 the Municipal  Sewage  Effluent for  Irrigation
 Symposium presents many nationwide aspects
 of this subject which attest to the validity of re-
 turning domestic liquid wastes to the land.28

 Nutrient Removal
    It  has been estimated that the fertility value
 of domestic effluent in terms of nitrogen, phos-
 phorous and potassium is in the neighborhood
 of $18.00  per acre foot.11 However, this figure
 can vary significantly. In order to effect the re-
 moval of nitrogen the waters must be applied to
 vigorously growing crops. The primary purpose
 of the crops in the renovation cycle is to remove
 the nitrogen added to the irrigation water. Once
 the nutrients are incorporated into the crops it is
 imperative to remove these crops from the irri-
 gated  area and thus remove the  incorporated
 nutrients.
   It has been found that phosphorous can be
 readily removed  by soils and crops  and the
 choice of crops can vary with the typical pat-
 terns of land use in the area. Although the crops
 only remove a portion of the phosphorous ap-
 plied,  many soils  have a high  capacity for ab-
 sorbing the remaining phosphorous.
   The nitrates contained in the effluent are not
 as  readily removed as  the phosphorous. The
 overall design  must ensure that these nitrates
 are adequately removed by crops and/ or soil mi-
 crobes so  that,the  renovated  water  does  not
 cause undue degredation of the groundwater.
   In high exchange capacity soils, potassium  is
 normally exchanged onto the soil complex and
 is  thus removed from the waste water.

 BOD
  Although a vigorous crop is necessary for the
removal of nutrients, the typical soil microflora
will greatly reduce the BOD in the waste water
as it trickles through  the system.  Only rarely
will the hydrologic environment not  renovate
waste water to the equivalent of secondary or
 biological treatment. Terrestrial disposal  has
 been used in lieu of secondary treatment.

 Special Considerations
   Sprinkler  irrigation has certain  unique ad-
 vantages  for dealing with waste water disposal.
 By sprinkling the soil at a low application rate,
 the soil moisture content can be maintained at a
 low level, as mentioned earlier, thus maintain-
 ing adequate soil  aeration for optimum plant
 growth and microflora activity. Some of the spe-
 cial design features required  for effluent  dis-
 posal  through  sprinkler irrigation  have  been
 pointed out.8 24 2S In essence,  for optimum  per-
 formance  it is  important to locate the  disposal
 area on the better quality lands where the water
 table depth is over 5 feet.  Storage facilities and
 sprinkle field design depend on effluent quality,
 volume, and climatic conditions. Prior to enter-
 ing  the   sprinkler  system,  domestic   sewage
 should be treated through a  secondary treat-
 ment facility and chlorinated to reduce potential
 health hazards.
   A number of legal  issues are involved in the
 application of  waste  waters through sprinkler
 systems.24  Under  wind conditions   the  spray
 from sprinkler systems is apt to drift a consider-
 able distance from the sprinkled area. One of
 the unanswered  legal and  health questions to
 date is the amount of border area required in
 order to eliminate potential health and environ-
 mental hazards from air-borne bacteria, virus
 and odors.
   Sprinkler irrigation has the  potential  of play-
 ing a unique role in the disposal of hot water
 from  large thermal power generating stations.
 The temperature of the spray reaches  the  wet
 bulb temperature regardless of the temperature
 of the irrigation water.27 Therefore, it  appears
 possible and practical to sprinkle with very hot
 water without casuing crop damage.

       Leaching with Sprinkler Systems
  Since sprinkler irrigation effectively produces
a one-dimensional soil moisture flow system as
described  earlier, it is uniquely adaptable  for
reclamation and leaching  purposes. Further-
more, by utilizing low application rates the re-
moval  of salts per unit of water applied can be
maximized. The system can be used for the rec-
lamation of new lands as well as for leaching the

-------
154    MANAGING IRRIGATED AGRICULTURE
salts which may build-up under other forms of
irrigation.


Reclamation
  In many cases, the reclamation of new lands
can most efficiently be achieved by the use of
sprinkler irrigation. Under sprinkler irrigation it
would not be necessary to remove the existing
vegetation and  disturb a delicate  soil  surface
during the initial leaching phases. Furthermore,
through  controlled  low application  rates the
amount of salts  removed per unit water applied
can be maximized to minimize the water table
build-up  associated with reclamation. After the
surface profile has been partially  leached the
native vegetation can be replaced by salt toler-
ant, densely  growing crops which may further
aid the leaching process. By applying more wa-
ter than the evapotranspiration demands, crops
can be grown which will produce some return
while the remaining soil profile is being reno-
vated.
Drip Irrigation
  With trickle or drip irrigation systems, water
is applied from point sources and redistributed
in the soil profile by capillarity and gravity. Drip
irrigation offers the possibility of applying water
frequently at short intervals to maintain the root
zone moisture at an optimum level for maximum
production.
  By irrigating frequently, the soil water within
the root zone may be maintained at high levels
to provide an adequate  invironment for plant
growth even if saline waters are used. With any
irrigation water accumulation of salts at the pe-
riphery of the wetted portion of the profile is to
be expected. Such a  condition will develop in
time depending upon the quality  of the water
used for irrigation. Where natural rainfall is not
sufficient to move these salts, some auxiliary
leaching is required. By periodic applications of
water through a sprinkler system the necessary
leaching can be achieved.  The  capacity of an
auxiliary  sprinkler system  used  for periodic
leaching could be quite small.  Under  normal
conditions it might only be necessary to apply a
single deep application of water once every sev-
eral years.
Subsurface Irrigation
  High water table and plastic seep line or emit-
ter type subsurface irrigation systems eventually
result in salt build-up in the surface soil. Where
natural rainfall is insufficient,  periodic leaching
with uniform applications of water to the  sur-
face is necessary as with drip systems.

Furrow Irrigation
  While  furrow irrigation  does not uniformly
apply water to the soil surface, under normal op-
erations,  the minor salt build-up in the beds be-
tween furrow is leached from time to time as the
fields are cultivated  and new furrows installed.
Of  unique  importance,  however,  is  the  salt
build-up  in the  beds during the seed germina-
tion period. Often seeds are planted just above
the water line so the upward movement of water
from the furrows to  the bed leaches the zone in
which  the  seeds are planted. This is  necessary
since many seeds and seedlings have a low salt
tolerance although  the  crop may have a high
salt tolerance once in a vigorous growth stage.
  The accurate construction  of furrows  and
planting of seed just above  the water line is dif-
ficult and  sometimes impossible. In order to
eliminate this problem,  sprinkler irrigation  is
often utilized for  the germination of delicate
seeds.  Once germinated, the crop can then be
furrow irrigated.

REFERENCES
   1. Alfaro, J.  F. 1971. Application of a Physi-
cal  Model Theory to Predict Salt Displacement
in Soils. Soil Science 112:364-372.
   2. Bernstein, L.  1967. Quantitative Assess-
ment of Irrigation Water Quality. Water Quality
Criteria, ASTM STP 416:51-65.
   3. Bernstein, L.,  and  H. E. Hayward. 1958.
The  Physiology of  Salt Tolerance.  Am.  Rev.
Plant Physiology 9:25-46.
   4. Bernstein L., R. A. Clark,  L. E. Francois,
and M. D.  Derderian. 1966. Salt Tolerance of N.
Co. Varieties of Sugar Cane. II. Effects of Soil
Salinity and Sprinkling on Chemical Composi-
tion. Agron. J. 58:503-507.
   5. Busch, C.  D., and F. Turner. 1965. Sprin-
kling Cotton with Saline Water. Progressive Ag-
riculture  in Arizona  17:27-28.

-------
                                                              SPRINKLER IRRIGATION
                                         155
   6. Carter, D. L., and C. D. Fanning. 1964.
Combining Surface and Periodic Water Appli-
cations for Reclaiming Soils. Soil Science Soc.
Am. Proc. 28:564-567.

   7. Christiansen, J. E.  1942. Irrigation by
Sprinkling. Bulletin 670, University of Califor-
nia, Berkeley, California.

   8. Cochran, R.  A. 1971.  Disposing of Hu-
man Sewage Effluent through Spray Fields and
Living  Soil  Filters,  presented at  Rain  Bird
Sprinkler Sales Meeting, Glendora, California.

   9. Ehlig,  C. F.  1961.  Salt Tolerance  of
Strawberries Under Sprinkler Irrigation. Proc.
Am. Soc. Hort. Sci. 77:376-379.

  10. Ehlig,  C. F.,  and  L.  Bernstein.  1959.
Foliar absorption of Sodium and Chloride as a
Factor in Sprinkler Irrigation.  Proc. Am. Soc.
Hort. Sci. 74:661-670.

  11. Eier,  D. D., A. T.  Wallace, and  R. E.
Williams. 1971. Irrigation and Fertilization with
Waste Water. Compost Sci. 12:26-29.
  12. Fry, A. W., and A. S. Gray. 1971. Sprin-
kler Irrigation Handbook.  Rain Bird Sprinkler
Mfg. Corp., Glendora, California.
  13. Harding, R. B.,  M. P. Miller,  and M.
Fireman.  1958. Absorption of Salts by Citrus
Leaves During Sprinkling with Water  Suitable
for Surface Irrigation. Proc. Am. Soc. Hort. Sci.
Sci. 71:248-256.
  14. Hart,  W. E. and W. N. Reynolds.  1965.
Analytical Design of Sprinkler Systems. Trans-
actions ASAE 8:83-85 and 89.
  15. Heermann, D. F. and M. E. Jensen.  1970.
Adapting Meteorological Approaches in Irriga-
tion Scheduling.  ASAE  National Irrigation
Sympsium Proceedings, Lincoln, Nebraska, pp.
00-1-10.
  16. Jensen, M. E. and D. F. Heermann,  1970.
Meteorological Approaches to Irrigation Sched-
uling. ASAE  National Irrigation Symposium
Proceedings, Lincoln, Nebraska, pp. NN-1-10.
  17. Jensen, M. E., D. C. N. Robb, and C. E.
Franzoy. 1970. Scheduling Irrigations using Cli-
mate-Crop-Soil Data. ASCE Journal of Irriga-
tion and Drainage Division, 96 (IR 1): 25-38.
  18.  Keller J. 1970.  Control of Soil Moisture
During  Sprinkler Irrigation. Trans. ASAE 13:
885-890.
  19.  Keller, J. 1970.  Design, Use and Manage-
ment of Solid Set Systems. ASAE National Irri-
gation Symposium  Proceedings. Lincoln, Ne-
braska,  pp. AA-1-10.
  20.  Keller, J.  and J. F. Alfaro.  1966. Effect
of Water Application Rate  on Leaching. Soil
Sci. 102:107-114.
  21.  Keller, J.,  et al.  1967. Ames Irrigation
Handbook. W. R. Ames Co., Milpitas, Califor-
nia.
  22.  King., L. G., R. J. Hanks, M. N. Nimah,
S. C.  Gupta, and R.  B. Backus. 1972. Model-
ing  Subsurface Return Flows in Ashley Valley.
Proceedings National  Conference on Managing
Irrigated Agriculture to Improve Water Quality,
May 16-18, 1972, Grand Junction, Colorado.
  23.  Marsh,  A. W.  1970.   Irrigation  Water
Quality  in Relation to Irrigation  Scheduling.
ASAE National Irrigation Symposium Proceed-
ings, Lincoln, Nebraska, pp. PP-1-8.
  24.  Meyers, E. A. 1970. Sprinkler Irrigation
for Liquid Waste Disposal. ASAE National Irri-
gation Symposium  Proceedings, Lincoln, Ne-
braska,  pp. y-1-10.
  25.  Meyers, E. A. and T. C.  Williams. 1970.
A Decade of Stabilization Lagoons in Michigan
with Irrigation as Ultimate Disposal of Effluent.
2nd International Symposium for Waste Treat-
ment Lagoons, pp. 89-92.
  26.  Pair, C. H., J. L. Wright and M. E. Jen-
sen. 1969. Sprinkler Irrigation Spray Tempera-
tures.  Trans. ASAE 12:317.
  27.  Pair, C. H., et al. 1969. Sprinkler Irriga-
tion. Sprinkeler Irrigation Association, Washing-
ton, D.  C.
  28.  Wilson,  C. W.  and  F. E. Beckett. 1968.
Municipal Sewage Effluent for  Irrigation. Pro-
ceedings of Symposium at Louisiana Polytech-
nic  Institute, Baton Rouge, Louisiana,  169 pp.
July.

-------
       Subirrigation  Studies in  the  High  and
                      Rolling  Plains  of Texas

                    C. W. WENDT, A. B. ONKEN and O. C. WILKE
                     Texas A&M University Agricultural Research and
                              Extension  Center at Lubbock
                                     Lubbock, Texas
ABSTRACT
  Results from the first year of a study involv-
ing the influence of subirrigation on soil water
conditions and ion concentration of soil extracts
are presented.
  Soil layering resulted in high soil water po-
tentials and retention.  With this condition sub-
irrigation can maintain high soil water poten-
tials for plant growth and yet leave water reten-
tion capacity for rainfall.
  Early in  the growing season, breakdown of
organic  residue  and  subsequent  nitrification
resulted in significant increases in nitrate-nitro-
gen  of- soil  water  extracts.  Nitrate-nitrogen
concentrations were lower in the subirrigated
plots than in the furrow- and sprinkler-irrigated
plots. Nitrogen was efficiently applied in incre-
ments through the subirrigation system.
  Calcium  and chloride  concentrations  were
lower in subirrigated plots than in sprinkler-
or furrow-irrigated  plots.   Concentrations  of
phosphate, potassium,  sodium and ammonium
were not influenced by  method of irrigation.
The quality of water applied influenced sulfate
concentrations and electrical conductivity more
than method of irrigation.
INTRODUCTION
  Irrigated  agriculture is the major consumer
of water in  the United States, With the increas-
ing  pressure  on the water  resources  of  the
nation for municipal and industrial  as well as
agricultural  needs, there is increasing concern
about the influence  of irrigated agriculture on
the quality of the nation's water resources.
  Water used in irrigated agriculture may be
stored in the soil, evaporated from the soil, con-
sumed by the crop,  or may  leave the irrigated
area as surface runoff or leachate which perco-
lates through the root zone. Water that runs off
and percolates through the root  zone may re-
turn to the source of the supply. Two major resi-
dues  in this water which create  water quality
problems  are dissolved mineral salts and plant
nutrients. Mineral salts tend to increase in  the
return flows due to the discharge of additional
minerals into the return flow and  concentration
of the minerals due to abstraction and consump-
tive use. Law1 notes in his  summary that  20-
fold  increases in mineral content have been
noted on the Colorado and Rio Grande Rivers.
  Intensification  of  crop  production  has  re-
sulted in increased applications of nitrogen fer-
                                           157

-------
158
MANAGING IRRIGATED AGRICULTURE
tilizer. In any area where water application is
applied in excess of that needed for crop pro-
duction, the potential of leaching of nitrogen
from fertilizers and nitrification exists.
  If irrigation water  could be more efficiently
utilized, less mineral salts would be applied and
less water would be available to return salts and
plant nutrients to the source through irrigation
                                          return flows, and the quality of waters receiving
                                          these return flows could be enhanced.
                                            The quality  of the waters commonly used in
                                          the High Plains and Rolling Plains of Texas is
                                          relatively good compared to that in other parts
                                          of the U.S. Commonly this water contains less
                                          than  1500 ppm total salts with less than 50 per-
                                          cent sodium salts (Table 1). However,  in certain
                                           TABLE 1

                         Analyses of water from typical wells in the High
                                  and Rolling Plains of Texas
                                                            High
                                                            Plains
                                                                   Rolling
                                                                    Plains
          pH

          Electrical Conductance (ECxlO6)

          Ion concentration (ppm or mg/1)
            Total salts
            Calcium
            Magnesium
            Potassium
            Sodium
            Carbonate
            Bicarbonate
            Sulfate
            Chloride
            Nitrate-Nitrogen

          Sodium adsorption ratio (SAR)

          Sodium percentage (SSP)
                                                      7.9

                                                    1101


                                                     627
                                                      68
                                                      47
                                                      7
                                                      36
                                                      0
                                                     309
                                                      50
                                                     134
                                                      2

                                                       .84

                                                      17.6
   7.5

2214


1360
  92
  67
   4
 217
   0
 425
 262
 244
  11

   4.17

  47.7
parts of the Rolling Plains a potential nitrate-
nitrogen pollution problem exists. The nitrate-
nitrogen content of 62 water samples  collected
from the  Seymour formation in 1962 varied
from 5-41  ppm with 39 exceeding the recom-
mended Department of Health limit of 10 ppm.
Nine towns use water from this formation and
none have an approved water supply due to the
presence of nitrate. Analyses made in  1968 and
1969 indicate that the concentration of nitrate
is continuing to increase.
  A project was initiated in 1970 to determine
if irrigated agriculture may  be contributing to
the  increasing  nitrate-nitrogen content of  the
ground water  and to evaluate the influence of
                                          irrigation  and fertilization techniques  on the
                                          potential pollution  of the underground water
                                          supply.  Irrigation systems  being  evaluated in-
                                          clude furrow, sprinkler, subirrigation and auto-
                                          mated subirrigation. Nitrogen fertilizer sources,
                                          various  moisture levels, and methods of fertil-
                                          izer applications are being examined.
                                            This paper deals  with  a portion of the data
                                          obtained from the  study concerned primarily
                                          with subirrigation.

                                                     Methods and Materials

                                            In early work at Lubbock, the major problem
                                          hindering the development of subirrigation sys-

-------
                                                               SUBIRRIGATION STUDIES
                                                          159
 terns was the clogging of orifices with sand and
 closing of punched orifices due to plastic flow.
 These problems  were reduced by  the develop-
 ment  of molded orifices4 which were installed
 in the subirrigation systems used for this study.
 The orifices have a diameter of .022 in. and are
 spaced 34  in. apart on the  0.5-in.  dia.  poly-
 ethylene laterals. The laterals are  40 in.  apart
 directly beneath  each crop row. The plots are
 220 ft long with header lines on each end. Flow
                rate is controlled by Dole valves. Water enter-
                ing the system was filtered with  100  or 300 v
                cartridge filters. The systems are operated for a
                short period at least once per month to  clean
                the orifices. These  systems have been  used for
                one year. No  clogging problems have  yet been
                encountered.
                  The field treatments at the Rolling Plains site
                are shown in Table  2. Each  plot is equipped
                with  soil water extraction  tubes at  various
                                           TABLE 2

                        Parameters being  evaluated at the Rolling Plains
                                   site (Knox County, Texas)
           Methods of Irrigation      Moisture Treatments    Fertilizer Sources
          Furrow
          Sprinkler
          Manually operated sub-
          irrigation
          Automated subirriga-
          tion
High moisture level
based on stage of
growth

High moisture level
based on tensiometers

Moderate moisture
level based on tensio-
meters

Constant high moisture
level based on tensio-
meters
                                                         NH3
     with "N-Serve"


Sulfur coated urea
Nitrogen solution
Control
depths from 0.5 to 30.0 ft and access tubes to
30.0 ft for making neutron probe measurements
at 2 locations in each site. Tensiometers were at
various depths from 0.5 to 10.0 ft at 2 locations
in each site. Cores for  physical and chemical
data were taken at 28 locations on the site at the
beginning «and  end of the season. Particle size
analyses  were made  using the Bouyoucos  hy-
drometer method.  Moisture retention data were
obtained  from pressure  plates.  Soil  water
samples were procured during the growing sea-
son  with  the  extraction  system  and analyzed
with an AutoAnalyzer for calcium (Ca++), sodi-
um (Na+), nitrite-N (NCQ, nitrate (NO^), am-
monium  (NHj), chloride (Cl~),  sulfate  (SOJ),
potassium  (K+), and phosphate  (POJ). Since
the nitrite-nitrogen in the soil  water extracts
was  present in  low concentrations and  in only
                a small number of samples, the analysis for the
                ion was discontinued.  The indicator crop was
                sweet corn and yield data were obtained at the
                end of the season.
                  Two plots of a prototype automated subirri-
                gation system  were installed at Lubbock  to
                eliminate automation problems prior to installa-
                tion  of an expanded system  in  Knox County.
                Each  plot was equipped with 4 switching ten-
                siometers in a circuit, whereby, if any two of the
                tensiometers increased to a preset tension level,
                a solenoid would open and apply water until the
                tension decreased on  1 tensiometer. An event
                recorder recorded the time water was applied.
                Work with the prototype has resulted in a sim-
                pler  and  less expensive automated system for
                Knox County. Soil water content and potential
                measurements  were  made at  selected  points

-------
160
MANAGING IRRIGATED AGRICULTURE
around orifices to determine  the  flow pattern
of the irrigation water.
            Results and Discussion

Soil Water Potential and Content
(Rolling Plains)
   The Knox  County  site, which had not pre-
viously been  irrigated, is classified as a  Miles
                                          loamy fine sand. As can be seen in Fig. 1, the
                                          clay content of the surface 12 in. of soil was
                                          fairly uniform  at 4 to 6 percent, except where
                                          some land leveling was necessary for furrow ir-
                                          rigation.  Below this the soil  contained alter-
                                          nating layers of higher and lower clay content.
                                            This layering effect influenced  the soil water
                                          potentials at field capacity. The soil water po-
                                          tential remained greater than -10 cb below 2 ft
                                          for most of the growing  season. The highest po-
 Distance, ft  250
                 500
750
1000
                                               1250
                               1500
1750
2000
Figure 1: Percent clay between the surface and the water table. (Miles loamy fine sand, Rolling Plains Field
Site)
tentials occurred in or above the layers of high-
est  clay content in  all irrigation systems. The
subirrigation system had an additional zone of
high  potential  around the  subirrigation pipe
(Fig. 2). As would be expected, increased water
retention  was associated  with the high poten-
tials. Pressure plate  and bulk density data indi-
cate that the soils retain 12 to 20 percent more
water by volume at -10 cb than -33 cb.
  Figure  3  shows the  relative soil water con-
tents  on a  subirrigated and sprinkler-irrigated
plot on various dates during the growing season.
The soil  water content is  related to the clay
content of the various layers.
  Most of the change  in water content in the
soil profiles between the  surface and  10 ft oc-
curred between May 10 and July 8, prior to crop
emergence.  Changes between July 8 and Sept.
                                          3 were small. For this reason,  as will be seen
                                          later, the highest concentrations of ions in the
                                          soil solution were generally located in the sur-
                                          face 5 to 6 ft and did not shift to lower depths
                                          of the soil profile.
                                            Effects  of layering in  soils have been pre-
                                          viously  reported  by Miller2 and  Miller and
                                          Bunger3.  They  point out that before water can
                                          move from  a finer-textured  layer to  a coarse-
                                          textured   layer, the  potential  must  be  high
                                          enough to allow water to enter the large pores of
                                          the coarse layer. Such effects are noted in these
                                          soils. The soils they described had only 1 layer
                                          while this soil had 3 to 4 porous layers.
                                            This layering effect could  be a major factor
                                          in maintaining  the water quality of  irrigation
                                          return flows. This  soil  retains water  at poten-
                                          tials of -10  cb and greater.  Many crops grow

-------
    0|-   Sprinkler Irrigated    -46 cb

    1
 4


 5


 6


 7


 8


 9


10
                      - July 19
                     Q - July 26
                  Furrow  Irrigated
                                                              SUBIRRIGATION STUDIES     161

                                                                            Subirrigated  -51 cb
                                                         - July 12
                                                        B - July 20
                                                                 A- July 8
                                                                 a- July 15
              -5   -10   -15  -20         0    -5   -10  -15   -20   -25          0    -5   -10   -15
             Potential,  cb                     Potential,  cb                   Potential,  cb
Figure 2: Soil water  potential in soil profile of different irrigation systems between water additions.  (Mile:
loamy fine sand, Rolling Plains Field Site)

                                                    well
Subirrigated
                                      Percent
                                       Clay
    30
                                           26

                                           ]]
  fi
  s
    20
            10       20      30       40
             Soil Water Content (percent by volume)

                           Sprinkler
                           Irrigated
                            a - May 10
                            A - July 8
                            o - Sept. 3
                    Percent
                     Clay
                                            5
                                           23
                                           13
                                           21
            10       20      30       40      50
             Soil Water Content (percent by volume)
  Figure 3; Maximum and minimum clay contents
  and changes in soil water content of sprinkler and
  subirrigated plots during the 1971 growing sea-
  son. (Miles loamy fine sand, Rolling Plains Field
  Site)
     at  lower  potentials.  Sufficient  irrigatior
water  could be added to such crops to  ade-
quately supply  their needs without  utilizing all
of the soil's water-retention capacity. Retention
capacity would still  be available  for a large
amount of rainfall so that a minimum of leach-
ing would occur with proper management.

Soil Water Potential (High Plains)
  The prototype of an automated subirrigation
system was installed in an Amarillo fine sandy
loam at Lubbock, Texas.
  Although  the subirrigation  system  was in-
stalled in March, the  automation  was not com-
pleted  until July when the sweet corn was at the
tassel stage of growth. Four switching tensiom-
eter in each of  two plots were  located 6 in. to
the side of orifices and were set at potentials of
-30 cb and -60 cb. The amount of water applied
through  the system to the 2 plots is shown in
Table 3. Between July 12 and July 22 the sys-
tem was inoperable  due  to the  necessity of
making changes in automation and Dole  valves.
  The -30 cb plot was relatively dry at the  time
the  system  was installed.  During  the  six-day
period from July 7 and July 12, more than 6 in.
of  water  were  added to the profile (Table 3).

-------
162     MANAGING IRRIGATED AGRICULTURE

                                           TABLE 3
                         Inches of irrigation water applied automatically
                                to subirrigation plots at Lubbock
Date
Time
Total
Hows
Water
Applied
(Inches)
Total
(Inches)
30 cb Plots 2 5 gpm Dole Valves
July 7-8
July 8-9
July 9-10
July 10
July 1 1
July 11-12


July 22
July 25-26
July 29
July 30
July 30
July 30-31
July 31
July 31
July 31
Aug. 3-4


7:30 p.m. -
8:30 p.m. -
9:30 a.m. -
4:00 p.m. -
7:30 a.m.
9:30 a.m.
1:30 a.m.
6:30 p.m.
12:30 a.m. - 9:30 a.m.
3:30 p.m. -


1:00 p.m. -
9:00 p.m. -
7:45 p.m. -
9:00 a.m. -
8:30 p.m. -
11:45 p.m.
9:45 a.m. -
1:00 p.m. -
8:30 p.m. -
11:00 a.m.


11:00 a.m.

30 cb Plots 2
ll:45a.m.
7:30 a.m.
8:15 p.m.
9:30 a.m.
9:00 p.m.
- 12:45 a.m.
10:15 a.m.
4:00 p.m.
12:00 p.m.
-11:30 a.m.


12
13
16
2V4
9
19'/2

2.5 gpm Dole
12V,
lO'/i
l/2
u
Vi
1
1A
3
m
24


60 cb Plots 2 5 gpm Dole
July 10
July 11-12


5:30 a.m. -
8:30 p.m. -


7:30 a.m.
11: a.m.

60 cb Plots 2
2
14fc

2.5 gpm Dole
1.12
1.18
1.49
.19
.84
1.82

Valves
1.06
.49
.02
.02
.02
.04
.02
.14
.16
1.12


Valves
.18
1.35

Valves






6.64











3.09
9.73



1.53

          Aug. 3-4
11:00 a.m. - 11:00 a.m.
24
1.12
                                                                        1.12
                                                                        2.65
As  evapotranspiration  potential  during  this
period did not exceed .5 in./day, it appears that
a portion of the water added  replenished the
profile.
  To obtain information concerning soil water
distribution, tensiometers  were installed at 4,
                          10 and 20 in.  from an orifice at 2 locations in
                          each plot at different depths. The system was
                          operated manually for 24 hours during which
                          1.12 in. of water were applied to each plot. From
                          Figs. 4 (low potential plot) and 5 (high potential
                          plot), it can be seen that the primary  area of

-------
                                                                         SUBIRRIGATION STUDIES      163
 4     10         20
  Distance,  in.
Aug.  3    7:30 am
                                     4     10         20
                                      Distance,  in.
                                    Aug.  3   11:00 am
                           4     10         20
                            Distance, in.
                          Aug. 4  11:00 am
                           4     10         20
                            Distance,  in.
                          Aug. 9   12:30 pm
Figure 4: Soil water potential in soil profile of plot with low initial soil water potential during and following ir-
rigation. [Values  on graphs in -cb,  o - tensiometer locations,  »- subirrigation pipe location. (Amarillo fine
sandy loam, High Plains Field Site)]
          4     10         20
          Distance, in.
         Aug. 3   7:30 am
  Distance,  in.
Aug.  3   11:00 am
 4     10         20
  Distance,  in.
Aug.  4   11:00 am
                                                                                 4     10         20
                                                                                  Distance,  in.
                                                                                Aug.  9   12:30 pm
Figure 5: Soil water potential in soil profile of plot with high initial soil water potential during and following ir-
rigation. [Values on graphs in -cb, o - tensiometer locations, »- subirrigation pipe location. (Amarillo fine sandy
loam, High Plains Field Site)]

-------
164     MANAGING IRRIGATED AGRICULTURE
potential  change following irrigation occurred
below 6 in. and above 36 in. In the low potential
plot,  the  largest  increase in potential occurred
at 12 in. followed by 24 in. Changes in potential
at 6 in. were small while those at 36 and 48 in.
were  negligible.  This indicates that  most of the
water applied to  this plot was stored at 12 and
24  in. and there was still retention capacity  in
the soil profile for  storage of rainfall. The corn
in the -60 cb plot showed visual signs of stress
and the yield was only 8500 ears per acre.
  In  the  high  potential  plot, which  yielded
17,000 ears  of corn per acre, the potential was
higher initially and less change in potential oc-
curred after the  irrigation  water was applied.
The crop was healthier and  apparently used
more of the water so that less decrease  in po-
tential occurred. As with the low potential plot,
there was little change in potential at 6, 36 and
48  in. This  soil  is underlain with a  sandy layer
at the 36-in. depth. It appears that this may be a
factor in  maintaining  a  high potential  under
field conditions. The  lack of potential change at
6 in.  suggests that  little water is lost to evapo-
ration.
  The automated subirrigation systems appar-
ently do have some of the advantages hypothe-
sized  in the initial project.  It is necessary  to
maintain a low potential in only a portion of the
profile to supply crops. This leaves room for soil
water  storage from rains.  Using  the  limited
content and potential data obtained, the authors
estimated  the water-holding capacity of the soil
at Lubbock following irrigations at  -30 and -60
cb. This information  is presented in  Fig. 6.
  The necessity to  maintain a high  potential  in
only a portion of the profile coupled with the
increase in water retention  due to  soil layering
indicates  that irrigation return flows  can  be
minimized and quality increased through better
water use efficiency in soils of this type.

Ion Concentration of Soil Water Extracts
  All of the ion concentration data were pro-
cured from  the  Rolling Plains Site located  in
Knox County. Breakdown of residue from the
previous dryland grain  sorghum crop and sub-
sequent nitrification  resulted in significant in-
creases  in nitrate  in the soil solution. Peak
concentrations occurred in  July. Nitrate-nitro-
gen in soil  water extracts at the  beginning  of
   Oi—
                                        J
    0        .5        1.0       1.5       2.0
         Estimated Water Storage Capacity, in.

 Figure 6: Estimated soil water  storage  capacity
 remaining in the top 4 ft of soil following subirri-
 gation at  low and  high  soil water potentials.
 (Amarillo fine sandy loam, High Plains Field Site)

the season   following  emergence  along  with
yields of unfertilized  plots from each irrigation
system are shown in Fig. 7. It  can be  seen that
nitrate-nitrogen  concentration  in  the  furrow
and  sprinkler  plots  was  considerably  higher
than that of the subirrigated plot. This was true
for similarly  treated  plots between irrigation
systems throughout the growing season.
  Some difficultly was experienced in obtaining
emergence on the subirrigated  plots when they
were planted on  the bed. Up to 8 in. of water
were applied to obtain emergence.  Some of the
plots  were replanted in a  furrow in  order to
place  the seed closer to the subirrigation pipe
to determine if  this  problem could  be elim-
inated,  and  emergence  was obtained  without
any problem with 2 in. of water or less.
  Significant  increases  in  nitrogen were  de-
tected  in fertilized subirrigated plots only when
fertilizer was applied through  the irrigation
system. This  was probably a consequence of
water from the subirrigation system moving the
banded nitrogen  away from the soil water ex-
traction tubes located in  the bed. However,  no
large concentrations of nitrate  were detected at

-------
                                                                SUBIRRIGATION STUDIES
                                                   165
            Sprinkler Irrigated
            Yield - 4375 ears/A
                             Percent
                               Clay
                                15
                                23
      0    10    20   30    40
              ppm NO,-N
    Furrow Irrigated
  Yield - 3250 ears/A
                  Percent
                    Clay
                     19

                     17

                     17

                     21

                     21

                     17
                                                              15
                                                   A - July 8
                                                       Sept. 3
0    10    20   30
       ppm NO,-N
                                                              15
                                                              13

                                                          i	11
   Subirrigated
Yield - 7125 ears/A
             Percent
               Clay
                10


                21


                25


                25


                26


                19
                                       A- July 8
                                       Q- Sept. 3
 10   20    30
    ppm N0-N
                                                  13
                                                  11

                                              i	13
Figure 7: Nitrate-nitrogen concentration of soil water extracts from various depths of unfertilized plots of dif-
ferent irrigation systems. [Concentration in irrigation water 6-8 ppm NO3-N. (Miles loamy fine sand, Rolling
Plains Field Site)]
depths greater than 4 ft under the subirrigation
system; therefore, it is  doubtful that important
amounts of nitrate were leached from the root
zone by water applied throughout the subirriga-
tion systems.
  Results of fertilization through the subirriga-
tion system are shown  in Fig. 8. It can be seen
that the injection of the nitrogen  solution in-
creased the nitrate-nitrogen content of the  soil
profile and the subsequent yield. The nitrogen
solution used contained 50 percent of its  nitro-
gen as  urea and  the other 50 percent as am-
monium nitrate;  however, increases in ammon-
ium ion concentration were not detected.
  The  highest  concentrations  of  ions  were
found immediately  above  the first or second
clay layer.  This  was true  not only  of nitrate
(Fig.  7), but other ions also (Na+ - Fig. 9). No
major differences  existed among irrigation sys-
tems  in the sodium  content extracted from the
soil profiles. There was a tendency for the con-
         centrations to  increase between the beginning
         and end of the season, but there was no general
         trend with respect to irrigation system  or mois-
         ture  level.  Except for a few points, the sodium
         concentration of water extracted from  the pro-
         file was  less than that of the irrigation water
         (84-127 ppm  Na+).
           Calcium concentration of extracts from the
         1- to 4-ft depths increased between the begin-
         ning and end of the season (Fig. 10), especially
         in the furrow- and sprinkler-irrigated plots. The
         sprinkler,  furrow,  and  subirrigated plots re-
         ceived 3, 15 and 5 in. of irrigation water, respec-
         tively, between the dates shown on Fig. 7. The
         subirrigation system apparently does not allow
         as much calcium to accumulate  in  the 1- to 4-
         ft  zone as do the furrow and sprinkler systems
         due to its unique flow pattern away from the
         orifice and the plane of sample extraction, while
         the other systems move surface  concentrations
         of ions into  the soil profile.  In this zone, the

-------
  166

    0

    1
MANAGING IRRIGATED AGRICULTURE
     August 17                     September 3
                                                                                September 27
 •M
 Q.
 Ol
                A - Unfertilized
                Q - Fertilized
                                      A -  Unfertilized
                                      Q -  Fertilized
                         A - Unfertilized
                         Q - Fertilized
                                                                                                   Yield
                                                                                                 (Ears/A)
                                                                                                   6,750
                                                                                                  14,500
         25     50    75    100         25    50    75    100   125          25    50    75    100   125
             ppm NO,-N                        ppm NO,-N                           ppm NO--N
                   «5                                O                                   3
Figure 8: Nitrate-nitrogen concentration of soil water extracts of unfertilized and fertilized subirrigation plots.
[Nitrogen solution was injected in the system of the fertilized plots on Sept. 2, Sept. 8 and Sept. 13 at the rate of
          Sprinkler Irrigated                     Furrow  Irrigated                      Subirrigated
                                                                                                  Percent
   1

   2

   3

   4
i«-
„•  5
o.
V
Q
   7

   8

   9

  10
                                 Percent
                                   Clav
      A - July 8
      o - Sept. 3
                                  13

                                  21

                                  21

                                  17

                                  13

                                  17
                                                        Percent
                                                         Clay
                                                           19

                                                           17

                                                           17

                                                           21

                                                           21

                                                           17

                                                           15
A - July 8
Q - Sept. 3
                                                          15
                                                          13
         50    100   150   200   250
                ppm Na
                                         i      .     i      ill
                                        50   100   150   200
                                             ppm Na
                                                                             A - July 8
                                                                             B - Sept. 3
25

26

19

13

^

i13
                                                                                  50    100   150   200
                                                                                        ppm Na+
 Figure 9:  Sodium concentration of soil water extracts of unfertilized plots of different irrigation systems. [Con-
 centration in irrigation water 84-127 ppm Na*. (Miles loamy fine sand, Rolling Plains Field Site)]

-------
                                                                  SUBIRRIGATION STUDIES
                                                                       167
          Sprinkler  Irrigated
                      Furrow  Irrigated
                            Subirrigated
   .
 &
     0 r-

     1


     2

     3
  .  5
    10
A - July  8
D - Sept.  3
                                  A-  July 8
                                  Q-  Sept. 3
            50    100   150   200
                    ^ ++
                ppm Ca
                    50
100   150
    _ ++
ppm Ca
200
50    100   150   200
      ppm Ca
Figure 10: Calcium concentration of soil water extracts of unfertilized plots of different irrigation systems.
[Concentration In irrigation water 70-96 ppm Ca^ . (Miles loamy fine sand, Rolling Plains Field Site)]

-------
168
MANAGING IRRIGATED AGRICULTURE
calcium concentration of the sprinkler- and fur-
row-irrigated plots generally  exceeded that  of
the irrigation water (84-127 ppm Ga  ) while
                                          the extracts from the subirrigated plots did not.
                                          The same trend was found with respect to chlo-
                                          ride content of the extracts (Fig. 11). There was
     Or    Sprinkler Irrigated
    10
                     A - July 8
                     Q - Sept.  3
                                    Furrow Irrigated
                                            Subirrigated
                                                     A- July 8
                                                     £3- Sept. 3
                                                                           A - July 8
                                                                           Q - Sept. 3
           50
          100   150
         ppm Cl~
200
50
100   150
ppm Cl~
200
100   150
ppm Cl~
                              200
Figure  11: Chloride concentration of soil water extracts of unfertilized plots of different irrigation systems.
[Concentration in irrigation water 81-97 ppm Cl~. (Miles loamy fine sand, Rolling Plains Field Site)]
a significant increase in the 1- to 4-ft zone in the
furrow- and sprinkler-irrigated plots with the
concentrations exceeding those of the irrigation
water (81-97 ppm Cl~).  In some of the furrow
plots there were significant increases  down to 9
ft. However, this was  not  a  general  trend over
all plots.
  The amount of sulfate in the extracts was in-
fluenced by irrigation  system and  amount and
type of water applied (irrigation or rainfall). The
sulfate concentrations  between July 8 and  Sept.
3 tended to increase in the  sprinkler- and fur-
row-irrigated plots (fig.  12). The increase in the
subirrigated plots was  small,  especially if the
plots were irrigated prior to the last sampling in
addition to  receiving rainfall. Subirrigated plot
(b), which received 4  in. of  irrigation water in
addition to the rainfall between  Aug.  17 and
                                          Sept. 3, did not show any changes in sulfate con-
                                          centration between the beginning and end  of
                                          the   season.   However,   subirrigated   plot
                                          (a), which received 3 in. of rainfall only between
                                          Aug. 17  and Sept. 3, showed a substantial in-
                                          crease  in  the  sulfate  concentration  of the ex-
                                          tracts from 1.5 and 2 ft.
                                            Sprinkler-irrigated  plot  (a)  received   only
                                          rainfall after Aug. 17 and showed a higher con-
                                          centration of sulfate on the last sampling date in
                                          the   top   2  ft  than  sprinkler-irrigated  plot
                                          (b) which received 9 in. of irrigation water plus
                                          rainfall, and showed  a substantial decrease in
                                          sulfate  concentration at 1.5 and  2 ft and a sub-
                                          stantial increase at the lower depths.
                                            The  furrow-irrigated  plots showed a similar
                                          trend. Furrow-irrigated plot (b), which received
                                          6 in. of irrigation water between Aug. 17 and

-------
                                                                SUBIRRIGATION STUDIES
                                                  169
           Sprinkler Irrigated
                  (a)
                      A - July 8
                      o - Aug. 17
                      Q - Sept.  3
   Furrow Irrigated
        (a)
   Subirri gated
        (a)
10


 0


 2


 4


 6


 8


10
            A - July 8
            o- Aug. 17
            Q- Sept.  3
        A- July 8
        o- Aug. 17
        B- Sept. 3
           100   200   300   400   500

                           (b)
                    A- July 8
                    o- Aug. 17
                    a- Sept.  3
                      i	i
100  200   300  400   500

                  (b)
                                                        A- July 8
                                                        o- Aug. 17
                                                        B- Sept.  3
100   200   300  400

               (b)
                                        A- July 8
                                        o- Aug. 17
                                        o- Sept.  3
          100   200   300   400
              ppm SO
 100   200   300   400
      ppm SO
100   200   300   400
     ppm SO
Figure 12: Sulfate concentration of soil water extracts of plots from different irrigation systems which received
different amounts of rainfall and irrigation water. [Plots in top graphs received rainfall while plots in bottom
graphs received rainfall plus irrigation water between Aug. 17 and Sept. 3. Sulfate concentration in irrigation
water 138-172 ppm. (Miles loamy fine sand, Rolling Plains Field Site)]
Sept. 3, showed some decreases at some of the
shallower depths with increases at lower depths;
while furrow-irrigated plot (a), which  received
only rainfall, still had high concentrations of sul-
fate at  0 to 4 ft and showed some decreases  at
the lower depths.
  The sulfate concentration at several depths  in
the soil profile in the sprinkler and furrow plots
exceeded that of the irrigation water, while the
sulfate  concentration  of the extracts from the
subirrigated plots exceeded that of the irrigation
water at only a  few depths.
  Electrical  conductivity  also  increased  with
rainfall  (Fig. 13).  Those plots which  received
only rainfall between  Aug.  17  and Sept. 3 had
higher conductivities than  plots which  received
irrigation water plus rainfall. This is not surpris-
ing since rainfall contains few if any soluble ions
while the irrigation water contains chlorine, cal-
        cium, sulfate and sodium. With this difference
        in water  quality, a difference in the concentra-
        tion  of some ions in the soil  solution and a
        change in conductivity may be expected. As pre-
        viously mentioned,  the  sulfate  concentrations
        were higher  at the  shallow depths following
        rains than following rainfall and irrigation.
          Typical profiles of phosphorus, potassium and
        ammonium   from  the  subirrigated  plot  are
        shown in Fig.  14. As would be expected,  there
        was little change during the growing season due
        to  treatment or  irrigation system. Phosphorus
        concentrations  ranged up to 5  ppm  on certain
        dates in some plots, but most of the phosphorus
        concentrations  of most extracts was 1.5 ppm or
        less with  the highest concentrations occurring in
        the upper 2 ft of the profile. With these  low con-
        centrations, phosphorus will never be a problem
        as a contaminant in the  ground  water  supplies.

-------
170
MANAGING IRRIGATED AGRICULTURE
           Sprinkler Irrigated
                                  Furrow Irrigated
Subirrigated
              o  - Rainfall plus
                  irrigation
              D  - Rainfall
              I. .   I
                                         o- Rainfall
                                            plus
                                            irrigation
                                         a- Rainfall
  10100 500  1000  1500  2000  2500    100  500
                            I  ,  lA .  . I .  .  I  .  . I.  .  I
                                                                                   o - Rainfall
                                                                                      plus
                                                                                      irrigation
                                                                             - Rainfall
                I
          Conductivity, y mhos
                                    1000  1500  2000  2500  100 500   1000  1500  2000  2500
                                Conductivity, y mhos            Conductivity, u mhos
Figure 13: Electrical conductivity of soil water extracts obtained on September 3 from unfertilized plots from
the different irrigation systems. [Conductivity in irrigation water 1040-1140 ju mhos. (Miles loamy fine sand,
Rolling Plains Field Site)]
No  measurable  concentration was  found  in
ground  water samples  procured  during  the
course of the study.
  The potassium concentration of the extracts
from the upper 3 ft of the soil profile was higher
than that of the irrigation water. The source of
this potassium was probably the exchange com-
plex of the soil because  the concentrations re-
mained  reasonably   constant  throughout  the
growing season.
  Ammonium was found  at various times dur-
ing the season in many of the plots at various
depths. There was no pattern as to the amounts
present as  influenced by treatment or irrigation
system.  However, the  data show that ammo-
nium may  move in these soils.

SUMMARY

  The layering in the soil profile produced con-
ditions  whereby  the soil  retained more water
than  would  be expected from  pressure plate
data. This  in  turn resulted in  most of the ions
being retained in or  immediately above the clay
layers  in the soil profile. Water  samples  ex-
                                          tracted  from  subirrigated  plots generally con-
                                          tained lower concentrations of nitrate, calcium
                                          and chloride. Sulfate concentrations and electri-
                                          cal conductivities were more influenced by rain-
                                          fall than by the method of irrigation. Phospho-
                                          rus and potassium concentrations were not  in-
                                          fluenced by irrigation system  or moisture addi-
                                          tions.
                                            Soil water potential data indicate that the pri-
                                          mary zone of water application from subirriga-
                                          tion  systems in the soils in this study is between
                                          6 in.  and 36 in. The subirrigation system thus
                                          appears to  have potential for maintaining the
                                          water quality of irrigation  return flows for sev-
                                          eral  reasons.  Concentrations  of certain ions
                                          such as chloride and calcium remain low in the
                                          root  zone due to the fact  that less water is ap-
                                          plied and the unique flow pattern of the system.
                                          Nitrogen can be easily applied to the system in
                                          small amounts  which will create  high nitrate
                                          concentrations  only in  the root zone  for plant
                                          growth. In layered soils, zones of high potential
                                          can be created which will be adequate for plant
                                          growth yet zones of low potential will remain in
                                          the soil profile for the storage of rainfall.

-------
                                                               SUBIRRIGATION STUDIES
                                                                                171
     Or-


     1


     2


     3


     4
    10
     A- July 8
     a - Sept. 3
\
A- July 8
D- Sept.  3
                      A- July 8
                      a- Sept. 3
0.5  1.0   1.5
 ppm P0°
                                     10    15
                                      20
      25
30
                                           ppm K
                                                                  ppm NH.
Figure  14: Phosphorus, potassium and ammonium concentrations of subirrigated plots. (Miles loamy fine
sand, Rolling Plains Field Site)
  In those areas where water of impaired qual-
ity is used, it is foreseen that a zone of high salt
concentration may accumulate in  the periphery
of the zone irrigated by the subirrigation pipe. It
may be necessary  to irrigate during or  immedi-
ately after rains to prevent these salts from mov-
ing into the root zone.
  Sweet  corn  was successfully grown on  the
subirrigation systems  at Lubbock  and Munday,
Texas  in  1971. Some  management  problems,
such as planting too  far from the subirrigation
pipe,  were  encountered but  solved.  The exis-
tance  of  fine-textured layers  underlain   by
coarse-textured layers creates zones of high  wa-
ter retention and low conductivity and  the pos-
sibility that wider  spacings of the  laterals could
be used  which would decrease  the cost of the
systems.

ACKNOWLEDGEMENT
  This study is supported by the Environmental
Protection Agency under Program 13030 EZM,
Dr. J. P.  Law, Project Officer, by  the Texas
A&M  University Vegetable Research Center at
Munday,  Tx,  Dr.  M. C.  Fuqua, Officer in
                                       Charge, and by the Texas Water Resources In-
                                       stitute, Dr. J. R. Runkles, Director.

                                       REFERENCES
                                          1. Law, J. P. and J. L. Witherow. Irrigation
                                       Residues  In. A Primer on  Agricultural Pollu-
                                       tion. Soil Conservation Society of America p 11-
                                       13.
                                          2. Miller, D. E. 1969. Flow and Retention of
                                       Water in  Layered Soils. Cons. Res. Report No.
                                       13. ARS, USDA.
                                          3. Miller,  D. E. and W. C.  Hunger.  1963.
                                       Moisture Retention of Soil  with Coarse Layers
                                       in the Profile. Soil  Sci. Soc. Amer. Proc. 27:
                                       716-717.
                                          4. Whitney,  L.  F. Plastic Orifice Inserts for
                                       Subsurface Irrigation.  Paper No.  68-758  pre-
                                       sented at  the 1968  Winter  Meeting of Amer.
                                       Soc. of Agricultural Engineers.

                                       DISCLAIMER
                                          Reference to commercial products or trade names
                                       is made with the understanding that no discrimina-
                                       tion is intended and no indorsement  by the Texas
                                       Agricultural Experiment Station is implied.

-------
            Irrigation  Return  Flow  Studies in

                          The  Mesilla  Valley

                                    P. J. WIERENGA
                                Department of Agronomy
                               New Mexico State University
                                           and
                                   T. C. PATTERSON
                         Department of Agricultural Engineering
                               New Mexico State University
 ABSTRACT
   Deterioration of the quality of the water in
 the Rio Grande is a major problem for water
 users in New Mexico  and Texas. From Otowi
 Bridge near Santa Fe, New Mexico to El Paso,
 Texas, a distance of 270  miles, the total dis-
 solved solids increases from 221 ppm to 787
 ppm, while the percent sodium goes from 25
 near Santa Fe to 52 at El Paso. The increase in
 salinity per mile is more than twice as great in
 the irrigated areas as in the non-irrigated areas.
 The greater increase in salinity per mile is due
 to the return of lower quality drainage water
from the irrigated areas to the river.  This paper
 describes a research project designed to deter-
 mine under field conditions rates of water and
 salt movement in the soil as affected by fre-
 quency and amount of surface water applica-
 tion. The effects of trickle irrigation  on return
flow quantity and quality are also determined.

 INTRODUCTION
  The quality of irrigation return flow repre-
 sents a major problem in the western United
 States. The  water of the  Rio Grande in New
 Mexico has  been reported as a classic example
 of water quality  degradation4 7 12 13. Table  1
 taken from  Wilcox12,  shows the variation  in
 composition of the Rio  Grande  water from
 Otowi Bridge near Santa Fe to El Paso, Texas.
  Table 1 shows a progressive increase in the
 concentration of total dissolved solids and per-
 cent sodium from the upper to the lower sam-
 pling stations. Only a small portion of land is ir-
 rigated along the  river  for 42 miles from San
 Marcial to Caballo Dam. From Leasburg Dam
 to El Paso, essentially all of the land is irrigated
 for a distance of 63 miles. The total dissolved
 solids in the river increased by 66 ppm along the
 area without irrigation and by 236 ppm  along
 the area with irrigation.  The increase per mile is
more than twice as great in the irrigated areas
 as in the non-irrigated area.
  The relatively large increase in dissolved sol-
ids  in  the river along  the  irrigated areas  is
mainly due to the concentrating effect of irriga-
tion. Plants take up water from the soil solution
and transpire it to the atmosphere, but they take
up very little salt.  Similarly water evaporates
                                           173

-------
174     MANAGING IRRIGATED AGRICULTURE

                                          TABLE 1
                         Variations in composition of Rio Grande water
                                 (weighted means, 1934-1953)
         Miles from
         El Paso
         Dissolved
         solids ppm
         Sodium %

         Land
         Irrigated
         Acres
Station
Otowi
Bridge
270
221
25
San
Martial
150
449
41
Elephant
Butte
130
478
43
Caballo
Dam
108
515
44
Leasburg
Dam
0?
551
44
El Paso
(1
787
52
80,000
from  the soil surface,  but the salts remain be-
hind,  causing an increased concentration in the
soil solution. With normal irrigation practices a
significant portion of the applied irrigation wa-
ter  percolates to the subsoil and groundwater
table. This percolating water carries with it salts
accumulated in the root zone, thereby prevent-
ing excessive salinization of the  upper soil pro-
file. Eventually this water affects the  quality of
underlying aquifers  or is picked up  by drains
and returns  to the main stream, resulting in in-
creased salinity of the river. The salt concentra-
tion of the  drainage water can  be as much as
two to five times higher than the salt concentra-
tion of the irrigation water10. Although the vol-
ume of drainage water is usually much smaller
than  the volume  of river water,  it is  obvious
from  Table  1, that recycling of irrigation return
flow has a detrimental effect on  the water qual-
ity downstream.
  The data  in Table 1 are for the years 1934-
1953.  Since  1951 the situation in  the Rincon and
Mesilla Valleys  in southern New Mexico has
worsened through the  addition of some  1800
wells. Starting in  1951  and continuing  through
1957  the Rio Grande project experienced a crit-
ical  shortage  of irrigation water.  During this
time alone, more than  1700 irrigation  wells were
drilled  to  provide  supplemental  water13. The
quality of the water from these wells is consider-
ably lower than of the water from the river. Fig-
ure 1, taken from Easier and Alary1, shows the
location in the Rincon and Mesilla Valleys  of 39
15,000    70,000
                                               Valley
                                                   El  RASC
                Figure 1: Map Showing the Electrical Conductivity
                in Micromhos at 25° C of Shallow Groundwater at
                Selected Sites in the Rincon and Mesilla Valleys,
                New Mexico

               selected shallow wells. Data from these wells,
               combined in four groups of wells along the river
               from north to south, are presented in Table 2.
                 According to the information from the Irriga-
               tion District  Office in this area, for  the years
               1951 through 1969, approximately 58 percent of

-------
                 TABLE 2

 Variation in composition of well water in the
  Rincon-Mesilla Valley, New Mexico (1967)

        No. of         Average         Percent
 Group  Wells   Conductivity (EC* 1Q6)   Sodium
1
2
3
4
10
10
10
9
1891
1849
2268
3417
46
51
57
73
RETURN FLOW STUDIES — MESILLA VALLEY     175

                     TABLE 3

        Average discharges and weighted mean
    conductivities from  1934 to 1963, for Leasburg
     dam, New Mexico and El Paso, Texas.  Data
    extracted from U. S. Salinity Report No. 113,
               byL. V. Wilcox, 1968
the water used for irrigation in the Rincon-Me-
silla Valley came  from the district canals, while
42 percent  was supplied from wells. Since the
groundwater  as represented by the 39 wells in
Table 2 averages 2320 micromhos/cm, while the
weighted mean conductivity of the river water
from  1934-1963 at Leasburg dam is 844 micro-
mhos/cm, it would seem that the use of irriga-
tion wells, starting on a major scale in 1951, is
adding to the salt problem.
  Because it is unlikely that irrigation with well
water is a great deal more efficient than irriga-
tion with river water, it may be  expected that
the use of well water will have an adverse effect
on the quality of the return flow. Table 3 com-
pares  the weighted mean electrical conductivity
of the Rio Grande water at Leasburg dam and at
El Paso, 63 miles downstream.  Data  are pre-
sented for the periods  1934-1950  and  for 1951-
1963.  Data for Leasburg dam after 1963 are  in-
complete. The average discharges at  Leasburg
dam and at El Paso are also presented. The last
column in Table 3 presents the increase in salin-
ity between Leasburg dam and El Paso for every
three  year period from 1934-1963.  Comparing
the quality  of the river water before 1951 with
the quality of the water from 1951 to 1963, it is
clear  that although  the total salinity has  in-
creased  somewhat, the  net increase in salinity
between  the two  measuring stations  remained
nearly the same.  Apparently the  increased use
of well water after 1951 had, up to 1963, very lit-
tle effect on the quality of the irrigation return
flow. This slow response to changes in irrigation
water quality is undoubtedly  due to the high
buffering capacity of most of the agricultural
Period Average
in Discharge in
Years 1000 acre
Weighted Mean
Electrical Conductivity,
ECxlO6
feet
Leasburg
Dam
1934-1936 698
1936-1939 742
1940-1942 1047
1943-1945 828
1946-1948 702
1949-1950 699
El
Paso
481
534
842
604
463
469
Leasburg
Dam
972
834
808
764
846
729
El
Paso
1351
1257
1021
1184
1254
1166
In-
crease
379
423
213
420
408
437
    1934-1950  791    571
827
1187   360
1951-1953
1954-1956
1957-1959
1960-1963
476
217
595
569
267
73
306
330
851
1304
813
832
1209
1505
1124
1287
358
201
311
455
    1951-1963  472    250
881
1237   355
    soils in the Mesilla and Rincon Valleys. How-
    ever, it  may be  expected  that eventually  the
    quality of the irrigation return flow, and thus the
    quality of the river water at El Paso, will be af-
    fected by the increased use of well water.
      The high  salinity of the water in the Rio
    Grande below El Paso has already resulted in
    extensive salt damage  to agriculture in  the  El
    Paso Valley, particularly in Hudspeth County,
    below El  Paso, where many farmers are discon-
    tinuing their  operations. Further  deterioration
    of the river water will  also have serious  affects
    on agriculture in the Mesilla and Rincon Valleys
    in New Mexico. In addition, the city of El Paso
    and the city of Juarez, located across the river
    from El Paso in Mexico, will in future years be-
    come more and more dependent on water from
    the Rio Grande for their municipal and indus-
    trial needs. These cities therefore have a strong
    interest in maintaining the quality of the river
    water.
      So far, no studies have been made to predict
    quality changes in the Rio Grande basin as a re-
    sult of agricultural practices  or municipal and

-------
176     MANAGING IRRIGATED AGRICULTURE
industrial developments within the  basin. For
sound long term planning of  the very limited
water resources in the Middle  Rio Grande ba-
sin, it is necessary to estimate  long  term water
quality changes resulting from changes in the
use of water. In the past much work has  been
done  on developing models  for predicting the
quality changes of water percolating through
soil profiles3 8 9. Such models are of great value
for predicting the quality of subsurface return
flow and for determining the effects of agricul-
tural practices on return flow quality. However,
few  of these  models  have  been  adequately
tested under actual field conditions.
  In cooperation witlrthe Environmental  Pro-
tection Agency a field plot experiment was initi-
ated by the New Mexico State University Ex-
periment Station to study the quality of irriga-
tion return flow. One of the  main objectives  of
this study is to  determine under field conditions
with varying degrees of  management  rates  of
water and salt movement, under  conventional
surface irrigation  as practiced in  the valley.
Similar determinations will be made  on plots ir-
rigated  with trickle  systems. Trickle irrigation
has been  used  successfully  for irrigation  of a
wide  variety of crops in Israel, Australia  and
New Zealand and is presently being  used on an
increasing scale in this country. Little is known,
however, regarding  the salt build up in the soil
profile around trickle lines or the quality of the
return flow under trickle irrigation.

             Project Description
  The project  is located at  the Plant Science
Farm of New Mexico State University near Las
Cruces, New Mexico. The farm is located about
eight  miles south and west of Las Cruces (Fig-
ure 1). The experimental area is located near the
farm headquarters and  is bordered on the east
by the 10 to 12 feet deep Del Rio Drain, which
serves  a large section of land  in the Elephant
Butte Irrigation District. In this area the Del Rio
Drain  runs  parallel to and approximately one
mile east of the Rio Grande. The soil at the site
is not uniform,  but differs in this respect  little
from  typical soils in  the  valley, which too are
highly variable. It consists in general of 35 cm of
clay loam over 20 to 30 cm of clay, over a vari-
able amount of silty loam. The latter changes
rather abruptly into medium sand. The depth to
the sand varies on this 2 acre field from 65 to
120 cm below the soil surface. Root penetration
into the medium to fine sand layer appears to
be almost negligible.
  In Figure 2 the field plot layout of the experi-
ment is presented. The experimental set up con-
sists of thirty 24 * 24 feet plots and six 20 * 60
feet plots. The 24 * 24 feet plots are irrigated by
surface irrigation and  the 20 x 60 feet plots by
trickle or  drip irrigation. Each plot has its own
water outlet, which is connected through a 2 or
4 inch underground pipeline to a 8 inch well just
outside the plot area. The surface irrigated plots
are separated from the surrounding soil by poly-
ethylene plastic to a depth of 2.5 feet below the
soil surface. Wooden boards of 1  x 12 inches, ex-
tending 8  inches  above  the soil surface and
covered with  polyethylene plastic, prevent sur-
face runoff from the plots.

Surface Irrigation
  Of the thirty surface plots, three are set aside
to determine the hydraulic characteristics of the
soil of this  study. Triplicate tensiometers have
been installed near the center of each of these
three plots at depths of 30, 60, 90,  120, 150 and
180 cm below the soil surface. In addition, two
neutron access tubes have been installed close
to the tensiometers. Following the procedures
described  by Van Bavel et. al." and Nielsen et.
al.5 the hydraulic characteristics of the soil  are
determined as a function of water content.
  The main treatment affects on the 24 x  24
square feet plots are frequency of irrigation and
amount of irrigation water applied. The time be-
tween  irrigations varies according to  soil type,
stage of plant development and climatic condi-
tions. In this investigation the plots will be irri-
gated on the basis of the amount of "available"
water in the soil profile, where "available" wa-
ter is defined as the amount of water retained
between 1/3 and 15 bars suction. Treatments in-
clude the  scheduling of an irrigation  when  25,
50 or 75 percent of the available water has been
depleted from the root zone.
  The  amount of water applied to replace  the
water depleted within the root zone is affected
by the efficiency of water application. In this in-
vestigation  surface  runoff  is prevented with
borders around  each plot and side seepage is
minimized by polyethylene  plastic  sheets to a

-------
                                              RETURN FLOW STUDIES — MESILLA VALLEY
                                          177
                          [ From Well
tf>
to
CO


"8 ffl 	 ffl 	 	
CVJ





o
10
CM





'O
IO
ft— I—1 "-» •—-"—•«—•
CM




14' 1 24' I0'1


24'
"8
"o
(VJ
o
"o


IO'




L ..„,.,.„ __ .
1
I
I
I
I
I
I
L






«—
~






Plot Boundary -
-1 «-i ' — i i-s>>— • i — i v











Redwood and Plastic










L
24' 1 IO1 24' 1 lO'l 24' 101 24' ICf 2
60' l, 3O' L 60'
f f





i


i
i




— i


1 — «' —




L






1
4' lltf 24' llOl 24' '
. 30' L 6O1
f
1
IL«J



L
T 24' V


•
— " — "— '  4" Pipeline
                                                                             -i — """  |" Pipeline
                                                                             -------  Trickle Headers
                                                                                   N
                            Farm Roadway
                         Del  Rio  Drain
  Figure 2: Field Plot Layout of Irrigation Return Flow Project at Plant Science Farm, New Mexico State Uni-
  versity, Las Graces, New Mexico
depth of 2.5 feet below the ground level. Thus
the field water application efficiency  is a func-
tion of the amount of water lost by deep percola-
tion only. Three field water application efficien-
cies have been selected for this experiment, e.g.
50, 75, and  100 percent, respectively. With 50%
efficiency half of the applied water is  lost by
deep percolation, while supposedly none is lost
with 100% efficiency.
  The main-treatment effects consist of combi-
nations  of three frequencies of irrigation  with
three field water application efficiencies,  result-
ing in a total of 9 treatments. Each treatment is
block randomized  with three replications per
treatment, resulting  in 27  plots for this part  of
the experiment. To determine  the water loss by
deep percolation,  measurements  are  made  of
the water content and the  pressure gradient be-
low the root zone of each  plot. From the pres-
sure gradient and knowledge  of  the  hydraulic
conductivity  and the water content  below the
root zone, the net flux in or out of the soil profile
is calculated, using Darcy's  equation. The qual-
ity of the water percolating  below the root zone
is determined by collecting samples from suc-
tion cups located below the  root zone. Although
changes  in the chemical composition of perco-
lating water are rather slow it is expected that
three years of filed data will show the effects of
frequency of irrigation and  of water application
efficiency on the quality  of percolating water.
Cotton will be grown on  the 24 x 24 feet plots
for the duration of the experiment.

Trickle Irrigation
  The effects of trickle irrigation on the amount
and  quality of percolating water will be studied
on six 20 x 60 feet square plots. Three of these
plots are irrigated to  maintain the soil-water
tension at 6 inches below soil surface at or be-
low  0.2 bars. On the other  three plots  the soil-
water tension at 6  inches is kept at 0.6 bars or

-------
178     MANAGING IRRIGATED AGRICULTURE
less. Irrigations are automated with an adjust-
able timing control to keep the soil-water ten-
sions at 6 inches at the desired levels.
  The amount of percolating water is again de-
termined from the pressure gradients below the
root zone as measured with tensiometers at two
depths, and the hydraulic conductivity of the
soil. Samples of the percolating water are taken
through suction cups placed below the root zone
of the crop. The electrical conductivity of the
well water used for  irrigation all plots is 1455
micromhos/cm. Thus a significant salt buildup
can  be expected away from  the trickle lines.
However, the  time required for the salt buildup
to occur  is unknown. A one year  study with
trickle irrigation of corn at New Mexico State
University showed no significant increase in salt
concentration in soil  samples  taken  in between
the rows, as well as  in the rows. To more  ade-
quately investigate this  phenomenon,  qualita-
tive  measurement are to be  made  of this in-
crease in soil salinity around the trickle  lines,
using recently developed salinity sensors6. From
9 to 12 salinity sensors are installed around each
of several trickle lines. This will allow contin-
uous monitoring of salt accumulation at various
grid points. The crop to be grown on the trickle
plots is cotton during the first year of the experi-
ment,  to be followed with lettuce,  onion and
possibly corn.
   In conjunction with the field experiment  a
record is  being kept of the quality of the water
in the Del Rio Drain adjacent to the plot area.
Two sampling stations have been established on
the  drain  for measuring the discharge and the
quality of the water at weekly  intervals.  The
first station is located adjacent to the plot area,
while the second station is 2.2 miles upstream.
The quality of the irrigation  water  used in the
area bordering the two mile section of the drain
is also monitored by periodically sampling the
surface and well water used for irrigation in this
area.

Expected Result
   The real interest in this  research project is in
the  quality and quantity of irrigation return
flow, as affected by  trickle and surface irriga-
tion. Trickle  systems have great potential for
 high  water-use  efficiency. The systems  can
 readily be automated and  may be programmed
to operate without excessive percolation losses
inherent with border and  furrow irrigation. At
present, the overall irrigation  efficiency  in the
Rio Grande Valley is around 40 to 50 percent.
Thus as much as 50 percent of the water used
for irrigation may be lost to the subsoil by deep
percolation. Part of this water is  used again by
pumping from wells, and part of it is returned to
the river as drainage return flow. However, the
quality of this water has degraded considerably
during its movements through the soil. It is ex-
pected that with trickle irrigation the quantity of
irrigation  return flow  can be greatly reduced.
What the effects will be on the quality of the re-
turn flow is uncertain. This project should yield
valuable information on the quality of the return
flow and on the changes in soil salinity resulting
from trickle irrigation. The project will also be
helpful in establishing management procedures
for trickle irrigation for the soil, water and cli-
matic conditions of the Middle Rio Grande Val-
ley.
  The surface irrigation treatments should yield
information on the amount and the quality of
water leaching from the surface plots. The ex-
periment may prove that possibly  less water
may be used for maintaining a favorable salt
balance than what at present  is thought to be
necessary for leaching of excess salts, based on
steady state flow rates. A  reduction in leaching
water will reduce the volume of drainage return
flow and thus  have a favorable effect on the
quality of the river water downstream.
  Water quality standards in New Mexico allow
agriculture to return  to  the  mainstream dis-
solved solids  present  in  the  irrigation  water.
However, in view of the increasing concern with
the quality of water in New  Mexico, it can be
expected that a closer look will be taken at the
practices of the  largest water user in the state,
than is done at present.


REFERENCES
   1. Easier, J. A. and L. J. Alary. 1968. Qual-
ity of the shallow groundwater in the Rincon
and  Mesilla Valleys, New Mexico and  Texas.
Open file report. U.S.G.S. Albuquerque, N.M.
   2. Bower, C. A. and L. V. Wilcox. 1969. Ni-
trate  content  of the upper Rio  Grande as in-
fluenced by nitrogen fertilization of adjacent ir-

-------
                                             RETURN FLOW STUDIES — MESILLA VALLEY     179
rigated lands. Soil Sci. Soc. Amer. Proc. 33:971-
973.
   3. Dutt, G. R.  1962. Prediction of the con-
centration of solutes in soil solutions for soil sys-
tems containing gypsum and exchangeable Ca
and Mg. Soil Sci. Soc. Amer. Proc. 26:341-342.
   4. Eldridge,  E. F.   1963.  Irrigation  as a
source of water pollution.  Journal Water Pollu-
tion  Control Federation. 35:614-625.
   5. Nielsen,  D.  R., J.  M. Davidson, J. W.
Biggar and R. J. Miller. 1964. Water movement
through  Panoche clay loam soil. Hilgardia 35:
491-506.
   6. Oster, J. D. and R.  D. Ingvalson. 1967. In
situ measurement of soil salinity with a sensor.
Soil  Sci. Soc.  Amer. Proc. 31:572-574.
   7. Scofield, C. S. 1940.  Salt balance in ir-
rigated areas. Journal of Agricultural Research.
61:17-39.
   8. Tanji, K. K.,  L. D.  Doneen and J. L.
Paul. 1967. Quality  of  percolating waters.  III.
The  quality of waters percolating through strati-
fied substrate, as predicted by computer analy-
sis. Hilgardia 38:307-318.
   9.  Thomas, J. L., J. P. Riley and E. K. Isra-
elsen. 1971. A computer model of the quantity
and chemical quality of return flow. Utah Wa-
ter Res. Lab. PRWG 77-1, pp. 94.
   10.  U.S.  Department  of the Interior. 1969.
Characteristics and pollution problems of irriga-
tion return flow.  F.W.P.C.A.  Robert S.  Kerr
Water Research Center, Ada, Oklahoma.
   11. Van  Bavel, C. H. M.,  C.  G. Stirk  and
K. J.  Brust. 1968. Hydraulic properties of a clay
loam  soil and the field  measurement of water
uptake by roots: I. Interpretation of water con-
tent and pressure profiles. Soil Sci. Soc. Amer.
Proc. 32:310-317.
   12. Wilcox, L. V. 1962. Salinity caused by ir-
rigation. Journal American Water Works Assn.
54:217-222.
   13. Wilcox, L. V. and W. F. Resch. 1963. Salt
balance  and leaching requirement in irrigated
lands. U.S.D.A. Techn.  Bull.  1290.

-------
             Irrigation  Scheduling in  Idaho
                                   MARVIN E. JENSEN
                               Agricultural Research Service
                                      Kimberly, Idaho
 ABSTRACT
   Scheduling the timing and amount of irriga-
 tion water applied has changed very little dur-
 ing the last two decades. Professional irrigation
 scheduling services  are  rapidly expanding and
 many  of the companies are using a computer
 program developed by the USD A.  This program
 uses meteorological, soil and crop data to pre-
 dict irrigation dates and amounts. Service com-
 panies modify the program to fit their needs. Pe-
 riodic field inspections are provided by trained
 technicians as part  of scheduling service.  This
 paper  briefly describes the development of the
 computer program and the current status of its
 application in Idaho.
                Introduction
  In Idaho, as in most irrigated areas in the
West, irrigation  scheduling had  changed very
little during the last two decades. In the mean-
time, there have  been significant advancements
in irrigation science  and technology. Growers
stand to gain direct economic returns from bet-
ter irrigation scheduling by increased yields and
in some cases,  lower  irrigation  costs. Irrigation
districts whose operating costs are borne by the
farm owners and managers regularly encounter
unnecessary direct  and indirect operating  costs
that  can be attributed to poor irrigation  prac-
tices.
   Irrigation scheduling involves several impor-
tant components  (1) timing, (2) the amount of
water applied, (3) the flexibility of the system,
and (4) the  ability of farm management to re-
spond  to irrigation needs.  But why has there
been little improvement in irrigation schedul-
ing? Probably the most obvious reason why the
amount of irrigation water has not been con-
trolled more precisely is that water  measure-
ment is not  common on most of the older irri-
gated areas. Water  measurement on each field
is not a simple task with surface irrigation. The
amount of water  delivered  to the  farm can be
measured with relative ease, but losses as seep-
age in the distribution system, deep percolation
near the upper ends of the field, and surface
runoff are generally not known to the farm man-
ager. Also, the farm manager is not interested in
measuring the amount of water applied if there
is no obvious direct economic benefit.  He is not
interested in expending additional funds, time,
and often being inconvenienced measuring wa-
ter just to collect data. Low cost water, obsolete
irrigation systems and high labor costs are also
important factors that have contributed to lack
of change in irrigation scheduling practices.
  Large  quantities  of data are available that
show significant  reductions in crop  yield or
quality if irrigations are delayed, but  few data
show the economic losses that occur as a result
of excessive  water  applications. The natural
tendency under these circumstances is to apply
                                            181

-------
182    MANAGING IRRIGATED AGRICULTURE
excessive amounts of water to be sure that soil
water is not limiting. Another very significant
factor is that the cost of irrigation water in most
gravity  projects  is  very  low.  Indirect  costs
caused by excessive irrigations such as yield re-
ductions, soil erosion, and additional nitrogen
requirements are not easily recognized or quan-
tified. Crop and  soil damages  encountered on
lower lying areas some times are a result of ex-
cessive use on the upper areas,  but these losses
are not  borne by the upper-area irrigators.
   Another significant factor has contributed to
 slow change  in  irrigation practices.  Until re-
 cently  research and  technical  service agencies
 have not recognized or thoroughly evaluated the
 scheduling problems faced by managers of irri-
 gated farms. They have not adequately evalu-
 ated the acceptability of  the usual scheduling
 procedures that have been promoted for many
 years. In addition, the farm manager often is not
 aware that yield reductions are caused by delay-
 ing an irrigation even though excessive amounts
 of water may be applied at the next irrigation.
 The lack of change in irrigation scheduling tech-
 nology  also strongly implies that the  mechanism
 for providing current irrigation scheduling infor-
 mation to the farm manager  has not been satis-
 factory. Irrigation water scheduling proposals
 must be economical. Expensive or excessive soil
 water monitoring or irrigation scheduling tech-
 niques  that do not pay for themselves in  terms
 of direct and indirect benefits are questionable
 unless subsidized. We also must not forget that
 there is no substitute for proven irrigation expe-
 rience;  any irrigation scheduling technique only
 supplements, but does not  replace experience.

  Professional Irrigation Management Services
   Irrigation  scientists and technologists  know
 how to optimize crop production by manipulat-
 ing irrigation practices, but these specialists are
 not  making  the  irrigation decisions on  every
 farm. The farm  manager  is faced with  many
problems involving soils, crops, plant nutrition,
and climatic considerations, and cannot become
an expert in all of these areas. Instead of follow-
ing the traditional approach of training  every
farm manager to become a specialist in irriga-
tion water management and collect his own data
to make scheduling decisions, we decided to de-
velop some basic tools  that would eventually
provide professional irrigation water manage-
ment services to the farm manager if he so de-
sired. This approach  required a technique for
bringing together scientific irrigation principles
into a single package that could be applied to
every field.  Also,  procedures  were needed for
the farm manager and  an experienced profes-
sional to  maintain  field-by-field  communica-
tions. The system had to be economical, practi-
cal and dependable, and had to use research and
climatological data that were available, or could
be obtained easily. The system also  had to be
dynamic, responsive to the changing needs of
the farm manager,  and backed up by field in-
spections by experienced and trained personnel.
  One alternative was to promote irrigation wa-
ter management  services  using  existing  tools,
but this had been tried to a  limited extent in
other  areas  and  in  general  was not  being
adopted very rapidly. A number of tools and in-
struments,  such as tensiometers, soil moisture
blocks, evaporation pans, and soil  sampling
augers and tubes have been available commer-
cially to farm managers and farm management
service groups  for many years. We decided to
develop a  different  approach, an  approach
which would be very palatable  to the farm man-
ager and at the same time provide the type of
data that he needed and wanted. Pursuit of this
objective resulted in the USDA-ARS-SWC Irri-
gation Scheduling Computer Program.

   Development of Computerized Irrigation
           Scheduling Technology
  A systematic approach based on climate-crop-
soil  interactions was needed  to  provide  esti-
mates of the current soil water status by fields to
the farm manager along with predictions of fu-
ture changes and irrigation requirements. With
the recent  advances in meteorological science
and  computer  technology the  adaptation  of
computers  to  irrigation water scheduling  ap-
peared to be an attractive  approach to modern-
izing irrigation scheduling.

Development of the USDA-ARS-SWC
Computer Program
  The USDA Computer  Program was devel-
oped cooperatively with farm managers and ser-
vice groups to incorporate their reactions during
the formulation of the  program. It was devel-

-------
                                                             IRRIGATION SCHEDULING
                                         183
oped as a tool for providing managers of irri-
gated farms with scientific estimates of irriga-
tion needs for each field. The computer program
utilized  simple mathematical models and basic
equations initially so that limited  input data
could be used. The various components of the
program can be replaced as more accurate rela-
tionships are  developed  through research and
applied  technology. Briefly, meteorological, soil
and crop data are used to  predict future irriga-
tion dates and amounts. The principles and pro-
cedures  have been described in a number of re-
cent  publications and are  not  included  in this
paper4 5 6 7 8.
  The basic components  of the  irrigation sched-
uling  program and  service were evaluated  in
southern Idaho in 1966 and 1967. The complete
computer program and management services
were evaluated in cooperation  with  farm man-
agers and the  Idaho Extension Service  in 1968
and  1969.  About 50 fields were scheduled  in
Idaho during these two years and a similar num-
ber were scheduled in Arizona by the Salt River
Project3. Several service  groups and companies
gained experience using this concept of provid-
ing irrigation  management services in 1970 and
1971.
  The U.S. Bureau of Reclamation, in its fourth
year of a pilot  irrigation water management
study, modified the  computer  program to pro-
vide either general irrigation forecasts for sev-
eral major crops, soil types, and dates of plant-
ing within an  irrigation district, or field-by-field
scheduling. The general service was made avail-
able to  54 farm operators  in southern Idaho in
1970. The field-by-field scheduling and monitor-
ing  service was  provided  for  68 fields  on 15
farms within the same area! The present USER
general  program predicts irrigation dates for
major crops arid soils, but the printout to each
farm is  selected from a combination of six crops,
four soils, and three planting dates.  This proce-
dure was evaluated on 76 farms totalling 14,000
acres in 1971. With this procedure, the farmer
must verify the  predictions  and keep his own
records. The  field-by-field  method, involving
field inspections by trained professional or tech-
nical staff members, was evaluated on 14 farms
totaling 5,000 acres in 1971. The USER will be
providing  the  general  scheduling  service to
about 24,000  acres and field-by-field scheduling
services to about 30 farms and 150 fields total-
ing about 9,000 acres in 1972. The Bureau's full
time staff will consist of two engineers, and an
agronomist in 1972. The A & B Irrigation Dis-
trict at Ruper, Idaho and  the Falls Irrigation
District near American Falls, Idaho, will each
provide an agricultural technician2.
  A commerical  irrigation  management service
corporation  was  formed by potato growers in
southern Idaho in  1971. This company will be
operating in four general  regions in Idaho in
1972. A college trained  agriculturalist will su-
pervise field monitoring and inspections in each
region. This company anticipates providing irri-
gation scheduling services to about 40,000 acres
in 1972.
  About three years ago  another commercial
service company began  an irrigation  manage-
ment  service  in southern Idaho using tensi-
ometers to indicate the soil water status. One or
more terisiometers per  station are installed in
the grower's field. Each station represents from
10 to 20 acres. These instruments are  read about
twice a week early in the season and every other
day during the peak water use period. The ser-
vice company installs, maintains, and reads the
tensiometers.  It  then interprets  the  data  and
makes irrigation recommendations to  the farm
manager accordingly.  This  company serviced
approximately 10,000 acres in 1971 and hopes to
expand  its  service to  15,000-20,000  acres in
1972.
  Another company began providing irrigation
and plant nutrition services in cooperation with
a  sugar company  in southern  Idaho in  1971.
This service company conducts soil and tissue
analyses for plant nutrients  and provides  gen-
eral irrigation scheduling and field-by-field in-
spections in a cooperative study to determine if
sugarbeet yields  and quality can  be increased.
This study involved 52  test fields in  1971. Field
inspections by this company frequently revealed
another primary irrigation  problem.  In addition
to the timing of an irrigation, nonuniform distri-
bution of water within a field that is surface irri-
gated can greatly affect production.

Development of a Communications System
   Under the field-by-field  scheduling service, a
better system for obtaining up-to-date feedback
data  involving  the   date  and  approximate

-------
184
MANAGING IRRIGATED AGRICULTURE
amount of water applied is still needed. Where
center-pivot sprinkler systems are used, service
companies probably will install their own rain
gages to measure rain and applied irrigation wa-
ter. These gages are designed  to minimize evap-
oration so that  they can be  read on a weekly
basis.

         Expected Improvements and
             Expanded Services
  As  quantatitive data are developed  relating
plant growth to other parameters in addition to
soil water  these can be  incorporated  into the
scheduling program.  Improved mathematical
models that predict plant growth based on mi-
croclimate,  plant  characteristics, plant densi-
ties, soil  water, and  nutrients will be added to
the program when they have been  fully devel-
oped and tested. Simple  calibration techniques
probably will be needed to relate the  calcula-
tions  to  the  specific site conditions. Improve-
ments  in techniques for estimating evaporation
from the soil surface will be  added. These im-
provements, when combined  with effective leaf
resistance to water vapor diffusion which varies
as plant  cover  develops, will improve  the esti-
mates when a partial plant canopy exists.
  Yield-soil-water-climate  functions   for  all
stages of plant growth are needed to predict the
relative adverse effects of delaying irrigations or
excessive irrigations, and to optimize the timing
of limited irrigations. Eventually,  procedures
will be developed to predict the distribution of
water throughout the length of the field by tak-
ing into  account the size of stream, the rate of
advance  through the field and the  duration of
the set.
  Most of the service companies now using this
computer program are operating in  areas where
salinity is  not a  significant problem, or where
excessive  water is being applied.  Under  these
conditions there is no salinity hazard and no re-
sulting  build-up of salts  in the   soil. If the
amount  of water  applied is  known or can be
measured and the quality of water taken into
account, then soil salinity changes  can also be
predicted. If the amount of water applied is not
measured,  we recommend that companies pro-
viding irrigation management services  monitor
the salt  concentration in soils where salinity
problems are suspected.  If salts increase to ad-
                                         verse levels then additional irrigations or larger
                                         irrigations can be scheduled to leach the salts.

                                         CONCLUSIONS
                                           Irrigation   scheduling  practices  have  not
                                         changed appreciably in most western U.S. irri-
                                         gated areas during the last two decades. Low
                                         water costs, unacceptable new scheduling tech-
                                         niques, and the fact that adverse direct and in-
                                         direct  effects of excessive  water  use are not
                                         readily apparent are major factors  that have de-
                                         layed  improvements in irrigation scheduling.
                                         Scheduling irrigations for each field by  using
                                         meteorological, soil, and crop data coupled with
                                         field  inspection appears to be  an economical
                                         and acceptable irrigation management service
                                         in the USA. A professional irrigation manage-
                                         ment service requires technical competence and
                                         a modern communication network. Reliable me-
                                         teorological data are also needed.  Collection of
                                         these data  requires technical competence and
                                         periodic calibrations of the instruments. The
                                         USDA-ARS-SWC  irrigation scheduling pro-
                                         gram using meteorological techniques and soil-
                                         crop data, is  enabling Idaho private firms and
                                         service agencies to provide improved manage-
                                         ment  services  while  gaining  experience and
                                         while the program is being refined.

                                         REFERENCES
                                           1. Brown, R. J. and J.  F. Buchheim, "Water
                                         Scheduling in Southern Idaho, 'A  Progress Re-
                                         port',  USDA,  Bureau of Reclamation." Pre-
                                         sented at the National Conference on Water Re-
                                         sources Engineering, ASCE, Phoenix, Arizona,
                                         January 11-15, 1971.
                                           2. Buchheim, J. F. "Irrigation Scheduling on
                                         the Minidoka Project, Idaho." A paper pre-
                                         sented at the  Irrigation and Drainage Specialty
                                         Conference,  ASCE  and ASAE, Lincoln, Ne-
                                         braska, October 6-8, 1971.
                                           3. Franzoy,  C. E.,  and  E. L.  Tankersley,
                                         "Predicting Irrigations from Climatic Data and
                                         Soil  Parameters."  Trans. Amer. Soc.  Agric.
                                         Engrs. 13(16):814-816, 1970.
                                           4. Heermann, D.  F.,  and  M.  E. Jensen,
                                         "Adapting Meteorological Approaches in Irri-
                                         gation  Scheduling for High Rainfall Areas."
                                         Proc.  ASAE Natl. Irrig. Symp.  Proc., Lincoln,
                                         Nebraska, p. 00-1-00-10. November 10-13, 1970.

-------
                                                            IRRIGATION SCHEDULING
                                         185
  5.  Jensen,  M. E.   "Scheduling  Irrigations
Using Computers." J.  Soil and Water Conserv.
24(8): 193-195, 1969.

  6.  Jensen, M. E. and D. F. Heermann, "Me-
teorological Approaches to Irrigation  Schedul-
ing." ASAE Natl. Irrig. Symp. Proc., Lincoln,
Nebraska, p. NN-1 to NN-10, November 10-13,
1970.
  7.  Jensen,  M. E.,  J. L.  Wright,  and  B. J.
Pratt.  "Estimating  Soil  Moisture  Depletion
from Climate, Crop, and Soil Data." Amer. Soc.
Agr.  Eng.  Trans.  14(5):954-959,  Sept-Oct.
1971.
  8.  Jensen,  M. E.  "Programming Irrigation
for Greater Efficiency." Proc. Intl. Symp.  on
Soil-Water Physics and Technology, Rehovot,
Israel, Aug 29 to Sept. 4,  1971.

-------
                Irrigation  Scheduling  in  the

                           Salt  River  Project

                                           by
                      E.  WIN KYAW and DAVID S. WILSON, JR.
                           Salt River Project Phoenix, Arizona
 ABSTRACT
   The problem of scheduling irrigation within
 the Salt River Project area is approached via
 two methods which determine soil moisture lev-
 els and the evapotranspiration rates for differ-
 ent crops and soils. These methods are:
   A.  Repetitive field checks by trained special-
      ists.
   B.  A computerized program described by
      Jensen  which  uses  either the  Jensen-
      Haise or the modified Penman equations
     for potential evapotranspiration. This pro-
     gram schedules predictions using climato-
      logical data,  estimated E, for specified
      crops, and the last irrigation date.
   These'two methods are based upon study and
 experimentation. They are explained and evalu-
 ated in this report to illustrate the feasibility of
 the two concepts.


 INTRODUCTION
  The Salt River Project's  Agriculture Section
 was formed in the summer  of 1965. It was cre-
ated in response to the need for a technical unit
which could work with individual farmers to ad-
vise them  as to the optimum use of irrigation
 water for any particular crop growing in any
 particular soil type. In other words, these men
 are charged with advising when to irrigate and
 how much water to apply during each irrigation.
  The need for this service is derived from the
 fact that the Salt River Valley in central Arizona
 is located in an arid zone which receives only an
 average of 7.7 inches of rainfall each year. Dur-
 ing many  years there is even less rainfall and,
 yet, the Valley, by means of a  reservoir storage
 system, supplemented  by  groundwater pump-
 ing,  supports  over  238,000  Project-member
 acres. In 1971 alone, more than  730,000 acre
 feet of water was applied to Project lands. The
 long-term  average  annual  reservoir storage is
 only 978,800 acre feet. On two different occa-
 sions, storage has almost been depleted. In any
 particular  year, water use  may exceed actual
 runoff. This comparison between annual appli-
 cation and  historic storage  shows  the need for
 efficient water use.
  From  its  beginning in  1965 with  two em-
 ployees serving 14,000  acres,  the  Agriculture
 Section has grown to eight full-time specialists
working with 116 cooperators farming 65,000
acres. The varieties of crops include alfalfa, cit-
rus, corn, cotton, pasture grasses, small grains,
sugar beets and vegetables.
                                           187

-------
188     MANAGING IRRIGATED AGRICULTURE
  The service rendered has also grown. In addi-
tion to regular field visits and inspection, an ex-
perimental computer program has been added
with hopes of eventually providing irrigation in-
formation to more  cooperators on more acres,
faster and more efficiently.
  When  requested,  plant tissue  and/or  soil
samples are taken for analysis to provide nutri-
ent information. The soils information is used to
help determine the best initial fertility program.
The plant tissue results are utilized to indicate
any needed changes in fertilization as the crop
grows to maturity.
  Other  services  rendered  include   irrigation
system efficiency evaluations  and soil profile
checks for root development and moisture pene-
tration.
  The Agriculture Section's services  are avail-
able without charge to any Project-farming en-
terprise which requests them. To date, the  pro-
gram has been received enthusiastically.

    Field Approach to Irrigation Scheduling
  The method used by specialists in  the field is
to check the soil moisture within the effective
root zone of a crop, using an Oakfield-type soil
probe. Samples are taken  at one-foot incre-
ments and  checked by feel  to determine the
moisture content. This can  be estimated quite
accurately with experience. After checking, the
specialist estimates  the soil's moisture depletion
and predicts the time interval (number of days)
before the next irrigation is necessary.
  The knowledge of available  moisture in dif-
ferent soil types, the nature of the growing  con-
dition of plants and roots, and the knowledge of
actual root distribution of plants at any particu-
lar irrigation time is the primary skill all  field
specialists must have. The technical  experience
and capabilities gained by thorough study of the
Salt River Project area has made the program
work more efficiently.
  Weekly or biweekly inspection  visits are  well
received.  The farmer  usually accompanies the
specialist during these field  checks. This offers
him an opportunity to see how the soil moisture
determination is made and promotes confidence
in  the specialist's prediction.  These joint  field
visits also encourage the exchange of ideas and
discussion of other field problems. Through the
specialist, the farmer learns  what is being done
by neighbors and other valley farms in the way
of fertilization, cultural, irrigation and other op-
erational activities.
  This team-approach provides the opportunity
for better management decisions. However, the
final decision must be the fanner's since the risk
is his.

 Computer Approach to Irrigation Scheduling
  Significant advances have been made in the
science of irrigation-timing in recent years;  yet,
for many reasons, it remains an art practiced
successfully only through years of experience.1
  Deviation from accepted methods of irriga-
tion, such as by calendar  or fixed rotation, to
another method  developed by scientists, may
not immediately be acceptable  to  many farm-
ers. Getting used to such  a development takes
time and a close working relationship  between
the technician and the farmer.
  The basic components of computerized irriga-
tion management service  were  evaluated in
1966 and 1967  in southern  Idaho.  The com-
puter program approach has been further evalu-
ated in 1968-1969 on about 50 fields in Idaho
and on similar numbers of fields in Arizona by
the Salt  River Project. (Franzoy and  Tanker-
si ey—1970). Formulas for short-term predictions
of evapotranspirations based on climatological
data were developed by Jensen-Haise2 and Pen-
man.3 These  were programmed and  computer-
ized for experimental use by the Salt River Proj-
ect. Adjustments, new ideas, and revision of the
program were involved during the course of trial
runs without any change in the concept of the
basic ideas developed by Jensen. During 1970, a
number of service groups and companies gained
experience in the use of the general concept of
irrigation scheduling.4
  The  Jensen program was rewritten by Tank-
ersley* and, during 1968,  1969, 1970 and 1971,
was tested on 2,162 acres located on 19 farms in
the Salt River Valley. The primary goal of this
program was to evaluate the applicability of the
Jensen theory by using short-term estimates of
potential evapotranspiration (Etp), and  local cli-
matological data to predict the actual consump-
tive use  or evapotranspiration (Et)  of a crop,

*E. L. Tankersley, Engineer, Association  Engineer-
ing, Salt River Project.

-------
                                                              IRRIGATION SCHEDULING
                                          189
with the use of an individual  crop coefficient,
Kc. In addition to the Jensen-Haise estimates of
potential evapotranspiration, a set of modified
Penman  equations for predicting the potential
evapotranspiration (Etp) of individual crops has
been used to determine which set of equations
best fits local conditions.
  For study purposes, the Project area was di-
vided into 5 different regions. (Figure 1).  When
the program  was initiated,  19  representative
fields were selected and assigned a code num-
ber.
 Figure  1: Geographic Regions Included in  The
 Irrigation Timing Study

  Representative crops  such as cotton, sugar
beets, barley, wheat, citrus, and maize, were se-
lected as initial crops in the study. The variety of
crops will be increased as the program is  per-
fected for operational use and realistic estimates
of Et are obtained.
  Studies of consumptive use,  both daily  and
yearly,  have been  conducted  by Erie in  Ari-
zona.5 Erie's data, derived from repeated gravi-
metric  determinations of  soil moisture deple-
tion, was used as a basis for checking the com-
puter prediction of potential  evapotranspiration
(Etp) and actual evapotranspiration  (Et) for the
various  crops. The predictions  have also  been
checked by field estimates throughout the study
period.
  Based on the  Jensen-Haise and Penman for-
mulas, the computer predicts an Etp which rep-
resents  the  upper  limit  or  maximum evapo-
transpiration that can  occur under given cli-
matic conditions for a  particular  crop. Then,
using the crop  coefficient  curves originated by
Jensen, an Et for a particular crop can be esti-
mated.  Because Jensen's curves were  developed
from data collected in several areas, some ad-
justments were made to make  the  curves fit
local valley conditions. Presently, Et predictions
on all crops are satisfactory.
   Data  received on  cotton and sugar  beets,
comparing the Jensen-Haise method, modified
Penman method and the actual  average  gravi-
metric analysis of Erie are illustrated  in Figures
2, 3, 4 and 5. As shown in these comparative
grphas, Et calculated by using the Jensen-Haise
and  Penman  methods, varies with climatologi-
cal changes.  This is  acceptable in view of the
actual changes that occur in consumptive  use.
Also the Jensen-Haise  method  of  prediction
shows closer agreement to Erie's  8-year average
results  on cotton than  the  Penman equation.
Penman's data showed slightly higher Et's most
of the time; therefore, preference was given to
the use of the Jensen-Haise method in the  Salt
River Project.
   In the process of implementing the computer
program, two types of data were used; meteoro-
logical data and field data.
                                                           *:* OXLECTCT J.T AIRPORT (PfOEMJl, JLl >
      	v	f-y	.
    \    X      SS    -->
                   Figure 2.
   i«U=i '-..• is": 7i_i-: '. . - -• f'S avtiACt
                yp ,•'•.:£ t;-«:s£
       i P-"  -    •~^.-« •••!!' ai.I- '>*«;
                   Figure 3.

-------
190
       MANAGING IRRIGATED AGRICULTURE
        ' -
       •to*- :- .* vs=tL*i>. -1 :»E'S *»'!.«.*,-•:
       ;f.S. -*-Ilit uSI CJhvI "O -KAT O* ,IS'_l\
       Ct A^tl*»i^-C:ONS F&.T.-C "UM 1ST: 3*
       L-Jku.-QUtt.Ctl. l*'a ;&-:.£.":.& 4: G. JEST. 4
        APR  fcUY
                  Figure 4.
       ~  I- ' -       - W :«•*.:;£
               - . -         i-

           A .
                 .,-.   ._.   AJC  - St">T
                  Figure 5.

  1.  Meteorological data consists of the follow-
     ing:
     A. Daily average temperature.
     B. Solar radiation.
     C. Dew point temperature.
     D. Wind speed for each day.
     E. Saturated vapor pressure.
     F. Vapor pressure at dew point.)
  (D, E & F are computed from air and dew point
  temperatures.)
  Data was collected by the National Weather
Service which is located at Sky Harbor Interna-
tional  Airport,  Phoenix. Some additional tem-
perature and precipitation data was supplied by
a number of farmers who telephoned reports to
the Project every morning. These farmers were
provided with thermometers and rain gauges by
the Project to ensure more accurate local clima-
tological data.
  2.  Field data  was collected by the field spe-
cialists for each of the  19 program fields.
     A. Date of the last irrigation.
     B. Optimum depletion  of water from the
        current root zone.
     C. Planting date.
                                                      D. Harvesting date.
                                                      E. Rainfall.
                                                      F. Estimated irrigation efficiency.
                                                   To  correlate   the  estimation  of  potential
                                                 evapotranspiration  (based  on  meteorological
                                                 and  field data) to actual consumptive use of a
                                                 certain  crop,  Jensen's  set  of  crop coefficient
                                                 curves were used.
                                                   Using this input data, the  computer prints out
                                                 a report containing:  1) Etp  by region; 2) Et for
                                                 each  crop  each  day;  3) estimated  number of
                                                 days before next irrigation;  4) amount of water
                                                 to apply at that particular time; and 5) estima-
                                                 tion of the next seven day's Et.  Figure  6 ex-
                                                 plains  the  field  numbering  code  used  and
                                                 sample printouts are  shown in Figures 7, 8, 9,
                                                 and  10. This  report utilizes the best moisture-
                                                 depletion   estimate  for the  print-out  period
                                                 based  upon the  type  of soil and extent of root
                                                 development.

                                                   Evaluation of the Two Methods (Field and
                                                                 Computerized)
                                                   Scheduling  irrigations with a computer is an
                                                 excellent supplement  to, but not a replacement
                                                 for,  regular field visits. Frequent monitoring of
                                                 moisture   disappearance   and   replenishment
                                                 through field  observations by experienced  per-
                                                 sonnel is an essential process for detecting vari-
                                                 ations in actual evapotranspiration and in main-
                                                 taining accurate predictions.1
                                                   When the program can be relied upon to pre-
                                                 dict  Et accurately, the computer printout can be
                                                 an aid to the  field  specialist or technician. Be-
                                                 fore going  out to the field,  he can refer to the
                                                 printout to see how the moisture level for a par-
                                                 ticular field and  crop should have been depleted
                                                 and how much loss in Et occurs per day. Fluctu-
                                                 ations in Et can occur daily with variation in the
                                                 climatological conditions. This review will help
                                                 the specialist decide more accurately how many
                                                 days  remain until optimum soil depletion oc-
                                                 curs. Another added advantage of computer use
                                                 is that it will enable the specialist to work more
                                                 efficiently with a greater number of fields. This
                                                 can be accomplished  since trips to the field can
                                                 be scheduled  as  the  computer-printed forecast
                                                 data indicates that  soil moisture is  nearing de-
                                                 pletion in the  root zone.
                                                   Regarding  operational purposes,  a general
                                                 picture of  the  irrigation  requirement of the

-------
                                                                IRRIGATION SCHEDULING
                                                                     191
            The following is the code identifying the fields and crops used in the consumptive
            use and irrigation scheduling study:
            A BCD E FG
            A
            BCD
            FG
            GENERAL
            CROP CODE
       The first digit represents a number assigned the specialist.
       The second through fourth digits are selected at random and
       represent the field number.
       The fifth digit  represents the region  in which the field is
       located.
           1 Peoria
           2 Tolleson
           3 Laveen
           4 Chandler
           5 Gilbert
       The sixth and seventh digits represent the crop grown on the
       field.  Each series of tens will represent a given group of
       crops. For example, all cotton will be represented by Tens, all
       grains by the twenties, etc.
            10
            20
            30
            40
            SPECIFIC
            CROPCODE
Cotton
Small Grain & Forage
Hay Crops
Oil Crops
            10
            11
            12
            20
            21
            22
            23
            24
            30
            40
Delta Pine Cotton
Acala Cotton
Pima Cotton
Wheat
Barley
Grain Sorghum
Corn
Oats
Alfalfa
Safflower
50
60
70
80
50
51
52
53
60
70
71
72
73
74
Citrus
Sugar Beets
Vegetables
Green Manure Crops
Grapefruit
Lemons
Oranges
Tangerines
Sugar Beets
Potatoes
Lettuce
Carrots
Onions
Sweet Corn
                                   Figure 6: Field Number Code
whole area can be predetermined on a unit-time
basis (weekly, for example). This can be a real
aid in.sequencing water deliveries within the ir-
rigation delivery system.

                  Discussion
  The past four years of study indicate that lo-
cal experience in  characterizing the  different
crops under variable growing conditions is very
essential.  Differences  in climatic  factors, soil
capabilities,  growth characteristics of  crops,
farming habits of growers,  agricultural opera-
tions, fertilization and  determination of timely
irrigation  require the  experience and skills  of
                          trained men making regular field checks to cor-
                          relate field  conditions and  use them in a com-
                          puterized prediction approach.
                            During the  course of the  study, it was discov-
                          ered that the  growth period to maturity varies
                          considerably in response to changing  climato-
                          logical conditions. Also the  methods for estimat-
                          ing the correct adjustment for crop coefficient
                          curves to meet local conditions and crops needs
                          careful  study.  Wrong adjustments for days to
                          effective cover  on crop coefficient curves will
                          produce erroneous Et values; either low or high.
                            For all of the crops  grown in the Salt River
                          Project area,  the Jensen-Haise predictive equa-

-------
REGION = TOLLESON
                                 RAIN      ESTIMATED         ESTIMATED
FIELD                 LAST  SINCE LAST  CONSUMPTIVE    DEPLETION SINCE  OPTIMUM  APPROX. DAYS APPROX AMT
   NO.      CROP      IRRIG    IRRIG    USE NEXT 7 DAYS      LAST IRRIG    DEPLETION  BEFORE IRRIG  TO APPLY
                                                                                                              (IN.)
5996260   SUGARBEETS 12/28/71

6621260   SUGARBEETS  1/26/72

6112260   SUGARBEETS 12/28/71
(IN.)
0.0
0.0
0.0
(IN.)
0.34
0.34
0.34
(IN.)
2.20
0.41
2.20
1 (IN.)
2.50
2.50
2.50
                                         Figure 7: Irrigation Timing Report for 2/3112
                                                                                                     3.39

                                                                                                     0.64

                                                                                                     3.67
o
i— »
o
53
TO
o
                                                                                                               o
                                                                                                               2
                                                                                                               o
 REGION = TOLLESON
  FIELD
    NO.

 5996260
 6621260
 6112260
SUGARBEETS AKC =
             ET =

SUGARBEETS AKC =
             ET =

SUGARBEETS AKC =
             ET =
                     1/27   1/28   1/29   1/30   1/31    2/1   2/2   2/3  2/4   2/5   2/6   2/7   2/8   2/9
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.07
0.94
0.07
0.94
0.07
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.04
0.94
0.04
0.94
0.04
0.94
0.06
0.94
0.06
0.94
0.06
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
0.94
0.05
                       Figure 8: Table of Daily Crop Coefficients and Consumptive use Values for 1/27/72 thru 2/9/72
                                                                                                                         73
                                                                                                                         m

-------
                                                             IRRIGATION SCHEDULING
                                                          193
REGION = TOLLESON
        CROP

    SUGARBEETS
    WHEAT
                     TOTAL CONSUMPTIVE               CONSUMPTIVE USE
                         USE FOR PAST                          FOR
                              WEEK           2/3    2/42/5    2/6   2/7   2/8   2/9
                              (IN.)
0.46
0.09
0.06
0.01
0.06
0.01
0.06
0.01
0.06
0.01
0.06
0.01
0.06
0.01
0.06
0.01
                     Figure 9: Consumptive use Table for 1/27/72 thru 2/9/72
REGION = TOLLESON

                                      VAPOR
                              SAT.    PRESS.                EVAPO-TRANS.  EVAPO-TRANS.
        AVG.  SOLAR AVG.   VAPOR, AT DEW  NET  HEAT  POTENTIAL    POTENTIAL
 DA TE TEMP.  RAD.   WIND  PRESS.    PT.    RAD.  FLUX (JENSEN-HAISE)   (PENMAN)

1/27/72  49.0   331.

1/28/72  47.0   311.

1/29/72  47.0   326.

1/30/72  51.0   314.

1/31/72  45.5   348.

2/ 1/72  44.0   264.

21 2/72  46.0   368.

21 3/72  42.0   320.

21 4/72  42.0   320.

21 5/72  42.0   320.

21 7/72  42.0   320.

2/ 7/72  42.0   320.

21 8/72  42.0   320.

2/ 9/72  42.0   320.

          Figure 10: Prediction of Etp based on Weather Calculations for 1/27/72 thru 2/9/72
3.8
4.7
6.3
5.9
6.8
7.1
6.2
6.3
6.3
6.3
6.3
6.3
6.3
6.3
14.0
13.6
13.6
15.1
13.0
11.9
13.4
11.5
11.5
11.5
11.5
11.5
11.5
11.5
5.4
4.0
4.0
3.1
1.4
2.7
-0.1
3.4
3.4
3.4
3.4
3.4
3.4
3.4
148.7
139.8
145.8
136.3
138.8
130.2
119.8
154.0
155.6
157.2
158.7
160.1
161.6
163.0
-11.7
-16.7
- 6.7
16.7
-14.2
-19.2
- 4.2
-15.8
-10.0
- 6.7
- 0.0
- 0.0
- 0.0
- 0.0
0.07
0.06
0.06
0.07
0.06
0.05
0.07
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.13
0.15
0.17
0.17
0.20
0.17
0.20
0.16
0.16
0.16
0.15
0.15
0.15
0.16

-------
194
MANAGING IRRIGATED AGRICULTURE
tions have given better estimates of Et than the
modified Penman  method. The latter's predic-
tions  were slightly high  for  all seasons.  An
added  advantage  in  using the  Jensen-Haise
method is  that less input information  is re-
quired.

CONCLUSION
  As  a result  of  the knowledge gained from
study and experience, accuracy in predicting ir-
rigation dates and the personal rapport  devel-
oped from working together as a team, the Proj-
ect's Agriculture Specialists  have gained  the
confidence of most of the farmers in the Valley.
These farmers have come to rely on the techni-
cal  advice in irrigation and water management
and the suggestions on soil and  plant fertility
provided by the specialists.
  A forecasting program built around Jensen's
concept for predicting evapotranspiration  has
been  initiated  as  the second  phase  of imple-
menting a practical agricultural water manage-
ment program in the Salt River Project area.
  The combination of field checks and comput-
erized information is intended to maximize the
efficient use of water by confining water  appli-
cation to just that amount needed to satisfy con-
sumptive use and  leaching requirements. This
will offer the farmer the opportunity to produce
crops less expensively through better and more
efficient management of water.
  Another  benefit of the agricultural  water
management  program  is the  potential  for im-
provement  in operational services, ie.,  actual
water delivery because the forecasting program
                                         can indicate water delivery demands as much as
                                         two weeks  in advance of actual need. This can
                                         be important when scheduling water  through a
                                         system of canals which serves 132,000  agricul-
                                         tural acres and 106,000 urban acres.
                                           In addition, preliminary  studies to correlate
                                         soil moisture to soil temperature, using remote
                                         sensory techniques,  are being  conducted  by
                                         NASA and other agencies.  In the future, if re-
                                         mote sensing becomes a practicability in irriga-
                                         tion scheduling, present computer programming
                                         may be adapted to use this additional source of
                                         data to  correlate  soil  moisture to  irrigation
                                         scheduling,  based  upon plant  crop require-
                                         ments.

                                         REFERENCES:
                                           1. C. E. Franzoy and E. L. Tankersley, "Pre-
                                         dicting Irrigations from Climatic Data and Soil
                                         Parameters,"  presented  at  the  1969  Winter
                                         Meeting,  A.S.A.E.
                                           2. M. E.  Jensen,  J. L.  Wright,  and  B. J.
                                         Pratt,  "Estimating  Soil Moisture  Depletion
                                         from Climatic, Crop, and Soil Data," presented
                                         at the 1969 Winter Meeting, A.S.A.E.
                                           3. H. L.  Penman, "The Physical Basis  of Ir-
                                         rigation Control," Proc. Intern. Hort. Congr. 13:
                                         913-924, London, England,  1952.
                                           4. M. E.  Jensen,  "Programming  Irrigation
                                         for Greater Efficiency," unpublished, May 1971.
                                           5. L. J. Erie, Orrin F. French, and Karl Har-
                                         ris,  "Consumptive  Use  of Water by Crops  in
                                         Arizona," University of Arizona Technical Bul-
                                         letin 169, reprinted August 1968.

-------
                Irrigation Scheduling Studies

            for Water  Quality  Improvement

                     RAY S. BENNETT and JAMES H. TAYLOR
                          Agricultural Engineering Department
                               Colorado State University
                               Grand Junction, Colorado
 ABSTRACT
   Irrigation  scheduling  has  resulted in  im-
 proved crop yields for the few areas in the West
 that have been studied to date. Irrigation sched-
 uling studies are underway in Grand Valley as
 a salinity control measure. By minimizing deep
 percolation losses resulting from irrigation,  the
 amount of salt pickup by the moisture move-
 ment through the soil profile and over Mancos
 Shale beds will also be reduced, thereby reduc-
 ing  the  salt  load  in  the Colorado  River.
   The hydro-salinity  model previously devel-
 oped for  the  Grand Valley  Salinity Control
 Demonstration Project will be used for evalu-
 ating the effectiveness of irrigation scheduling
for salinity control.  A proposed study is  de-
 scribed for  developing  more  sophisticated
 models regarding moisture  movement through
 the'soil profile and  the associated  chemical
 quality changes.  The development of such a
 model would  significantly enhance  the pre-
 dictive ability for evaluating proposed salinity
 control measures.
INTRODUCTION
  The  most significant improvements in con-
trolling irrigation return  flow quality will po-
tentially  come  from  improved  on-the-farm
water  management.  Irrigation practices  on
the farm are the primary source of present re-
turn flow quality problems.
  In order to maintain  an irrigated agricul-
ture, a salt balance  must be maintained  in
the root zone. This requires that salts applied
to the land by irrigation water must be leached
through the root zone. Thus, the  leaching re-
quirement represents the minimum quantity of
deep percolation losses which must be allowed
to sustain  irrigated agriculture. Therefore, if
high irrigation application efficiencies were  to
be achieved, there would still be some deep
percolation losses.
  Many studies have shown that in most irri-
gated areas the amount of water applied and
the timing  of this water  application  are quite
random. For example, oftimes when the farmer
finds that  his  field is dry, he will irrigate, but
the irrigation application  may be more than is
really needed by the crop. Thus a two-fold prob-
lem occurs where  the plant  has already been
stressed because of the  field being  too  dry,
which means  that  the yield  has already been
reduced. The  second problem is due to  more
water being applied than  was really necessary.
In extreme cases,  this might even lead to a
problem of reduced aeration of the soil.
                                          195

-------
196     MANAGING IRRIGATED AGRICULTURE
  One of the more interesting areas of water
management control presently being explored
is  that of optimum irrigation scheduling.  The
purpose of  irrigation scheduling is to advise a
farmer when to irrigate and how much water
should be used.  Primarily, a farmer relies on
visual  indications of crop response to decide
when to irrigate, or he may have to irrigate  on a
fixed water  rotation system. Irrigation sched-
uling is  geared towards  taking  soil moisture
measurements,  along with computing potential
consumptive use for the crops being grown, to
determine when  to irrigate  and  the quantity
of water to be applied.
  The reason  that irrigation  scheduling  has
become successful is because the measurements
are  being  made  as  a   service  to  farmers.
This has saved the farmer the effort of going
out and making these same measurements  him-
self at a time when he  is concerned with many
farm operations.  Because of the busy schedule
of the farmer, and the  difficulty he might  have
in the initial  interpretation of the data, the
problem of  irrigation scheduling has met  with
little  success in the past. The efforts in Idaho
and  Arizona look extremely  promising and the
farmers are  claiming a  significant benefit from
irrigation scheduling.  Yields  have  been in-
creased due to the fact that  water was applied
when needed rather than after, the crops  were
stressed.  Another benefit to the  farmer from
this  program is that he can anticipate the dates
when  irrigation  is  to  be accomplished.  This
allows him to schedule  irrigation along with the
other duties that must be performed on the farm
and  relieves  him  of the responsibility of decid-
ing exactly when  is the  best time to irrigate.
  The limited amount  of irrigation scheduling
which has taken place to date has not been con-
cerned with  controlling the quality of irrigation
return flows. Thus, there is a real need for  such
a demonstration project. Also, it should be rec-
ognized  that irrigation scheduling  has many
positive economic  benefits to farmers. At the
same time, this particular control measure over-
comes certain  institutional constraints  regard-
ing improved water use efficiency on the farm.
   Irrigation  scheduling is a  positive economic
incentive for improving water management be-
cause of increased crop production. This tech-
nique  for  improving irrigation  practices  has
the advantage of partially overcoming the nega-
tive  aspects  of western  water  laws.  Although
the irrigation scheduling reported  to  date does
not cite any reduction in total water use, this
reduction  will take place as more experience is
gained by the farmer regarding  soil moisture
management in the root  zone of his croplands.
Even so, the farmer may still wish to divert his
full water right. Again, if the farmer could re-
ceive an economic return by being allowed to
transfer the saved water, then  the  improve-
ments in  irrigation water use  efficiency could
be brought about at a much faster pace.

               Planned Studies
  For the  last three years,  the  Agricultural
Engineering  Department of   Colorado  State
University has been conducting studies aimed
at reducing the salinity of the Colorado River.
The primary objective is to demonstrate that an
improvement in mineral quality of the Colorado
River can be achieved by controlling the inflow
of  highly  mineralized  water from  aquifers
underlying the irrigated  farm  lands  in  Grand
Valley.  The  initial grant,  which  was recently
completed, evaluates the effectiveness of canal
and lateral lining in reducing seepage  losses,
thereby lowering  the piezometric head which
causes displacement of  highly saline ground-
water into the Colorado River. At the  present
time, these studies are being  extended to in-
clude an evaluation of the effectiveness of two
other salinity control measures,  namely  irri-
gation scheduling  and tile drainage, which will
also lower the piezometric head of the ground-
water aquifers.
  Secondary  objectives  are   to  demonstrate
that irrigation scheduling can be effective  in
achieving greater  water use efficiency on the
farm along. with  increased crop  yields, while
tile  drainage will  be shown to be effective in
reclaiming lands  now  experiencing  low agri-
cultural productivity  due to high groundwater
tables resulting from canal and lateral  seepage
losses, as well as deep  percolation  losses, on
higher lands. For  those lands incorporating tile
drainage,   irrigation  scheduling  will also  be
practiced  in  order  to  minimize  the drainage
problem.
  Water  entering the near-surface aquifers in
Grand  Valley  displaces  highly  mineralized

-------
                                                      IRRIGATION SCHEDULING STUDIES     197
waters from these  aquifers into the Colorado
River. In any area where the water is in pro-
longed contact with soil, the concentration of
mineral  salts  tends toward a  chemical equi-
librium with the soil. In  Grand  Valley, as in
many other areas, high equilibrium salinity con-
centrations  are known to exist in the near sur-
face aquifer. The key  to achieving a reduction
in salt loading is  to lower the groundwater
levels, which  will  result in less  displacement
of water from the aquifer into the Colorado
River. This can  be accomplished by reducing
seepage  losses through canal and  lateral lining,
which is presently  being demonstrated,  or by
reducing deep  percolation  losses by improved
on-the-farm water management. Present studies
advocate irrigation scheduling  as  a salinity
control  measure which  will achieve improved
farm water management.  Also, a combination
of irrigation scheduling and tile  drainage will
be demonstrated in the  lower-lying high water
table agricultural lands.
  The primary  local  benefit will be increased
crop yields. Other water quality benefits to the
local area will be derived  from the lowering of
the  groundwater table, which will increase the
productivity of local  agriculture in areas pres-
ently having a very high water table, as well as
reduce  non-beneficial evaporative  losses of
water through phreatophytic vegetation.  Also,
the  value of lands presently suffering from high
groundwater tables will be increased. Addition-
al local  benefits will be obtained by elimination
of nuisance problems  such aSjSewer infiltration,
basement flooding, and localized swamps which
lead to  public health  problems associated with
the  production of mosquitoes.
  The principal study area  in  Grand Valley,
which has been used  for demonstrating the ef-
fectiveness  of canal  and  lateral lining  in re-
ducing  the  salt load entering the  Colorado
River,' is also being  used to demonstrate the
effectiveness of  irrigation scheduling and tile
drainage as salinity control measures. The ad-
vantage  in  continuing to utilize this study area
is that the  hydrology is already known. In ad-
dition, there has been considerable expenditure
of  funds in both equipment and personnel for
instrumenting   this  particular  demonstration
area. The wealth of available information pro-
vides a  strong basis for evaluating the effects of
either irrigation scheduling or tile drainage as
salinity control  measures in comparison with
lining the irrigation conveyance and distribution
system.
  At least two fields will be selected for study-
ing tile  drainage. These  fields  will be located
approximately 1 mile  north  of the Colorado
River. In this area, the soils are very tight and
high  groundwater  levels already  exist.  As  a
result, many  of  the lands  near  the river have
suffered  economic  damage due  to low agri-
cultural  productivity. Many of these same lands
were among the most productive in the valley at
the turn  of the century. Also, in this region some
drainage problems  occur as a result of a leak in
the  cobble aquifer with consequent  upward
moisture movement. Due to these difficulties,
designing  a tile  drainage system  can  be very
costly.
   In order to effectively demonstrate the  ad-
vantages of tile  drainage to a local fanner,  it
becomes essential  that heavy  consideration be
placed upon costs.  Therefore, the plan of attack
will be  to thoroughly investigate a number of
likely fields, say four, which would benefit from
tile drainage.  A one-year pre-evaluation of these
fields will be  undertaken to provide detailed in-
formation for designing a tile drainage system.
These investigations will have  the benefit of
minimizing the construction  cost of installing
tile drainage for  each field.
   Due to high groundwater levels and tight soils
in  the region for  which tile  drainage  will be
demonstrated, the  reclamation  of these lands
will  require very  careful farm  water manage-
ment. Therefore, the fields  which will be select-
ed for installing tile drainage will also become a
part of the irrigation scheduling studies.
   Four  farms have previously  been investigated
in the principal  study area as a part of the ir-
rigation efficiency  evaluations.  Two of these
farms have been studied more intensively than
the other two.  These  four farms will be in-
corporated into  the  evaluation  of irrigation
 scheduling as a salinity control measure. These
farms, along with the  fields being investigated
for  tile  drainage, will serve as the base for col-
lecting detailed information on  salinity control
 benefits. During the second year of study, the
 irrigation  scheduling  program will be greatly
 expanded to  include large blocks of land in the
 principal study area.

-------
 198
MANAGING IRRIGATED AGRICULTURE
    All of the irrigation scheduling to date, which
 has  been  quite  limited, has  been  concerned
 primarily with  the timing of  water delivery.
 This demonstration project  is concerned with
 minimizing the quantity of water used,  as well
 as timing, along with minimizing deep percola-
 tion losses because of the consequent high rates
 of salt  pickup in  Grand Valley.  As a  conse-
 quence,  the effort required to evaluate irriga-
 tion scheduling as a salinity control measure is
 considerably  greater than  previous  irrigation
 scheduling efforts.

           Water Quality Evaluation
   A  hydro-salinity  model,   which is  dis-
 cussed in a later paper, has been developed for
                                         the demonstration area in Grand Valley.  This
                                         model will allow an evaluation of any proposed
                                         salinity control measure.  However,  this model
                                         describes the present physical conditions, which
                                         results in one severe limitation. The model as-
                                         sumes that a 50% reduction  in ground water
                                         flows would result in a corresponding 50% re-
                                         duction in salt pickup. If deep percolation losses
                                         were significantly reduced, the  reduction in salt
                                         loading reaching  the Colorado River could be
                                         determined  if field measurements were collect-
                                         ed over a sufficiently long time period.
                                           The hydro-salinity  flow system depicted  in
                                         Figure 1 can be divided into an examination of
                                         the water and salt flows  even though the salt
                                         flow is  located within  the water transport net-
                                            :
                                                                                      Stream
                                                                                       System
                                             oncos
      Figure 1:  Schematic Representation of Flow System in The Grand Valley Demonstration Area
work.  There is,  however, one  portion  of the
salinity system, which is  the salt  pickup from
soils and subsurface contacts,  that has been
determined by  independent analysis for incor-
poration in the model.
  The  computation of the water budgets is the
first essential in  hydro-salinity modeling  be-
cause salt flows  and water quality in general
are dependent on the magnitude and  distribu-
tion of water in  the  hydrologic system. In all
but the most limited cases, water budgeting re-
quires  some  adjustment to formulated data to
compensate   for   instrumentation  limitations,
data accuracy, and interaction among variables.
A  comparison of mass balance  and direct cal-
culation methods  is made on ground water flows
                                         as a basis for model refinement. It seems justi-
                                         fied to conclude that this procedure gives an ad-
                                         ditional degree of confidence to the results.
                                          The first step in computing the budgets is to
                                         evaluate the distribution of the water before it
                                         reaches the root zone. Water is initially diverted
                                         from the two.rivers into the various canals from
                                         which  a small  percentage is lost from  seepage,
                                         some is spilled  into washes or drains to regulate
                                         capacity, and the  bulk  is diverted into  the field
                                         lateral system.  In the lateral network,  some of
                                         these flows are lost by seepage and the rest is
                                         applied to  the cropland. In the Grand Valley
                                         irrigation  systems, a  large proportion of  the
                                         field applications eventually  wind up on  the
                                         drainage  system  as  field  tailwater, with  the

-------
                                                     IRRIGATION SCHEDULING STUDIES    199
remaining portions being  placed in the root
zone, along with precipitation, for crop use.
  Once in the root zone, opportunity exists for
consumptive use by crops. Under favorable con-
ditions, the  water utilized  from the root zone
will  approximate the crops' potential demand.
The  potential evapotranspiration from cropland
surfaces  in  the  Grand Valley has been com-
puted using the  modified Blaney-Criddle meth-
od. The excess water in the root zone that can-
not be stored percolates through the soil profile
until reaching the ground water table. Since the
water table is not always at the bottom bound-
ary of the root zone, a region of flow between
this  boundary and the water table should be
considered.  For purposes of this study, this
region (soil  profile)  has  been assumed com-
pletely saturated and  has been ignored initially.
This limitation can only be overcome by con-
sidering  unsaturated  flow and  the associated
chemical changes resulting from  ion exchange,
precipitation, etc.
  The ground water additions are comprised of
canal and lateral seepage and deep percolation
from  excessive  irrigation.  Occasionally,  the
amount of precipitation is sufficient to account
for some inflows to the ground water system. A
certain proportion of ground  water is either in-
tercepted by the drainage  system or transpired
by  phreatophytes. In this model, it has been
assumed that phreatophytes use  water at their
potential  rate and  use all  precipitation that
falls on the  phreatophyte acreages.  Because
of the fluctuating nature of the ground water
additions, the storage within  the ground water
basin  also varies significantly throughout the
year. In budget computations, a negative stor-
age  change  simply  indicates  that more flows
are  leaving  the  subsurface  basin  than  are
entering. The remaining water is forced to flow
towards the river by the hydraulic gradients
acting on the water from elevation differences
in the  local topography.  These flows  are re-
ferred to in the model as the total ground water
outflows and are then used to evaluate the per-
formance of the entire model.
  Although the salinity flow system  generally
follows the water flows,  there are three addi-
tional somewhat complicating factors that must
be considered. The first of these is the ion ex-
change process  which occurs  in  the root zone.
Since the scope of this project did not allow for
such an extensive examination, this factor has
been ignored. It should be emphasized that this
information is very  important to understanding
salinity  problems in agricultural areas. The
The second  aspect which must be considered is
the quantity of salts actually imparted to the
water from  the  soils themselves.  The distribu-
tion of this "salt pickup" quantity is not known
and was assumed to occur only in the ground
water basin. This conclusion is obviously false,
but does  not  interfere with the  scope  of this
modeling effort. Again, the knowledge of where
the salt is being added is of the utmost impor-
tance to salinity control planning. (For example,
if it was known that little salt is actually added
in the root  zone, then the interception of deep
percolation  by a system of field tile drainage is
a feasible salinity control alternative.) The third
and final  component of the independent vari-
ables  affecting  salt movements  is related  to
the salt changes that occur as a result of stor-
age changes within the ground water  basin.
When a large influx of water is added via deep
percolation  or  conveyance seepage, this water
has not as yet  picked up much salt due to the
short time it has been in contact with the soils.
As a  result, the  large storage increases in the
spring and  early summer result  in only small
salt flow  to the ground water basin. On the
other hand, when the large storage  decreases
occur in the late fall and early winter, a rela-
tively  large salt depletion  occurs  from the
system.  This  particular parameter   regarding
the salinity flow has been handled  in the model.

      Proposed Evaluation Improvements
  The greatest  single technological need at the
present  time for the subject area of  irrigation-
return flow quality is the development  of pre-
diction  techniques  which  will  describe the
quantity and quality of subsurface return flow.
The  real critical problem is defining the vari-
ability  in  subsurface  return  flows   for  large
areas, such as an  irrigated valley or a  large
portion of the valley (e.g., Grand  Valley).
  In  studying large areas,  a balance must be
reached  between the  sophistication  of the
model and  the  cost of collecting field  data.
There are still  a number of limitations in our
ability to make  accurate field measurements.

-------
200    MANAGING IRRIGATED AGRICULTURE
These  models  must  be capable  of  predict-
ing changes in the quantity and quality of sub-
surface  return flows under a variety of water
management alternatives.  To  fully evaluate
chemical quality changes,  the  models  should
be  capable  of  handling precipitation and ex-
change  reactions  which  take  place as  the
moisture moves through the soil profile.  These
transformations  alter  the ionic balance of the
chemical constituents  in solution and are very
important  in describing the quality of irriga-
tion return flow.
  The development of such a model for Grand
Valley  would answer the  question:  "If  deep
percolation  and seepage losses  are  reduced by
a particular amount, what is the corresponding
reduction  in  salt ~ loading  in  the Colorado
River?" To begin  with, detailed studies  must
be  undertaken  to  delineate  the relationships
between moisture  movement   and  chemical
quality. Then, the  results  can be applied to a
much larger area, such as the  demonstration
area in Grand  Valley. A proposal  for accom-
plishing the initial  detailed studies is described
below.
  The initial studies would be undertaken in an
intensive  study  area  located  in  the Grand
Valley  west of Grand Junction, Colorado. An
area of approximately 40 acres will be  leased
for the study. It is contemplated that the area
will be near the Government  Highline  Canal
from which the water supply will be obtained.
This intensive  study area will  be divided  into
about 100 plots each 100 feet square (10,000
sq. ft.)  as shown in Figure 2.  Both timing and
amount of water application will be varied.
  The amount and chemical quality of irriga-
tion water will  be monitored to determine the
system  input.  Rainfall  and temperature data
will also  be collected.  Solar  radiation, wind,
and humidity data will be obtained from the
Grand  Junction   weather  station,  which  is
located only  a few miles from the intensive
study area.
  Drains will be installed across both the upper
sides of the plot area. They will be used to inter-
cept any natural discharge water which might
otherwise get into  the plot area. This natural
drainage water will be monitored for chemical
quality and quantity.
  Each plot will have an individual drain across
each of the lower sides of the plot. Details  of
J_
100'
T~
—










100










,^100 plots (MCh 1












/







/



























































OtfiiOO'l
\
Direction of
slept of shad
]Allol(a filKr |
( strip to stobilii* j
} humidity j
                       BucxJH of pipes conviyng
                       «ff lo*nt from individual plots
Waltr mtOMiring station
(or (Maturing quantity - — "
and quality of affluent



from MCh drain J
I
I
                       Drain carrying oil •ffkrtnt
                       to natural drain
                                       I
     Figure 2: Layout of Intensive Study Area

these drains are shown in Figure 3. The site
will  be selected so that shale is approximately
10 feet below  the surface:  A trench will be
excavated into the shale.  In this trench will be
laid  a  sheet  of polyethylene  film  which  will
extend vertically a distance of about 5 feet.  The
purpose of this  film is to  stop unsaturated flow
across the drains. Small plastic drain pipes will
be  laid  in  the trench  and  covered with  a
gravel filter.
  Soil samples will be collected throughout the
root zone and below the root zone to shale.  The
first  purpose of the sampling will be to deter-
mine the type of ions present at all levels in
the  soil profile  and  to  determine   their con-
centration.  This will  provide a reference point
for comparing  future measurements. Another
purpose of  this sampling will be to determine
soil  fertility.  From  the  analysis of the  soil
samples, the  amount and  types of fertilizers
needed to bring the soil up to optimum balanced
fertility levels will be determined.
  Four different  crops  will be  used  in  the
tests, namely alfalfa,  barley,  corn, and pasture.
Each will be irrigated with a variety of irriga-
tion   treatments.  In  addition,  two levels of
nitrogen fertilizer will be applied to the corn.
This  will  demonstrate the  effect of irrigation

-------
                                                      IRRIGATION SCHEDULING STUDIES     201

                                                                 Ground Surface
  Gravel
  Filter
                                                                                      Approx. 10'
              Drain Pipe
                                                  Shale
                           Figure 3: Plot Corss-section with Drain Details
practices  on nitrogen efficiency and  will also
provide useful information on the movement of
nitrogen  fertilizer  into drainage  effluent.  It
is  anticipated that  for  some of the treatments
where excessive water is applied, an  excessive
application  of nitrogen will almost  eliminate
the yield reduction by making up for the amount
of  nitrogen which  is leached out of the  soil
profile. It is hoped  that this demonstration will
show that there is also an additional yield reduc-
tion because of excessive irrigation water app-
plication and that the cost of the added nitrogen
fertilizer will provide incentive for more effi-
cent farm irrigation water management.
  Soil moisture levels in each plot will be care-
fully  monitored using soil  moisture  resistance
blocks. In addition, neutron probes will be used
to  check  moisture  levels throughout the soil
profile.
  Soil solutions will be extracted from various
levels of the profile and analyzed to determine
the types  of ions present and their concentra-
tion. These data, together with the information
on  the amount and type of water leaving the
drains as well as changes noted in soil samples,
make  it possible to  determine where the salt
pickup is  taking place in the soil profile under
the various treatments.
   In most field research projects, the moisture
and salt movement in the soil profile is moni-
tored, which allows a prediction of the amount
of  moisture and salts  reaching the ground
water table. This project will have an advantage
in that the outflow will be monitored as well as
the inflow  and  movement through the  soil.
Thus, the model  which will be used to describe
the movement of moisture and salts through the
soil profile can be verified.
  The construction of the  drainage system for
each plot affords an excellent opportunity to
obtain undisturbed  soil samples that can be
used in the  laboratory to determine the rela-
tion between saturation and capillary pressure
during both imbibition and drainage. The devel-
opment of these of these relations, along with
field measurements of either  moisture content
or capillary pressure,  allow the moisture  move-
ment to be described. Relations between satura-
tion and capillary pressure have been developed
for numerous soils.
  The primary  advantage  of undertaking this
study is the development of a prediction  model
describing  moisture  movement and associated
chemical quality, thereby allowing more accu-
rate predictions  of  the  effects   of  proposed
salinity control measures.  Additional benefits
resulting from this study would be a knowledge
of fertilizer use efficiency and the development
of crop production and crop damage functions
for economic evaluations.

-------
        The  Role  of Modeling in Irrigation
                       Return Flow Studies
                               ARTHUR G. HORNSBY
                                      Soil Scientist

                                          and

                                 JAMES P. LAW, JR.
                         Robert S. Kerr Water Research Center
                            Environmental Protection Agency
                                    Ada, Oklahoma
ABSTRACT
  Modeling can play a critical role in finding al-
ternatives  to present water management and
cultural practices  to  improve  and control
salinity levels in irrigation return flows. The use
of  systems  analysis   techniques,  sensitivity
analysis,  and  optimization  procedures  can
greatly enhance  the  usefulness of modeling.
A generalized  conceptual model is presented
to demonstrate the application of systems anal-
ysis to irrigation return flow systems. Several
applications of systems modeling are suggested
with on-the-farm water management cited as
having the greatest potential for salinity  im-
provement.   The  importance  of prediction
modeling in resource planning and in bringing
about institutional  changes  needed to  ensure
the success  of salinity control  measures is
established. Systems models for irrigation re-
turn flow quality can be integrated into multi-
purpose river basin models.
INTRODUCTION
  Irrigation return  flows are the  result  of a
multitude  of  natural  flow  phenomena  and
humanly  controlled management practices or
malpractices.  Water  diverted  for  irrigation
seeks  the lowest  gravity  potential  possible
either on the surface or underground. The ir-
rigator uses this characteristic to his advantage
to distribute and manage water for the irriga-
tion of crops. Likewise, he uses the same driving
force for  removal of drainage water and flush-
ing salts  from  the  soil  profile to maintain a
satisfactory salt balance in the root zone.
  The movement of water and  salts in soils is
influenced additionally by soil  properties  such
as texture, bulk density, water content, hydrau-
lic conductivity, and exchangeable ions. These
parameters  may  vary considerably from one
location to another within  an irrigated  area
further complicating the flow  behavior.  The
loss of water from  the surface  by evaporation
                                          203

-------
204    MANAGING IRRIGATED AGRICULTURE
and  from the root zone  by transpiration not
only interrupts the flow behavior but also con-
centrates salts in the remaining soil water. The
characterization  and prediction of the quality
of irrigation return flow waters are considerably
more difficult than quantity, although the two
are closely related. Both are dependent upon a
large number of variables which may be inter-
related to some degree.
  To visualize  this  complex system and the
multitude  of interactions  which  take  place
within it at some instant in  time is taxing even
to the analytical mind of the scientist. Not only
must he understand the basic laws of physics,
but he must  bejcognizant of the hydraulic and
physical-chemical behavior  of  water and salts
moving through the soil. Accurate  prediction
of the  effect of some imposed  management
practice would indeed seem impossible.
  Perhaps the best solution to this dilemma is
to consider the  overall system to be composed
of several discrete subsystems  which  can be
individually  identified  and examined.  Such
subsystems can be simulated by physical models
or mathematical models to elucidate their be-
havioral responses to  management practices
and  other input variables.  These  simulation
models  allow examination of the processes oc-
curring in the system  and  their effect on the
quality  of the drainage waters before capital
outlay for construction  or  implementation  of
management  practices. With this prior knowl-
edge  of the  expected response  of the system,
more judicious planning can be made.
  The Utah  State  University  Foundation  re-
port3 has recommended that:
   A comprehensive,  reliable  model should be
 developed of salt movement through soil profiles
 in relation to  changes in quantity and composition
 of solutes, soil characteristics and irrigation, and
 other management  practices. Parameters must
 be identified  that will permit rapid evaluation of
 field situations and predictions of the quality and
 quantity of return flow water under a variety of
 conditions.
   Systems methodologies for data collection and
 analysis should be developed for use in assessing
 the balance between water and salts applied to an
 area and the  quantities leaving as return flow.

  The need  for such models has been recog-
nized by  others and  developments  have been
made in this regard1. The availability of high-
speed electronic computers has  permitted  re-
searchers  to examine  in  detail the  complex
interrelationships of these systems models.
  This paper is meant to be an introduction to
the  models  and   terminology  of  prediction
modeling  as applied  to irrigation return flow
studies. Other papers in these Proceedings will
be addressed to  specific problems and problem
areas of irrigation return flows.

             Simulation Models
  The use  of simulation  models to  examine
water   resource  problems  has  greatly  in-
creased the productivity of researchers in this
field. Freedom  from  real time and  physical
constraints  allows  the  researcher opportunity
to examine in depth the behavior of a particular
system  and the processes acting therein in a
minimum of time. Simulation models may be of
various forms. Those relating to irrigation have
been physical models and mathematical models.
Physical  models are  generally  scale  models
which contain  the  pertinent characteristics of
the object being simulated. Examples of physi-
cal models are packed  soil columns and sand
tanks which may be used to investigate ion ex-
change  processes   and  drainage phenomena
respectively.  Surface   conveyance  structures
are studied as  models  also.  For instance,  the
conveyance and control structures of the Cali-
fornia State Water  Project were evaluated with
scale models. Operational  and  design faults
were identified and corrected before construc-
tion began, thus avoiding costly reconstruction.
  Electric analog  techniques  can be used to
simulate certain physical processes.  The close
analogy  between flow  of water in soils and
flow  of electricity  in circuits provides an  ex-
cellent means of simulating a process occurring
in nature. The electric analog models use re-
sistance  and capacitor  networks to  simulate
drainage  patterns   and  flow  behavior. Their
chief advantage  is their dynamic  nature in that
a measurement can  be made at any point of the
system at any time  and the measured value will
be up to  date. The major disadvantage is that
to solve  a  large problem a  large  resistance-
capacitor  network  must be used. The size of
problem and size of networks are directly re-
lated 1:1.

-------
                                                                           MODELING
                                         205
  Mathematical models are representations of
physical and/or chemical processes in the form
of  mathematical  equations.  The  behavior of
these processes is known  to  follow the  basic
laws  of physics and chemistry which  can be
expressed  mathematically.  Since  these  pro-
cesses may be dependent  on time, space, tem-
perature  and/or  concentration gradients the
mathematical equations are likely to be  com-
plex and difficult to  solve. The usual procedure
is  to make certain simplifying  assumptions
which will allow easier solution of the equa-
tions. Use  of these assumptions requires  some
judgment since some error may be introduced
by   the  simplification procedure.  The  end
product is a mathematical expression which can
similate the  process in question  with limits
of error which can be tolerated.
  Mathematical models may be separated into
two broad categories based on the  nature of the
process being modeled. A "stochastic process"
is generally defined as the dynamic  part of prob-
ability theory in which a collection of random
variables is studied  from  the  point of view of
their interdependence and limiting behavior.
A stochastic process is one which develops in
time  in a  manner  controlled  by  probabilistic
laws. Some  examples of  stochastic processes
are Brownian motion, radioactive decay, growth
of  a bacterial  colony,  and  a summer rain
shower.  Studies  of  reservior  management,
rainfall-runoff relationships, and  stream  flood
stage predictions  rely on stochastic methods of
analysis.  Models of the  above processes are
referred to  as stochastic models.
  The second  category is that  of determin-
istic models. These models do not contain prob-
ability mechanisms  and are  developed  using
known physical laws and  principles to derive
the .governing equations. Deterministic models
are theoretically  valid only if they include all
the variables significantly affecting the  output.
Since deterministic  models  simulate  the  in-
dividual  processes  taking place  within the
system, this approach leads to a clear definition
of  each  process  and its  contribution  to the
system as a whole. In deterministic theory, the
correct prediction will simply state the value of
the observable result that will be obtained. The
accuracy of the prediction depends on how well
the  model represents the  components of the
physical  system. Modeling of the components
in irrigation systems has generally been deter-
ministic  in nature.  Conveyance, infiltration,
consumptive use, ion exchange, and drainage
are some of the components which  have been
modeled  in this manner.
  Recently, the concept of systems analysis has
been applied to water resource planning  with
great success. The basic tenet of systems analy-
sis  is that the system being modeled can be
broken  down  into  subsystems  which can be
more  easily handled and that  the  subsystems
can be related to one  another  in a  systematic
manner.  A subsystem  in turn  can be further
broken down  to consider the  individual  pro-
cesses  taking  place within. Each  individual
process  can be  modeled mathematically  and
related to  other processes taking place simul-
taneously.
  Consider for  example the general process
model presented conceptually in Figure 1.  The
model  would  have associated  with it "i"  in-
put  variables,  "j"   internal  variables,   "n"
model  parameters,  and  "k" output variables.
Each individual process occurring within the
system has these characteristics.  More specifi-
cally, if  one  considers the  process of ion ex-
change  in a  soil profile the  input variables
might include concentration and ion composi-
tion of the inflowing waters; internal variables
might include  ion exchange rates and pH  rela-
tionships;  model  parameters  might  include
cation exchange capacity and porosity; and the
output variables might  include  concentration
and ion composition  of the effluent waters.
The process model then simulates ion exchange
based on the variables and parameters known to
affect the process.
  Other  several  process  models are  closely re-
lated and dependent upon one another and can
be  collected into a group called a  submodel.
Figure 2  shows such a  submodel for the  soil
system which presents conceptually the process
models  included as well as their interdepen-
dence.  The output of one process model is the
input to  another process model. The submodel
has the characteristics of input,  internal,  and
output variables and model parameters just as
the process models, however they  are some-
what more complex since they include the in-
formation necessary for all  the process models

-------
206
MANAGING IRRIGATED AGRICULTURE
      input
   variables
                     inferno/
                    variables
  model
parameters
 output
variables
                         Figure 1: Conceptual Diagram of a Process Model
                SOIL SYSTEM SUBMODEL

Ion
Eichong*
Proem
MwM


N.
/
Dissolution
Prtclfllottor,
Proem
Modol



k /
Moistur.
Proem
ItoW
/ \.



Evop.
Tronspirotiofl
Proctss
ttodtf
I/
k


Nitrogen
Tronsformotlofl
Proems
Uooil

 Figure 2:  Conceptual Diagram of a Soil Subsys-
 tem Submodel Comprised of Process Models
contained in the submodel. The soil system sub-
model  should contain all the process models
needed to define the soil system and to simulate
the processes that are of interest.
  Similarly, several submodels which are close-
ly related can be collected to form a systems
models. Figure 3 shows such a collection for an
                                         irrigation systems model.  The submodels are
                                         interdependent in that they share input—output
                                         variables. The irrigation systems model contains
                                         all the characteristics of the submodels as well
                                         as the individual process models, since each of
                                         the  submodels  was  developed  from process
                                         models. The response of the irrigation systems
                                         model to some input variable is the net inte-
                                         grated response of all the process models con-
                                         tained by the systems model to that input infor-
                                         mation.
                                           The real power of systems analysis in the sim-
                                         ulation procedure is the capability to model a
                                         complex situation by systematically considering
                                         the individual processes which are active within
                                         the system in such a manner that  the response
                                         of  the system reflects the effect  of  the  indi-
                                         vidual processes. In addition, since each process
                                         model or system submodel is complete within
                                         itself, they can be used to simulate that process
                                         or  subsystem independently  of the remainder
                                         of the system.
                                          The irrigation systems model presented here
                                         could be  considered  complete within itself or
                                         as a part of a much  larger systems model for a
                                         river basin or water resource management sys-
                                         tem. The management  of water  resources re-
                                         quires consideration of many variables which
                                         can  be handled more easily and accurately by
                                         a systems analysis approach.
                                          Use of  a concept  termed sensitivity analysis
                                         allows the processes occurring within a system
                                        significantly enhance  the quality of the drainage

-------
                                                                            MODELING
                                          207
   Input
 Variables
                                                                                J
                                                                                       Output
                                                                                      Variables
          Figure 3: Conceptual Diagram of an Irrigation Systems Model Comprised of Submodels
to be evaluated as to their relative impact on
the total behavior  of the system. Thus,  the
sensitivity of the system  to a  change in some
part of it can be analyzed. This concept can be
used also to evaluate the effect of some input
variable on the response of some process  oc-
curring within  the system or  on the total re-
sponse  of the system. Sensitivity analysis is a
tool available to  determine  the sensitive por-
tions  of the system and thereby  call the  at-
tention of the researcher to the areas needing
closer evaluation.
  Optimization  techniques have been develop-
ed which can be used to determine the "best"
or "optimal" solution to a  simulation model
based on certain imposed constraints. In simple
terms,  if several possible changes  in  a system
can cause nearly the same result, optimization
procedures can  choose the most advantageous
change.  Optimization procedures  are usually
used in reference to some goal or constraint
such as reduced salinity level or maximum eco-
nomic  benefit. In  the  more general river basin
models,  the constraints may be power genera-
tion, flood control,  recreational use, irrigation
supply, or quality control. In areas where con-
junctive use  of water takes place, these pro-
cedures can be  used for  efficient management
of water resources.
Application To Irrigation Return Flow Studies
  The irrigation return flow system can be sub-
divided into three major subsystems: (a) water
distribution, (b) farm water  management, and
(c)  drainage and water  removal. The  water
distribution subsystem  consists  of  the  con-
veyance structures from the point of diversion
to the turnouts on individual farms. The  farm
water  management  subsystem begins at this
point and terminates at the point where surface
return water leaves the field and where percola-
tion water leaves the plant root zone. The drain-
age and water removal subsystem completes the
system by  conveying  drainage waters and sur-
face return waters to the receiving streams.

Water Distribution Subsystem
  The planning of adequate  reservoir storage
for an irrigation project requires an estimate of
the irrigation demand that will be placed on the
reservoir, as well as other functions for which it
will serve, i.e., flood control,  power generation,
etc. The design, construction,  and performance
of  irrigation  conveyance structures  can  be
evaluated  with simulation  methods in  order
to produce the maximum benefit for the re-
source investment required for installation. In
situations  where canal lining can be shown to

-------
208     MANAGING IRRIGATED AGRICULTURE
water, as well as conserve water, these benefits
should be taken into account when considering
the economic benefits  and costs of a project
during planning stages.
  Since  changes in quality  may be subtle in
intensively farmed  areas  where a wide variety
of management  practices  are  in use, prediction
models should be  sufficiently sophisticated to
be capable of evaluating the  different processes
occurring within the system  in order to estab-
lish cause and effect relationships.
  With such detail  in the models, the processes
which contribute significantly to water quality
degradation can be identified and management
schemes can be generated to alleviate or control
the degradation. Prediction models are needed
which will provide for rapid and accurate exam-
ination of novel management schemes to re-
duce  salinity  in irrigation return  flows.  The
greatest single advantage of prediction model-
ing is the capability of examining alternatives
to our  present management  practices  without
actually  implementing them  in the field.  This
permits  rapid  screening  of  potential  control
measures as to their expected response in the
system.

Farm Water Management
  On-the-farm  water  management has   been
cited  as  perhaps the area of  greatest potential
for improving  salinity levels in return  flow
waters.2  Since  changes  in   current  manage-
ment  practices may not be reflected as immedi-
ate changes in the  quality of the return water,
expected long term improvements and benefits
must  be  assessed  by  prediction  techniques.
Knowledge  of the  manner in which water be-
haves  in  soils  has permitted development  of
prediction  techniques for water  movement  in
soil. These techniques can then be used to esti-
mate  the amount and timing of irrigation appli-
cations resulting in increased water-use  effi-
ciencies. Such increases in efficiency can  lead
to  better  management  of salinity in  return
flows since excessive leaching can be avoided.
  Models  which   include  optimization   pro-
cedures  can  predict   the  best  management
schemes for reducing  salinity levels  and im-
proving yields as well as  simulating water and
salinity flow in  the system. Optimization  pro-
cedures allow the system  to  be simulated  with
constraints  on salinity levels while  maximizing
crop yields. Potential management schemes can
thus be generated and evaluated by this techni-
ques.
  The usefulness of recently developed meth-
ods  of application of irrigation water in re-
ducing or controlling  salinity in drainage water
should  be evaluated  before  they  are recom-
mended for widespread use. The  high cost  of
installation of subsurface and trickle irrigation
systems may make their use restrictive unless
comparable benefits can be  shown  to accrue
as a result  of their  installation.  Simulation
modeling  can be used to evaluate their useful-
ness both in reducing salinity in the drainage
waters and  in  improving yields as a result  of
better  water management and increased  effi-
ciencies.
  Likewise, the effects of irrigation scheduling
on  both  yield  increases and drainage water
quality can be assessed. Scheduling can be  used
to maximize yields and at the same time used to
avoid overirrigation with the resultant excessive
leaching of the soil profile. This is of particular
interest in areas where the soils are  high  in
residual salts or where  there are  underlying
saline deposits.
  In regions where the soils are not high in resi-
dual salts or underlain by  saline deposits, the
primary causes  of salinity changes are due  to
consumptive use and salt precipitation and dis-
solution reactions. To anticipate the degree  of
change, models have been developed to simu-
late  consumptive use, precipitation-dissolution
reactions,   ion   exchange  reactions,  nitrogen
transformations,  and  water movement. These
models are used to predict the quality of drain-
age water eluting from the soil profile as well
as the distribution of salts within the soil profile.
Individual ions and/ or total dissolved solids can
be considered.
  In areas where the quality of applied irriga-
tion water is marginal with respect to specific
ions, models which simulate the quality changes
resulting  from  ion  exchange  reactions  and
precipitation-dissolution reactions  in the soil
can  be of considerable  aid  in managing  such
waters. With simulation models of the physical-
chemical behavior of water and salts in soil, the
probable effects of irrigation with the marginal
waters can be determined before their use.  This
is especially helpful  in managing high sodium
and/or high  bicarbonate  waters where precipi-

-------
                                                                            MODELING
                                          209
tation of CaCO3 may  give  rise to increased
sodium "hazard."

Drainage and Water Removal
   Drainage studies have long been conducted
using simulation techniques, however, only re-
cently have quality  aspects  been  considered
from the standpoint  of minimizing salinity in
the  return  flow  water.  Sand  tanks,  electric
analog  systems,  and mathematical  simulation
have been used to examine the theory of drain-
age. Based  on the technology developed here,
prediction  models can be used to evaluate the
salinity level  of  drainage  water which results
from  consumptive  use, ion exchange,  salt
pickup, and mixing phenomena which occur as
a  result  of  irrigation practices. In areas where
poor drainage has resulted in  high salt levels in
the soil profile, planning for  subsurface drains
can  be enhanced by use of prediction techni-
ques which not only handle the hydraulics but
also the  quality aspects  of the proposed drain-
age system.
   Some  areas contain naturally saline mineral
formations  underlying  the  soil  profile  which
contribute  readily to salt  pickup in drainage
water.  Simulation models  with optimization
techniques  can be used in planning interceptor
drains which would  prevent applied  irrigation
water from  flushing  salines out of the marine
deposits. Depth,  spacing and  size of the drains
can be evaluated  as to their effectiveness in the
control of salinity problems  using  simulation
modeling before  construction begins. This as-
sures greater success of  the proposed reclama-
tion project.
  The length of time required  for a unit volume
of water and salt to move through the soil pro-
file and saturated aquifer to a drain is referred
to as "travel time." In  assessing the  effects  of
irrigation management practices on the quality
of the stream receiving return flows, an estimate
of the travel time is needed and can be obtained
by prediction modeling.  Thus, one  can predict
how soon after changing a management prac-
tice that changes will occur in  the quality of the
drainage waters.

   Application To River Basin Management
  Irrigation return flows  are often a  major
portion of  the stream  flow  in some western
streams,  thus their  management may  greatly
 affect the river basin system in which they are
 located.  In river basins such as the Colorado
 where salinity levels are of prime importance
 in the development  of new areas and mainten-
 ance of  present project areas, simulation is a
 useful tool in managing  the  total  water re-
 source system. The  benefits of salinity control
 measures implemented in a given area may ac-
 crue to some downstream users. If an equitable
 sharing of the expense  of upstream improve-
 ments in control measures is to be  found,  then
 some mechanism  for identifying the benefits
 accruing to both areas must be developed. Sim-
 ulation models  can be used  to  assess  these
 quality  improvements and arrive  at an  ade-
 quate estimate of benefits accrued to each area.
  In management of a river basin, irrigation
 may be only one of  several resource uses which
 are considered in water resource planning. The
 relative  importance  of  quality  degradation
 caused by irrigation return  flows may dictate
 a shift in water resource uses to comply  with
 salinity standards. To date, most planning of
 this  nature has relied  upon seat-of-the-pants
 estimates  of the salinity contribution that can
 be attributed to agriculture.  Simulation can be
 used to ascertain that which is caused by agri-
 culture and that which is caused  by  natural
 processes and other sources.
  In some states water laws and water duties
 are structured such that they may be the chief
 deterrent  to establishing salinity control mea-
 sures in that area.  If institutional changes must
 be brought about  to ensure  success of salinity
 control measures, simulation  models  can be
 very helpful in generating  possible  alternatives
 to present practices. Subjecting these potential
 alternatives to economic analysis by optimiza-
 tion procedures  can result in better  insight of
 the probable success of the proposed changes.
 Changes  in water laws are  not likely to  take
 place unless the quality benefits of doing so are
 considerable.   Prediction models used for this
 purpose  must necessarily  be  comprehensive
 and  reliable in projecting  the benefits of  sug-
 gested changes.

 SUMMARY
  An introduction to the terminology and ap-
 plication   of  systems analysis  and  prediction
modeling as used in irrigation return flow stud-
ies has been  presented to  acquaint the reader

-------
210     MANAGING IRRIGATED AGRICULTURE
with the tools used by researchers to solve com-
plex water resource problems.
  Systems  analysis has  greatly increased the
usefulness of prediction modeling in water re-
source studies. Application of these  methods
to irrigation return flow problems offers a more
rapid and economical evaluation  of proposed
salinity  control  measures than is possible by
field studies. Field studies and demonstrations
should not be abandoned in lieu of prediction
modeling, but rather they should be  used to
verify potential control measures generated by
the prediction models. The irrigator will prob-
ably not be using prediction modeling himself,
however, the  information developed by such
models can be useful in solving salinity control
problems resulting from management and cul-
tural practices. Prediction modeling is a tool to
be  used in evaluating  alternatives to present
practices in an efficient  and economical man-
ner.
  Prediction modeling can be used not only to
evaluate salinity control measures for  water
delivery, farm water management, and drainage
and water removal subsystems, but also to as-
sess the economic benefits accrued by both local
and downstream  users as a result of these mea-
sures.
  For management of river  systems involving
conjunctive  use  of water resources, prediction
modeling can play an important role in planning
the system to maximize  benefits. In situations
where institutional changes are needed to as-
sure success of salinity control measures, pre-
diction modeling can provide insight into the
longterm benefits that can be expected.
  The greatest single advantage  of prediction
modeling is the capability of evaluating alterna-
tives to present cultural and management prac-
tices before they are instituted in the field.
REFERENCES
    1. Hornsby, Arthur G., Prediction Modeling
for Salinity Control in Irrigation Return Flows,
EPA, Robert S.  Kerr Water Research Center,
Ada, Oklahoma (1972)
    2. Skogerboe,  G. V.,  and Law, J. P., Jr.,
Research Needs  for Irrigation  Return  Flow
Quality  Control,  EPA,  Robert S.  Kerr Water
Research Center, Ada, Oklahoma (1971)
    3. Utah State University Foundation, Char-
acteristics and  Pollution Problems of Irrigation
Return Flow,  EPA (FWQA),  Robert S.  Ken-
Water Research Center, Ada, Oklahoma (1969)

-------
        Modeling Subsurface  Return  Flows

                                  GORDON R. DUTT
                       Department of Soils,  Water and Engineering
                                  University of Arizona
                                    Tuscan, Arizona
  The current concern of the public over the en-
vironment has lead to  enactment of legislation
and the setting of research goals for the mainte-
nance or improvement of water quality.  Not
only are  scientists and  engineers asked to de-
velop practices to improve  water quality, but
also to assess the impact on the environment for
water projects being planned for the future. In
the past, judgement based on experience  and
measured parameters have been used by compe-
tent scientists and engineers to estimate the ef-
fects of water projects on ecology. However,
these judgements are subject to  error  and can
possibly be based on a  more quantitative basis.
  Among the  more complex problems is the
prediction of the effects of irrigated agriculture
on water quality.  Methods are needed to  pre-
dict: 1) the short and long term salt concentra-
tions,. including nitrogen salts, in drainage ef-
fluent from agricultural areas, 2) the chemical
composition of water moving down into ground-
water aquifers from irrigation  projects,  and
3) the impact on water quality in an area under
study for irrigation. A provising approach is the
development of conceptual  models of the dy-
namic soil-water  system. Such models can be
used  for predicting  management  practices
which minimize pollution under economic con-
straints.
  It is not the purpose  of the paper to present a
detailed account  of any particular model, but
rather to report on the "state of the technol-
ogy." Detailed discussions have been reported
elsewhere1 5'15.

   Simulation Models for Predicting Solute
     Changes for One-Dimensional Flow
               Through Soils

  Over the past decade, the author and col-
leagues have addressed themselves to the devel-
opment of models for predicting the quality of
percolating  water.  The foundation  of these
models was the  development of computerized
numerical solutions predicting from initial non-
equilibrium conditions, the equilibrium solute
composition  for soil-water systems at different
moisture contents2.  The numerical solutions in-
volved more than one chemical reaction (ion ex-
change Ca   and Mg  , and solubility of a salt,
CaSO4-2H2O) and  were  based on  thermody-
namics,  ion  exchange  equations,  and  Debye-
Huckel theory. This early work was verified at
moisture  contents much higher than those en-
countered under irrigated  agriculture, but was
later shown  applicable to soil  systems in the
field moisture range4.
  Using the above  procedures and a finite dif-
ference method, a model was developed to pre-
dict the changes in  solute composition in an ef-
fluent from a saturated soil during one-dimen-
sional flow. Since the chemical reactions consid-
                                           211

-------
212     MANAGING IRRIGATED AGRICULTURE
ered were limited, the above model was not
applicable to real field situations but the model
did demonstrate that such models were feasible.
  In essence the procedure was simple. Water
was considered to move along a path normal to
finite soil segments  and the chemical changes
were calculated for each aliquot of water as it
passed through successive soil segments. Due to
the complexity of the soil system these calcula-
tions could only be reasonably performed by a
digital computer. Modification and extension of
the basic procedure and extension to systems of
interest in the  biosphere have been performed.
To  date these  improvements can be classed  in
two groups,  one dealing  with moisture move-
ment and ones  dealing with chemical changes.

Water  Movement
  At the offset the simplest method which was
of interest and applicable to certain soil systems
was considered, i.e., piston displacement of wa-
ter. This  procedure  has been shown to be  of
value when considering the movement of consti-
tuents  in water which interact with the soil ma-
terial during saturated flow3-6. Obviously, there
are  soil systems of interest  in which moisture
changes occur  and in which the  solutes only
slightly interact with the soil material. When the
primary concern is in solutes which are rela-
tively inactive with soil material and when mois-
ture contents are fluctuating, the dispersive ef-
fect of soil material becomes a first order inter-
action. To account for these dispersion effects,
Tanji, et al14, and Terkeltoub16 have considered
machine mixing, i.e., a portion of the soil water
to be held by the soil matrix and mixed with suc-
cessive increments of water entering the matrix.
  The  most  commonly encountered  situation
under irrigation agriculture is a zone of unsatu-
rated zone below the water table. Within the un-
saturated zone  moisture movement and content
is a function of soil properties and water added
or removed. A model of the unsaturated zone
above a water table or a region of constant mois-
ture has been prepared5. This model basically
utilizes  the finite difference scheme  developed
by Hanks and Bowers9. In addition, a moisture
sink term to  allow for moisture loss  by evapo-
transpiration has been included. As  with the
models of Tanji14 and Terkeltoub16 machine
mixing is utilized.
 Chemical Changes
  The chemical constituents and reactions con-
 sidered to date have been governed by two fact-
 ors: 1) effect of the constituent or reaction on
 other solute species present and 2) the import-
 ance of the constituent on problems of interest.
 To date, only neutral to alkali  soils have been
 considered. This limitation of soil alkalinity is
 related to 2) above. Early models3-15  consid-
 ered exchangeable Na+,  Ca+ , and Mg+  ; solu-
 ble salts of the exchangeable cations; ion pairs
 of CaSO4 and gypsum.              2
  Additional SO^2  ion pairs  of Mg  and Na
 have been considered11,  as well as CaCO35-8-10
 and boron12. To facilitate the application to field
 problems, methods have been developed  to cal-
 culate   the exchangeable  cations  mentioned
 above5. This last method substantially reduces
 the laboratory analysis required  to perform cal-
 culations. In addition, methods have been pre-
 pared  for predicting the electrical conductance
 in the  system10. All  of the above interactions
 have been accounted for by taking advantage of
 the fact that  the chemical reaction rates  are
 greater than the rate  of water movement. Thus
 equilibrium relations can be used  to integrate
 the dynamic case.
  With some constituents of interest, the rates
 of reaction are slow in comparison with water
 movement. Such is the case with microbiologi-
 cal transformations  of nitrogen. In such cases
 the reaction rate,  which may be a function of
 several soil-water factors, e.g., moisture content
and concentration of constituents,  must  be  de-
termined. Thus the distribution of constituents
including the moisture must be found as a func-
tion of time. By incorporating unsaturated flow
equations,  and  the  aforementioned soil reac-
tions  which may  be considered as successive
equilibria, a procedure was developed to calcu-
late time dependent rate equations. In turn, it
becomes  possible  to  calculate  the  integral
changes throughout the soil-water system  for
the nitrogen constituents being  considered.  To
date, the microbiological transformation treated
in the  above manner consider the  constituents
NO3, NH4, urea, and organic-N5.

Applications
  To date applications of one-dimensional mod-
els  to field situations on a research basis have

-------
                                               MODELING SUBSURFACE RETURN FLOWS     213
included predicting and thus verifying the qual-
ity of percolating water14, predicting the quant-
ity  of gypsum and leaching water required for
reclamation7 16,  predicting the salinity in the
root zone10 and predicting the nitrogen entering
ground water5.
  It has been  shown that it is feasible to  con-
sider 1) moisture additions (irrigation  and  rain-
fall)  and  movement;   2) evapotranspiration;
3) nitrogen transformations including hydroly-
sis of urea-N,  immobilization-mineralization of
ammonia-N or organic-N and immobilization of
nitrate-N; 4) changes  in the solute concentra-
tion of soil-water due to oin exchange,  solubility
of gypsum and lime (CaCO3) and dissociation of
certain  ion pairs; 5) nitrogen uptake by crops,
and 6) additions of soil amendments such as ni-
trogen fertilizers and gypsum. Models of the
type discussed here could be of value in devel-
oping management practices for  reducing pol-
lution by agriculture.

    Predicting Solute Changes for Two and
            Three Dimension Flow
  As discussed  elsewhere  in this publication,
lines of one-dimensional flow can be joined to-
gether to consider the two-or three-dimensional
case, where the system now becomes a field or a
basin. The U.S.  Bureau of Reclamation has ap-
plied a stochastic hydrology model with a soil-
water simulation model, considering  saturated
flow. This hybrid, three-dimensional model has
been  tested in  Utah.  In  addition, the  same
agency  has  a  simulation model for predicting
solute  in water from  tile  drains.  This latest
model  considers chemical  changes  occurring
above  the  water table5  mentioned above and
chemical changes  and dispersion occurring in
the saturated  zone. Another interesting hybrid
model of a basin has been reported by Thomas,
et al17.  This  model utilizes the principles dis-
cussed earlier and a general hydrologic model.

SUMMARY
  Computer simulation  modeling techniques for
predicting the solute  composition of water in
soil-water systems  have  been developed and
verified on a research basis. The concentrations
of NH4  and several alkali metals and alkaline
earth salts including NOj  have been  predicted
for soil systems,  for one-, two-, and three-dimen-
sional  flow. Application to several systems of
interest  concerning reclamation and quality of
agricultural return flow water have been made.
Other  applications   include incorporation  of
these  soil-water  simulation models  with  other
hydrologic models to form simulation models of
irrigation projects of groundwater basisn.
  It would appear, on the basis of the simula-
tion models thus far developed, that the tech-
nique could be  used to  develop management
techniques for reducing or minimizing  the  salt
problems and/or pollution normally associated
with irrigation agriculture.
REFERENCES
   1.  Dutt, G. R., "Quality of Percolating Wa-
ter No. 1. Development of a Computer Program
for Caclulating the Ionic Composition of Perco-
lating Waters," Water Resources Center Con-
tribution No. 50. Univ. of Calif, (1962).
   2.   Dutt.  G. R., "Prediction of the Concen-
tration of Solutes in Soil Solutions for Soil Sys-
tems Containing Gypsum and Exchangeable Ca
and Mg." Soil Sci. Sox.  Amer,  Proc. Vol. 26,
(1962).
   3.  Dutt, G. R., "Effect of Small Amounts of
Gypsum  in Soils on the  Solutes of Effluents."
Soil Sci. Soc. Amer. Proc., Vol. 28, (1964), 754-
757.
   4.  Dutt, G. R. and Anderson, W. D. "Effect
of Ca-Saturated Soils on the  Conductance and
Activities of Cl", SO", and Ca++, Soil Sci. Vol.
98, (1964), 377-382.
   5.  Dutt,  G.  R., Shaffer, M. J. and Moore,
W. J., "Computer Simulation  Model of Dy-
namic Bio-Physiochemical Processes in  Soils,"
Tech. Bui. No.  196 of the Univ.  of Ariz. Agr.
Expt. Sta., (1972), In Press.
   6.  Dutt,   G. R.,   Terkeltoub,  R. W.  and
Rauschkolb, R. S., "Predictions of Gypsum and
Leaching  Requirements  in  Sodium-Affected
Soils," Soil Sci. In press.
   7.  Dutt, G. R. and Tanji, K. K., "Predicting
Concentrations  of Solutes in  Water Percolated
Through a Column of Soil," Jr. Geophys. Res.,
Vol. 69 (1962), 3437-3493.
   8.  Dyer, K.  L., "Effect of CO2 on the Chem-
ical Equilibrium of Soil Solution and Ground-
water," Ph.D. Dissertation,  Univ. of Ariz., Tuc-
son, (1967).

-------
214    MANAGING IRRIGATED AGRICULTURE
   9. Hanks, R. J. and Bowers, S. A.,  "Num-
merical Solution of the Moisture Flow Equation
for Infiltration into Layered Soils," Soil Sci.
Soc. Amer. Proc.  Vol. 26, (1962), 530-534.
  10. Oster, J. D. and McNeal, B. L., "Compu-
tation  of  Soil Solution Variation with Water
Content for Desaturated Soils," Soil Sci. Soc.
Amer.  Proc. Vol. 35, (1971), 436-441.
  11. Tanji, K. K., "Solubility of Gypsum on
Aqueous Electrolytes as Affected by Ion Asso-
ciation and Ionic Strength," Environ. Sci. Tech-
nol, Vol. 3, (1969), 656-661.
  12. Tanji, K. K., "A Computer  Analysis on
Leaching  of Boron from  Stratified Soil Col-
umns." Soil Sci., Vol.  110, (1970), 44-51.
  13. Tanji, K. K., Doneed, L. K.,  Ferry, G. V.
and Ayers, R. S., "Computer Simulation Analy-
sis on Reclamation of Salt-Affected Soils in San
Joaquin  Valley,  California,"   Soil Sci.  Soc.
Amer.  Proc. Vol. 36, (1972) 127-133.
  14. Tanji,  K. K., Doneed, L. D. and Paul,
J. L., "III. The Quality of Waters Percolating
through  Stratified Substrate as  Predicted by
Computer  Analyses,"  Hilgardia,   Vol.  38,
(1967), 319-347.

  15. Tanji,  K. K.,  Dutt, G. R., Paul, J. L.
and Doneed,  L. D., "Quality of Percolating Wa-
ters II. A Computer Method for Predicting Salt
Concentrations in Soils at  Variable  Moisture
Contents," Hilgardia, Vol. 38, (1967), 307-318.

  16. Terkeltoub,  R. W. and  Babcock,  K. L.,
"Calculation  fo the Leaching Required to Re-
duce the Salinity of a Particular Soil Depth Be-
neath a Specified Value," Soil Sci. Soc. Amer.
Proc., Vol. 35, (1971) 411-414.

  17. Thomas, J. L., Riley, J. P. and Israelsen,
E. K., "A Computer Model of the Quantity and
Quality of Return Flow," PRWG  77-1, Utah
State Univ., Logan, Utah.

-------
            Modeling Salinity in the Upper

                       Colorado  River Basin
                                   M. LEON HYATT
                           National Field Investigations Center
                   Environmental Protection Agency Denver, Colorado
 ABSTRACT
   Increased use of water resources produces
 quality changes (greater  salinity o.r total dis-
 solved  solids) that affect downstream users of
 the resource. Models  of river systems can be
 used as efficient management tools to minimize
 harmful effects on these users.  The discussion
 concerns a computer simulation model that de-
 scribes the salinity in the Upper Colorado River
 Basin.  Using mathematical expression, the sa-
 linity model describes only the most fundamen-
 tal and basic processes in the basin, but has
 value  in  assessing  the quality  changes  that
 might result from contemplated development at
 a particular location within the river system.
   Successful modeling and management of the
 salinity system requires a definition of the Basin
 hydrology that  is accomplished through water
 budgeting'procedures. Available data of histori-
 cal water flows and salinity concentrations for
 agricultural, municipal,  and industrial uses, are
 employed to verify a salinity model. Develop-
 ment of a salinity model is illustrated using the
 White River subbasin which is verified by com-
paring  computed and observed output values.

 INTRODUCTION
   Greater use of the Colorado River Basin wa-
 ters is  creating a difficult  situation in terms of
the water resource quantity and quality. Unfor-
tunately, increased utilization of the existing re-
sources, through use and reuse of water for irri-
gation and by industries, concentrates and adds
non-degradable materials that produce a degen-
eration of water quality. For example, the natu-
ral inorganic salts, leached out by percolating
waters from the rocks and soils within a water-
shed, are concentrated through the consumptive
use of water in irrigation. The problem is further
aggravated by the leaching effects of excess irri-
gation waters as they percolate through the soils
of irrigated areas.
  This degradation of water quality from inor-
ganic  salts—generally referred to  as total  dis-
solved solids (TDS) or, more commonly, as sa-
linity—results from various uses, such as irriga-
tion and industrial; nevertheless, the  extent to
which each  use contributes to the salinity load
remains  a controversial issue. Promotion of
complementary uses and the reduction of con-
troversy and competition for the existing water
resources of the Colorado River Basin are neces-
sary aims of good management. In this river ba-
sin each upstream  use of water has some effect
on its quantity, quality, and on the timing of
flow occurring at downstream points. The prob-
lem of  quantitatively  assessing these  down-
stream effects is a difficult one. The cause of
this is the complex interrelation and variable na-
                                           215

-------
216    MANAGING IRRIGATED AGRICULTURE
ture of the many different processes that occur
at the  same time. In other words, the extent of
downstream effects, such as those caused by in-
creased diversions  at  a specific location,  de-
pends  upon the dynamic characteristics of the
system and  the water use patterns that prevail
within the Basin. However, many of the factors
affecting the system are  subject to manipula-
tion and regulation, and through proper man-
agement criteria,  efficient use  of  the water
resources of the Basin can be achieved.
  Since water resources within the Upper Colo-
rado River  Basin are now nearing full utiliza-
tion, application of sound management princi-
ples, based upon consideration of various alter-
natives, is  becoming  increasingly  important.
Sound  water resource  management  requires
thorough and thoughtful  planning in harmony
with overall public goals and needs. Water is an
important key to successful development of the
area. Effective planning must consider the vari-
ous consequences of water resource manipula-
tion. This objective can be adequately met only
by rapid and accurate quantitative  assessment
of various  possible  management alternatives.
These  might involve  such variables as water-
shed treatment, weather modification, increased
urbanization,  salinity  control  measures,  and
changes in irrigation  practices  within a basin.
Consequently, as pressures upon the  river basin
resources increase, more sophisticated methods
are required for planning and management pur-
poses.  The advent of modern, high-speed, com-
puters  for modeling entire water resource sys-
tems provide a technology to use as a tool for
the management of water.

            Modeling Background

Basic Concept
  Development of a  specific and quantitative
description of a river basin including its  salinity
dimension is a difficult problem. This is due to
the complex and variable  nature of the many
different processes that occur at the  same time
within  the total basin. The problem is, therefore,
first to describe in mathematical terms the vari-
ous processes that  occur and then  develop a
realistic approach for combining these relation-
ships  into  models that faithfully describe the
system. Measurements,  or  observations,  are
made of the model and its characteristics when
subjected  to  conditions  similar to those con-
fronted by the prototype.  These models allow
for easy and quick examination of alternatives
posed  by planning and management changes
within the prototype system (a real-world river
basin) being modeled.
  The  suggested approach in the  development
of a model that describes salinity in the  Upper
Colorado  River Basin  is to consider  initially a
model  that is macroscopic in nature in  which
only the most fundamental and basic processes
are described. Such a model is not refined or so-
phisticated in terms of describing the specific
details and intricate processes that actually oc-
cur. The vast area of the basin is subdivided into
relatively  large, yet descriptive  spatial  units,
and within each of these the basic relationships
are defined. Only the important model parame-
ters are used.  These variable  parameters use
average values in the model solution.  As  a gen-
eral rule, availability of the data needed  by the
model  dictates the size  of spatial unit selected.
Temporal resolution is obtained by selecting a
specific time  increment  over which average
values of time-varying parameters are used. De-
velopment of such a  model  quickly  highlights
data limitations and needs. The model is capa-
ble of examining various alternatives for the Ba-
sin as  a whole,  or between its subbasin units,
but may not  be able to pinpoint specific  results
at exact locations.
   Once a model of this nature is developed, def-
inition  of the system can be improved by reduc-
ing the magnitudes of the spatial units and time
increments. The improved model  is then capa-
ble of solving the same basic relationships as its
predecessor as well as solving many  additional
problems and more complex processes that re-
quire detailed description.
   For example, considerable interest is now evi-
denced in the phenomenon occurring within the
soil profile relating to the precipitation, solu-
tion, and exchange reactions of  various ions.
These reactions are functions of the composition
and concentration of salt in the applied irriga-
tion water, soil properties, irrigation practices,
and of other related characteristics.  Adequate
description  and simulation  of the reactions
within  soils requires a small spatial increment.
Similarly,  precipitation  interception  rates and

-------
                                        MODELING SALINITY —COLORADO RIVER BASIN
                                          217
changing snowpack temperatures are processes
that can be included only in models based on a
small time increment.
  Additional improvement of the definition of
the salinity system can be accomplished through
development  of  other models.  A  model  is
needed  that  can  route salts,  both collectively
and individually, through a  reservoir  system.
Such a model  would require consideration of
salt deposition, stratification, and solution from
the confining boundary. A model is needed to
predict the salt load of water originating from
thundershower  activity, common to the basin.
Similarly, a model, which adequately estimates
quantitatively,  as  well  as  qualitatively,  the
chemical composition  of  water originating on
watersheds,  would  be useful where  there are
now deficiencies  in water  quality data. This
model would relate such variables as  flow rate,
vegetative cover, evapotranspiration,  geologic
characteristics,  slope of watershed, vegetative
cover, time, and flow path. Each of the models
outlined above would have smaller spatial defi-
nition and time increments. They would greatly
improve the definition of specific processes con-
tained within the general model covering the en-
tire salinity system.
  The ultimate in modeling would employ con-
tinous time and spatial definition, although the
practical limitations of this approach are obvi-
ous. For the most  part,  the complexity of a
model designed to describe the salinity within a
basin depends upon the magnitude of time and
spatial  increments  used in the  model. These
should be consistent with the kinds of questions
that might be asked of the model.
  Once the model adequately describes the hy-
drologic and salinity aspects of the basin, other
dimensions should  be added to the model. All
water management  alternatives need to be con-
sidered  in-terms of existing legal, political, and
institutional  constraints and  should  be evalu-
ated through a particular set of social objectives,
economic and otherwise. Such a comprehensive
model is, of course, a future probability. Model-
ing can, however,  aspire towards  such lofty
goals.

Use of Computers for Modeling
  Modeling or simulation of different systems
can be accomplished in different ways. The ap-
plication of electronic models, those that use
high-speed computers,  to simulate  river ba-
sin salinity is  a new technique. This procedure
involves the use of a computer in order to syn-'
thesize a mathematical model of the prototype
system. Mathematics  become the link between
the computer model  and the prototype.  Time
and space scales  consistent  with  the require-
ments of the  problems and data availability are
readily selected.
  Traditionally, electronic computers have been
divided into two general classes, namely, digital
and analog. The digital computer employs a se-
quential procedure to perform step-by-step op-
erations, at high speed and great accuracy, on
combinations of discrete data. The analog com-
puter solves problems through electronic  com-
ponents that  behave in a manner analogous to
the dynamic or time-varying  prototype  system.
A more recent development is the  hybrid  com-
puter.  It combines the desirable characteristics
of both the digital and analog  computer, both al-
lowing for a high level of efficiency in computer
simulation and combining  high speed with dy-
namic accuracy.
  Advantages  of computer simulation  include
an opportunity to evaluate proposed modifica-
tions at a  minimum expense and effort; nonde-
structive testing of the system; no time loss or
inconvenience  in the prototype system while im-
provements or modifications  to existing works
are considered; and considerable  insight into
the physical relationships and properties of the
system  receiving  consideration.   Computer
modeling, or simulation, is valuable in the proc-
ess of  investigating and improving mathemati-
cal relationships. In this respect the computer is
a valuable tool not only for its calculating poten-
tial, but also for its ability to  yield optimum so-
lutions. Modeling is also ideal for investigations
of the sensitive places in the  system.

               Salinity Model
  Salinity in the Upper Colorado River Basin is
dynamic in nature and is totally dependent upon
the quantity of water available to transport salt
loads. Hence,  to describe the  salinity in  the Ba-
sin requires an accounting of the physical, hy-
drologic quantities at various points  within the
basin. Using the concept of continuity of mass,
water moves or is routed through the system in

-------
218
MANAGING IRRIGATED AGRICULTURE
the proper relationship to both space and time.
In the salinity system salt moves by dilutation
and  mixing of water flows. Thus,  the salinity
system is  dependent upon the hydrologic  sys-
tem. The  salinity level of a  given water  dis-
charge indicates the mass rate of salt movement
with that  flow segment. Thus, the hydrologic
and salinity systems are related through the sa-
linity concentration  at  a given rate  of water
flow. The rate of salt input to a system is esti-
mated  from the salinity concentration levels of
all hydrologic flow inputs. If consideration  is
given to storage changes  and influences pro-
duced by water uses within the system, the mass
balance concept provides an estimate of the re-
sultant TDS concentration at the outflow of the
system.
  An analog computer model  is developed to
describe the salinity system in  the Upper Colo-
rado River Basin. To provide spatial resolution
the Basin is divided into 40 subareas or sub-
basins; each is modeled separately. Then, the
subbasin models are linked into a single  model
of the  entire Upper Basin (Figure 1).  The sub-
basin models that correspond to the  numbered
basins  in Figure 1 are given in Table 1. In gen-
                                        eral, subbasin boundaries are established on the
                                        basis of the availability  of water salinity data.
                                        Only the valley floors are included within the
                                        modeled  area, with both  gaged and ungaged
                                        tributary inflows of water and salt from the sur-
                                        rounding drainage areas being represented by
                                        either  observed or estimated input quantities.
                                        The time increment selected  for the model is
                                        one month.  Therefore, time-varying  quantities
                                        are expressed in terms of mean monthly values.
                                           Appropriate historical data from USGS gag-
                                        ing stations  are assembled and used as meas-
                                        ured inflows of water and salt loads to a given
                                        subbasin. Unmeasured water  input  quantities
                                        are estimated  through correlations  based on
                                        precipitation, snowmelt,  and gaged streamflow
                                        rates. Salt flow rates are estimated  by associat-
                                        ing a salinity concentration with each waterflow
                                        component. It is assumed that no salts  are car-
                                        ried by precipitation quantities. Input flows are
                                        routed, delayed, increased, or abstracted within
                                        a  given  system by means  of diversion; return
                                        flows; deep percolating flows;  municipal and in-
                                        dustrial  activities; exports or imports; evapo-
                                        transpiration  and  salt loading from point and
                                        diffuse from natural sources; municipal and in-
                                          TABLE 1
                   SALINITY SUBBASIN MODELS WITHIN THE UPPER
                                COLORADO RIVER BASIN
                 Salinity
             Subbasin Model

         New Fork River Basin
         Green River, above
         LaBarge, WY

         Green River, above
         Fontenelle Reservoir

         Big Sandy Creek Basin

         Green    River,   above
         Green River, WY

         Blacks Fork River Basin
                        Model Number
                        from Figure 1

                               1
                               4

                               5
       Salinity
   Subbasin Model

Colorado River, above
Glenwood Springs, CO

Roaring Fork River Ba-
sin

Colorado River,  above
Plateau Creek

Plateau   Creek   Basin

Gunnison River,  above
Gunnison, CO

Gunnison River,  above
No.   Fork   Gunnison
River
Model Number
from Figure 1

     21
      22


      23


      24

      25


      26

-------
                                         MODELING SALINITY —COLORADO RIVER BASIN

                                  TABLE 1 (CONTINUED)

                  SALINITY SUBBASIN MODELS WITHIN THE UPPER
                                COLORADO RIVER BASIN
                                                          219
                Salinity
             Subbasin Model
Model Number
from Figure 1
       Salinity
   Subbasin Model
Model Number
from Figure 1
          Green   River,   above
          Flaming  Gorge   Dam

          Little Snake River Basin
          Yampa   River   Basin
          Greer    River,   above       10
          Jensen, UT

          Ashley   Creek   Basin       11

          Duchesne River,  above       12
          Duchesne, UT

          Duchesne River,  above       13
          Randlett, UT

          White River Basin             14
          Green   River,   above       15
          Ouray, Utah

          Price River Basin              16
          Green    River,   above       17
          Green River UT

          San Rafael  River Basin       18
         Colorado  River,  above       19
         Hot  Sulphur  Springs,
         CO
         Eagle River Basin
      20
dustrial loads; and of irrigation pickup by leach-
ing soils.
  The evapotranspiration rate is estimated from
an empirical relationship that is dependent upon
surface  air temperature, latitude, available soil
               Uncompahgre River            27
               Basin

               Gunnison  River,  above        28
               Grand Junction, CO

               Colorado  River,  above        29
               CO-UT State Line

               San  Miguel River Basin        30
               Dolores   River   Basin       31

               Colorado  River, above       32
               Cisco, UT

               San  Juan River, above       33
               Arboles, CO

               San  Juan River, above       34
               Archuleta, NM

               Animas River Basin           35
               San  Juan River, above       36
               Farmington, NM

               LaPlata River Basin           37
               San  Juan River, above       38
               Shiprock, NM

               San  Juan River, above       39
               Bluff, UT
Colorado River,  above       40
Lee's Ferry, AZ

 moisture, and  crop species.  No  abstractions
 from the salinity flow system occur through the
 evapotranspiration process.
    The rate of natural salt pickup is determined
 from an  empirical  relationship based on  the

-------
220    MANAGING IRRIGATED AGRICULTURE
                Figure 1: Upper Colorado River Basin Showing Salinity Subbasin Models

-------
                                         MODELING SALINITY — COLORADO RIVER BASIN
                                         221
degree of interchange between surface and sub-
surface flows within the Basin. In addition, an
annual salt balance (no storage) within the agri-
cultural system is assumed.
  Water and salts that flow through the system
are changed  both spatially and temporally as
water with its accompanying salt load moves
through  a  hydrologic system and  as  storage
changes  and  abstractions occur.  Resultant  re-
sponse or output functions of the model repre-
sent the integrated effects of the many physical
and chemical processes that occur  within the
Basin.
  Change in flow of water and salts is deter-
mined  by mathematical expressions describing
the various system relationships (mentioned in
the preceding paragraph).  These expressions,
synthesized into a program for the electronic
analog computer, contain certain coefficients or
model  parameters that are evaluated and fixed
during the  model calibration procedure. Under
this procedure, data for a given subbasin are in-
put to  the computer. The model coefficients are
then adjusted until observed and computed out-
put functions closely match. A model was cali-
brated for each subbasin by matching observed
and computed functions for water and salt over
a period of 24 consecutive months. So far as is
possible, the  calibration period was  selected to
represent a wide range of flow conditions.

   Illustrative Example of a Salinity Model
  Outlined in this section is a brief, non-mathe-
matical explanation of the processes involved in
developing  a salinity  model. In actuality, these
processes are really described by mathematical
expression  that relate the different parameters
and incorporate them into  a model. However,
for the sake of simplicity, the different processes
are discussed in general terms; a more rigorous
explanation is found elsewhere.1
  The  development and verification procedure
that is followed in obtaining a salinity model of
the Upper Colorado River basin can best be il-
lustrated by referring  to a specific example. The
subbasin selected for  illustrative purposes is the
White  River  system.  Each  river system differs
slightly from one to another, but the basic pro-
cedure in the development of a model for a sub-
basin is illustrated well by the White River sub-
basin.  Development  of other subbasin  salinity
models followed the same logic, as outlined in
the following paragraphs, for the White River
model.
  The White River lies primarily in west-central
Colorado, with water flowing westward to the
point of confluence of the White and Green Riv-
ers in east-central  Utah.  The  river basin is
bounded, on the north, by the Yampa  River, and
on the south, by the  Colorado River. Livestock
production, irrigated agriculture (primarily feed
and  forage  crops,  comprising  about  29,000
acres), and production of oil and natural gas
provide the economic base of the area. The 1970
population in the subbasin is about 5,200 with
more than one  half the persons living in the
towns of Meeker and Rangely. The elevation of
the  river  system  varies from  approximately
5,000 feet at the outlet (assumed to be the U.S.
Geological Survey (USGS) monitoring site lo-
cated near Watson, Utah) to about 12,000 feet in
the upper reaches of the headwaters. Because of
this elevation difference, the climate is variable
with wide temperature  extremes in the settled
areas. The average  annual  frostfree period for
cropland varies from 60 to 125 days. Similarily,
the average  annual  precipitation varies from  9
to 30 inches.

Model Development
  The first step  in developing a salinity model
of the White River is to define the model bound-
aries. The boundaries  include all the subbasin
area  located between  the inflow and  outflow
points of the system.
  The inflow of water and salt to the system are
taken at the USGS monitering points  on the
White River near Buford, Colorado (no. 9-3028);
South Fork White River at Buford, Colo. (no.
9-3040); Big Beaver Creek  near Buford, Colo.
(No.  9-3041);  and  Coal Creek near Meeker,
Colo. (No. 9-3043). These stations were gener-
ally  located upstream of the  agricultural lands
and  other human activities of man in the sub-
basin. Salinity data are available and relatively
complete at both input and output points for the
years 1964 and 1965, and these data are used in
verifying the White  River subbasin model. Wa-
ter salinity data are  lacking for both Big Beaver
and Coal Creeks. However, because of the topo-
graphic  and geologic  similarity of these drain-
age areas  to those of South Fork of the White

-------
222    MANAGING IRRIGATED AGRICULTURE
River, salinity concentration measurements for
the South Fork,  near  Buford,  are assumed to
apply to both creeks, Big Beaver and Coal. The
measured inflow,  per month, of water and salts
to the White River Basin, for the years 1964 and
1965, are in Table 2. These values are given in
acre-feet, and the  average salinity concentration
in mg/1, is obtained by weighting, on a flow ba-
sis, the various water and salt inflows.
   Monthly ungaged inflows of both water and
salt to the basin were estimated by a correlation
technique that considers  three  hydrologic pa-
rameters, namely a gaged tributary inflow—the
White  River, near Buford, Colorado—times a
proportionality factor; precipitation times a pro-
portionality factor;^ and the rate of  snow melt
times a proportionality factor. The rate of snow
melt is determined from the available energy
and quantity of  precipitation stored as snow.1
The monthly ungaged inflow quantity,  com-
puted by the model and given in Table 2, is esti-
mated primarily from the first correlation tech-
nique outlined  above.  The assumption is made
that the average salinity concentration of water
entering the White River  basin from unmeas-
ured inflow sources equals that of measured in-
flows.
   Recorded  precipitation  and  temperature
values  for the years  1964 and  1965 were aver-
aged using data  collected from climatological
stations located at Meeker and Little Hill, Colo-
rado. The values  used in the model are given in
Table 2. Precipitation values are used  in the
model as a water inflow. Air temperature is the
parameter used in the model as a criterion for
establishing the form of precipitation and as an
index of the energy available for the snowmelt
and evapotranspiration process.
   The evapotranspiration  relationship used in
the model [Blaney-Criddle] requires, in addition
to mean monthly temperature, the monthly per-
centage of annual daylight hours and a monthly
crop growth stage coefficient.2 The crop distri-
bution pattern used  in the model for the irri-
gated area is 29  percent clover, 27 percent al-
falfa, 34 percent  pasture and meadow hay, and
10 percent grains. The acreage of phreatophytes
within the basin is estimated to be equivalent to
a concentrated  stand of 3,800 acres.
   The actual evapotranspiration rate computed
by the model is equal to or less than the poten-
tial rate computed. As the moisture content of a
soil is reduced by evapotranspiration, the mois-
ture tension which plants must overcome to ob-
tain sufficient water for growth is increased.
When the plants begin to wilt because soil mois-
ture is a limiting factor, an empirical linear rela-
tionship is assumed  between available soil mois-
ture quantity and  actual transpiration  rate.
Thus, the actual evapotranspiration rate is re-
duced to something less than the potential rate.3
The amount of  water transpired by crops and
phreatophytes, as calculated in the model, is
given in Table 2.
  The quantities of irrigation water diverted to
the agricultural  lands during  1964 and  1965
were obtained by distributing a yearly volume of
water according  to  monthly canal or ditch rec-
ords that generally represent the water distribu-
tion patterns of the area. Both the yearly volume
of water diverted and the individual representa-
tive ditch  records  were  obtained  through the
State  Engineer and the appropriate Irrigation
Division Engineer  in  the  State of Colorado.
These quantities  are listed in Table 2.
  Transfers of water either to or from the sub-
basin (imports and exports) and groundwater
pumpage are not significant in the White River
drainage.  Therefore, these processes were not
included in this particular model.
  It is postulated that the natural saltload  is
contributed to the waters of the Upper Colorado
River Basin through an interchange process be-
tween surface and  subsurface waters.  Influent
flow from the main stream enters the ground-
water basin and, under conditions of  equilib-
rium, an  equal  volume of effluent flow enters
back into  the stream in a lower reach of the
channel.  The salinity concentrations of the ef-
fluent waters are assumed  to be equal to  those
of the groundwater. In this study the rate of in-
terchange flow for  each subbasin of the Upper
Colorado was expressed empirically as a per-
centage of the flow rate at a particular point in
the main surface channel. For the White River
system this relationship was based on the sur-
face flows at Watson, Utah. The salt load from
the interchange  phenomenon is computed as a
function of the  percentage interchange,  basin
outflow, and groundwater salinity concentra-
tion. Total  basin outflow rates for both  water
and salt and for  the salt load contributed by nat-

-------
                                        TABLE 2




AVERAGE MONTHLY HYDROLOGIC FLOWS AND SALINITY LOADS WITHIN THE WHITE RIVER SUBBASIN
Surface
Year
and
Month
1964
Jan
Feb
Mar
Apr
May
June
July
Aug
Sept
Oct
Nov
Dec

1965

Jan
Feb
Mar
Apr
May
June
July
Aug
Sept
Oct
Nov
Dec
Inflow, a-f
White R. So. Fork
near White R.
Buford at Buford

6600
6980
7650
8540
38530
49900
20490
12980
9560
9780
8960
9310



8710
7600
7870
10750
45000
71340
34290
18250
14340
10710
9810
8410

4790
5170
6260
6810
39790
62620
15070
9560
6980
6520
6240
6370



6530
6650
6080
7760
33160
80610
23810
11790
9810
6520
6240
6370
Big
Beaver
Creek

111
138
202
704
5250
1440
50
39
23
46
177
182



103
229
385
1261
6511
2707
254
244
306
159
280
158
Coal
Creek

86
92
139
311
1050
770
108
119
94
86
101
111



86
119
206
441
1890
1190
285
165
131
119
148
98
Ungaged
Input
corre-
lations

4000
4300
4700
6400
24400
31000
12500
7900
5800
6000
5500
5800



5200
4600
4800
23500
39600
51600
21700
11200
8700
6500
6000
5100
Weighted
Input Precipi-
salinity tation
rug/ 1 inches

197
192
186
177
117
103
164
171
177
177
174
200



190
181
197
175
120
127
129
177
190
167
171
186

0.66
0.56
1.34
2.72
1.63
2.07
0.79
2.08
0.59
0.22
1.50
2.20



1.10
0.89
0.77
1.14
2.94
1.98
2.74
0.90
2.00
0.30
1.32
2.99
Evapotranspiration
Temper- Crops Phreato-
ature phytes
°F a-f

18.1
19.3
26.3
40.9
51.9
58.5
68.8
62.7
55.5
47.9
31.0
25.3



25.4
24.3
27.1
42.9
50.4
57.3
65.3
62.2
50.5
48.7
38.1
25.5

0
0
0
3000
9800
10400
17500
10700
6200
1900
0
0



0
0
300
3700
7100
10700
13900
11200
5700
4200
800
0

0
0
100
500
1300
1800
2800
2000
1200
700
200
0



0
0
100
500
1200
1900
2400
1900
1000
700
300
0
Salt load
Canal from natu-
diver- ral sources
sions tons

0
0
0
12000
32000
57000
23000
9500
12000
7500
0
0



0
0
0
16000
57000
62000
30000
25000
13000
0
0
0

13100
12900
15300
15900
23600
24300
16000
15000
12300
13800
13600
14200



15700
14700
18000
17900
26500
31900
23400
17800
17600
18100
16600
16700
Water Salt
outflow Outflow at
Watson Watson,
Utah, Utah,
a-f tons

17100
18760
26040
30320
84210
99350
30220
24680
15700
20060
20350
21540



23350
22390
33460
35100
98050
157700
69380
31910
32520
33560
27800
26580

15800
16050
21570
26760
34820
33370
19190
20140
12900
15580
16670
20450



21150
19430
31030
33220
45840
59840
45580
23520
23220
22780
19890
21940






o
O
r
3
C/3
r
3
H
*%
|
n
o
0
73

o
o
2
Ffl
73
93
CO
HH
z;

«

-------
224
MANAGING IRRIGATED AGRICULTURE
ural sources, as estimated by  the interchange
process, are given  in Table 2. The actual value
of the salt load contributed from natural sources
requires further research  to  define adequately
its magnitude.
  Several  other single-valued  parameters are
required in the  model, such as soil moisture ca-
pacity, the rate of snow melt, various correlation
coefficients,  and delay times for  movement of
different flow components in the  ground water
system. These  parameters,  once  determined
through a trial and error verification process,
are assumed fixed for a given subbasin.
  The  appropriate routing of inflow water and
salt through the system in relation to loss, and
additions, which occur over space and time, re-
sults in a flow at the outlet point (Watson, Utah)
that is both surface and subsurface. Active net-
work delays on the computer simulate the long
transport times  necessary for groundwater flows
and deep  percolating waters from irrigation re-
turn flows to be routed to the outflow gaging
station.

Model Verification
  Computer output from  the verified model of
the White River subbasin is shown in Figure 2.
(The model  was calibrated with 1964 data and
                                         tested over  the  12 month period of 1965.) A
                                         comparison  is made between computed and ob-
                                         served values for both water discharge in acre-
                                         feet and salt discharge in tons for both years at
                                         the monitoring station near Watson, Utah. On
                                         an  annual  basis the differences between the
                                         computed and observed values  of discharge do
                                         not exceed  about three percent for water and
                                         seven percent for salt. On a monthly basis dis-
                                         crepancies are slightly higher,  but  in  general,
                                         good  fits were obtained for both the water and
                                         salt discharges. As a point of interest the agree-
                                         ment  between the computed and observed salt
                                         outflow  rates for  the  White  River subbasin
                                         proved to be among the least accurate of those
                                         obtained for all  of the subbasins of the Upper
                                         Colorado River  drainage.
                                           Additional output functions, from the White
                                         River basin  model, that illustrate the kind of in-
                                         formation available are shown by Figures 3 and
                                         4. Time variation of the soil moisture level in the
                                         plant root zone of the basin is illustrated in Fig-
                                         ure 3.  Snowmelt produces the sharp  rise in
                                         soil moisture storage during April. In early May
                                         the capacity is reached. Irrigations  and rainfall
                                         are sufficient to  virtually maintain this level
                                         throughout  the remainder of the year.  In sub-
                                         basins,  where adequate supplies of irrigation
                                                Computed
                                               . Observed
                              J1 J ' A  ' S '  0 ' N '0 ' J  ' F 'M ' A 'M ' J  ' J ' A '  S '0 'N 'D
                                                 Dj   F

                Figure 2: Computed and Observed Monthly Discharge of Water and Salt

-------
                                         MODELING SALINITY —COLORADO RIVER BASIN
                                                                   225
       2-
        1 -
            JAN '   FEB  ' MAR ' APR  '  MAY  '   JUN  '  JUL'  AUG  '  SEP  '  OCT  '  NOV  '  DEC
water are  not available, soil  moisture levels
might not  reach capacity during the irrigation
season.
  Deep percolating waters from the agricultural
lands are believed to move through the ground-
water basin, eventually  to  appear  as  effluent
flow in the main surface channel of the  sub-
basin. The discharge function for the salt  load
carried  by these waters, as computed by the
model  of the White River basin for  1964, is
shown in Figure 4. These flows were delayed by
  MAK    AfK    MAT    JUN    JUL     AUli    btr

Figure 3:  Available Soil Moisture in Agricultural Area
                         3.5 months (3.5 seconds computer time) from
                         the time of percolation to the time of outflow
                         from the groundwater system.

                                     Salinity Model Results
                            The general procedure outlined in the preced-
                         ing  section for the development  of  a  salinity
                         model of the White River subbasin was repeated
                         for each of the 40 subbasins considered in this
                         study (Figure 1). The subbasin models are then
                         linked into a single model of the entire Upper
            JAN  '  FEB
          Figure 4: Rate of Salt Movement with Deep Percolating Waters from Agriculture Area

-------
226    MANAGING IRRIGATED AGRICULTURE
Colorado River Basin.1 Some general results ob-
tained from the salinity models are given in the
following paragraphs.
  The  average  annual  water discharge rate
from the Upper Basin, modified to 1965 condi-
tions, is  estimated  to  be 10,900,000 acre-feet.
The average annual virgin flow at Lees Ferry is
about  14  million acre-feet  per year of  which
about 10,500,000 acre-feet of water is measured
directly, as to it's source of  origin, at points on
tributaries to the main stem  rivers.
  The  average annual salt discharge from the
Upper Basin above Lees Ferry is estimated to be
8,600,000 tons. This salt discharge produces an
average salinity concentration  at Lees Ferry of
580 mg/1. Within  the Upper  Basin it is esti-
mated  that average salinity concentration at
Lees Ferry is increased by about 20, 45, and 115
mg/1 by reservoir evaporation, consumptive use
by phreatophytes, and by agricultural evapo-
transpiration,  respectively, from what is would
be with no concentration of salts by these con-
sumptive depletions.
  Salinity  measurement of tributary streams
within  the various subbasins of the Upper Basin
measure about 1,700,000 tons  of the total aver-
age annual salt outflow of 8,600,000 tons re-
corded at Lees Ferry. On the basis of these fig-
ures, only 20 percent of the  average annual salt
flow from the Upper Basin is measured with re-
spect to its area of origin. This limitation of
available salinity information explains in part
the difficulty encountered in simulating the salt
flow system. The simulation  models indicated
that the remainder of the total average annual
salt load originates from the following sources:
(1) 1,100,000 tons from ungaged tributary  in-
flows; (2) 1,500,000 tons from pick-up of salt by
irrigation return flows;  and (3) 4,300,000 tons
from natural diffused  and point sources  within
the system. It is recognized  that the magnitude
of salt load contributed from the  latter two
sources  requires additional research to ade-
quately define this portion  of the salinity sys-
tem.

SUMMARY
   Many of the factors that affect the quantity
and quality of water resources of a  river basin
are subject to management or regulation. Opti-
mum water use, assessment of management al-
ternatives, and logical criteria for regulation and
administration  of  the  water  resources  are
needed. With  increasing pressures  upon the
Colorado  River Basin water resources more so-
phisticated models  are required for planning
and management purposes.
  A general model of the salinity flow system of
the Upper Colorado  River Basin is proposed
and synthesized on an electronic analog  com-
puter. The basis of the model  is a fundamental
and logical mathematical representation of the
various salinity and hydrologic flow processes.
The salinity model is macroscopic in nature and
describes only  the basic  salinity system within
the basin. The model is not refined or sophisti-
cated  in terms of describing some of the specific
details and intricate  processes that actually oc-
cur. Additional improvement of the definition of
the salinity system can and should  be accom-
plished  through  the  development  of   other
models  that describe these complex processes.
Such models could then be incorporated into the
general salinity model to improve both the spa-
tial and temporal definition of the flow of salts
in the Upper Basin.
  Inadequate data frequently restrict the ability
of a model to define a system. In this study more
complete data would enable refinement of the
major salinity processes  within the model and
thus provide one that more accurately simulates
the salinity flow system. The level of complexity
of a model that describes a salinity  system de-
pends upon the fineness in time and spatial data
that are used in the model.
  Modeling the salinity  of the Upper Basin is
accomplished by  tracing  the average monthly
mass rate  of water and salt through the various
paths  in the hydrologic  system. The Basin is
subdidived into 40 smaller units or subbasins. A
salinity model is developed for each subbasin by
routing stream inflows, diversions to irrigated
lands, evapotranspiration, municipal and indus-
trial uses,  return flows, changes in ground water
storage, and stream  outflows  through the sys-
tem. The salinity concentration associated with
a given segment of water flow determines the
volume of salts that are transported by that par-
ticular flow segment. Unmeasured quantities of
water and salt are  estimated  through various
correlation techniques. Salt loads from  point
and diffuse natural sources are determined from

-------
                                       MODELING SALINITY —COLORADO RIVER BASIN
                                        227
an empirical relationship. The resultant output
from the model represents the integrated effects
of the many physical and  chemical  processes
that occur within the basin. Verification of the
model  is accomplished  by  comparing the  re-
corded outflow from the basin with that  com-
puted by the model. The 40 subbasin models are
then linked into a single model of the entire Up-
per Basin.
  The development and verification procedures
that are followed for subbasin model are  illus-
trated using the White River system (in western
Colorado). Each of the 40 subbasin models were
developed in much the same manner. Some  re-
sults from the general salinity model of the Up-
per Colorado River Basin are  given.

ACKNOWLEDGEMENTS
  This study was performed at the  Utah Water
Research Laboratory, Logan,  Utah  for the Fed-
eral Water Quality Administration under P.L.
89-787,  Demonstration Grant Number 16090-
DUV,  Contract  Number  12-14-100-9715(41).
Assisting in the study were J. Paul Riley, M.
Lynn  McKee,  and  Eugene  K.  Israelsen.

REFERENCES
  1.  M. Leon Hyatt, "Analog Computer Model
of the Hydrologic and Salinity Flow Systems
Within The Upper Colorado River Basin." Ph.
D. Dissertation. Department of Civil Engineer-
ing, College of Engineering, Utah State Univer-
sity, Logan, Utah, July, 1970.
  2.  Harry F. Blaney and Wayne D. Criddle,
"Determining Water Requirements in Irrigated
Areas  from   Climatological   and  Irrigation
Data." Technical Paper No. 96, Soil Conserva-
tion  Service, U.S.  Department of Agriculture,
February 1950.
  3.  J. Paul Riley, D. G. Chadwick, and J. M.
Bagley,  "Application  of Electronic  Analog
Computer to Solution of Hydrologic and River-
basin Planning  Problems: Utah  Simulation
Model II," Utah Water  Research Laboratory,
Utah State University,  Logan,  Utah, 1966.

-------
                    Hydrologic Modeling of

                         Ashley  Valley,  Utah
                                  ROBERT F. WILSON
                                  Bureau of Reclamation
                             Engineering and Research Center
                                    Denver, Colorado
 ABSTRACT
   A  mathematical  model for predicting  the
 mineral quality of irrigation return flow water
 is presented.  The model has been tested using
 data  collected from  an  existing  irrigation
 project.
   In concept the model incorporates the use of
 deterministic  and/or probabilistic inputs  and
 demands and in turn  measures the system's
 responses or  yields  under a  multitude of vari-
 able  systems, operation  criteria  and  design
features.  Two of the five computational blocks
 are described in detail, the data analysis sub-
 model and the simulation submodel. The data
 analysis submodel is designed to measure  the
 worth or information content of all input data
 sets.  The simulation model has been designed
 using a  nodal  scheme  which allows many
 configurations in the simulation  of  the whole
 water resource system.
   Representative  graphical  output from   the
 use of the  Vernal Unit data is included.  The
 "goodness of the prediction" results  somewhat
from  repeating data to fill in missing years or
 months.

 INTRODUCTION
   The objective  of  this study is to  provide a
 technique for predicting  and assessing more
precisely than now possible the mineral quality
of irrigation return flow. The means for accom-
plishing this objective will be through the use of
mathematical   models  and  high-speed com-
puters. The mathematical model  is primarily a
mathematical formula or expression attempting
to duplicate conditions encountered on an irri-
gation project.  The study utilizes data from ex-
isting irrigation projects in order to verify the
technique. If it is possible to duplicate or near-
ly duplicate existing  conditions with the model
then confidence is established and the method
can  be  used,  with a small  amount  of water
quality data, to predict mineral quality from re-
turn flow.
  The next step would be to predict mineral
quality of return flow under modified manage-
ment to see if improvements could be achieved
through irrigation scheduling or improved ir-
rigation systems.
  The Vernal  Unit in northeastern  Utah was
chosen  for the  initial  modeling effort.  The
selection was  made  partially on  the  basis  of
available data and the fact that the inflow and
outflow  salt loads and water quantities from the
project could be easily measured. Considerable
water quality and quantity data were collected
in this area prior to completion of a storage re-
servoir and continuing through the beginning
of project operation.
                                           229

-------
230     MANAGING IRRIGATED AGRICULTURE
  Initial  runs on the computer using  existing
Vernal data  are encouraging and the possi-
bility of success in predicting salt loads in return
flow now seems to be assured.
  The  most important requirement in  a  study
of this type is to have  reliable and  consistent
data, especially flow and quality and consump-
tive use. Application of the model to the Vernal
data turned  up several  inconsistencies in the
basic data, the most serious being the estimated
consumptive  use and the determinations of the
chemical exchange in the return flow waters.

            Description of Model
  The  mathematical model  is  being designed
within   the   context  of  systems  engineering
utilizing  general aspects of  systems  theory,
probability, and mathematical statistics. It also
utilizes computer science, engineering, mathe-
matics  and numerical analyses.
  In concept, the model incorporates the use of
deterministic  and/or probabilistic inputs,  de-
mands,  and  in  turn measures the system re-
sponses or yields under  a multitude of variable
systems operation criteria and design features.
The  systems model  under this concept  consists
of five fundamental computational blocks which
could be considered as primarily submodels of
the overall system  model. Each primary  sub-
model can be used independently or collectively
to provide flexibility in the design of an overall
model  for a particular system objective and al-
ternatives thereof.   The  five  computational
blocks  or submodels are listed below in the se-
quence pertinent to the flow  of the complete
model as conceived:
  1.  Data analysis submodel
  2.  Simulation submodel
  3.  Sensitivity and impact analysis submodel
  4.  Optimization   submodel  utilizing linear
     programming  theory, i.e.,  the  ability  to
     maximize  or minimize a linear objective
     function (dual simplex algorithm) on the
     basis of pertinent constraint equations
  5. The best plan submodel utilizes a modi-
     fied form of dynamic programming theory
     (Buras, Bellman, et al.)
  Some  of the  primary submodels have been
used to  study the Vernal Unit water resource
system. The  system study which  was a  con-
junctive use analysis has been completed using
historical  data for the period  January  1958
through  December 1962.  All inputs and re-
sponses for this study were  in a monthly time
frame.
  The objective of the Vernal study was to use
a model in predicting changes in capacity and
the associated water quality distribution of the
aquifer and also the quality distribution of the
water as surface effluents from the system. The
water quality distribution was reported in terms
of concentrations  of calcium, magnesium, sodi-
um and potassium combined,  sulfates, bicar-
bonates, carbonates, chlorides, and in total salts
as tons per acre.  The prediction of the system
responses was compared with the historical data
(drain data and aquifer data), both quantity and
quality  distributions, as a measure of the reli-
ability of the  model.  With the  exception  of
widely divergent comparisons of aquifer capaci-
ties, all other comparisons appeared adequate.
  It was concluded from the study of this his-
torical period that the divergence in aquifer
capacities was an  accumulation of errors perpe-
trated throughout the system analysis.
  Each of  the primary submodels described
above is further subdivided  into sets of minor
submodels  or subroutines.  Each  of the  sub-
routines  was designed  to  be used  either in-
dependently or collectively in any  of the  parti-
cular  primary submodels.
  The  following   description  of  the  various
functions contained  within  the model  as sub-
routines will focus primarily on how the model
could be used for the  purpose of studying an
irrigation scheme  and the prediction of quality
distribution throughout the conjunctive use sys-
tem as affected by such an irrigation scheme. It
must  be recognized, however, that the model
has the capability of handling simultaneously
multipurpose conjunctive use water resource
systems which will accommodate the operation
of the system for flood control, municipal and
industrial requirements,  power  requirements,
salinity control and recreational requirements.

Data  Analysis Submodel
  Initially all available historical data pertinent
to the water resource system would be collected
and compiled. These  data would consist  of
surface flows and  associated quality distribu-
tions,  temperature, solar radiation, irrigation
data such as acreage under  irrigation, types of

-------
                                                             HYDROLOGIC MODELING
                                         231
crops being irrigated, acreages under nonbene-
ficial use, soil data such as depth of soil column,
degree of saturation, quality distribution,aqui-
fer characteristics, etc.
  Hydrologic  processes  may  be deterministic
or stochastic (probabilistic) or a combination
of both. A process where a definite relationship
exists between the hydrologic variable and time
is called deterministic. In this case, the function
defines the process for all time and each obser-
vation in  succession adds no  new information
concerning the process.  The  opposite of this
case is called stochastic (probabilistic) and is
represented all or in part by a random mecha-
nism. Generally, hydrologic data are a combina-
tion process  and the data analysis  is directed
toward isolating these components.
  The hydrologic and climatological processes
(historic  data sets as part of input to system
study) will in all cases be treated first as a dis-
crete time series. The time interval can be daily,
weekly,biweekly,  or monthly. It will then be
assumed  that the series is first order stationary
(mean) and second order stationary (variance).
The next important assumption made is that the
series is homogeneous in time or that each event
has the same recurrence interval at all times. A
test is made at this point and if the series is
found to  be nonhomogeneous in time  or  the
non-homogeneity exists at a single point (event)
i.e., the criteria of equal  probability is violated,
an  attempt will be  made to remove  the non-
homogeneity. If the removal is not possible, the
series  decomposition  of these  data will  be
terminated and all statistics will be saved  for
further  treatment in obtaining  a  transfer  of
filter function for these data in a  subsequent
portion of the data  analysis submodel. On the
positive  side  the  series  decomposition  will
continue.  When  criteria of homogeneity  are
violated, such violation may be caused by man's
intervention or, for that matter, any natural
phenomenon.  Natural changes or  phenomena
include among other things trend  and period-
icity.  Changes due to man's intervention may
include the effects of dams,  diversion works,
watershed management, and/or other causes.
  The subroutines used  in  the  data analysis
submodel for the evaluation of the deterministic
components are:
  1. Fourier analysis for the detection of and
     subsequent removal of the cyclic or peri-
     odic  characteristics  of the first moment
     (mean) and the second moment (variance)
     of the series.
  2.  Autocorrelation  and spectrum analyses1
     subroutine. This is  also a multiple-entry
     subroutine  and  at  present handles  the
     spectrum of the  first through third order
     linear  autoregressive  models.   A  simple
     modification  to  this  subroutine  can be
     made to accommodate higher than- third
     order autoregressive models.
  3.  A complete regression analysis  for simple
     linear and  simple  curvilinear  as  well as
     multiple   linear  (step-wise)2   with  the
     inclusion of 10 or so transforms  and ac-
     ceptance and rejection criteria on a basis
     of "P1 levels.
  Minor subroutines used in  evaluation of the
stochastic (probabilistic) component of the time
series are:
     a. A  ranking routine which   ranks  the
      series  in ascending or  descending order
      of magnitude, and then divides the rank
      series  into  cells of equal frequency or
      equal probability.
     b. A routine that computes the frequency
      function,   the  cumulative  distribution
      function  for  the  standardized  normal
      distribution where the  integration  for
      the  cumulative  distribution function is
      obtained  using  Simpson's rule.  A  sec-
      ond entry into this routine is  utilized to
      determine  the  nonstandardized  vafiate
      with  a simple  transform as  a function
      of  the  standardized variate with  the
      computed   mean   and   variance,   i.e.,
      mean   not  equal  zero  and variance
      not equal one.  A  third entry into  this
      subroutine  will compute the  inverse of
      the  cumulative   distribution function
      which simply finds the value of the vari-
      ate for a given probability and again in
      this case a simple transform from the
      standardized variate to the nonstandard-
      ized  variate is performed. Several al-
      gorithms are available  in the literature
      for  the computational accomplishment
      of  those  factors considered in the sub-
      routine.
  4.  A  multiple-entry subroutine  that  treats
     the   Poisson's,   Exponential  (single  or
     double  branch),  Beta,   two   and  three

-------
232    MANAGING IRRIGATED AGRICULTURE
     parameter  gamma  functions.  The  sub-
     routine contains all the necessary approxi-
     mations to the  inverses of the integrals of
     these functions.
  5. A  subroutine  which  is actually  a third
     degree  spline   fitting  function  which
     would  allow a  fit  through  each of the
     points  of  the  cumulative  distribution
     functions and integration of same, and in
     turn with  the use of the simple regression
     analysis can be transformed  to the stan-
     dardized   or   nonstandardized  normal
     deviate. Transforms  for all of the func-
     tions as described above have been design-
     ed in terms of. the  standard  or nonstan-
     dardized variate.
  6. As  the time  series  decompose  in this
     manner,  the  residuals  which  represent
     the  final unexplained variance of a totally
     decomposed  series  are  also  examined.
     This is done by a subroutine which com-
     putes  the  Chi square  value for the com-
     parison of the fitted and computed distri-
     bution functions. A  critical  Chi  square
     value  is assigned at the discretion of the
     analyst, which in turn is a function of the
     overall study objective.
  To summarize the capabilities  of the  data
analyses  submodel it can be said that within
the  structure  of the various subroutines all
historic data (input to model) can be evaluated
as (1) a series,  homogeneous (or homogeneous
if nonhomogeneity can be removed), preserving
first and second order statistics, with probabi-
listic (stochastic) preserving third and fourth
moments, i.e.,  shewness and kurtosis such that
larger sample periods can be generated from the
smaller historic sample period, and (2)  transfer
of information from  one point in the system to
another point  in the system by  (a) filter func-
tion, (b) transfer function, both of which can be
deterministic, probabilistic, or  a combination
of both.  Further, all pertinent statistics will be
preserved with  respect to mean, variance,  co-
variance, and the classical cumulative distribu-
tion functions (including multivariate, probabi-
listic). It is now obvious that the data  analysis
submodel is designed to measure the worth or
information content  of all  input data sets and
such measures and related statistics will be used
extensively in  the sensitivity (statistical)  and
impact submodel.
Description of Simulation Submodel
  The simulation submodel has been designed
using a nodal scheme.  Each node  has all the
characteristics of the simulation submodel as a
whole  and each node is an entity. The nodal
concept  allows  many  configurations  in  the
simulation of the whole water resource system;
consequently,  it  provides maximum flexibility
in the design of  alternate schemes of operation
for the system without major modification to the
overall system simulation. A node can represent
the simulation of subbasin within the  whole
water resource system or it can simply represent
a  point  of  measuring  responses for a  single
dimension within the system.
  The number of nodes that can be accommo-
dated within  the simulation submodel is depen-
dent  upon the  capabilities of the computer
system  being used for  the   analysis.  Most
modern computer systems such as the CDC
6000 to 7000 series can  handle  as many as 100
nodes without overlays or segmentation.
  The simulation submodel, like the data analy-
sis  submodel,  is  composed  of several  sub-
routines or minor  submodels,  as shown in the
flow diagram, (Figure 1).
  Any of these subroutines can be  used in-
dependently or collectively, as is the case with
those in the data analysis submodel.
  1.  Control  Subroutine — This particular sub-
     routine is used in conjunction with a set of
     input  control  cards  which designate the
     sequence of each node in  the system and
     the  features of each node. The sequence
     assigned each node provides the key as to
     the direction of flow within water resource
     system.  The control  subroutine also sets
     the indices  for entry or nonentry of con-
     straint equations pertaining to any parti-
     cular node.
  2.  Run of River (out of kilter) Subroutine —
     This  subroutine  is a  multiple-entry sub-
     routine  which purposely  sets  the whole
     system out of kilter or balance  at the start
     of each "ith" iteration through the simula-
     tion submodel and at the end of each itera-
     tion maintains and  updates the delta flows
     (quantified) required as entering or leav-
     ing each node to  balance the water re-
     source system.

-------
TO'
3
 a
><
 Q.
 i
I/I
w
I
o1
C/3
o
C/3
cr

I
                 SURFACE STORAGE, CANAL
                  RIVER, PIPELINE AND
                    TREATMENT PLANTS
                     SYSTEMS MODEL
                                               QUALITY DISTRIBUTION MODEL
                                                                  TOTAL  AVAILABLE WATER
                                                                        SUPPLY
                                               QUALITY DISTRIBUTION  MODEL
PERCOLATION PONDING MODEL
                                                                                                                   PRECIPITATION MODEL
                                                                                                           (EVAPJ
                                                                                                         .  MODEL  J
                                                                                                                                     RETURN FLOW AND
                                                                                                                                       LAG MODEL
                                                                                                                       QUALITY DISTRIBUTION MODEL
                  MODIFIED DUTT QUALITY
                    MODEL PERCOLATION
                  THROUGH SOIL COLUMN

UTION MODEL

k
i
'
t
OTHER
MODE
1
\
i
f EVAP "
^ MODEL
                                                                                                   CONSUMPTIVE AND CROP
                                                                                                     IRRIGATION MODEL
                                                                                                                        NONBENEFICIAL
                                                                                                                    CONSUMPTIVE USE  MODEL
                                                                                                                                 4-
    GROUND WATER  STORAGE
    AND AQUIFER SYSTEMS
          MODEL'
                                               QUALITY DISTRIBUTION MODEL
                                                                                                                                                  INTER-NODAL
                                                                                                                                                  TRANSFER OR
                                                                                                                                                  SYSTEM LOSS
                                                                                                                                                     r
                                                                                                                                                           AS OF 6-30-71
                                                                                                                                                                                    a
                                                                                                                                                                                    o
                                                                                                                                                                                    *
                                                                                                                                                                                    o
                                                                                                                                                                                    o
                                                                                                                                                                                    o
                                                                                                                                                                     O
                                                                                                                                                                     o
                                                                                                                                                                     m
                                                                                                                                                                     r
                                                                                                                                                                     M
                                                                                                                                                                     2!
                                                                                                                                                                     O

-------
234     MANAGING IRRIGATED AGRICULTURE
  3.  Reservoir  Storage Subroutine — In this
     subroutine an index is obtained from the
     control subroutine as to (a) mathematical
     relationships   or  area  and  capacity  as
     functions of reservoir levels; also, from an
     index obtained from  the  control  sub-
     routine the computation enters (b) a sub-
     routine to compute evaporation (multiple
     entry) as a fixed value, as a  function of
     temperature and  precipitation (determin-
     istic), or as  a multivariate  probabilistic
     distribution    function   of   temperature
     (ambient)  and precipitation at a pseudo
     random  probability level. With the use of
     another  index obtained from  the control
     subroutine, the computation enters (c) a
     multiple-entry  subroutine  to   simulate
     water quality  changes at  10 strata in the
     reservoir where the upper boundary of the
     first  strata is  the updated reservoir  level
     from  the  last iteration  and the lower
     boundary is fixed at a point of the lowest
     outlet elevation. The quality  distribution
     of the water leaving the reservoir will be
     the average  quality  distribution  for all
     strata above  the  reservoir level obtained
     at the completion of this at "ith" iteration.
     Constraint  equations  pertinent   to  the
     reservoir operation  will  be  obtained  as
     indices from the control subroutine.
  4.  Irrigation  Demand  Subroutine  —  This
     subroutine  will also  be a multiple-entry
     subroutine  based upon  indices obtained
     from the control subroutine where the ir-
     rigation  demands  will be used  as  con-
     straints  or  as variables which  include
     losses in the  transmission added to  the
     consumptive use requirement  as obtained
     from  the   consumptive  use  subroutine
     which can be  the Blaney-Criddle,  Jensen-
     Haise (deterministic models) or a multi-
     variate  distribution   of  solar radiation,
     precipitation, and crop species (beneficial
     or nonbeneficial) and updated by number
     of acres irrigation  (indices  from  control
     program).
  5.  Return Flow Subroutine — This is also a
     multiple-entry  subroutine  where  return
     flow is simply the difference between the
     irrigation  demand and consumptive  use
     (with or without a time lag). Where the
   return flow can be directed to an inter-
   nodal transfer (surface or a ponding facil-
   ity) or is a loss to the system, the quality
   distribution of the  return flow  can be
   deterministic   or  probabilistic  as  distri-
   bution  obtained  from historic  data of
   drain outflow, quantity and quality wise.
   A  constraint  equation will enter into the
   subroutine to determine  violations of
   salinity standards, and at the "ith"  plus
   one  iteration  the  irrigation  demand will
   be reduced to meet the  salinity standard
   and in this case the reduction will not be
   considered as a shortage  in maintaining
   the ideal irrigation demand.
6. Percolation Subroutine — This is a multi-
   ple-entry subroutine where an index as ob-
   tained from  the  control subroutine will
   indicate a lateral transmission at a parti-
   cular level in the soil column and the lag
   time required as simulated in the ponding
   facility  which is really a function of the
   flow rate  (vertical) through  the  soil col-
   umn. For  example, if it takes 2 months for
   a unit of water to percolate from the sur-
   face  down through the soil column to the
   aquifer, the water to be percolated will be
   held  in the  ponding  facility for the 2-
   month period and the simulation  to the
   soil column will take place during the next
   time  interval. All water held in the pond-
   ing facility will be percolated through the
   soil  column  as  a  stepped-piston  effect.
   The percolation subroutine is the original
   Dutt model with the  use of modified al-
   gorithms.  Figure  2, is a  diagram of this
   part of the program.
7. Aquifer Subroutine — This is also a mul-
   tiple-entry  subroutine where  an  index
   obtained from the control subroutine first
   updates the aquifer quality and  quantity
   wise  with  any lateral internodal transfers.
   At present, the quality distribution model
   for this transfer is a simple linear function.
   Constraint equations for  each of the aqui-
   fers  will be obtained  as an index  from
   control  subroutine. No internodal trans-
   fers (lateral or vertical) will be made at the
   "ith" iteration until after the aquifer  has
   been updated and considered as being in a
   static state. This update will also include

-------
                                                            HYDROLOGIC MODELING
235
SURFACE PONDING FACILITY
^««
QUALITY DISTRIBUTION MODEL
0
u
o
i-j
o
o
pi
PM
M
co
w
pi
CJ
1-1
u
E-"
CO
CO
INTER-NODAL TRANSFER OR
SYSTEM LOSS
i
* . '
y
^ \
\
•f
•tf ' •
•>\
<*
X
*• '/'
/ 1
\^-
. "* '/
^
§
pt,
0
w
pi
^ ^



r
^
^
^
' . .^i //'^* ^/X i'^tf ' ^-^/'/ ~ * I i £*
s\
•/ INTER-NODAL TRANSFER OR
/. SYSTEM LOSS AT ANY
s INCREMENT IN PERCOLATING ^
>' COLUMN
s
,/
^ /
V
v/
V.
<^ *PISTON DISPLACEMENT FROM
\V ONE INCREMENT TO NEXT LOWER
'^ INCREMENT
^ GROUND WATER STORAGE
AND AQUIFER SYSTEMS MODEL
V
QUALITY DISTRIBUTION MODEL
AS OF 6-30-71
INTER -NODAL TRANSFER OR .
SYSTEM LOSS
Figure 2: Detailed Diagram of Dutt's Percolation Model as Contained in Simulation Submodel

-------
236    MANAGING IRRIGATED AGRICULTURE
     all waters as percolation through the soil
     column. At  the present time the model of
     quality  distribution  for  the  percolated
     waters and  those  in the aquifer is simu-
     lated with the use  of a simple linear func-
     tion.
  As mentioned  earlier, all of the subroutines
in the simulation  submodel can  be used in-
depently  or   collectively.   Therefore,   with
the exception of physical demand  constraints,
these subroutines can be modified  at will with-
out  interruption  to the  logical flow  of the
subject model or violation of the overall concept
of the simulation submodel. Such modification
will also not violate the  concept of the complete
systems model.
  The computer  output and input  formats for
the  simulation submodel  can  be printed  or
graphical CRT representations. The output can
also be modeled to set varied report objectives.
Input and  output subroutines will  be designed
as per the desires of the analyst and in all cases
will  not be included as an integral part  of the
systems model.

Irrigation Return Flow Simulation Model
  In  addition to  the mathematical prediction
model thus far described,  an irrigation  return
flow simulation model has been developed that
will permit prediction of return flow patterns for
lands drained  by tile drainage systems. This is
a modification of a model that will be described
later by Dr.  Gordon  Dutt. Both quantity and
quality variations are predicted by the  model.
Where desired,  these  predictions   will  be in-
corporated into the overall system model.
  In  this  separate program  the  process  is
treated as deterministic and the movement of
water is traced through the soil system, starting
at the soil surface and ending at the drains.
  As presently utilized, this later  "Dutt" pro-
gram involves or requires the use of three other
programs. The Dutt program itself consists of
two  programs, one an unsaturated  flow model
and the other a chemistry model.
  The three associated programs are:
  1.  An irrigation scheduling program to obtain
  inputs for the unsaturated flow model.
  2.  A saturated  flow model that transfers the
  percolation to the ground water to a drainage
  system.
  3. A drainage system design program that
  adjusts the output of (2) above into a drainage
  outflow pattern.

  The irrigation scheduling  program was de-
veloped to predict timing and amounts of irri-
gations and is used in an applied research study
to improve irrigation efficiencies. Climatic data
(solar  radiation,  temperatures,  wind  move-
ment,  humidity, and  precipitation),  soil data
(moisture  holding  capacity), and crop data
(types, planting date, and harvest date) are used
as input to the program. The program computes
a running balance of soil moisture and  sched-
ules timing  and amount  of irrigation. Con-
sumptive use is adjusted for soil surface wetting
and the drying of soil profile.
  In  the  Dutt  program, infiltration into the
unsaturated soil zone,  the removal of water by
evapotranspiration processes with depth, and
the  movement  of water  under  unsaturated
conditions are simulated by the numerical solu-
tion of a nonlinear form of Darcy's Law. This
form  arises  when permeabilities  and tension
heads  are taken as functions of the moisture
content.  Saturated flow is  treated  by com-
bining a steady-state  potential  flow drainage
solution with a transient solution to calculate
the water  table buildup due  to intermittent
recharge and  the corresponding drain dis-
charge. The  steady solution  is  merely used to
obtain the relationship between flow down a
finite number of stream tubes.
  The assumption is made that  flow rates and
consumptive use  are  independent   of  water
quality. Consequently, quality  considerations
are treated  separately once  water movements
are predicted.  Processes modeled include ion
exchange as  well as  precipitation  and solution
of ammonia, calcium,  sodium, magnesium,  bi-
carbonate,  chloride,  carbonate,  and  sulfate in
both  the  saturated and unsaturated systems.
Only base soils are treated. In addition various
nitrogen transformations  are  also   simulated
under   aerobic   conditions.   These   trans-
formations  are  relatively  slow  and must  be
treated as kinetic. Nitrogen uptake by plants is
included.

  A large number of inputs are required  by the
model.  For  a given group  and  site,  those
include:

-------
                                                                HYDROLOGIC MODELING
                                          237
   1. System Geometry
     Depth to barrier
     Depth to drains
     Drain spacing
     Drain filter design
   2. Unsaturated Characteristics
     Conductivity-moisture  content  relation-
       ship
     Diffusivity-moisture content relationship
   3. Saturated Flow Characteristics
     Permeability
     Specific yield
     Porosity
   4. Initial Conditions
     Soil extract  analysis  for both  saturated
       and unsaturated soils
     Initial moisture contents
     Initial water table position
     Initial quality of ground water in storage
     (These are  not  strictly necessary.)
   5. Water Inputs
     Irrigation applications and dates
     Precipitation applications and dates
     Semimonthly consumptive use
     Root distribution
   6. Quality Inputs
     Irrigation water chemical  analysis
     Chemical   fertilizer  applications,   dates,
       analyses  and application technique
     Organic   fertilizer   applications,   dates,
       analyses, and application technique
     Plant uptake of nitrogen
     Temperature data in soil profile

   Inputs  under  Item (5)  are determined  by
using climatic data, moisture holding capacities,
and crop  information when irrigation sched-
uling is practiced.
   When combined with  the saturated  flow and
the drainage system  design programs the out-
put  includes  moisture  contents,  water table
positions,  flowlines, and drain discharge for
flow. Quality of the  soil water can be monitored
at the water table or at the drains. Accumulated
pickup by the drains as well as yearly rates can
also be  computed.  Output  includes  both the
usual  printed  listings  and graphical  results
(microfilm) produced by a cathode  ray tube
plotting  system under control of the  digital
computer.
   During  applications  of the  model  to the
 Vernal Unit, several inconsistencies in the basic
 data were found. The most serious was the esti-
 mated consumptive use and the explicit deter-
 mination of  the chemical  exchanges  in the
 return flow waters. The inconsistencies  in data
 concerning  consumptive use were dampened in
 the  overall systems  analysis  by allowing the
 inconsistencies to  accumulate in the  aquifer
 storage  facilities.  All  effort to obtain explicit
 or deterministic  analyses  of the chemical ex-
 changes in the return flow waters was discarded
 in favor of statistical inference which  measures
 the chemical exchanges on a probabilistic basis.
 It was  found that the  use of statistical in-
 ference  enables the prediction of return flow
 chemistries, constituent by constituent, at about
 92.5  percent  level.  To  obtain data  for the
 statistical inference study, it was assumed that
 all waters  available  for  diversion,  with the
 exception of extremes, were applied to irriga-
 tion and further, the measured aquifer chemis-
 tries  from  each  node (drain  outflow) rep-
 resented the  chemistry of  the return  flows.
 Even though there was some significant differ-
 ence in  the distribution functions, constituent
 by constituent, it was found that if the distribu-
 tion function obtained for  the  chemical con-
 stituent  of highest concentration  was used, all
 other  constituents  could be estimated  with  a
 simple  transform with  respect  to the fitting
 parameters.   This  technique  was  justified
 because  of the low  sensitivity of those  con-
 stituents of lesser concentrations. It was found
 that not only had the daily sampling fluctua-
 tions been dampened by the longer time period
 of monthly  reporting, but also in several cases,
the  supposedly  observed  data had  been ob-
tained   for  missing  periods by simply  using
 the mean as the expected value. This particular
manipulation  would  also  account for the in-
 consistency  in the distribution  function for the
lower concentrated constituents.
  At the conclusion  of  the above described
analyses and with the use of statistical inference
techniques,  several simulations were  made of
the  Vernal  Unit using  the  conjunctive use
model. In each of these simulations, parameters
describing  the allotment of inflow waters to
each of  the nodes were manipulated  until the
system was  in balance hydraulically or quanti-

-------
238     MANAGING IRRIGATED AGRICULTURE
                       Figure 3.
                                                         •*;•      -«z

-------
                                                              HYDROLOGIC MODELING
                                         239
tatively. The last of these simulations was one
that compared predicted aquifer capacities with
those  as  computed for the historic period 1958
through  1962. In each  of the nodes, with the
exception of one, it was found that a high rate
of  divergence   existed  between  the  aquifer
capacities as observed and those  that  were
predicted with the use of  the model. The diver-
gence was  expected  because the inconsisten-
cies of consumptive use had been accumulated
in the aquifer.
  No effort  was made  to simulate waters per-
colating through the soil columns. This simula-
tion was not required because of the very low
sensitivity to the overall  objective as provided
by this type of simulation. Also,  the  rate  of
change of chemistry in  the aquifer, time period
by time  period, was  not significant.  Subse-
quently,  each of the analytical  results of the
Vernal applications with  the use of the model
depicted graphically by photographing on 35-
mm film the image of the graph as projected on
a cathode ray tube. Examples of the results  of
the graphical analysis are  shown in Figure 3.
   From  these  several  applications, it  can  be
concluded that  the total  objective in studying
the Vernal  area with the historic time sets had
been  satisfied. It was  further concluded that
the on-going sampling format of data in the
Vernal Unit would have to be changed to render
a  more  meaningful  predictive  model  for the
Vernal area. Some of the  expected ramifications
of the data presently being collected with the
changed sampling format are: (1) a lower level
of predictability with the use of statistical in-
ference because of the  impact of the true sam-
pling fluctuations, (2)  a  higher degree of con-
sistency in  the  estimate  of  consumptive use,
and (3) the elimination  or at least a considerable
reduction in the divergences of the experienced
and predicted aquifer capacities.

Model Testing with Additional Data
   Data  were collected on the Vernal Unit dur-
 ing the  1971 irrigation season and will be col-
lected again during the 1972 irrigation season.
The model will be tested using these data.
  It is also planned that data collected by the
Geological Survey on  the Cedar Bluff Unit in
cooperation with the Kansas State Department
of Health will be  used to test the model  and
determine if the concepts  are  valid under  a
different set  of geologic  and hydrologic con-
ditions. A substantial amount  of data has been
collected over a period of about 6 years and this
covers the entire period since  irrigation started
on the Cedar Bluff Unit.
  Data collected on the Grand Valley Project
right  here in Grand Junction will be analyzed
to determine if it would be feasible to test the
model under  these conditions.
 CONCLUSIONS
   Although model testing is not complete and
 development of all the mathematical submodels
 is not complete, it appears  at this point that a
 satisfactory model has been designed to predict
 the mineral quality of return flow from irriga-
 tion  projects.  Completion  of  the  submodels
 will extend capability  to impact analysis, opti-
 mization and best plan.
   The implication for water resource projects
 is that farm operation could be designed to use
 the least amount  of water,  return the  smallest,
 amount  of salt  to the river  and  permit the
 farmer to obtain the greatest  possible return
 from his farm.
 REFERENCES
   1.  Southworth, Raymond W., shown in Rals-
 ton and  Wilf, second  printing, January 1962,
 pp. 213-220.

   2.  Tuckey, Dr. John W., as shown in Ralston
 and  Wilf, second printing, January  1962,  pp.
 191-203.

-------
         Modeling  Subsurface  Return  Flows
                            in  Ashley  Valley

               LARRY G. KING, R. JOHN HANKS, MUSA N. NIMAH,
                     SATISH C.  GUPTA and RUSSEL B. BACKUS
                                  Utah State University
ABSTRACT
   One possibility for optimizing the control of
the quality of irrigation return flow is to prop-
erly  manage  the  application  of irrigation
water. Such management depends upon knowl-
edge  of water and salt movement through the
root zone of the  crops. Two models are pre-
sented for flow of water and salt through the
soil with extraction of water by evapotranspira-
tion. One model was designed for use as an irri-
gation management tool while the other model
was initially intended to provide a detailed un-
derstanding of the water and salt flow through
the soil.  Field testing of the models indicates
that the best management model will probably
result from  a combination of the two present
models.. Timing of irrigation has been tested
as a management  variable.  With all other con-
ditions the same, the model predicts that as the
time interval between irrigations increases the
season totals of salt removed from the root zone,
salt remaining in the profile, and water required
for leaching tend to level off. However, the in-
terval between irrigations has a significant ef-
fect upon when the salt is  removed during the
season. The results indicate that  managing irri-
gation for control of return flow quality requires
good  control of depth and timing of irrigations.
INTRODUCTION
  The  recent (February  1972) session of the
Federal-state Enforcement Conference on the
Colorado River in Las Vegas, Nevada, re-em-
phasized the problem of salinity in the Colorado
River Basin. The considerations of the session
were based mainly upon a report by Regions
VIII and IX of the United States Environmental
Protection Agency (EPA) entitled "The Mineral
Quality Problem in the Colorado River Basin."1
  In 1960, the  Colorado River  Basin Water
Quality  Control  Project  was established  and
charged with the responsibility for identifying
and evaluating the most critical water pollution
problems in the Basin. As a result of early Proj-
ect  investigations, salinity was identified  as  a
pressing water quality problem. The report1
summarizes the results of the salinity investiga-
tions begun in 1963. Included in the conclusions
of the report are the following:
  1. Salinity is the  most  serious water quality
    problem in the  Colorado River Basin.
  2. Salinity concentrations  in  the  Colorado
    River system are affected by two basic pro-
    cesses: (a) salt   loading,  the  addition of
    mineral salts from  various natural  and
    man-made sources;  and (b) salt concen-
    trating,  the loss of water from the system
                                          241

-------
242     MANAGING IRRIGATED AGRICULTURE
     through evaporation, transpiration, and
     out-of-Basin export.
  3.  Salinity control in  the  Colorado  River
     Basin may be accomplished by the alterna-
     tives of: (a) augmentation of Basin water
     supply, (b) reduction of salt loads (includ-
     ing  improvement of irrigation and drain-
     age  practices); (c) limitation of further de-
     pletion of Basin water supply.
  This paper deals with the possibility for con-
trolling the quality of irrigation return flow by
proper management  of  the application  of ir-
rigation water. Such management depends upon
knowledge of water and salt movement through
the root zone of the crops. Two different models
were  developed- for  describing flow  of water
and  salt  through the  soil with extraction of
water by evapotranspiration. The two are called
"simplified" and "detailed" models for the pur-
poses of  this paper. Each model is discussed in
considerable detail below.
  The simplified model was intended to provide
a tool for irrigation management. It was formu-
lated to  require  a small amount of computer
time and a minimum of field data as input and
to allow consideration of a wide range of factors
affecting  the quality of irrigation  return flow.
It was expected that the model would predict
gross effects rather than  the detailed ones. The
simplified model will be updated and improved
by considering factors determined  to  be  neces-
sary for description of water and salt movement
adequate for irrigation management.
  The purpose for the  detailed model was to
understand the specifics  of simultaneous water
and salt flow through the crop root zones. This
model will help in the analysis  of field measure-
ments and serve as a basis for  determining how
much detail must be incorporated into the sim-
plified model to obtain an adequate tool for irri-
gation management for controlling the quality
of drainage effluent.
              Simplified Model

Description
  The simplified model is based upon the con-
cept of temporarily storing salt in the soil pro-
file and leaching only when necessary to prevent
the salt from  becoming too concentrated. The
root zone of the crop is divided into "n" layers
of equal thickness.  A constraint  is set  on the
allowable concentration in each layer. The cal-
culation begins by adding to the top layer suffi-
cient water to fill the available soil water reser-
voir for the entire root zone. The computer pro-
gram calculates the movement of water and salt
from layer to layer through the profile. If the
concentration constraint  of any layer  is ex-
ceeded,  then  the computer program iterates by
adding more  water until the constraint is satis-
fied. After the proper depth of water to be ap-
plied in a given irrigation  has been determined,
the final conditions existing in the profile (after
passing  of the time interval between irrigations)
become the initial conditions for the next irriga-
tion. Thus, the entire irrigation season  is cov-
ered in one computer run.
  The  crop is assumed to extract  pure water
leaving the salt behind in the soil moisture. The
water and salt are assumed to move downward
through  the  soil — lateral and  upward  move-
ment  are not allowed.  The model  uses a so-
called "field capacity" concept; i.e.,  after an ir-
rigation, downward  drainage of  water  occurs
until the  soil reaches the upper  limit of field
moisture, after which all downward movement
of water ceases.  The upper limit  of field mois-
ture must be determined for each soil layer
under conditions to be encountered  in the field.
For layers near and  including the water table,
the upper limit will  be  the  saturated moisture
content.  Whenever the term "field capacity" is
used in this paper it means the upper limit of
field moisture as defined above.
  Figure 1 shows a typical layer of soil with the
symbols defined as follows:
  DX is the thickness of the layer.
   ET is the total depth  of water extracted from
      the layer by evapotranspiration.
    de is the  depth of water entering the layer.
    0o is  the  initial moisture content  of the
      layer.
   0fc is the moisture content when the soil is at
      field capacity.
    0f is the  final moisture content existing in
      the layer at the end of the irrigation in-
      terval.
    di is the depth of water leaving the layer.
  Ce, Co, Cfc, Cf, and Q denote the correspond-
      ing concentration of salt in the water.

-------
                                          ET

                                          I

g
1 '

1
(
J
c
c
J

•o
fc
r

1
e
\
t
€
J

0
[
fc
'
>f
1 	
  Figure 1: Typical Layer of Soil in the Root Zone
  in Simplified Model

  The movement of water is treated in a fairly
simple manner. Since drainage from a particu-
lar  layer  ceases  when the  soil  reaches  field
capacity, the depth of water leaving the layer is
equal to the depth entering minus the depth re-
quired to  bring the soil layer to field capacity.
After the  soil reaches field capacity, the only
loss  of water from the layer is by evapotrans-
piration. Thus, in general,  the final moisture
content of the layer is determined from the dif-
ference between the amount of water at field
capacity and the amount extracted  by  evapo-
transpiration for the period between irrigations.
  The movement  of salt is more difficult to de-
scribe.  Since  it is assumed that evapotrans-
piration removes salt-free water from the layer,
the mass of salt in the  layer does not change as
the moisture content is reduced from 0fc to 0f;
however, the salt becomes more concentrated in
the water  remaining in the layer (Cf>Cfc). A
mass balance on salt of any layer can be written
as
cod0 =
                                cfdf
in which d0  and df are depth of water corre-
sponding to 00 and Of, respectively. All terms in
Eq. 1, except C\ and Cf, are known or can be
determined. The final concentration, Cf, is not
allowed to  become greater than a certain value
(concentration constraint) and can be solved for
only if GI is known.
  Using soil samples from the field, a laboratory
method of calibrating the soil to estimate Ci was
developed.2-3  This  calibration technique  was
based upon a leaching factor, LF, defined as
                                          MODELING — ASHLEY VALLEY    243


                                                      C1<*1
                                                           LF =
                                                                          [2]
                                 It  was  concluded from  the laboratory data
                                 (See Table 1) that the measured leaching factor
                                 could  be described as a function of dj and 00.
                                 From  the measured leaching factor  and Eq. 2,
                                 Cj can be determined and from Eq. 1, Cf can
                                 then be found. The leaching factor  function is
                                 contained in a subroutine, thus only the subrou-
                                 tine needs changing to model a different field
                                 soil.

                                                  TABLE 1

                                 Summary of results from laboratory technique
                                 of soil calibration to determine leaching factor
                                         (LF) for Hullinger farm soil
Effluent
Ratio
(di/dfc)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.25
1.50
1.75
2.00
2.25
2.50
2.75
3.00
Leaching Factor (LF)
60 = 0.05 0o = 0./5 6o = 0.20
0.201
0.275
0.303
0.365
0.412
0.432
0.461
0.495
0.520
0.541
0.586
0.633
0.681
0.719
0.741
0.763
0.787
0.798
0.118
0.192
0.261
0.312
0.359
0.393
0.441
0.461
0.485
0.519
0.572
0.621
0.662
0.702
0.733
0.757
0.771
0.793
0.084
0.143
0.195
0.272
0.303
0.353
0.394
0.422
0.462
0.483
p.553
0.606
0.652
0.681
0.713
0.735
0.762
0.776
                                Input Data
                                  The data necessary as input to the computer
                                for the simplified model include:
                                  1.  The  number  and  thickness of the  soil
                                     layers.
                                  2.  The concentration of salt in the irrigation
                                     water.
                                  3.  The length of the season and the interval
                                     between irrigations in days.

-------
244    MANAGING IRRIGATED AGRICULTURE
  4. The number of days between the initial soil
     sampling and the first irrigation.
  5. The moisture content and concentration of
     salt in each layer at  the initial  sampling
     date.
  6. For each soil layer the fraction of total
     evapotranspiration,  the upper and lower
     limits of field moisture content, and the
     maximum allowable concentration of salt
     in the soil moisture.
  7. The daily evapotranspiration for the entire
     irrigation season.
  8. The leaching factor  function (as a subrou-
     tine) for the particular soil.

Output Data
  A selection of the desired  output  is made
from the following:
  1. All input data.
  2. Immediately prior to the first irrigation
     and after each irrigation interval, the salt
     existing, the moisture content, and the salt
     concentration, respectively, for each layer,
     and the total salt existing in the profile.
  3. For each irrigation,  the depth of water ap-
     plied, the amount of salt leaving the root
     zone,  the amount  of leaching water, the
     leaching requirement in percent of applied
     water,  and the evapotranspiration for the
     interval following the irrigation.
  4. The totals  for the entire season  of depth
     of irrigation water applied, amount of salt
     leaving the root zone, amount of leaching
     water,  leaching  requirement, and  evapo-
     transpiration.

Results of Field Tests
  The model required verification or  testing in
the field.  An  approach  to field  testing might
have been to run the model for the entire season
to produce an irrigation schedule and conduct
the field work according to this schedule. Main-
taining a  strict  schedule in the field would  be
difficult. The irrigation water was not available
on a strict demand nor was its quality accurately
predictable. An accurate forecast of  the  daily
evapotranspiration  was  not   available.  The
above approach to field testing was abandoned.
Instead, the field work was scheduled according
to a reasonable overall plan and necessary data
for testing the model were collected. The opera-
tion of the computer program was changed so
that both the salt concentration and depth of
irrigation water were input for each irrigation
and the flow of salt and water was simulated
irrigation by irrigation through the season. The
iteration of applying water to maintain salt con-
centration of the soil moisture below a pre-de-
termined level was suppressed.
  The model was tested using field data col-
lected during 1971 on the Hullinger farm near
Vernal, Utah. The crop  was alfalfa which was
seeded with oats as a nurse crop in the Spring of
1970. Because of a relatively light stand of al-
falfa, additional seed was applied  without de-
stroying the  existing alfalfa early in the Spring
of 1971. In order to assure good germination of
this reseeding, rather frequent irrigations were
applied to the entire field until the first cutting
in June. The data for field testing the  model
were gathered for the rest  of the season. The
salt and moisture contents of the soil profile
were determined on June 22. These data fur-
nished the initial conditions for computations
with the model. Additional field samples were
taken on August 3 and  September 9. Figure 2
shows some results of the field tests of the sim-
plified model. Each  of the  measured values  is
the  average  of  measurements  at six  separate
locations. Complete detail of the field tests will
be  included in a research report which is pre-
sently in preparation.
  The results for  moisture content shown in
Figure 2 are reasonable  since the soil profile
was  considered  homogeneous  in the calcula-
tions. The model is not limited to homogeneous
soil, but further testing would have to be con-
ducted.  For  instance  a  larger  value of field
capacity at depths  of 1.5 and 3.5 feet would re-
sult  in better agreement of calculated and mea-
sured moisture contents.
  The electrical conductivity (EC)  data  com-
parison  of  Figure  2  is  not as encouraging.
Rather  poor agreement  is shown between cal-
culated  and  measured results. Reasons for the
disagreement can  be either inaccuracy of the
measured data  or  inadequacy  of the  model.
Both of these reasons are discussed below.
  Table 2 shows a comparison of the measured
and computed amount of salt (expressed as tons
per  acre) in the  soil profile  when the field
samples were taken.  To  convert electrical con-

-------
                                                MODELING — ASHLEY VALLEY    245
               Computed
           Measured
                 •—Sample I
                  -Sample 2
                 •-Sample 3
    0


  Or
a.
9
O.I                0.2
      Moisture  content,  9
0.3
0.4
             _L
          1
    0246
                               EC, mmho/cm
 Figure 2: Comparison of measured and predicted moisture content and electrical conductivity profiles for sim-
 plified model in 1971 for alfalfa

-------
246     MANAGING IRRIGATED AGRICULTURE

                                        TABLE 2

               Comparison of measured and computed amount of salt in the soil
                  profile when field samples were taken (Sample 1 - June 22,
                        Sample 2 - August 3, Sample 3 - September 9)

Depth
(ft)
0-1
1-2
2-3
3-4
Sample 1
Measured Computed
(T/Ac) (T/Ac)
0.72 0.70
0.67 0.57
0.56 0.56
0.69 0.69
Sample 2
Measured Computed
(T/Ac) (T/Ac)
0.82 0.60
0.94 0.97
1.52 0.92
1.07 0.81
Sample 3
Measured Computed
(T/Ac) (T/Ac)
0.74 0.40
0.50 0.90
0.71 1.31
0.93 1.00
          TOTAL
                      2.64
2.52
4.35
3.30
2.88
3.61
ductivity of the soil solution to equivalent tons
per acre of salt, one mmho/cm was taken as
equal to 640 ppm of salt.4 This constant factor
was used for all measured data and in the com-
puter program. Table 3 shows the irrigation
water and imported salt during the periods be-
tween field samples. Between sample one and
sample two a total of 9.83 inches of irrigation
water containing 0.88 tons of salt per acre was
applied. The model showed a discharge of 0.10
ton per acre  during the period giving a mass
balance on the salt (2.52 + 0.88 - 0.10 =  3.30).
The measured data showed a net  gain of salt
during this period of 0.83 ton per acre (2.64 +
0.88 + Gain = 4.35). The only possible source for

                 TABLE 3

       Irrigation water and salt imported
       between June 22 and September 9
              by sample period

             Depth      Cone.       Salt
              (in)     (mmho/cm)     (T/Ac)
                 Samples 1 to Sample 2
             3.00        1.27       0.28
             3.00        0.98       0.21
             3.00        1.53       0.33
             0.83        1.00       0.06
             9.83
TOTAL
  TOTAL
   0.88
               Sample 2 to Sample 3
            4.00        0.94        0.27
            3.00        0.86        0.19
            7.00                   0.46
  this net gain would have been upward flow of
  moisture from the water table. If we assume
  that the EC of the water in the saturated zone is
  3 mmho/cm, which is about the  value at 4.5
  feet depth as shown in Figure 2, the net upward
  movement of water from the water table would
  have had to have been 3.8 inches. More likely
  the water below the water table had EC equal to
  about 1.8  mmho/cm giving an estimate of the
  upward flow of 6.4 inches. Similar calculations
  performed for the  period  between sample two
  and sample three show a loss of water and salt
  from the root zone to the  water table. The net
  salt loss would have been 1.93 tons per acre in
  order to have 2.88 tons per acre remain at the
  time sample three  was  taken. This would have
  meant  a downward percolation of 14.8 inches
  of water at an EC of 1.8 mmho/cm or 8.9 inches
  if EC = 3 mmho/cm.  However, only 7 inches
  were  applied  as   irrigation  water between
  sample two and sample three.
    The above discussion points out some incon-
  sistencies in the measured salt content of the
  soil profile.  Perhaps these  are sufficient to ex-
  plain the differences shown in Figure 2. How-
  ever, the  simplified model has some assump-
  tions which are also subject to question. Opera-
  tion  of the  computer program has shown the
  results to  be very sensitive to the  leaching fac-
  tor. Better methods for obtaining  this function
  are needed.  The model neglects upward move-
  ment of soil moisture and long-term downward
  movement. The root extraction pattern is diffi-
  cult to obtain directly  in the field. More  field
  data on bulk density  and the upper limit of

-------
                                                          MODELING — ASHLEY VALLEY
                                                247
 field  moisture as functions of depth in the soil
 profile  might  improve  agreement  between
 actual conditions and  those predicted  by the
 model.
   Although good field verification of the sim-
 plified model has  not  yet been obtained, the
 model is expected to predict general trends and
 influences of various factors upon water and salt
 movement fairly well. Use of this model  to pre-
 dict effects of certain factors on the quality of
 irrigation return flow is discussed later  in this
 paper.

               Detailed Model

 Description
  This model considers the details of soil water
 and salt flow in small increments of depth and
 time.  The general water flow equation, which is
 solved by a numerical approximation,  for  each
 increment of time and space is
         at
                              +  A(z)
[3]
where 6 is water content, H is hydraulic head, z
is depth, K is hydraulic conductivity, t is time
and A(z) is root extraction. This model is quite
general  and  will account for unsaturated or
saturated flow either up or down or in combina-
tion.
  The  salt flow component of the model con-
siders mass flow of salt, exchange and precipita-
tion  and solution of CaCO3 and  CaSO4. The
mass flow  is computed from a solution of the
water flow equation after which corrections are
made for exchange,  precipitation and solution.
  As presently used the model does not consider
hysteresis or layered soil although both of these
have been considered earlier.5 6 Further assump-
tions are made that the soil properties, primarily
the hydraulic conductivity-water content rela-
tion  and the pressure  head-water  content rela-
tion do not change with time (there is no change
in soil structure).
  This model also  requires  some assumption
regarding  the partitioning of potential evapo-
transpiration into potential  transpiration and
potential evaporation directly from the soil.  At
present  this  partition is  done  rather  crudely
based on an estimate of percent of cover of the
plant.
Input Data
  The input data needed are as follows:
  1. Hydraulic conductivity-water content and
     pressure  head-water content tabular data
     covering  the range of water content to be
     encountered during the period of interest
     (basic soil property).
  2. Plant water potential  below  which the
     plant wilts and the actual transpiration will
     be less than potential transpiration (basic
     plant property).
  3. Air dry and saturated soil water contents
     (basic soil data).
  4. Root distribution with depth (active roots
     for adsorbing water) for the period.  At
     present the  model has  no provisions for
     changing this with time  (basic  plant prop-
     erty).
  5. Water  content-depth tabular data at the
     beginning (initial conditions).
  6. Chemical composition-depth tabular data
     at the beginning (initial conditions). This
     involves knowledge of the chemical analy-
     sis of the important chemical species. At
     present we consider Ca, Mg,  Na cations
     and Cl, SO4, and CO3 anions.
  7. Potential  transpiration   and   potential
     evaporation  rate or potential irrigation or
     rainfall rate  as  a function of time for the
     period (boundary conditions).
  8. Chemical composition of the irrigation or
     rain water (boundary condition).
  9. Presence  or  absence of a  water table at
     the bottom  of  the  soil  (boundary condi-
     tion).

Output Data
  The  type of output data that is available is
almost unlimited. Consequently, a selection of
the  desired data is made from the following:
  1. Soil water content and pressure head vs.
    depth and time during the period.
  2. Chemical  composition of the soil solution
    vs. depth and time during the period.
  3. Estimated  evaporation  and transpiration
    as functions of time.
  4. Water flow into the water table or up from
    the water table as a function of time.
  5. Chemical  composition of the water going
    into the water table or up  from the water
    table as a function of time.

-------
248
MANAGING IRRIGATED AGRICULTURE
  6. Estimated plant water potential as a func-
     tion of time.

Results of Field Tests
  The model was tested in the field at Vernal,
Utah, in  1970 and 1971. Figure 3 shows a com-
parison of water content-depth profiles at the
end  of a 9-day run in  1970  on oats seeded to
alfalfa. No actual measurements of the root ex-
traction function were  made so Figure 3 was
          O.I
                  Water   content
                0.2    0.3    0.4
                           05
                              Root depth*30cm
                              \
   160 -
                         \
 Figure 3: Comparison  of water content profiles
 as predicted and measured for oats in 1970

computed using the assumption that the roots
were in the top 30 cm. The data of predicted and
actual  measurements of soil water content are
quite   close.  However, this  agreement  using
estimated root distributions in 45 and 60 cm
depth of root zones was not as good.
  Figure 4 shows a comparison of potential,
actual  and predicted ET using the three root
depth  distributions.  The 30 cm root extraction
model  predicted 4.9 cm cumulative ET which
was 0.4 cm less than actual (measured) ET. The
45 and 60 cm root extraction models predicted
5.8 cm cumulative ET  which was 0.5 cm more
than actual. Thus, it would  appear from these
data that the true root distribution depth was
                                            70
                                            6.0
                                            5.0
                                            4.0.
                                            3.0
                                            2.0
                                                    1.0
ET


I
                                                                                1
                        1
                                                       Potential
                                                               Actual
                                                                        30cm
                                                                                45cm
                                                                            -Predated ET-
                                           Figure  4: Comparison of actual, potential, and
                                           predicted evapotranspiration using three different
                                           root distributions for oats in 1970

                                          between 30 and 45 cm assuming other variables
                                          used were correct.
                                            Figure 5 shows  a comparison of actual and
                                          predicted upward flow from the water table. At
                                          the end  of the 9-day period actual  upward flow
                                          was 2.1  cm which compares with  2.2, 2.3 and
                                          2.7 cm predicted by the 30,  45 and 60 cm root
                                          extraction  depths, respectively.   From  these
                                             30
                                           S

                                           I to
                                                   Actual
                                                            30cm
                                                                      45cm
                       60cm
                      (Rt depth)
                                           Figure 5:  Comparison of actual and predicted up-
                                           ward flow from the water table for oats in 1970

-------
                                                         MODELING — ASHLEY VALLEY
                                         249
data the 30 cm  root  extraction depth appears
to give the most  correct values.
  Much more data were collected in 1971 with
alfalfa as the crop. Figure 6 shows a comparison
of water content profiles as measured and pre-
dicted. The data show best agreement on July 7
and June 22  which were several days after any
water addition.  The  poorest  agreement is  on
June  29 and  July 10  which are right after irri-
gation.
  Figure 7 shows a plot of water content as a
function of time  for the  predicted and mea-
sured values. The agreement is very good with
no apparent "growth" of errors. Because of  ex-
perience in 1970 with water flow up from  the
water table, considerably more irrigation water
was added in 1971 to get more downward flow.
Consequently,  the  actual evapotranspiration
and potential evapotranspiration were the same
for this year. Thus, the agreement may be par-
tially due to the "forcing"  of the water removal
by evapotranspiration to be the same as mea-
sured.
   However, there were no such constraints on
the water flow up or down from the water table.
Figure  8  shows  the water flow there  as com-
pared to the measured data  at several times
during  the year. The data show some difference
between the  predicted and measured values of
up to 4 cm. There  does  not seem to be  any
growth of errors with time.
   In  summary,  regarding  the water flow data,
the model seems to predict well. It appears to
sufficiently describe  the system that subtle in-
fluences,  like water  flow up  instead of down,
from a  water table are detected.
   The  data  on  chemical composition  of  the
water within the soil and drainage water are not
as satisfying. The field data showed very little
change in chemica.1 composition of the water in
the soil or drainage  water during the normal
course of the season. Consequently, it was diffi-
cult to  really test the model since nothing very
different was occurring. Figure 9 shows the pre-
dicted depth and salt concentration of the drain-
age water during a period  late in 1971.
   Field attempts were made to alter the chemi-
cal composition of the irrigation water in several
instances in 1971. However, most of the changes
were small  and  it  was  difficult  to  measure
any effect in the field. Figure 10 shows  a com-
parison of measured and  predicted data for a
problem with dry salt added to the surface layer
before irrigation. The predicted salt concentra-
tion profiles  right after irrigation and  several
days subsequent to irrigation are also given. The
predicted data show the salt build-up at the sur-
face due to evaporation and the increase in con-
centration  due  to  root  extraction  of water.
Figure 10 also shows the predicted salt concen-
tration and that measured from electrical con-
ductivity of water extracted 324 hours after irri-
gation. The agreement is  fairly good with evi-
dence that the concentration peaks are different
by  15 cm. This difference  is within the error of
the field measurements, however.

          Implementation of Results
  The best model to use for managing irrigation
to control quality of drainage water is probably
a  combination  of  the two models presented
above. It is believed that the simplified model
can adequately evaluate  the effect  of certain
parameters upon the quality of drainage water.
Figure  11 shows the  results predicted for a
hypothetical  problem  with irrigation intervals
of 20 and 8 days, respectively. For a period of
121 days, irrigation water with constant EC =
1.5 mmho/cm was used and the EC of the soil
moisture was not allowed to exceed 6  mmho/
cm. Leaching  water  was applied  only when
needed to maintain the EC of the soil moisture
at  or below  this level during any interval be-
tween irrigations. The  121-st day was  included
to make a valid end point for comparison. An
irrigation occurred on the 121-st day  with  no
subsequent ET. Thus all  computer runs for the
different  intervals  between  irrigations began
with the same moisture deficit and ended with
the soil moisture at field capacity. For the 121-
day period, intervals between irrigations of 2,
3, 4, 5, 6, 8,  10, 12, 15, and 20 days were used.
A 24-day interval dried the soil below  the wilt-
ing point somewhere in  the  season  and hence
was not acceptable.
   Figure  11 shows the trend of longer  intervals
between irrigations releasing salt  from  the root
zone earlier  in the season and over a longer
period of time than the  shorter intervals. This
trend was evident by considering all the inter-
vals but space does not permit inclusion of re-
sults for each interval here. Figure 12 shows the

-------
250
MANAGING IRRIGATED AGRICULTURE
                   Moisture   Content  by  Volume   6

                O.I       0.3      0.5     O.I      0.3
                                                     0.5
         40
         80-


      o 100
     &


     1
   140
        180

                                Predict
                                 ensured
                   June 22,1971       '      June 29,1971

            End of first crop, beginning  of second  crop

                O.I      0.3     0.5      0.1      0.3     0.5
      o.
      0>
      O
      O
      V)
   40




   80


  100




  140.




  180
                                Fred id ed
                               Meosur >d
                  July 7,1971
                                      July  10,1971
      Figure 6: Comparison of predicted and measured water content profiles in 1971 for alfalfa

-------
                                                          MODELING — ASHLEY VALLEY    251
0.4
0.3.
02.
0.
0.4
03.
0.2
 §  0.1
o
 w  0.4
   0.2]
   O.I
          9 at 105  cm  depth
          9 at 75 cm  depth
       6  at 30  cm  depth
                                     May 13-Sept. II, 1971
                                     •+*7*-
      0     300    600     9OO    1200     1500   1800    2100    2400   2700   3000
                                       Time  hours
Figure 7: Comparison of predicted (solid lines) and measured (dots) water content at three depths for alfalfa in
1971
              300
                           900
2100
2700
                                          1500
                                     Time   hours
Figure 8: Comparison of predicted (solid curve) and measured (dots) upward flow for alfalfa in 1971

-------
252
MANAGING IRRIGATED AGRICULTURE
         10.

          8
      i
      i
      I  2^
      °  O.
                               200
                                             400
                                      Time   hours
                                                                 600
                              200
                                                                600
                                             400
                                      Time    hours
Figure 9: Predicted drainage and salt concentration of the drainage water for a period in 1971
results at the  end of the  121-day  season  as a
function of the interval between irrigations. The
very short intervals do not require  much leach-
ing water and  store a maximum amount of salt
in the profile during the season. As the interval
increases the  leaching water reaches a maxi-
mum after which it drops off slowly. Coincident
with the  leaching water maximum, the salt re-
moved from the root zone tends to level off as
does the total water applied.  The salt remaining
in the profile continues to decrease although not
very significantly. The important aspect from a
management standpoint is that there is a  con-
siderable range of irrigation frequencies  over
which  seasonal results do not change signifi-
cantly. However,  the time  of salt  discharge
during the season is affected by different irri-
gation frequencies.
  The  above example was produced with the
simplified model. The detailed model or the best
                                        management model could be used to study these
                                        same effects.
                                          The above  results indicate possibilities for
                                        control of drainage water quality as it leaves the
                                        root zone.  On a valley-wide basis  two possible
                                        alternative  modes  of  management might be:
                                        (1) Irrigate different farms at different frequen-
                                        cies so as to maintain a nearly constant quality
                                        of drainage effluent during the season; or (2) Ir-
                                        rigate all farms at  nearly the same  frequency so
                                        as to peak  the salt discharge at some convenient
                                        time. This  peak of salt discharge  might be di-
                                        verted to an evaporation basin or other suitable
                                        salt sink in areas where man-made drainage sys-
                                        tems are extensive.
                                          Considerable further work is needed in order
                                        to  predict  the fate of the water after it leaves
                                        the root zone until it arrives as return flow in the
                                        stream or  groundwater body. Study must be
                                        given to factors such as residence  times,  travel

-------
          Salt Concentration   me/liter

                40        80      120
     10



    20



    30



    40



    50



    60



f.  70


Q  80



    90



   100
 o
          y
            •s^...	
                 O/
                               ••**:-.	
After       ^,  V
irrigation      A'"
            ••***"  /

  >	:^
                                                        Salt Concentration     me/liter

                                                               40        80       120
                                                E
                                                o
 10



20



30



40



50



60
                         £  70
                         a.


                         S  80



                            90



                           100
                   Predicted

Measured     s^x^ 324 hours

from  solution-
conductivity       "v
                                                                                                 o
                                                                                                 a
                                                                                                 m
                                                                                                 r
                                                                                                 3
                                                                                                 o
                                                                                                 C/i
                                                                                                 a
                                                                                                 m
Figure 10: Predicted salt concentration profiles after dry salt was added to the soil surface just before irrigation as compared to measured salt concentration

profile in 1971

-------
o

6


1 4
^
o.
«
•.
1
«


-


-

i


O
3
Ob
6
5'
•R 4
Q
g 2
9 r\


mt
-

eu











aays











8 days









1
09





•





wee











ri|











priori




















Salt in — >
profile (left
scale)

\
*— Total water applied (in.)
« — Leaching
water

t-ET [
•

« — Salt re-
moved
(right scale)












between irrigations












•



i






•



•



•















}



•

•





^











.


2
2
0.6 1
53
g
S
0.4 §
o
0
jb
0.2 §
g
a
(0 m

9

i






1

*-* 	 n



:
\

06 i
a.
•
0.4 |
|
0.2
o
20
40
80
100
                                      60
                                     Time, days
Figure 11: Results from simplified model for two different intervals (8 and 20 days) between irrigations
120

-------
     8-
«o



o


                                             a
                                            •o
                                                                                                                    
-------
256     MANAGING IRRIGATED AGRICULTURE
paths,  degree  of  mixing with  groundwater
underflow before full evaluation can be made of
managing irrigation to achieve water quality
control. It is evident from the work on control
in the root zone that very sophisticated irriga-
tion systems  capable  of  precise  control  over
depth and timing of irrigation are  necessary.
Such systems do now exist but are rather expen-
sive.7  For full  benefit considerable man-made
drainage systems  having short residence  times
and capable of flexible control are also deemed
necessary.

CONCLUSIONS
  It is concluded from the research work  that:
  1. Models  can  be  developed which  ade-
     quately  describe the simultaneous  flow
     of water and salt through the crop root
     zone for  use in  managing irrigation for
     control of drainage water quality. The best
     model for management will be a combina-
     tion of the detailed and simplified models
     described above.  The models need further
     development and field testing.
  2. Precise control of depth  and timing of irri-
     gations is necessary for control  of  root
     zone effluent  quality.
  3. The interval between irrigations affects the
     timing of salt discharge from the root zone
     during the season.
  4. There is  a considerable range of irriga-
     tion frequencies over which total seasonal
     effects are not significantly different.

ACKNOWLEDGMENTS
  This  research work was supported and  fi-
nanced by the U.S. Department of the Interior,
Federal  Water Quality Administration   (pre-
sently the United States Environmental Protec-
tion Agency), under  grants WP-01492-01(N)1
and 13030 FDJ, and the Utah Agricultural Ex-
periment Station, Utah State University, Logan,
Utah.


REFERENCES
  1. United  States Environmental Protection
Agency.  1971. The mineral quality problem in
the Colorado River Basin.
  2. Rasheed, H.  R.  1970. Irrigation manage-
ment to control the quality of return flow.  PhD
Dissertation, Utah State University, Logan.
  3. Titavunno,  P. 1971. Quality of leachate
from Vernal project soil as affected by certain
conditions in the soil  and applied water. M.S.
Thesis, Utah State University, Logan.
  4. Richards,   L. A.   (Ed.).   1954.  Diag-
nosis and improvement of saline and alkali
soils. Agriculture  Handbook  No. 60, United
States Department of Agriculture.
  5. Hanks, R.  J., A.  Klute, and E.  Bresler.
1969. A numeric method for estimating infiltra-
tion, redistribution, drainage,  and evaporation
of water from soil. Water Resources Research,
5(5): 1064-1069.
  6. Bresler, E. and R. J. Hanks. 1969. Numeri-
cal method  for estimating simultaneous flow of
water and salt in unsaturated soils. Soil Sci. Soc.
Amer. Proc., 33(6):827-832.
  7. Keller, J.,  J. F.  Alfaro,  and L. G. King.
1972. Water quality aspects of sprinkler irriga-
tion. Proceedings National Conference on Man-
aging Irrigated Agriculture to Improve Water
Quality, Grand Junction, Colorado, May 16-18,
1972.

-------
               Surviving  with  Salinity in the

                   Lower  Sevier River  Basin

                                W. ROGER  WALKER
                              Sevier River Commissioner
                                      Delta,  Utah
                                         and
                                 WYNN R. WALKER
                          Agricultural Engineering Department
                               Colorado State University
ABSTRACT
  The maximum utilizution of the water supply
of a river system requires that a minimum basic
supply must have equal priority along the sys-
tem. Shortages must be spread over the system
as far as possible and development  will take
place only to  the extent that  the shortages are
tolerable. The distribution and extent  of excess
reservoir capacity coupled with flexible diver-
sion rights dictate the ultimate utilization of the
water resources of a river basin. These factors
are  now becoming  an  integral  part  of water
quality management  in the Sevier River Basin.

  Water quality, deteriorating downstream on
the  Sevier River with continual reuse of the wa-
ter,  indicates that the practical limit of the use of
the  return flow is between 40% and 50% of the
total diversions for any year.  The regions most
affected  by salinity in the basin adapted to that
condition by implementing an agriculture based
on alfalfa seed production with grain  rotations
to add nitrogen and organic matter, and provide
a leaching opportunity for controlling salinity
levels in  the soils.
INTRODUCTION
  The Sevier River Basin, shown in Figure 1, is
one of the major drainage systems in the Great
Basin which has had no seaward drainage since
the period when prehistoric  Lake Bonneville
was at its highest level. The river originates in
southern Utah in the 7,000 to 10,000 foot divide
between the Great  Basin and  the  Colorado
River Basin.
  The  Sevier River begins  as  two principal
forks. The south fork originates near the town of
Hatch, Utah in Panguitch Valley, while the east
fork begins in Plateau Valley some twenty miles
east. The two forks join just above Piute Reser-
voir, then the river  flows  northward approxi-
mately eighty miles until it is joined by the San
Pitch River and  enters Sevier Bridge Reservoir.
From the Sevier Bridge Reservoir, the  river
courses  westerly through  what  is known as
Leamington Canyon, and then southwesterly
about forty miles to  Gunnison Bend Reservoir
two miles west of Delta, Utah. Before the com-
plete development of the river occurred, it ex-
tended another  twenty-five miles southwest to
                                          257

-------
258     MANAGING IRRIGATED AGRICULTURE
                                 SEVIER LAKE
                                    BASIN
               !
             _  ARIZONA
                       LOCATION MAP
     Figure I: The Sevier Lake Drainage Area

what was its natural end, Sevier Lake,  a total
length  of  approximately two  hundred  fifty
miles.
  Of the entire water supply in the basin origi-
nating as precipitation, only about one percent
is not consumed within the basin; the remainder
is ground water and surface outflows. In fact, in
only two years since 1922 has any appreciable
quantity of water escaped the irrigation system.
A comparison of streamflow data indicates that
between 40% and  50% of the diverted water is
from irrigation return flows from lands upstream
and  on  the  tributaries.  For example,  in the
twenty-four mile stretch between the gaging sta-
tions at Sigurd and Gunnison, the total tributary
and  surface inflows were 59,000 acre-feet, while
the  outflow into Sevier Bridge Reservoir was
103,000  acre-feet.   Thus,  44,000 acre-feet were
added in the  reach from ground water  return
flows. This point  illustrates the nature  of the
Sevier River  System as being a series of tubs of
which the river is the overflow.
  The agricultural use of the Sevier River ac-
counts for approximately 25% of Utah's irrigated
acreage, but  the river's flows are less than 10%
of the supply in  the  Great Basin.  In  order to
operate the system in such an effective frame-
work, it is necessary to have complete control of
the system throughout the basin. This is accom-
plished both  by the location and the excess
capacity  of  the reservoirs along the river, as il-
lustrated in Figure  2.  The  available storage
capacity  of the major reservoirs in the Sevier
River  Basin is currently 424,000  acre-feet, of
which only  about 45% is utilized in an average
water year.
                               SEVIER RIVER BASIN
                                   UTAH
       Figure 2: Sevier River Flow Network

                Water Rights
  It should be obvious to the reader that in the
total  developed  river system, the water quality
of the flows reaching the lower users is usually
highly degraded.  In addition, the management
of  salinity, for  example,  is  highly dependent
upon the allocation of water within the system.
In the Sevier River Basin, the storage rights that
developed to divide an inadequate storage water
supply are now the mechanisms for maintaining
an agriculture in the Delta, Utah area by provid-

-------
                                                            SURVIVING WITH SALINITY
                                          259
ing a means of regulating and controlling salin-
ity.
  Development of the waters of the Sevier River
began almost  simultaneously  throughout  the
length of the basin.  In January of 1860, a group
came to what is now Deseret, Utah to build a
dam and excavate a canal to the lands to be irri-
gated. During  periods  when  the temperatures
were  often 20°  below  zero, brush  had to be
burned on the land to take the frost out of the
ground so the wheelbarrow  canal  excavation
could  proceed. Many problems were encount-
ered  by these early settlers  when numerous
flood periods destroyed the brush and rock di-
version dam or when an Indian threat appeared.
However,  by 1890  nearly all  of the lands that
could be irrigated by these direct diversions had
been developed and Gunnison Bend Reservoir,
the first permanent  storage facility on the Sevier
River, had been constructed. Although the water
supply  had  at times been  excessive,  the late
1890's brought a series of very dry years in
which a time came  when no water reached the
town of Deseret with which the local farmers
could mature their crops. By then, the Supreme
Court  of the United States had affirmed the de-
cisions of  the State  Supreme Court that "first in
time was  first in right," but there was no  en-
forcement agency to maintain such a concept so
the priority had little effect. As a result, an up-
stream user felt no  obligation whatsoever to re-
lease water down to a lower user and so each di-
verted all  the water possible at his headgate. In
typical  western  reaction, the  water users of
Deseret organized an armed force to go up the
river  and  destroy the dams of the subsequent
appropriators; however, by  the time the group
had  returned,  the  dams had been replaced.
Many of the communities upstream have their
own legends of how they protected their diver-
sions, but  in reality probably one or two danger-
ous eye to eye confrontations were  enough to
end this type of adjudication.
  In about 1899, the Deseret Irrigation Com-
pany  initiated  legal action against  upstream
users to gain an equitable solution to the  ques-
tions  of water rights. In most cases, the early
testimony  was  so  conflicting that  the courts
placed most direct flow rights on common prior-
ity to be prorated when the supply was insuffi-
cient to meet the listed  right. In the  meantime,
 people began to look at the possibilities of utiliz-
 ing the water that had been going to waste dur-
 ing the winter months.

   In  1902 the  site for Sevier Bridge Reservoir
 was filed upon and construction started that fall.
 The average flow during the  month of October
 1902  was only seven second feet of water and
 this low flow was a prime factor in the success of
 the construction. The scheme  to finance the dam
 at Sevier Bridge was unique. A man and team
 was paid at the rate  of $2.50 per day. Payment
 was in stock of the Deseret Irrigation Company
 at a par value of $5.00 per share. Then each fall
 the Deseret Irrigation Company levied an  as-
 sessment  of $5.00  per  share.  Consequently,
 whenever a stock holder could spare the time, he
 loaded his provisions on a wagon and went up to
 the dam to work.
  After some  adjudication, the people realized
 that the funding necessary to  legally settle every
 claim in the system would be  much greater than
 the water was worth, so regional  committees
 were formed to work out differences. The plan
 that eventually evolved is  known as the  Cox
 Decree and  represents one of the truly remark-
 able,  mostly voluntary, arrived at  agreements
 between individual water users in the state. The
 specific mechanisms of water  distribution in the
 Sevier River system are extremely complicated.

  Because the system of water rights along the
 river is the  criterion for  managing the system,
 previously from a quantity  standpoint and now
from a quality  standpoint,  it is best to relate a
 specific example.  Presently,  the  Sevier River
 network is divided  into two zones which are sep-
 arated at  what is  known as Vermillion Dam.
 Thus, Sevier Bridge  Reservoir is the  principal
 reservoir in  the lower zone  and Piute Reservoir
 is the principal one serving the upper zone.  In
the settlement  of the disputes among water
 users as to certain rights, Sevier Bridge Reser-
voir was awarded the first 89,280 acre-feet in the
 river,  with Piute  Reservoir getting  the second
40,000 acre-feet. Subsequent quantities, if avail-
able, were divided so both reservoirs fill at the
 same time. In addition, the solution allowed  all
physically able  rights to place their water on a
call system.  This meant that all of the users  on
the main stem from below Piute Reservoir to the
end of the river were able to use the two main

-------
260     MANAGING IRRIGATED AGRICULTURE
reservoirs as credit banks to regulate water in
the most beneficial manner.
  The West  View Irrigation Company,  which
operates between the two  principal reservoirs,
has the right to 24 cfs when the accumulation
between Vermillion Dam and Delta is 295 cfs,
and a weighted percentage when the accumu-
lations are less. This diversion must  be made on
a daily pro-rata basis between March 1 and the
15th day of April. However, from April 15 to the
10th of October, this pro-rata right can be accu-
mulated and credits given to the  company; or in
other words, during this period the  water is on
call. After October 15, they have a right to 30 cfs
for fall  irrigating that is not subject to the pro-
rata.  The formulation of this right  was based
upon historical records which showed the com-
pany  had  on occasion diverted 24 cfs in March
and 30 cfs in the fall, but had not continuously
diverted the water. Since the conditions under
which the company  would  divert  this  water
could not be known until the very day it was to
be taken,  circumstances such as a cold  spring,
unseasonal snow  storms,  or a  wet  fall  which
would make the users in the company wish not
to irrigate, give vested rights to  other users for
the water not diverted. As a result,  a very seri-
ous  problem arises. The original participants
eventually pass on so the  personnel  actually
computing the water allocations are the only
people that  see the day to  day operation  and
thus understand it. Then, when a new genera-
tion of users,  government  agencies, or study
groups  desires  to  determine a  right, the Cox
Decree  must be consulted. In the  case noted
above, they  would read that West View Irriga-
tion Company is entitled to 24 cfs without fully
understanding  the  limitations  thereof.  Obvi-
ously, there are many conditions under  which
the company cannot use  24 cfs, so  these inter-
ested  parties seek a way to change the operation
to make more  beneficial use of the water and
more completely use  the right as stated  in the
Decree. The official policy of the State of Utah
is that transfers must be allowed  if they will lead
to a higher and more beneficial  use. But  in the
Sevier River Drainage, all the water is used, so
that water that is transferred for beneficial use
at one  point is simply decreasing a beneficial
use at another. On a totally diverted river sys-
tem,  the only change possible is the point of
consumptive use.
                Water Quality
  The final scheme for deriving an  equitable
distribution of water among users in the Sevier
River Basin came about through regional com-
promise. The obvious catalyst was the fact that
the potential  use  demands  far  exceeded  the
average annual  supply. However, the compli-
cated  and  interrelated  system "accidentally"
saved the lower Sevier River area, and the Delta
area in particular,  from complete  devastation
from  high  salinity  concentrations in the irriga-
tion water.
  The head waters of the Sevier River are of a
suitable quality for most uses, but the concentra-
tion of total dissolved solids drastically increases
as the flows proceed downstream. The concen-
tration varies from  about 60 ppm in the head-
waters, to about 1700 ppm in Sevier Bridge Re-
servoir, to over 2000 ppm in the Delta area. The
increase is  not a uniform one, but is influenced
by the ratio of surface and ground water inflows
in different reaches of the river. In general, the
water quality of the main  stem flows changes
from  a calcium carbonate to sodium chloride
type waters by the time they reach Sevier Bridge
Reservoir.
  The identification  of the quality of return
flows, surface inflows, and the role of the reser-
voirs  in managing water quality in the system
can be best illustrated by considering some 1964
data collected by the U.S.  Geological Survey1.
Data from  the gaging station at Vermillion Dam
(Figure 3) shows the water releases from the up-
per zone satisfying  the first storage priority of
Sevier Bridge (89,280 acre-feet) have a salinity
concentration of approximately 500 ppm. Later
in the season, the water quality of the flows in a
short  reach below  Vermillion Dam rises  to
above  900  ppm, indicating the  effect of irriga-
tion return flows. In addition, the tributary in-
flows in this reach,  composed of small flows in
Lost Creek and Salina Creek combined with re-
turn flows  from the  irrigated  acreage  along
these  streams, have had salinity concentrations
as high as  10,000 ppm. In the reach just down-
stream from Salina  Creek, three irrigation com-
panies divert water. Their rights are such that
the total river flows are often diverted at each
point, so that the diversions amount primarily to
return flows. The specific conductance of these
flows ranges all the way to 3250 micromhos/cm,
indicating a danger to agriculture in these areas.

-------
                                                             SURVIVING WITH SALINITY
                                          261
               USG.S Goge
 Figure 3:  Sevier River System  Between Piute
 Reservoir and Sevier Bridge Reservoir

The significance of the location of the largest re-
servoir  with  a first  priority being  towards the
end  of the system  is now apparent.  Without
some high quality water  being released down-
stream for dilution during part of the irrigation
season in the dry years, it is doubtful that the ir-
rigation system in this reach could maintain the
present type  of agricultural production. The
value of the flexibility of their rights is also ap-
parent in that during dry periods when salinity is
most  severe,  much more  early  spring  and late
fall water is used which is usually of acceptable
quality. In this manner, the irrigated acreage in
this area is able to maintain acceptable  salt con-
centration in  its soils.
  Also  without the high  quality water released
from Piute Reservoir during the dry years to mix
with the return flow water accumulated in Se-
vier  Bridge Reservoir, agricultural  production
downstream   from  Sevier  Bridge  Reservoir
would be seriously affected.
  Although salinity  is a  problem to irrigators
above  Sevier  Bridge Reservoir,  their  soils are
usually well drained so leaching is no problem.
Below this reservoir, such is simply not the case.
The outlet of Sevier Bridge Reservoir is charac-
terized  by  a  specific conductance  of 2300
micromhos/cm and indicates  the water quality
which is available to most of the irrigated acre-
age downstream. At the  end of the river,  the
specific conductance may reach a level of 4000
micromhos/cm depending upon the operation
of the river. The  storage within Sevier  Bridge
Reservoir, however, does allow the quality to
change from a very high salinity  and high  so-
dium  hazard to a medium to high salinity and
low to medium sodium hazard because  of  the
mixing  of winter  flows  with summer  return
flows. When this mixing is sufficient, the  re-
leases from Sevier Reservoir, along with  the in-
flows  to the river from various points, can main-
tain a specific conductance in the Delta area to
around 2000 micromhos/cm, although this may
also vary  with  other agricultural diversions  be-
tween the reservoir and Delta.
  It is now being realized that the excessive  res-
ervoir capacity at the end of the river  can be
used  in   conjunction  with  the  present  water
rights to  manage the water quality of  waters
being  used for irrigation.

    Use of Saline Water in the Delta Area
  The Delta area is the major area in the Sevier
River Basin where the salinity problem is large
and universal. The soils are generally of a heavy
clay nature and of low permeability.  The  irri-
gated  acreage has ranged  between 120,000 acres
during the  high  flow years for the 1920's to
about 40,000 acres in the  1930's. This large fluc-
tuation in irrigated acreage was not a result of
local  planning, but simply an effort  to meet the
challenges of each new year as they became nec-
essary. Prior to about 1910, a large  growth pe-
riod  was  experienced in  which  no  realization
was made that drainage might become a prob-
lem. A tile  drainage system was constructed to
alleviate the large salt accumulations in the soils
and the high water tables. The funding  for the
drainage system was provided by bonding when
inflation  rates  were very high  and  the  repay-
ment  unfortunately came with decreasing farm
prices and eventually  depression dollars. Even
with the almost totally devastating financial cri-
sis that was encountered with the drain funding,
probably  the worst blow came when  most of the

-------
262    MANAGING IRRIGATED AGRICULTURE
drains failed because of poor designs. The solu-
tion was painful but obvious; open drains were
dug throughout the area to solve the problems.
  By 1935 when the lowest flow year on record
had occurred, a pattern had been developed for
survival under these adverse conditions. In ex-
cess of 90% of the irrigated crop land remained
in alfalfa because alkaline conditions caused by
either high water tables or spreading the water
supply too thin made alfalfa the only reliable
crop.  Thus, the farmers enlarged their alfalfa
seed operation whenever the water  supply be-
came  too short. For the Delta area,  alfalfa has
proven  to  be a remarkable plant in that  well
rooted young alfalfa in the heavy clay soils will
produce a seedxrop for several years with no ir-
rigation  and will grow in alkaline  slick areas
that barely  support harvesting machinery.  Evi-
dence thus  supports  the theory  that with the
meager  rainfall the alfalfa plant  is able to ex-
tract from the capillary zone above the water
table enough moisture to produce a seed crop. In
fact, alfalfa that will grow  under these condi-
tions while  not giving the largest yields is the
most certain to produce a crop of alfalfa seed.
One other practice when water is short is to irri-
gate alfalfa once and then let it go to seed; or to
cut a  light crop of hay, creating a dust mulch,
and then let it go to seed. The salt buildup under
these programs is extensive; however, by leach-
ing irrigations on the remaining acreage during
a grain crop rotation for a couple of years, most
of the farmers manage to stay in business. This
indicates than an annual leaching irrigation  is
not required on all these saline  soils but  that
adequate leaching of a part  on a rotation basis
can be  a practice to manage a deficient water
supply of minimum quality. Studies have been
made that suggest the most economical course
would be to shrink the acreage to fit the water
supply,  but the people making the studies just
don't have the same perspective as the farmers.
In the first place, who knows what next year will
bring, and if the farmer can  stretch his water
supply one  or two years he might  be able to
catch up later. Secondly, how does the outlying
farmer pull up stakes and move into a shortened
system each drought cycle? Is there  a place for
him and who puts up the  capital?  One other
reason for not abandoning the tighter soils  is
that they are the  more consistent alfalfa  seed
producers since the salt seems to act  as an ef-
fective control on the plant growth for seed pro-
duction. Nature aids in one more way—the salt
buildup encourages an increase  in  the  alkalia
bee, which is probably the best pollinater avail-
able for alfalfa.
  Of course, this practice of little or no leaching
could not go on forever and during the thirties
the Delta  area was in serious economic trouble.
The crop  production here was only  53% of the
state average2. In 1936,  the water supply cycle
started trending up and 1942 filled all the reser-
voirs on the Sevier River. It was an opportune
time  to give the land a good leaching, but new
problems  developed which restricted this prac-
tice.  The  expensive  tile  drainage system  had
failed in many places and this emphasized one
fundamental fact, that leaching attempted with-
out adequate  drainage further  aggravated the
adverse salt conditions.
  Starting in the late  1950's, the  inevitable dry
cycle started again.  However, in a few years it
became evident that history would not repeat it-
self as unit  crop production continued to  in-
crease instead of decline. During the period of
1965 to  1967, observation farms were estab-
lished to determine the production per acre-foot
of water applied, and an especially well man-
aged farm was selected to show what could be
done. The results were twelve tons of corn silage
per acre-foot of water applied and three tons of
alfalfa hay per acre-foot. Some  land  that  had
been recommended for abandonment by various
studies was now yielding one hundred bushels of
small grains and crops of five hundred pounds of
alfalfa seed per acre. Hindsight gives  some ex-
planation  for the improvement.
  The Abraham and  Deseret Irrigation Com-
panies in the area have a right to the first 10,000
acre-feet made below Sevier Bridge Reservoir
during the winter months. The effective fluctuat-
ing capacity to hold this water, Gunnison Bend
Reservoir, was only around 3,000 acre-feet. Con-
sequently, the Central Utah Water Company
would divert the good quality make in Leaming-
ton Canyon as early as possible, the residue and
downstream accumulation going to satisfy the
10,000 acre-feet.  The specific conductance of
this water is, as mentioned, between 3,500 and
4,500 micromhos/cm. The practice was to issue
small dividends  whenever it  was necessary to

-------
                                                           SURVIVING WITH SALINITY
                                         263
create more capacity and this very poor quality
water was then delivered to lands of these two
companies  which are furthest  out, lowest, and
least able to tolerate the salty water. In  1960,
D.M.A.D. Reservoir of 11,000 acre-feet capacity
was constructed to enable Deseret and Abraham
to place this winter water on call. Also at this
time, the four lower companies started a pro-
gram of tapping the  deep aquifers for irrigation
wells and pumping this water into the D.M.A.D.
Reservoir.  By careful  management, the reser-
voirs were completely emptied at the end of the
irrigation season, and because of the increased
capacity, an agreement was reached to  store
Central Utah's water in D.M.A.D. We now had
water coming into  D.M.A.D.  and Gunnison
Bend reservoirs of 1600 to 1700 micromhos/cm.
The irrigation companies now were able to start
the season with 14,000  acre-feet of water of a
quality nearly as good as that entering the lower
zone. From the standpoint of water quality man-
agement, more  reservoir capacity at the end of
the river system  could be economically justified
to dilute and mix the water supply. Some accom-
modation of poorer quality water would  then be
possible with the increased capacity.
  Extensive canal lining commenced and the
well drilling continued. These  projects have in-
creased and stabilized the low base water sup-
ply. It might be of interest to note that canal lin-
ing has continued at an increased rate during the
recent wet cycle. The farmers have demanded
these projects for water table control and drain-
age benefits. Improved  water  quality  was not,
however, one of the  planned objectives of these
projects.
  The visible evidence of improving the heavy
clay soil  by mechanically shattering deeply has
been before  the  farmers since  the tile  drains
were installed. After more  than  fifty years and
long after they  had ceased to function, the
course  of the tile drains can be traced and in
some of the tight Abbott clay  series there is no
crop production whatever  except over the old
tile line and here it is good. The outlines of
dumps of manure on the salty spots are discern-
ible more than twenty-five years later by the in-
creased growth. These conditions spelled out
what was needed to produce good crops on very
tight soils with a saline water supply, level the
land so that the water can be controlled to make
an even application of a given quantity, mechan-
ically  breaking the soil as  deeply as possible  to
accelerate the leaching  process, and a program
to encourage the  formation of humus for free-
sodium tie up to improve crop production.
  The farm operators have always known that
the heavy clays were rich  in all plant nutrients
but nitrogen. The reclaiming of alkaline spots by
adsorbing the free-sodium by  humus has long
been practiced, but it took the  arrival of the 100
hp +  tractors, the heavy applications of nitro-
gen, the planned large crop  residue manage-
ment, and  water  supply-water quality manage-
ment  to put it all  together.
  The Delta area has come to the point where
they are now past  the trial,  error and coinci-
dence  method. The information  necessary for
good  planning was probably always available,
but the insistence of the farmers on the privi-
lege of making their own mistakes resulted from
a combination of the scientific community being
unable to "sell" their expertise and  the means
not being at hand to apply the knowledge. It ap-
pears  that water quality, water supply, and farm
management are finally coming together to sta-
bilize  agriculture for this locality.

REFERENCES
   1. Hani, D. C. and Cabell, R. E. 1965. Qual-
ity of surface water in the Sevier Lake Basin.
U.S.  Geological Survey and the  Utah Depart-
ment  of Natural  Resources. Basic Data Report
No. 10.
   2. Hahl, D.C. and Mundorff, J. C. 1968. An
appraisal of the quality of surface water in the
Sevier Lake Basin,  Utah.  U.S. Geological Sur-
vey and the  Utah Department of Natural Re-
sources. Technical Publication No. 19. 43 p.

-------
        Water  Right  Changes to  Implement
           Water Management Technology
                                G. E. RADOSEVICH
                              Colorado State University
                                Fort Collins, Colorado
ABSTRACT
   Water in the western U.S.  is allocated and
distributed according to the concepts and rules
of the prior appropriation doctrine. This doc-
trine was developed during the mid-1800's as a
solution to conflicts arising between those com-
peting for available stream flows. Few changes
have occurred in the substantive or procedural
aspects of this  law, resulting in  a lag  between
water management technology and the law's
capability to incite implementation of these de-
velopments. The added problem of water qual-
ity not included in the appropriation  doctrine
complicates the management scheme.
  The principal legal concept in the appropria-
tion doctrine which has served as the constraint
to effective adoption  of more efficient  water
practices within the irrigation flow system is the
property right in water, a valid and protected
right incurring  to the  benefit of its holder.  To
potentially resolve the water management prob-
lem in the West, recommendations in changing
or reinterpreting the doctrine governing the ex-
ercise of this right are suggested. Three  basic al-
ternatives exist: do nothing, provide incentives
within the legal framework, or enforce the  law
strictly. The recommendations will lead to  im-
plementation of water management technology
in an attempt to solve the problem existing at
the state, interstate, and international level cog-
nizant of environmental consideration,  the irri-
gated land sector of the economy, and other cur-
rent and potential users placing demands upon
available supplies.

INTRODUCTION
  Water is undeniably one of the  most crucial
resources known to man; his very existence de-
pends upon an  available supply for his con-
sumption and the needs of his activities which
perpetuate his existence.  How he uses  those
supplies within his reach will  determine, to a
great extent, his future and the future of genera-
tions to come.
  In the semi-arid western United States, we
have reached the apex of progress with natural
stream flows. (Figure  1) The majority of all wa-
ters have been appropriated to specific uses at a
time when certain other needs are just material-
izing and the factor of quality in water is becom-
ing a major consideration. Projected water de-
mands within the national basins require quan-
tities of quality water that do not exist. There-
fore, it is incumbent upon all users to introduce
the most efficient management practices on
available supplies.
  This paper explores the institutionalization of
a legal  and organizational  system that has di-
                                         265

-------
266
MANAGING IRRIGATED AGRICULTURE
              Tn« numtMTt inow tn« incftn of w«t»c
              tnat could covif the ground i( til UM
              wattr Ihit fiowrt from «i« twin in «n
              
-------
                                                               WATER RIGHT CHANGES
                                          267
mons became overloaded and the need was rec-
ognized to develop some means to regulate the
use of this valuable resource. In social terms,
there was a willingness for each to give up a lit-
tle  so  that all could have more; in economic
terms there was a willingness to internalize the
costs of the externalities created through the use
of this common resource.
  The system that emerged was simple and di-
rect. The principles that developed held:
  (1) There had to be a diversion from the nat-
      ural  stream.
  (2) The  water had to be applied to beneficial
      use.
  (3) When completed there was created in the
      appropriator a water right.
  (4) This right insured the holder of continued
      use for the quantity appropriated.
  (5) There attached an appropriation date as
      of the date of application to beneficial use
      to govern the appropriator's relationship
      with other water right holders.
  In every state,  with time, an administrative
agency was established to administer the water
provisions  and ensure the distribution of water
according to priorities. Here again, the early set-
tlers demonstrated a  willingness to internalize
costs associated with perfecting a system of wa-
ter  use that provided them with the needed se-
curity to protect their own investment.
  Consistent with developing water codes at the
state level, the need to perfect the water  right
was also voiced. Consequently, it was decided
there would be adjudications of water rights in
which  the  right holder would receive evidence
of his right. Due to the fugitive nature of water
as compared to that of land and other objects,
this water  right is described as a unsufructuary
right; that is, only a right to use the water once it
is captured. Once the  water is under the control
of the appropriator, the  corpus of the water it-
self becomes his personal property for the use so
appropriated. In  the famous case  Coffin vs.
Lefthand Ditch  Company,  the  court stated,
"water in the various streams thus acquires a
value unknown to moister climates. Instead of
being a mere incident to the soil, it rises when
appropriated to the dignity of a distinct  usu-
fructuary estate or right of property	the
right to water in this  country by priority of ap-
propriation thereto, we think it is and has al-
ways been the duty of the national and state
governments to protect."2
  Once the  user  meets  the appropriation re-
quirements of diversion, application to benefi-
cial use and compliance with state statutory re-
quirements, a property right having four basic
characteristics is created. First, the right is for a
definite quantity of water.  This is determined
according to the use to which the water will be
put.  Second, the right is for a definite use. In
general, most rights are appropriated for single
uses where a direct flow right is sought. Where a
storage right is requested, the use may be for a
multitude of beneficially recognized uses. Third,
there is an established point of diversion from
which the water will be acquired in the exercise
of the right.  And, fourth, the water right  is
usually for a time period depending upon the
nature of use. For example, a municipal water
right  may be for a yearly diversion, whereas an
agricultural  direct flow right  may only be ex-
ercised during the normal growing season.
  Because a  protected water  right exists does
not infer the absolute security  in exercising that
right in an unreasonable fashion. There are doc-
trinal limitations that vary  from state to state
which operate as constraints to the right holder.
The  appurtenancy doctrine, tying a water right
to specific land for which the water was appro-
priated and subsequently prevents transfers to
other lands, is in effect in five states.3 Duty of
water  concepts  range from specific statutory
limitations of the amount of water that is rea-
sonably necessary for a  particular purpose or
use,4  to  allowing  administrative discretion in
deciding what is an adequate supply.5
  In  addition to specific doctrinal limitations,
there are direct statutory restrictions to the ex-
ercise of a water right, such as, Colorado Re-
vised Statute Section  148-7-8 providing "during
the season it shall not be lawful for any person
to run through his irrigation ditch a  greater
quantity  of water than is absolutely necessary
for irrigating his  land and  for domestic and
stock purposes; it being the intent and meaning
of this section to prevent the wasting and use-
less  discharge  and  running  away  of  water."
Further limiting this right, the Colorado court
held in 1893 that no one is entitled to a priority
of more water than he has actually appropriated
nor for more than he actually needs. Again,  in

-------
268    MANAGING IRRIGATED AGRICULTURE
 1912 the court reiterated this limitation by limit-
 ing the appropriator to that amount  of water
 that is reasonably necessary to irrigate  his lands
 regardless of the quantity appropriated.
   The system of  property rights in water serves
 a dual purpose. While permitting right holders
 to use this resource, it likewise restricts the use
 and action by others.6  In this way, each man's
 private use of the resource is determinable un-
 der the law.
   Another unique feature  of the appropriation
 doctrine is that  a  water right acquired there-
 under may be lost if the appropriator  does not
 exercise it for a specified period of time. There
 are two general methods by which this  right can
 be lost and the wateTs appropriated thereunder
 made available for appropriation by others. The
 first is through the abandonment of the right by
 the  holder. This requires  the  nonuse and the
 intent not to use. Both of these elements must
 be present before a right can be dissolved under
 this theory. More common, however, is the for-
 feiture theory which is usually statutory in na-
 ture requiring only nonuse for a specified period
 of time. It immediately becomes apparent that if
 intent does not have to be shown the dissolution
 of the right is much simpler.
   This is an  abbreviated discussion of the legal
 solution to the water management and  utiliza-
 tion  problems that occurred  nearly a century
 ago.  In  general,  it allows  for the use  of water
 wherever the user could show a beneficial use
 could be made. It  did  not  restrict  it,  as under
 the eastern riparian doctrine, to lands adjacent
 the water course.
   Many irrigation  systems began  dotting the
 map throughout the western states. Transmoun-
 tain diversions have transported the  water to
 areas  where  uses could be made  but natural
 suppliers were unavailable.  Technology and the
 legal means of acquiring a secure right to a pre-
 dictable and dependable supply  of water have
 permitted nearly unfettered distribution  of wa-
 ter here and  there throughout this semi-arid re-
 gion.

      The Solution Becomes The Problem
   For nearly a  century, agricultural  develop-
 ment has taken place in the seventeen western
 states subject to the provisions of water alloca-
 tion and distribution  proclaimed by  the  state
 water codes.  Individuals have  governed  their
activities in accordance with their right to use
appropriated water and have made investments
according to the  stability and quality  of their
right. Irrigation systems developed establishing
social communities where its members were the
direct recipients of the economic base made pos-
sible  through  the describable use  of  this re-
source. The primary and secondary benefits of
agricultural development has had  its  impact
throughout the nation in meeting the needs of
consumers.  States  have established  relation-
ships with  neighbors over interstate  streams
through negotiation or litigation based upon the
principles of their water law. The federal gov-
ernment has upheld the right of states to govern
the use of water arising within their boundaries
according to the  laws, customs and decisions
that the states developed.
  More basic to agriculture is the fact that its
economy rests on security of water rights, parti-
cularly in an era when economic maximization
and optimization principles question the validity
of traditional allocations over more contempo-
rary productive uses.
  Given the state of the situation, how has the
water right solution of the  1870's become the
problems  of the  1970's? We can examine the
"problem"  at  two  levels —state  and  inter-
state— considering both the water quantity and
quality aspects.
  The real problem of water rights  at the state
level begins with their issuance. The administra-
tive and  adjudicative  mechanism was in its in-
fancy when confronted with the task of deter-
mining and awarding water rights to applicants.
Adequate means of accurately measuring diver-
sions  were not  developed or used and, conse-
quently, guesstimations were made on  the di-
verted flow according to a number of measure-
ment techniques.
  Few questioned the quantity allocated the ap-
plicant because  sufficient water remained in the
streams to meet subsequent  appropriators. The
amount decreed was  not as important as the
amount of water  remaining  in the stream after
diversion or return flows. Eventually, however,
the tabulation  of  water rights far exceeded the
actual flows. In the mean time, appropriators re-
ceived valid enforceable water rights which may
later be exercised to their full capacity if benefi-
cial use could be shown. This discrepancy is
called paper water  rights — the  difference be-

-------
                                                                WATER RIGHT CHANGES
                                             269
tween the  quantity  decreed  and  that  which
could be beneficially used  or historically di-
verted. It is difficult, if not impossible, to effec-
tively distribute  water or develop a  manage-
ment water plan at the state level  when such a
discrepancy exists.
  A second problem  arises from the nature of a
water right itself which contains two  elements
inhibiting the implementation of efficient water
management practices. The first arises over a
fear of loss through abandonment or forfeiture
of the right through the failure to exercise it. As
a result, appropriators divert  as much of their
total allotment as possible to protect their right
resulting in overapplication of water and subse-
quent diminution in quantity flows. The second
problem pertains not to the fear of the right but,
rather, engendered from the security  of a  pro-
tectable  property interest  in the continued di-
version of  the decreed  quantity of water and
certain limitations upon the exercise of the right
that bind current uses to past practices.
  Irrigators adjust their operation according to
the water supply available and conditions under
which they can exercise their  right. If their ap-
  propriation is sufficient to allow delivery of wa-
  ter needed to produce the intended crop at the
  point of application, no incentive is provided to
  reconstruct diversion facilities for  transmission
  canals.  Likewise,  if the supply is  adequate or
  even  over abundant, close management in the
  application is not encouraged. Return flows may
  be due to over application, mismanagement or
  seepage  benefiting  downstream appropriators
  dependent upon this source of supply but detri-
  mental  to the  operations  of the system  within
  the state boundaries (see Figure 2). Transfer re-
  strictions found in certain  states prevent the uti-
  lization of water  on more productive lands or
  uses.  These  restrictions may be a  part  of the
  right  itself tying that appropriation to specific
  lands, doctrinal concepts, such as the area of or-
  igin, or purely organizational impediments and
  red tape discouraging such transfers.
     The epitome of the problem is represented by
  the decision in Salt  River Valley  Water Users
  Association  vs. Kovacovich.7 The  Arizona Su-
  preme Court was  presented with the task of de-
  ciding whether or  not an owner of land having a
  valid pertinent water right may, through water
                               PRECIPITATION
INFLOW TO
 CANALS
                  EVAPORATION
                  FROM (iAMALS

        UPSTREAM                / 7
                                                                                          LA?:0
                                             APPLIED TO
                   RIVER    DIVERTED FOR     IRRIGATED LAND
                   FLOW       IRRIGATION
              GROUNDV/ATER
              CONTRIBUTION
                                                                                     IRRIGATION
                                                                                     RETURN' FLO'.V
  SURFACE RUNOFF
FROM MON-IRRIGATED
       LAND
         IND. &
         WASTES
                                                                            DOWNSTREAM
                   NATURAL
                    INFLOW
                        Figure 2: Model of the Irrigation Return Flow System

-------
270    MANAGING IRRIGATED AGRICULTURE
saving practices, apply the water thus saved to
immediately adjacent lands  in  his ownership
without applying for a right to use the salvated
waters under the state water code.8 In this case,
the defendant improved certain  of the  ditches
and concrete lined others- The water that was
normally  lost through seepage or evaporation
was applied to 35 acres of adjacent land. The de-
fendants did not increase  their diversion  from
the river. In analyzing the situation, the court
stated
   "It was argued that decision of this issue in fa-
   vor of appellants  (Salt River Valley Water
   Users Association) would result in penalizing a
   person who, throughjheir  industry, effort and
   expense, engaged in water saving practices...
   This court is of the opinion that water saving
   practices entered into by  appellees (irrigators
   Kovacovich  and Ward) not only  resulted in
   conservation of water but also other benefits to
   appellees such as weed and vegetation growth
   control along such irrigation ditches and reduc-
   tion of time and cost of maintenance of such
   ditches. Certainly any effort by users of water
   in  Arizona tending toward conservation and
   more economical use of water is to be highly
   commended. However, commendable practices
   do not in themself create legal right.

   ... In an effort to  achieve some degree of
   order. . . our court, through a series of deci-
   sions developed and applied what we today re-
   fer to as the doctrine of beneficial use.

   This court is of the opinion that the doctrine of
   beneficial use precludes the application of wa-
   ters gained by water conservation practices to
   lands other than those to which the water was
   originally appurtenant. ..  Beneficial use is the
   measure and the limit to the use  of water. The
   appellees may only appropriate the amount of
   water from the Verde River as may be benefi-
   cially used in any given year upon  the land to
   which  the water is appurtenant  even though
   this amount may be less than the maximum
   amount of the appropriation. . .  Any practice,
   whether  through water saving procedures or
   otherwise whereby appellees may  in fact  re-
   duce the quantity of water actually taken  in-
   ures to the benefit of  other water users and
   neither creates a right to use the waters saved
   as a marketable commodity nor the right to ap-
   ply the same to adjacent property  having no ap-
   purtenant water right."9
   It is evident from this decision that Arizona
provides no incentive to implement water man-
agement technology in  individual irrigation op-
erations. Fortunately, the holding in Kovacovich
is not the general rule. It is generally held that
waters salvaged  through man's improvements
are subject to appropriation  and  use by  the
party saving them.10  However, the  cases estab-
lishing the general rule are early cases decided
in a period of time when agriculture was  king
and the multitude of other uses were not placing
such high demand upon water supplies. The re-
sult obtained in Arizona could be decided in
other appropriation  states upon a strict  con-
struction of the beneficial use concept and defi-
nition of an appropriation.
   On an individual basis, the water right prob-
lem may not appear significant. Taken on the
aggregate at the  local and state level, however,
the impact of this right deserves considerable
attention.
   Turning  to the problem of water quality at the
state  level  we find that quality has never been
an element of a water right under the appropria-
tion system.  The appropriation  doctrine devel-
oped in an  atmosphere  where the significant
feature  of water was  its  abundance or  lack
thereof. Consequently,  the only guarantees that
the appropriation system offers right holders is
the right to a specific quantity of water divert-
able in order of priority.  Cases in a number of
jurisdictions  have decided  the right of appro-
priators to a usable  quality of water but with
few exceptions these  are early decisions  and
their holdings subject to question under revised
and amended water codes. The issue depends in
part upon  whether the  injured party is a down-
stream senior or junior  user. In the first instance
the general rule is stated in Right v. Best: "it is
an established  rule in this state that an appro-
priator of  waters of a stream,  as against upper
owners with  inferior rights  of use, is entitled to
have  the water at his  point of diversion  pre-
served in its natural state of purity,  and any use
which corrupts the water so as to essentially im-
pair its usefulness to the purpose to which he
originally  devoted it  is an  invasion of  his
rights."11 This rule is applied in Arizona, Colo-
rado  and  Utah.  In each instance,  the injured
party was adversely affected by the action of an
upstream mining operation. Where water law

-------
                                                              WATER RIGHT CHANGES
                                         271
theory is deficient, the common law doctrines of
nuisance,  trespass,  negligency and equity pro-
vide the remedy to protect the property right in-
jured. In Suffolk Gold Mining and Milling Com-
pany vs. San Miguel Consolidated Mining and
Milling Company, the Colorado court held "that
all property rights are subject to the very equita-
ble principle 'use your property so as not to in-
jure others.' (translated) This is a principle both
of morals and of law and there is no principles of
absolute right  to property  which is not mea-
sureably subject to this condition."12
   In a recent Colorado  case the Supreme Court
upheld the right of a senior downstream appro-
priator whose  domestic water supply was pol-
luted by  an upstream fishery. Again, this  case
was decided upon the property right concept in
an appropriation.
   The next question is whether  a  senior  up-
stream appropriator can diminish the quality of
water  since downstream  junior appropriators
must take the  stream as  they find it.  There is
substantial case authority  indicating  that the se-
nior has such a right. However, in Colorado the
right of the senior to pollute has been denied
and the riparian  rule of reasonable use applied
instead.13
   In spite of well defined rules set forth in the
preceding cases, the real issue  is  whether a
downstream irrigator can prevent an upstream
irrigator from exercising his right to  use the wa-
ter for irrigation purposes  where through his use
the  quality of  water reaching the downstream
user has been impaired. No cases have  decided
this question.  The issue  is  becoming  increas-
ingly important due to the salinity  concentra-
tions  in  certain streams  reaching  levels  ad-
versely affecting  plant growth. Because of this
deficiency in the appropriation doctrine, it is
commonly felt that other means  of alleviating
the problem must be devised. The nature of the
salinity problem is not confined to state bound-
aries as is the law and concepts that have con-
tributed to the  problem.
  Technology has succeeded in developing fea-
sible solutions  to reducing the salinity  problem
arising from  both  man-made  and   natural
sources, but the law has been slow to encourage
or  direct  implementation.  The technical possi-
bilities can be divided into two categories: mea-
sures to increase the water supply and measures
to reduce salt  loading in the streams.14 With
respect to agriculture, the water supply can be
increased  through conservation measures de-
signed to reduce the consumptive use of water
or implement improved irrigation practices. To
reduce the amount of salt entering the stream
from  irrigation return flows, proper land selec-
tion, canal lining, improved irrigation efficiency,
proper drainage, and treatment  or disposal of
return flows is  recommended. It  is noted in the
government report that various factors such as
economic feasibility,  lack  of research and legal
and  institutional constraints  limit the present
application of  certain technical possibilities to
the salinity control problem.
  When examining the total picture of quantity
and quality management in the West, the prob-
lems  facing the individual state seem minimal.
To complicate  the basin problems, divisions of
interstate stream have  been made by  compact,
judicial' decision or congressional  allocation.
These solutions have neglected to include qual-
ity as an  element  in describing  the rights and
obligations of states on interstate streams.
  The question then becomes, what is the role
of agriculture and  how has the use of water for
this purpose contributed to the basin problem?
Agriculture, being the  largest diverter of water
in the Colorado  River Basin is likewise the
greatest contributor to the  problem due  to
stream flow depletion and salt loads introduced
into  the stream by return  flows.  In addition to
the salt concentration in remaining waters as a
result of stream flow diversion for irrigation, out
of basin diversions substantially affect the prob-
lem.  The total  export of water from the Colo-
rado River Basin amounts to 6,569,000 acre feet
per year,  almost  one-half the  total Colorado
River flow.15 Having identified the major legal
problem, the next  section  will consider ways in
which the law can be changed or re-interpreted
to promote the efficient and effective use of wa-
ter through the implementation of technological
innovations.

  Water Right Changes to Implement Water
          Management Technology
  In  the preceding discussion the villain  of the
western United States water problem has been
described as the property right concept in water
created through the appropriation doctrine. It is

-------
272     MANAGING IRRIGATED AGRICULTURE
                            Wuln ol urow ;XO3e*1iOn«4 tl
                            l«Ugmltro«iQuiin.lK
-------
                                                                WATER RIGHT CHANGES
                                          273
encourage maximum  use of available waters
through incentives  where possible,  strict en-
forcement  of the  law  where  necessary, but,
more importantly, through the  personal  initia-
tive of water users and a recognition of their so-
cial responsibility.
  Recommendations to resolve the water quan-
tity and quality problems of river basins must, in
reality,  examine the river system as one hydro-
logic  unit regardless of state boundaries. In so
doing, the modifications and reinterpretations
of the law must be designed to accommodate
the overall goal  of  efficiency and  effectiveness
in water management. This will require coordi-
nation  and cooperation  by  state  and federal
agencies  presently  working  with water re-
sources management. Although irrigation pollu-
tion  control  is  theoretically designed to  be
covered by pollution control  laws at  the state
and federal level, this still does not provide a
solution extensive enough to reach the  principal
cause of salinity loads. For this reason the water
right  laws must be examined  and  necessary
changes made to meet this goal.
  Focusing upon the agricultural uses  of water,
yet cognizant of the multitude of other uses in
river  systems, the first step is to structure the
water right around three sublevels of the irriga-
tion return flow system to integrate quantity and
quality constraints in formulation of changes in
the exercise of the appropriation. The three sub-
levels are (1) the water delivery system, (2) the
farm  and (3) the water removal system.18 In this
manner, the  various individuals and organiza-
tions  holding water rights or transporting waters
can effectively be included in any water man-
agement scheme.
  In  certain  instances a farmer may,  through
the exercise of his right, divert from the stream
through his own conveyance system and apply
the water to the land, gathering tail water, seep-
age or bypass water into a drainage ditch and
discharging it  into a water course. At each level,
the rights to the use  of that water depends upon
the water right held. The degree of efficiency in
the farmer's  operation will depend upon the
amount and dependability of the right, at which
point the diversion is measured and the physical
and legal possibilities of capturing and reusing
the irrigation return flows before they  reenter a
water course.
   In other cases,  ditch companies or irrigation
and/or drainage districts may own and operate
the delivery and/or water removal system. The
company or district may transport  the  water
owned by the company, district or farmer to the
latters  headgate.  Losses through the  delivery
system  may or may not be  charged against the
farmers right, share in the company, or rented
quantity. Likewise, the irrigation  and drainage
district  may be able to decrease losses in  its re-
turn flow system in transporting the drained wa-
ter back to a stream. The law must be conscious
of  quality  implications  in each  sublevel since
each irrigation return flow system  is affected by
upstream or effects downstream systems.
  Three general alternatives  are available  to
promote the  implementation of water manage-
ment technology in the irrigation return flow
system.  The first is to do nothing with the legal
and institutional constraints and  depend upon
the operations of the market place  and social re-
lationships of the user.  This obviously will be
time  consuming,   uncoordinated,  misdirected
and the results unpredictable.
  The second alternative is to devise a system
of  direct legal incentives designed,  within  the
present  water rights system, to encourage a so-
cial consciousness  regarding water use among
agricultural and other water users. Conceptual
and structural  changes or modification in  the
law are recommended for  adoption  conjunc-
tively.
  The third alternative is to stiictly  apply  the
water law concepts against the exercise of a wa-
ter right and let the burden lie where it will.
  I submit the laws are nearly sufficient in their
present  state to resolve the legal and institu-
tional phase  of the problem. Reinterpretation
will bring  into  focus  the water  law doctrines
with  technology and  the public needs. Many
concepts now postulated, if placed in the proper
perspective and if applied, would promote more
efficient water management.

           Conceptual Alterations
  The concept that is at the heart of the appro-
priation  doctrine  and which  could make the
most substantial impact in solving the legal and
institutional constraints is the concept of bene-
ficial use. This is a very nebulous concept which
defines  the measure and the  limit of a  water

-------
274    MANAGING IRRIGATED AGRICULTURE
right. In general, beneficial use pertains to noth-
ing more than the reasonableness of the diver-
sion according to the use to which the water is to
be applied.  At present, what is a beneficial use
for  acquiring  a water  right may depend on
whether that particular use is one recognized by
the  state constitution  or statutes. The concept
must be  conceived and  directed  not only to
types of uses but to the nature of the use on the
farm with respect  to the users needs. More im-
portantly, this concept must be viewed with re-
spect to the users  responsibility to other down-
stream users and the public interest.
  The concept could be used to implement wa-
ter management technology. Irrigation schedul-
ing  and careful water  control could  be encour-
aged through an  interpretation  of the concept
as prescribing to the most advanced technolog-
ically feasible  management  program "with re-
spect to on farm use.  Conventional methods of
water application using ditches and borders and
in some  cases sprinkler  systems have the ad-
vantage of being economically inexpensive but
conversely place a  great strain on the water bud-
get  due to  losses  through  evapotranspiration.
Where  feasible, subsurface application  or the
trickle  method could  be  encouraged  which
would have the effect  of reducing the quantity
of water applied and salt loads in return flows.
  A major change in the nature of a water right
that would  serve to protect the interests of the
right holder and later water users would be to
add the element of water quality. In so doing,
the right holder would have the same assurance
and likewise liability in the use of diverted wa-
ter  within the priority system for quality  pur-
poses as  he now  has  for quantity flows.  This
change  would  be  instrumental in encouraging
practices  to treat  or  dispose of highly saline
waste waters and encourage the proper applica-
tion of water on the farm.
  Significant  to  five appropriation  doctrine
states is  the appurtenancy concept which ties
water to land. This concept breeds  inefficiency
by promoting irrigation of certain lands that are
not  as  productive as  other available lands be-
longing to the right holder or other land  owners
wishing to purchase water rights. Elimination of
this concept would allow the landowner to make
a proper selection of land that would yield the
highest  agricultural returns to his operation.
  States following the rule in Kovacovich case
discussed earlier regarding the use of salvaged
waters could greatly benefit if that decision was
distinguished or overturned and instead the rule
set  out in Reno vs. Richards adopted.19 Reno
held: "if one, by his own efforts, adds to the sup-
ply  of water in a stream, he is entitled to the wa-
ter which he has developed even though and ap-
propriator with more senior priority might be
without water. The reason for this rule is the ob-
vious one that a person should reap the benefits
of his own efforts, buttressed by the view that a
priority relates only to the natural supply of the
stream as of the time of the appropriation."
  Another  concept  existing in  certain  states
pertains to the right of an appropriator to recap-
ture his waste water. A liberal rule was set out
in Binning vs. Miller granting the right to recap-
ture waste water still on the original land and to
reuse it on that land.20 A more acceptable rule in
light  of the  concept that return flows inure to
the  benefit of downstream appropriators such
that the transfer of a water right can not exceed
the  consumptive  use diverted  was  set  out in
Lason vs.  Seely21  adding an additional  element
that if the waste water has re-entered a natural
stream so that the appropriator has lost control
of it, no right of recapture would exist. Colorado
applied both the doctrine in Reno and Lason to
salvaged and developed waters and would ap-
pear to extend the requirements to waste and
seepage water as well.22 A change in the laws to
reflect the above  stated rule  would permit  im-
plementation of a pump back system for tail wa-
ter running off the end of a field.  In addition to
allowing the farmer a right to reuse water that
has been through his operation,  it would also
serve as a stimulus to be more careful about ap-
plying fertilizers, pesticides and other pollutants
on his fields or overapplication of water causing
salt concentration in his runoffs since the water
would again be placed on his own land to the
detriment or benefit of his operation.
  A final  doctrinal impediment in the  exercise
of  water  rights in  the transfer  restriction  of
rights within an irrigation system to other uses,
or outside of the basin. This constraint may exist
in the substantive water law or as a result of the
organizational and administrative system of the
state. There are few states that prevent the sale
and transfer of water rights from within or with-

-------
                                                                WATER RIGHT CHANGES
                                          275
out the present uses. States restricting transfers
rely upon the appurtenancy concept to prevent
such  shifts. However, the law should be modi-
fied or changed to reflect state encouragement
in the renting, leasing, transferring or selling of
water rights to  other uses and places so long as
the vested  rights  of others  are  protected. Al-
though there are no restrictions on  the transfer
of water rights  in Colorado, the organizational
red tape — delay and expense — acts as an im-
pediment. Changes in the administrative and ju-
dicial system  should be made to facilitate ex-
changes of  water rights. Recognition of such a
right  and a change in the concept of beneficial
use to include  recreation, aesthetics, fish  and
wildlife and other beneficial uses would serve to
nullify the fear of loosing that portion of the
water right not exercised by permitting the
transfer of the unneeded portions to other uses
within the system.
  Removing these  rigidities  in the  law to give
the right holder greater freedom and flexibility
will eliminate many of the irrigation problems
perpetuated by  the appropriation doctrine.  Ag-
ricultural  users are  subject to constraints that
other users  are  not, which is frequently passed
over when comparing the use of water for agri-
culture to other uses.
  A substantive change in the water  laws affect-
ing the administrative organization  of the state
that should be enacted to enable a  greater de-
gree  of cooperation  between the state agency
and water users and at the same time permit the
state  to concentrate on the development of a de-
sirable state water plan is to  enact legislation
permitting the state water resources agency or
other public organizations the right to acquire
water through appropriation,  purchase,  aban-
donment or condemnation. The significance can
be  seen  at  the state and interstate level  by
granting the state greater freedom  in carrying
out its responsibilities and  negotiating agree-
ments with its basin users and states.

            Structural Alterations
  Certain structural changes are recommended
which would encourage water users  to take ad-
vantage of the conceptual changes suggested
above. Already in certain states, irrigation dis-
tricts  are used  to  circumvent  constraints im-
posed by the appurtenancy concept and transfer
 restrictions. In Wyoming and Arizona, for ex-
 ample, water rights can be transferred to an irri-
 gation district which then has the power to shift
 the water within the system according to the de-
 mands of the irrigators. In Colorado, ditch com-
 panies operate to rent and transfer water within
 their system in order to avoid the  cumbersome
 organizational impediments.
   The problem is that these practices are on a
 very limited scale. What is needed is a means of
 allocating and reallocating  water within the ir-
 rigation system by an organization  cognizant of
 the needs of water users within the system, the
 state water development plan and the basin and
 international impacts.  Suggested is  the develop-
 ment of a centralized state brokerage system to
 operate as a market center for the exchange and
 sale of water rights or renting of water avail-
 able under the rights held.
   This brokerage system could  be  organized as
 a  public or private institution. It would permit
 water  users to divert only that amount of water
 necessary for their operation without fear of los-
 ing the unused  decreed quantity and  lease  or
 rent the difference to other users. Hence, there
 would be an economic incentive to implement
 the most  efficient water management practices
 in their operation in an attempt to reduce the
 necessary  quantity of water applied.
   A brokerage system created as a  public entity
 could  be  established in the  Office  of the State
 Engineer  or water planning and resource  de-
 partment of the state. This office or subdivisions
 in the  various basins within the state would list
 all available water for rent, lease,  exchange or
 sale. The location of available waters will deter-
 mine the  impact  upon other  vested  rights but
 the responsibility for delivery and protection of
 such other rights would rest upon either the wa-
 ter right  holder  or water  acquirer.  Uniform
 prices  of units of water could be established or
 the available water could be  transacted to the
 highest bidder. The adoption of such a system in
 state  organization  would  require  changes  in
 agency laws to permit this type of activity. Like-
wise,  it  would  be  imperative  that the state
 should have the power to purchase, condemn or
 receive water  rights in the  name of the state.
This would allow the state to take action against
appropriates who refuse to  implement efficient
practices,  acquire their unused rights and retain

-------
276
MANAGING IRRIGATED AGRICULTURE
them for future use while renting or leasing the
water during the interim. A percent of the trans-
acted price would be retained for the operation
and  maintenance expense of the brokerage sys-
tem.
   If a brokerage system was to be established in
the private sector of our economy, it is recom-
mended  that  the  organization  created  be  re-
quired to submit to the standards of a nonprofit
corporation and subject to direction and assis-
tance from  the state water agencies. The func-
tions of this private entity would be the same as
those above described. However, it is envisioned
that problems would be created in establishing
such an institution. With time many people will
ignore the nonprofit aspects of the agency and
create similar profit corporations. Secondly, the
redistribution of water and water rights may not
be in the best interest of state and interstate ba-
sin resource management.
   The final alternative to implementing water
management practices  is  to enforce existing
laws  strictly.  Certain  rigid  rules,  if executed
would decrease quantities  diverted.  Applying
the concept of beneficial use the court in City
and County of Denver vs. Sheriff stated "what
is beneficial use after all is a question of fact and
depends   upon  the   circumstances  in  each
case."23 In Ericson vs. McLean the New Mexico
court said that  "an excessive diversion of wa-
ter through waste cannot be regarded as a diver-
sion to a beneficial use within the meaning of
the constitution."24 And a 1917  Utah case held
"beneficial use contemplated in making  an  ap-
propriation must be one that inures to the ex-
clusive benefit of the appropriator and subject
to his complete dominion and control."25 When
excessive water is  lost  through canal-seepage,
evapotranspiration or  waste, the  state would
modify the right to reflect only the  water  ap-
plied beneficially.  Thus, the duty of water for
the use appropriated would reflect the most ef-
ficient means  of transporting and applying wa-
ter. Applying the statutes and rulings in the var-
ious states may serve to prod irrigators into im-
plementing  more  efficient practices  into their
operation  through fear of losing their water
right.  Socially this is undesirable. To be effec-
tive in enforcing  the  law re-evaluation of the
water  rights now held  would be necessary to
free up paper or waste  water rights. Every ap-
                                         propriation doctrine state has a sufficient mech-
                                         anism either through abandonment or forfeiture
                                         to attack a water right not being exercised ac-
                                         cording to the law.
                                            I am of the opinion that strict enforcement of
                                         water law doctrines should be implemented only
                                         in cases  where an  obvious discrepancy occurs
                                         between the water right decreed and the histori-
                                         cal diversion. A more favorable approach would
                                         be to enact the incentive recommendations; the
                                         same results  obtained  through  strict  enforce-
                                         ment would be attained with  less cost, conflict
                                         and adverse public  relations. The State  of Colo-
                                         rado is now  in the process of tabulating water
                                         rights  under a 1969  law  as a first  step in the
                                         eventual  abandonment  of paper  and waste wa-
                                         ter rights. The publication of this first tabulation
                                         cost the state $90,000 and it is questionable how
                                         many  more  publications  must  take  place to
                                         eliminate errors in the original  tabulation. Utah
                                         has been going through abandonment proceed-
                                         ings on a sub-basin scale in order to accurately
                                         assess  the amount  of water diverted by water
                                         users.  It has taken approximately 23  years to
                                         perfect the  abandonment procedure.  Three
                                         years ago they felt the system was sufficiently
                                         developed to  permit a reasonable proceedings
                                         without undue expense and delay in the remain-
                                         ing basins. In the  meantime,  vast amounts of
                                         state money and time have been consumed. Ap-
                                         propriators  in the  unadjudicated areas are at-
                                         tempting to validate their rights through the ap-
                                         plication of the decreed quantity  of water or sell
                                         to users who could utilize the  entire right.
                                            The question is whether we should follow the
                                         letter of the law or be pragmatic. Economic and
                                         social consideration should be weighted before
                                         embarking on such a policy. Would it have cost
                                         less  to permit farmers to sell excess water rights
                                         to other users? The cost to the public of such a
                                         program may warrant the outright purchase of
                                         paper and waste water rights by the state with
                                         much less adversity.

                                                        Why the Push?
                                            The time is now upon us to  critically  and seri-
                                         ously evaluate the allocation and use of water in
                                         the  United  States  and particularly  the western
                                          states at every level of influence and connection
                                         with this resource.  It is essential not only to de-
                                          termine the degree of protection  and security in

-------
                                                                WATER RIGHT CHANGES
                                          277
existing rights and performance of duties under
current law and practices but also to assess the
impact  of environmental  considerations upon
the  water  picture.  Environmental challenges
have primarily been the concern of federal gov-
ernment and affected state and private organi-
zations,  but the  trend  emanating from  this
movement cannot be  ignored either  by water
users or those responsible for its allocation, dis-
tribution and adjudication. The emphasis is no
longer  confined to the  jurisdiction  of state
boundaries.  Rather,  basin,  regional,  national
and  international considerations in  both water
quantity and quality are becoming an integral
factor for  state and interstate water agencies.
  So why the  push? What does all this  mean?
Let's take  the  Colorado  River Basin for ex-
ample. Each state has adopted a water code for
allocation  and distribution  of waters  arising
within its boundaries. In addition, the  Colorado
River  has  been subject to compacts,  court de-
crees and congressional allocations  in the divi-
sion of its  waters among  the respective  states.
Interstate allocation is only  quantity  oriented.
The treaties with Mexico obliges  the  United
States to deliver certain quantities of water and
only recently, through threat of a conflict, have
negotiations partially resolved the quality prob-
lems in the basin.
  Water users  in the various states have appro-
priated water  according  to  the state's  system
with little concern for the effect upon any user
outside  their immediate  system. These water
users still have little  reason to  be concerned
with  other users  unless  their  own rights are
being  affected.  With  present and  projected
shortages  of water and   diminution  of water
quality in  the  Colorado River there are  few, if
any, right holders who should not be concerned,
regardless of their location within the  basin. Al-
though secure in their own use, the next highest
authority may  have an obligation  which,  if met,
will dramatically effect   the individual  user.
Someone must meet the deficits if delivery com-
mitments are not met.  The question is, who?
Someone must alter his practices if the quality
of water is decreasing to the stage where down-
stream appropriators  can no  longer exercise
their rights to  the use so  appropriated. Again,
the  question is who —the  state  or the water
right holder?
  Initially the state will be responsible to other
states for its water uses, but the obligation will
not stop there. State water officials will be com-
pelled to examine state water budgets and pos-
sibly be required to enforce the law strictly, re-
sulting in unwanted abandonments or forfeiture
proceedings.
  At this point, the field of environmental law
and recent  court decisions  must be  consulted.
Many  new  concepts and doctrines  are  being
adopted  due to the assessment of man's  rela-
tionship  with  his  surroundings. A particularly
important case that could have a significant im-
pact upon the salinity problem in the Colorado
River and  the obligation of states and water
users was recently decided by the Tenth Circuit
Court of Appeals. In Texas vs. Pankey the court
held that Texas  had a right under federal com-
mon law to prevent the application  of insecti-
cides by defendants in the state of New Mexico
in .such a manner to damage the water of the
Canadian River Watershed  in Texas.26 Citing
the Tennessee  Copper Company case which rec-
ognized the right  of one state to abate a nui-
sance occurring in another adversely affecting
the former the tenth Circuit Court stated
  the source or basis itself for such a quasi-sover-
  eign ecological right in a state's position in a
  general political union was not discussed, the
  right  apparently was regarded as  having ex-
  isted in the common law and as being entitled
  to remedy within common law principles. . .

  Federal common law and not the varying com-
  mon law of  individual states is, we think, en-
  titled and  necessary to be recognized as a basis
  for dealing in  uniform  standard with the en-
  vironmental  rights of a state against improper
  impairment of sources outside its domain.27
  The implication of this case in handling the
water quality  problem of the Colorado River is
clear where  it  can be demonstrated that a lower
basin  state's  rights  have  been  damaged  by
another state or actions occurring therein.

SUMMARY
  Man cannot rely upon technology to solve all
his problems nor ignore the true relationship he
has with his environment and others within his
realm of effect. The water problems of the west-
ern United  States typify the predicaments  of a

-------
 278    MANAGING IRRIGATED AGRICULTURE
 society that has relied upon a developed  legal
 and institutional system to solve many of its past
 problems in governing man's relationship  with
 water. The solution of the 19th Century has be-
 come the problem of the 20th.
    We are no long justified in isolating our con-
 siderations to the immediate locale or region but
 must lift our sights to evaluate the entire impact
 of our actions. We can no longer hide behind nor
 rely upon,  with impunity, the security provided
 by a century old concept that met the require-
 ments  of that day. By the same token, we  can-
 not disregard the established  system  and those
 who dedicated their livelihood to its features.
    Cooperation,  coordination  and   pragmatic
 planning among water-users,  state officials, re-
 gional  bodies and governmental entities to de-
 vise solutions to the current,quantity and qual-
 ity problems are essential. Concentrating upon
 the agriculture sector in the western  states, we
 note the present practices in use of water and
 the legal system which permits and perpetuates
 these practices, the advanced level of technolog-
 ical innovation  to' improve water management
 practices within the irrigation return flow  sys-
 tem and  the obvious discrepancies between the
 scientific improvements and  the legal impetus
 to  implement the same. Changes in the exercise
 of  water rights to induce or require more  effi-
 cient uses of this  resource must and can be
 made either through voluntary action in which
 all  parties involved  benefit or through drastic
 legislation, judicial intervention, or strict admin-
 istrative application  of current rigid water con-
 cepts in which few parties involved are signifi-
 cantly benefited. It is not too late to unify  and
 create a  flexible  solution to this problem —
 flexible and pragmatic so as not to itself become
 the problem as time changes.

 ACKNOWLEDGEMENT
  "The work  upon which this publication is
 based was supported in part by funds provided
 by  the  United  States Department of the Inte-
 rior as  authorized under the Water Resources
 Research Act  of 1964, Public  Law 88-379"

REFERENCES
    1. Hardin,  Garrett, "The  Tragedy of  the
Commons," Science, Vol.  162, p. 1243, 13 Dec
1968.
    2. 6 Colo. 443, (1883).
    3. Neb., Nev., Okla., S.D., Wyo.
    4. Wyo. Stat. § 71-216 & S. D. Session Laws
 § 61.0126- 1 c.f.s./70 a. N. D. Rev. Code 61-0411
 to .0421 - 2 a'/a.  Neb. Rev. § 46-231  - 1 c.f.s./
 70 a. nor more than 3 a'/a.
    5. Nev. Stat. § 533.070.
    6. Trelease,  Frank J.,  "Policies For Water
 Law:  Property  Rights, Economic  Forces,  and
 Public Regulations," 5 Natural Resources Jour-
 nal 1 at p. 10. May 1965.
    7. 411  P2d 201, (Ariz. 1966).
    8. Salvaged  water are defined as those  wa-
 ters "saved by improvements made to the chan-
 nel of a stream; they are  waters that otherwise
 would  be  lost  by  seepage  or  evaporation."
 Hutchins,  W., Selected Problems in  Western
 Water Law, Misc. Pub. No. 418, U.S.D.A., p.
 348, 1942.
    9. Supra, note 7.
   10. Clark, R.  E., Water & Water Rights, Vol.
 1, p.  342, Allen Smith & Co., 1967.
   11.  121  P2d  702, (Cal.  1922).  See Meyers,
 C. J., Functional Analysis  of Appropriation
 Law,  p. 19, National Water Commission Report,
 NTIS, PB  202 611, 1971,  for discussion of wa-
 ter quality considerations  under the appropria-
 tion doctrine.
   12. 48 P 828 at 832, (Colo. 1897).
   13. Supra, note  11 Meyers, p. 72.
   14.  The  Mineral  Quality Problem  in  the
 Colorado River  Basin, Appendix C, Chap, III,
 US. E.P.A., 1971.
   15.  Ibid, Appendix A, p. 5.
   16.  Trelease, Martz, Moses.
   17.  Young, Gaffney, Hartman.
  18.  Skogerboe, G. V., and Law, J. P., Jr.,
 Research Needs For Irrigation Return Flow
 Quality Control, Ag Engineering,  C.S.U. and
 EPA, Project 13030 FVN,  Aug. 1971.
  19.  178 p2d  81, (Ida. 1918).
  20.  102 P2d 54, (Wyo. 1940).
  21.  238 P2d 418, (Utah, 1951).  See Trelease,
 Water Law, West Pub. Co.,  1967, p.  100.
  22.  Leadville Mine Dev. Co.  v Anderson,  17
P2d 303, (Colo. 1932). Also see Pikes Peak Golf

-------
                                                            WATER RIGHT CHANGES    279
Club v. Kuiper, 455 P2d 882, (Colo. 1969). Fora       25. Lake  Shore Duck Club v.  Lake View
discussion see Meyers & Tarlock,  Water Re-     Duck Club,  166 P2d 309, (Utah 1917).
sources Management, Univ. Casebook Series, p.       26  2 E R C  1200  (1971)

                                                 27' Ibld"  P'
            836, (Colo. 1939).
  24.  308 P2d 983, (N.M.)

-------
                  Institutional  Influences  in
             Irrigation  Water Management*

                               WARREN L. TROCK
               Department of Agricultural Economics and Rural Sociology
                                 Texas A&M University
ABSTRACT
Among the many institutional influences that af-
fect the development and use of water and land
in the Lower Rio Grande Valley are (1) a prolif-
eration  of special districts,  (2) inappropriate
water management policies, (3) uncertainties in
water rights and (4) numerous, competing gov-
ernmental entities involved in planning for and
administration of water resources. For more ef-
ficient  management of water used in irrigation
of crops it is recommended that small, underde-
veloped irrigation districts be consolidated, and
that systems  rehabilitation be  accomplished.
Water  meters should be used to measure  deliv-
eries of water to reduce the incidence of over-ir-
rigation and extend the supply. To overcome re-
sistance to  changes in district organizations and
managerial policies, widespread and intensive
educational programs,  emphasizing the benefits
of more efficient water management,  will be
necessary. To provide for desirable levels  of in-
vestment in water supply and distribution sys-
tems, uncertainties of rights must be eliminated.
To promote allocation of water so that it is most
productively used,  rights should be negotiable.
A change in water laws may be necessary. De-
velopment  of drainage systems  to solve prob-
lems of flooding, salinity, etc.  will be facilitated
by cooperative efforts among state, federal, and
local  agencies having responsibility for flood
control and drainage. Integration of water sup-
ply, drainage and flood management systems
appears  necessary.  This  will require consider-
able redevelopment  of facilities and reassign-
ment of responsibility for control of systems.
  Located in the southernmost part of Texas is
a highly developed urban, industrial and agricul-
tural region called the Lower Rio Grande Val-
ley. It includes  the land  and water of most of
three  counties and covers about  3,000 square
miles  of  land. It is one  of three areas  in the
United States capable of  producing citrus fruits
and certain vegetables, and large quantities are
grown annually. In recent years, about 800,000
acres of land have been cultivated, of which ap-
proximately 80 percent were irrigated.  The irri-
gated  areas, most of which are contiguous, con-
stitute one of the largest single concentrations of
irrigated land in the state.
  While the valley is very productive  of crops
such as cotton, vegetables, citrus fruits and grain
sorghums, it suffers from problems relating to
land and  water use. These  problems include
(1) periodic shortages  of water, (2) inefficient

This study has been supported by the Office of Wa-
ter  Research, U.S.  Department of Interior, under
P.L. 88-379, Project B-025-TEX and the Water Re-
sources Institute, Texas A&M University
                                           281

-------
 282     MANAGING IRRIGATED AGRICULTURE
use  of existing water supplies, and  (3) inade-
quate  drainage. Physical conditions  of  supply
and use of water are, of course,  important to
these problems. The flow of the river has varied
from more than 4 million acre feet to less than
one million acre  feet annually. Water available
per  irrigated acre is often insufficient for opti-
mum production of important irrigated crops.
Obsolete facilities in some of the irrigation dis-
tricts contribute to inefficient water usage. Un-
lined,  open  canals allow  seepage,  excessive
evaporation, and  loss of water to weeds and
bushes along the ditches. Drainage problems are
aggravated by too frequent, excessive applica-
tions of water, naturally high water tables and
man-made obstacles to nalural drainage. These
conditions are  important to land and water use
and they are getting the attention of engineers
and to soils scientists. But of equal importance
are institutions that affect water allocation, use
and conditions of use. They are significant to the
efficiency with which  we use  water  and they,
therefore, deserve the  attention of economists,
political scientists, lawyers and others who can
deal with them.
   Institutions have been described as "well es-
tablished social structures within which men do
collectively  the things which  seem  right and
proper, in regard to some fundamental  interest
in life."1  With respect to  the  problems of re-
source use in  the Valley we are  interested in
those social mechanisms which function to di-
rect and control resources and  economic activi-
ties. They include such diverse things as water
rights, commodity markets, traditional farming
methods, water districts and government pro-
grams. Such institutions are the product of polit-
ical acts, court decisions, customs, common law,
tradition and other social phenomena. They are
complex; they tend to be lasting and  inflexible;
but they are creations of men and can be altered
to suit our purposes.
   An important institution affecting  the devel-
opment and use of water in the Valley is the wa-
ter right. Rights  in Texas  have been  developed
under the Spanish, Mexican and Texas civil law.
During the reign of the Spanish and Mexican
monarchs, rights to divert and use water from
the Rio Grande were extended to settlers in con-
nection with the  land  grants made. Since 1889
water use permits have been granted upon ap-
plication by the state of Texas. Today all verifi-
able grants and permits are recognized,  regard-
less  of their  origin,  and they are employed to
claim and take water from the river.
  An unfortunate  characteristic  of the  early
grants and permits was the lack of specification
of the amount of water involved. Reference in
the grants was to the land area to be irrigated.
By the middle of this century authorization to ir-
rigate three-quarters  of a million  acres of land
had  been extended, while water in the river for
irrigation was ordinarily sufficient for only 600-
650 thousand acres. A court action to adjudicate
rights was  necessary. Rights were classified as
to their nature and extent and were made spe-
cific as to amount.
  Certainty in the water right is of course quite
necessary to optimum development and efficient
use of water. An individual who is uncertain of
rights tends to make less than optimum invest-
ments in his water supply. He is somewhat fear-
ful of loss and so he spends just enough to ac-
quire that water for  which he has a  claim. As a
user, the uncertain rights holder tends to exploit
his source, "taking  it while he can." The short
term planning horizon of this individual  is espe-
cially evident where he holds rights to an under-
ground water supply. He sees himself in compe-
tition for the water in the common pool, and he
uses it in a manner so that he is sure to obtain
his "share." An inefficient use of such water is
often the result.
  In the Lower Rio Grande a second institution
that  is quite  important to water development
and  use is  the  special district. In  the  three-
county area there are 34 irrigation districts, four
drainage districts and a few fresh water, naviga-
tion  and other districts. These all operate to
serve their members in various, necessary ways,
but  the profusion of districts, the variations in
purpose  and  activity,  and  the  competition
among the districts  do not contribute  to effi-
ciency in development and allocation of existing
supplies. In a study of the irrigation districts we
found them to vary  significantly in  their physi-
cal  facilities, their  capabilities for  service to
members and  in the operational efficiency of
their systems. Some districts can respond to re-
quests for water deliveries quite promptly, i.e. in
a few hours. But others, because  of obsolete
equipment  and poor management, required two

-------
                                                            INSTITUTIONAL INFLUENCES
                                          283
or three days for delivery of water. Some dis-
tricts had storage space for water, lined canals
and  ditches, underground pipelines and modern
gates, weirs, etc. necessary for efficient handling
of water.  Others had a pump on the river,
earthen canals and ditches  choked with weeks
and trees, and leaky, rusty equipment difficult to
operate.  Differences  in operational efficiency
have been found to be significant.  Recently re-
habilitated systems deliver a half foot more wa-
ter to customer-members than do older, obsolete
systems in a typical, irrigation season.2
  Benefits from rehabilitation of districts, and in
many instances from consolidation of districts to
provide for larger operational units, are appar-
ent. Our studies showed rehabilitation to be fea-
sible, at  least for the systems now serving  the
presently  irrigated  acreage.3 Reorganization of
districts,  to  increase  average  size, was  also
found  to  be  economic. Annual operating  costs
for districts of 40-50  thousand acres are signifi-
cantly lower than those for 10 thousand acre dis-
tricts. But two factors act as deterrents to district
reorganization and rehabilitation. First, operat-
ing costs  of districts, which ultimately  are paid
by members via service charges, are small when
compared with other farm operating costs. Sec-
ond, and perhaps more significant is the effect of
consolidation of districts on the  control  and
management of district operations.  With 34 irri-
gation districts now  active in  the  region, 170
farm operators participate, as district directors,
in policy and management decisions in their dis-
tricts.  Any reorganizational plan which would
reduce the number of districts would also reduce
the number of district managers and the number
of farmers on district  governing  boards. Reorga-
nization would certainly lead to a centralization
of power  in district management that would be
opposed by many.
  Associated, with the facilities and the  manage-
ment of irrigation  districts  in the  valley is  the
practice of unmeasured deliveries  of water to
farmers. This practice or policy is a product of a
time when the water resource  was  plentiful—
when there was  little use of  water in  the Rio
Grande by Mexico and less than a half million
acres of land irrigated in the Lower Valley. It
was then not necessary to measure and conserve
water.  But unmetered deliveries  have led  to
farm management problems, and they are now
critical to the efficient use of water which is no
longer plentiful. A problem of land management
in the valley is heavy clay subsoils. Drainage is
poor and salinity is a problem. Too frequent and
excessive applications  of irrigation water  com-
plicate this  problem.  Increasing attention  must
be paid to moisture needs of crops to avoid im-
proper applications of water. Measures of water
delivery by means  of meters would greatly facili-
tate soil and crops management.  Improving the
efficiency  of use  of  water on crops will  also
"stretch" the water resource so that it  may be
used to irrigate greater acreages of land. Benefi-
ciaries will be existing rights holders and water
users who will produce a greater output per unit
of water, and those who process and merchan-
dise agricultural produce from irrigated land.
  Another problem of institutions in the valley
is to be found in  the  customary  "pricing" poli-
cies  for  water.  Water is  not  purchased  from
some autonomous water supply agency. It  is ac-
quired, via the exercise of individual rights, and
by means of a water distribution agency (the ir-
rigation district) which is  owned and managed
by rights holders. The "price"  of water  is  thus
simply the cost of delivery of water to users. So,
only rights holders can obtain and use water and
they use it as they see fit. Now when water is ac-
quired (delivered to the  farm)  at no more than
$10 per acre foot,  it can be used for irrigation  of
almost any  crop which can be grown in the val-
ley. In such circumstances, water is used  quite
indiscriminately, on crops which produce low re-
turns to this increasingly scarce resource as well
as crops which produce high returns to water.
There is little incentive for water rights holders
to allocate this  resource  among alternative uses
so that returns  are maximized. When we com-
pared the existing mix of crops on presently irri-
gated land to a  mix of crops which would  make
efficient use of water  and associated resources,
we found significant differences.4 The result is a
regional net  income  which  is  considerably
smaller than that which is possible.
   A fairly obvious solution to  this  problem is a
negotiable water right. Rights  must be released
from the existing tie to land.  A  change or clarifi-
cation of water law to permit purchase and sale
of rights, or lease of the right to an annual incre-
ment of water, would facilitate exchange so that
water would be employed in higher value uses.

-------
284    MANAGING IRRIGATED AGRICULTURE
Farmers holding rights to water who are unable,
for various reasons, to  make the most produc-
tive use of them, would find it much to their ad-
vantage  to sell all  or a portion of their water
rights. Individuals not having rights, but able to
employ water in uses that produce  high returns
to the resource, could acquire water via the pur-
chase of the right.
  Negotiable  rights are a desirable solution  to
problems of allocation  of water whether it is a
surface  water resource  in  south  Texas  or a
ground  water resource  in  the high  plains  of
western  Kansas  and  eastern Colorado. Where
rights to water are tied to land in specified quan-
tities, there are ridigities in water uses which re-
sult in less than, efficient allocation of this re-
source.
  Drainage problems in the valley,  exclusive of
those created by inadvisable irrigation practices,
are a result of high  water tables, a relatively flat
and featureless  land without well established
drainage ways and barriers to movement of wa-
ter  across the  land which include highways, rail-
roads, floodways, etc. The institutions which are
important to the problems of drainage are a pro-
liferation of  organizations with plans to  solve
the  problems, numerous  other  organizations
with more or less authority and responsibility for
drainage and an information system which oper-
ates most  ineffectively to produce an under-
standing of problems and possibilities for solu-
tions. Since the turn of the century, some private
or  public organization has been studying the
drainage problems  of the valley. There  are nu-
merous plans  for controlling the route and flow
of the river,   managing runoff which at times
reaches flood  proportions, and providing relief
from high water tables causing problems of sa-
linity.  Responsibility for positive action in flood
control and drainage of lands rests with four  or
five established drainage districts, the Interna-
tional Boundary  and Water Commission, and
numerous irrigation districts which have been
given responsibilities for the provision of drain-
age within their boundaries. But the boundaries
of the drainage districts and the irrigation dis-
tricts do not coincide and their facilities are not
integrated. Many  districts have no  outlet for
their  drainage systems  except  for  ill-defined
drainageways  which may be closed by  a high-
way,  railroad or floodway.  The International
Boundary and Water Commission has provided
for the routing of high water flows in the river,
through floodways and arroyos that have outlets
in the Gulf.
  Several,  comprehensive  flood  control  and
drainage systems have been proposed  by federal
agencies, private  engineering firms and district
organizations. But effective means for informing
the potential beneficiaries of these  systems and
for discussion of and comparison of these sys-
tems have been lacking. As a result, valley resi-
dents are confused, angry and resentful  of what
they see to  be interference with "their" prob-
lems. When efforts  to obtain appropriations  or
sell bonds are made, they fail for lack  of local
support.  It is thus quite evident that attitudes,
past experiences, suspicion, disorganization, etc.
are factors more important to drainage problems
than design of drainage systems.
  These  few  institutions affecting land and wa-
ter use in the Lower Rio Grande serve  to illus-
trate the point. Institutions are significant fac-
tors affecting land and water use. We must learn
to understand the nature and influence  of insti-
tutional  arrangements,  the  social  mechanisms
by  which we control and  direct resources for
productive uses, so that we may change them,
modify them, use them to achieve our ends.
REFERENCES
  1. Renne, Roland R., Land Economics, Har-
per and Brothers, New York, 1947.
  2. Texas Water Commission,  Water Supply
Limitations on Irrigation from the Rio Grande in
Starr, Hidalgo, Cameron and Willacy Counties,
Texas, State of Texas.
  3. Gray, Roy and W. L. Trock, A Study of the
Effects of Institutions  on the Distribution and
Use of Water for Irrigation in  the Lower Rio
Grande Basin,  TR-36, Water Resources Insti-
tute,  Texas  A&M  University,  College  Sta-
tion, Texas.
  4. Thenayan, Abdullah, Organization of Agri-
cultural Resources in  the  Lower Rio  Grande
Valley of Texas with Limiting and Non-Limiting
Water Supply,  an unpublished dissertation, De-
partment  of Agricultural Economics and Rural
Sociology, Texas A&M University, College Sta-
tion, Texas.

-------
   Sociological  Considerations  in  Irrigation
     Water Management:  Facing  Problems
                 of Water  Quality Control
                                EVAN VLACHOS
                             Colorado State University
                              Fort Collins, Colorado
 ABSTRACT
   The paper attempts to present a systematic
framework for an examination of social dimen-
 sions involved in the analysis of water resources
 systems  and in problems of irrigation return
flow quality control.  Throughout the presenta-
 tion, emphasis  is placed on the institutional
 constraints as well as major socio-demographic
factors affecting the development and future
 use of water resources.
  A major argument for an understanding of
problems of water  quality,  especially  in the
 Western  United States  is the  changing cir-
 cumstances and the consequences  of four in-
 terrelated trends: the continuous population
 increase and in-migration  to the West,  the in-
 creasing urbanization and metropolitanization
 increasing industrialization, and, finally, com-
pounded problems from ecological perturba-
 tions.
  In delineating problems of water quantity
 and quality,  a systems  approach is utilized
 which attempts  to integrate physical and social
parameters and to  incorporate major com-
ponent parts of irrigation systems designed to
 achieve  maximum productivity. In  addition,
major systems functions are also elaborated
and  water quality control problems  are ex-
amined in connection with various criteria of
efficiency, effectiveness, and efficacy.
  The conclusion of the exposition rests heavily
on the significance  of cultural practices, the
social forces  of resistance to change,  and the
intricacies and limitations in  the interpretation
of the  western  water law. Finally, a  number
of suggestions are provided  concerning water
quality control in the context of an  urgent need
for an understanding of the larger social trends,
social awareness,  and integrated  social  and
technical considerations.

INTRODUCTION
  The present paper attempts  to  delineate  a
number of sociological considerations involved
in any  discussion of a  water management
system.  Emphasis is placed throughout the
following text  on irrigation systems in the
Western United States. The key problem neces-
sitating the present exposition is the increasing
degradation of water quality as a result of both
natural and man-made pollution. More, specifi-
cally, special  attention is focused on the prob-
                                        285

-------
 286     MANAGING IRRIGATED AGRICULTURE
 lem of salinity in many irrigated valleys in the
 Western United  States. It has been recognized,
 that the problem of salinity is closely associated,
 not only with the natural carry of various rivers
 in the area, especially the Colorado River, but
 also accentuated, if not aggravated, by present
 irrigation  practices  and  attitudes  concerning
 water use in the West.
   The challenge for the social scientist, at the
 present time, is  not so much the provision of
 exact answers on how to implement collective
 and individual changes, or incorporate attitudes
 leading  to  new  practices  that would alleviate
 present problems of water management. More-
 over, what is needed at  this time is a basic
 framework  of understanding the larger social,
 cultural,  and  political  millieu  within  which
 technical solutions can be implemented.  At the
 same time, irrigation, or any water system, have
 to be  understood in the context of larger re-
 gional and national trends which influence both
 the present  status as well as  future solutions of
 problems of water degradation.
   The core argumentation of the present work
 centers around the following main points:
   1. Larger demographic changes in the West-
     ern United  States necessitate reconsidera-
     tion of present and potential major water
     development systems.
   2. Concerted  research is needed in order to
     provide  a  better understanding of  the
     cultural environment, social practices, and
     organizational arrangements within which
     present water systems operate and techno-
     logy is applied.
   3. In order to develop both research priorities
     and recommendations  for  action,  it is
     imperative that a framework be developed
     which   incorporates  both  physical  and
     social parameters of any water system. We
     should  be able to  provide an integrated
     scheme in which the dimensions of both
     the technical and human conditions,  as
     well as their  interrelationship  can  be
     analyzed.
   The  essential argument, then, is that the use
of water in  any  irrigated agricultural society
must  be socially  controlled through  a set of
institutions known as an irrigation system. It is
then assumed that the way in which water sup-
plies, patterns of water distribution and water
reclamation practices are regulated in a given
irrigated agricultural society will depend largely
upon the nature and structure of its irrigation
system which  like any other system is essential-
ly dependent upon the larger socio-cultural en-
vironment,  and  the  specific ecological  cir-
cumstances of a given region.
  One  of the  most  widespread  expressions
today is the general phrase, "consideration of
human   factors,"  or   "institutional  aspects
of.. ." in any major  project.  It seems almost
ironic that such an important dimension as the
social and human aspects of any  system is re-
legated   to  a  residual  category   of  concern,
which  denotes mostly our inability to exactly
define  the non-physical dimensions of a given
system. The  problem, of course,   here is that
there  is  not only the engineers' or the practi-
tioners' fault that they have not considered the
social dimensions of any given project, but at
the same time the state of the art, as well as our
knowledge of social dimensions,  is, to say the
least,  limited and the  general propositions are
at most tenuous. As more and more concern is
expressed about the  technological imperative,
environmental disruption and  the future dire
consequences   of  man's   interference   with
the biosphere, social scientists are increasingly
called upon to  provide  recommendations and
to  answer  long-standing questions  on  the
human side of any project. Unfortunately, social
scientists  are  ill-prepared to give   the  kind of
answers that are urgently needed at the present
time. More than anything else, the difficulty of
providing answers rests on the general lack of a
cognitive  map that succinctly tries to describe
the dimensions  of the social  system and the
interconnection  of each  component  part  with
other types of environments.
  What we should attempt to do, in view of the
lack of precise  data  and limited   research on
environmental questions, is  to provide  some
generalizations and a  systematic  scheme of
incorporating  social and physical dimensions
in any  water  resource  system. What follows,
therefore, is only an introduction to a problem
that has only recently come to our  attention.
Water  quality degradation  due  to  irrigation
practices, although occurring also  in the past,
had never the urgency and immediately that is
becoming apparent in  the changing social and

-------
                                                       SOCIOLOGICAL CONSIDERATIONS
                                          287
economic conditions of  the Western United
States. Most of the work in the past was primar-
ily preoccupied with increasing crop yields and
improving the structure of distribution systems
with very little attention to associated problems
of water quality degradation. The  limited pre-
cious commodity  of water  will force a new
thinking in which projects will be evaluated not
only in connection  with the general attempt  of
increasing  national  welfare  and  growth, but
also  in connection with much more  complicated
trade-offs and larger questions of social policy.
The  survival of this part of the country, as well
as of the nation, will depend on a proper mix of
technological feasibility and social considera-
tions which increasingly will  become the major
elements for future water  use development and
for solutions to increasing problems  of  water
degradation.
  Perhaps  another  qualification   should   be
added.  Many practitioners and engineers feel
that  the title   "sociologist"  is  adequate  to
at least guarantee  or  provide  an answer  to
any  question  that  incorporates  the  adjective
"social." At  the same time, social  scientists
blithely have been playing the role  of  social
critic of new Jeremiahs,  in  providing blanket
statements  and generalized accusations against
the  "manipulators  of the environment", "the
engineering mentality", and  "the despoilers  of
man's habitat." It is a happy event,  indeed, that
a golden mean is attempted in  an increasing
number  of projects and  research,  and an in-
creasing dialogue  between  technicians  and
social scientists is being established. Hopefully,
the result of such  coming-together will be  an
increasing  sensitivity of  the  technical limita-
tions and the physical constraints in any kind of
a system, and, on the other hand, a recognition
of the  qualitative aspects  of life and of the in-
tangible dimensions involved in any type of a
physical structure in the  midst  of any human
community.
  To sum  up,  then, the essential point  of the
above  lengthy  introduction  we  might under-
line  again  the  need  of both; first, an under-
standing of  the larger  trends  affecting the
development  and future use of water resources
systems  in the  country,  and, second, of the
pronounced need for an integrated framework
that  will bring together technical,  social, eco-
nomic,  political,  and  cultural considerations
in any attempt to solve the complex problems
associated with natural and man-made degrada-
tion of the environment.

    Irrigation In The Western United States
  Irrigation  enterprises are  associated  from
early historical  times with  every civilized  so-
ciety. Today almost 7,000 years after the begin-
ning of irrigated agriculture in Mesopotamia,
irrigation  systems continue  to be  built  and
provide the basis for national wealth and power
for  many  countries. At  the  same time,  the
present  level  of agricultural  engineering  and
organizational skills are better understood in
the context of the larger social and economic
environment. In all  countries,  complex irriga-
tion systems are associated with  larger social
political changes of the economy and culture of
a  given  region. At  the same  time,  not only
irrigation  projects,  but any larger water  re-
sources  development has  been  associated with
efforts for local growth and stability and  re-
gional and  national  impacts  by  providing a
diversified  basis  for choice commodites  and
solutions to the basic problems of community
survival.
  As societies become much  more complex
and  diversified  and   demands   continuously
increase and expand in  scope and  intensity,  the
use of scarce water resources and  the preserva-
tion of  the natural environment become much
more important concerns in the  planning and
evaluation  of water  resource developments. In
any water resource  development three major
problematic situations give rise to  a continuous
re-examination of the parameters  of any water
use system:
   1. Continuously   changing  economic  and
social conditions (such as increasing population,
rapid urbanization, and  industrialization).
  2. Adverse   environmental   circumstances
stemming either from  ecologically fragile  en-
vironments, i.e.  from natural  sources  of eco-
logical pollution, or man-made perturbations,
especially  the  increasing  environmental  dis-
ruption from the discharge of  effluents,  the
fouling  of the air, the misuse of the land, and
the despoliation  of the water supply.
  3. The   strong   presence   of  institutional
constraints, result of long  historical and cul-

-------
288    MANAGING IRRIGATED AGRICULTURE
tural practices embodied in laws and doctrines
and of  traditions  reflecting  the  norms  and
practices of a given society and community.
   Irrigation is practiced  on about  ten per cent
of the total crop land in the United  States, yet
it produces approximately  25 per cent of the
total value of all crop production. As in other
regions of the country,  irrigation  contributes
to  strengthening many  facets of  the region's
economy in that it creates employment oppor-
tunities and provides for a diversified  variety
of choice commodites. The nation has abundant
water resources,  but the distribution and timing
of the  water  resources  differ in  the various
regions of the country, from season to season,
and from year to year. In various studies it has
been repeatedly  indicated  that water would
play a major role for the life and development
of the nation. An essential point, however, made
throughout a  number  of diversified studies is
the central hypothesis and projection that as the
nation grows, the limitation of water and related
land resources available  to competing regions
becomes more important from a national point
of view. The survival  of the  country and the
meeting of major national  goals will be very
much 'dependent on efforts of comprehensive
planning and regional development,  as well as
integrated policies of proper resource develop-
ment.  At the  same time, balanced population
and balanced economic  growth  become essen-
tial  ingredients  for a successful  equation of
future survival.
   Perhaps it is  easier to talk about the West
demographically  since analysis  follows estab-
lished  lines  of State boundaries.  It becomes,
however, much more complicated to  talk about
the West  from a water  development point of
view. The Mountain and Pacific divisions are
comprised of a  number  of regions formed by
natural water runoff districts not  necessarily
coinciding with  present  administrative bound-
aries. It would be a fair generalization, however,
to  say  that perhaps with  the  exxception of
the Columbia-North Pacific region  and some
parts of the California and Missouri regions, the
area of what  is  known  as the Mountain  and
Pacific States  (that is the  States  west  of  and
including  Montana, Wyoming, Colorado,  and
New Mexico) is expected to have  severe water
shortages by the year 2020.
  The Western United  States  has also been
steadily moving from an economy dominated
by  agriculture  and  the  resource  producing
industries of forestry and mining, to an economy
increasingly shaped  by  urban growth,  manu-
facturing,  and  service  and  recreational  in-
dustries.  Although the quantity of the water
being used  by agriculture still seems to increase,
more and more urban  oriented considerations
will  be influencing  the nature  of water  de-
velopment  in the coming years in  this  part of
the nation.
  The anticipated  water shortages in the West
are not only  a  result of an expected reduction
in the overall supply of water, but also a con-
sequence of  the  following  national  and  re-
gional trends.
  1.  Increasing population, especially the con-
tinuous internal migration to western  states.
Indeed while  the population increased by about
14 per cent during the  last decade, 1960-1970,
most  of  the  states in the  west increased sub-
stantially above the  national average.  Table  1
shows the  population changes in the Western
United States  during the  last  three decades,
where, with the exception of Wyoming and New
Mexico, most of the States have been experi-
encing recently  high rates of population growth.
  2.  Another important regional trend is  the
increasing urbanization and metropolitanization
and  the increased demand for  municipal ser-
vices with a resultant conflict between farm and
non-farm water usage. As the urban centers in
the West  continue  to grow,  the  most pro-
nounced picture of population movement in the
Western  United States  is  the increased con-
centration  of population around metropolitan
cores.  The 1970 census has already indicated
that the fastest  growing metropolitan areas and
central cities  in the  nation are located  in  the
western regions. Between  1960 and 1970  the
metropolitan  population of the Pacific division
increased by 4.8 million or 27 per cent while the
metropolitan  population of the Mountain divi-
sion  grew by 1.2 million or 34 per cent. As a
matter of fact,  the proportion of the Mountain
division  population  living  in  metropolitan
areas rose  from 51 per cent in 1960 to  57  per
cent  in 1970 (although this percentage  is still
among the lowest in the nation). In contrast,
the population  of  the Pacific division is more

-------
                                                       SOCIOLOGICAL CONSIDERATIONS

                                          TABLE 1

                Population changes in the Mountain and Pacific states, 1950-1970
                                          289
Region
Mountain
Montana
Idaho
Wyoming
Colorado
New Mexico
Arizona
Utah
Nevada
Pacific
Washington
Oregon
California

1950
5,074,998
591,024
588,637
290,529
1,325,089
681,187
749,587
688,862
160,083
15,114,964
2,378,963
1,521,341
10,586,223
Years
1960
6,855,060
674,767
667,191
330,066
1,753,947
951,023
1,302,161
890,627
285,278
21,198,044
2,853,214
1,768,687
15,717,204

1970
8,283,585
694,409
713,008
332,416
2,207,259
1,016,000
1,772,482
1,059,273
488,738
26,525,774
3,409,169
2,091,385
19,953,134
Per Cent
Change
1950-1960
35.1
14.2
13.3
13.6
32.4
39.6
73.7
29.3
78.2
40.2
19.9
16.3
48.5
Per Cent
Change
1960-1970
20.8
2.9
6.9
.7
25.8
6.8
36.1
18.9
71.3
25.1
19.5
18.2
27.0
highly concentrated in metropolitan areas than
elsewhere in the nation, having  surpassed the
Middle Atlantic division during the last decade.
  With the exception  of Montana, every state
in the West (excluding Alaska and Wyoming
which have no standard metropolitan statistical
areas) had  metropolitan increases of 20 per cent
or more. States with the fastest growing metro-
politan areas in the nation are Nevada, Arizona,
and  Colorado,  the  last with  a  metropolitan
growth rate of 33 per cent, placing the State  in
fourth place nationally. Thus, by late 1960, we
have in the region not only major metropolitan
concentrations, but the beginnings of the emer-
gence  of megalopolitan concentrations such  as
the Front Range megalopolis in Colorado (en-
compassing almost 80 per cent  of the total
population  of  the  State),  the  Wasatch  Front
megalopolis in Utah, and other megalopolitan
formations such as the Sante Fe-Albuquerque
emerging   megalopolis,  the  Phoenix-Tucson
conurbation, and of course the vast strip cities
in California, particularly the  San megalopolis
between San Francisco and San Diego. Accord-
ing to the  1970 population count, neatly 24
million reside in four large urban regions—the
California, Puget Sound, Metropolitan Arizona,
and Colorado urban complexes.
  Similarly,  there seems to be general  agree-
ment  in various studies and projections that the
greatest  percentages of  all future population
growth  will  be in  the  western third  of the
nation. It is expected that the West will increase
its present national share of population from 17
per cent  to 22 per cent by the year 2000. Thus,
in  most  western   states,  three interrelated
trends will be crucial in the solution of emerging
problems of water supply and use: flight from
the countryside and  abandonment  of  small
towns,   increased   metropolitanization   and
urban sprawl,  and  total  population  growth
from  both  natural increases and continuous in-
migration.
  3. Increasing  industrialization in  the  West
which not only effects the total volume of water
use, but also the quality of water supply. Major
industrial concerns  have moved, for example,
in formerly sparsely  industrialized areas of the
Mountain  states such  as IBM  and  Kodak in
Colorado, and Litton and Sperry Rand in Utah.
Needless to say, one can only indicate the con-
tinuous growth of major industrial concerns in
the Pacific  States, especially California.
  4. Another major  trend in the region, as well
as in the rest of the nation,  is the  increasing
concern  with  ecological  perturbations  and
ecological mismanagement and the consequent-
ly increased  requirements  and  costs for  pol-
lution control which  will affect both agricultural
and non-agricultural uses.  This is particularly

-------
290    MANAGING IRRIGATED AGRICULTURE
important in the case of the western states, es-
pecially the Mountain division, where a fragile
ecological  environment and limitation of re-
sources compound problems of  pollution. By
the year 2000, it is estimated that the amount
of water flow needed for pollution abatement in
the West will be approximately 2/3 of all  con-
sumptive uses. In other words, while in the past
the uses to which water could be put, essential-
ly predetermined the  character of the develop-
ment. Now, and increasingly more in the future,
water resources will be employed as a control-
ling commodity for shaping future patterns of
population distribution and industrial location
in the West.
  Both the number of inhabitants and the  spa-
tial dimensions of the future large urban masses
in the  country are truly impressive. In view of
the continuous trends of growth and the prob-
lems of urban sprawl and metropolitan decay, a
national growth policy has been also proposed,
advocating dispersal  of the population away
from the heavily congested  urban regions.  The
West, relatively sparsely populated and despite
its present  rapid growth rates offers many op-
portunities for meeting proposed national policy
(such as, e.g. new towns of various sizes). Re-
sources, however,  and  the relatively fragile
ecological  environment  of many states   are
restrictive  conditions  as   to  the  maximum
volume of  future  inhabitants,  the  choice  of
future  locations, and the  flexibility in the  use
of such limited commodities as water.
  From the above four trends, it becomes ap-
parent  that new, different,  and  expanded  de-
mands for  water will be generated in the West-
ern United States. However, the  urgency for a
comprehensive   water  development  policy
depends not  only on past and present trends of
population increase, urbanization, industrializa-
tion, and ecological awareness, but also on other
factors  complicating the physical  and  technolo-
gical aspects of water resources  planning and
use.  A major problem is the numerous govern-
mental jurisdictions, each with specific respon-
sibilities for  water conservation  and  manage-
ment. Contrasted to the usually unified govern-
ment unit managing water  supplies in most of
the nations in the world,  the American federal
system  divides  power between national (Fed-
eral government) and the  States.  The last dele-
gate powers to several types of local authorities,
including counties,  cities, districts, and special
administrative units. Thus, upon the numerous
river basins of the  Western United States and
in addition  to the  national government repre-
sented  by 17 states, there are to be found over
14,000  units of local government,  all with var-
ious responsibilities for determining the alloca-
tions of water  for  specific uses. At  the same
time, segments of  private enterprise  vie with
publicly owned and operated  enterprises, such
as,  for  example, in the case of hydroelectric
power generation.
  When we look at the problem of water man-
agement from a macropoint  of view,  we have
also to  acknowledge the larger difficulty where
each State is obligated under either a compact,
a court decree, or  a judicial allocation  to ap-
portion  the  water  of  the  interstate stream
between States. Two problems are immediately
associated with this fact, one of quantity and
the other of quality. Thus, although liability for
water  distribution  rests  with  the particular
State, there is no effective or efficient mecha-
nism for proper transfer of this obligation to
the people  using the water. At the same time,
through the use (or misuse) of his water rights,
the individual holder is effecting the quality of
the stream,  either through misapplication (re-
sulting in salt concentration), or by taking the
water out of the system and thus maintaining
(and in some  cases  increasing)  natural salt
pick-up.  As water supplies become more fully
utilized,  the importance of irrigation  return
flow quality will be of even greater significance
in the over-all water management and develop-
ment in a green basin.
  What we have, then, in the West is a compli-
cated  system of demographic,  administrative,
and natural water districts which, in addition to
natural overlaps, create a multitude  of prob-
lems in jurisdiction  and use. At the same time,
given the open character of the water systems,
we have major ties of every kind of a major
basin   with  surrounding  water systems.  For
example, the Colorado Basin  is  not a  self-
contained system but has many ties  with sur-
rounding areas, such as the provision of water
from the Colorado  to the Great Basin through
the Central Utah  Project,  water from the
Colorado Basin to  the Missouri basin through,

-------
                                                       SOCIOLOGICAL CONSIDERATIONS
                                          291
for  example,  the  Denver  water withdrawal
system,  the  Fryingpan-Arkansas  transfer  of
water, the San-Juan Chuma transfer of water to
he Rio Grande basin, the transfer of water to
the  Central  Arizona  project,  and  finally the
major transfer  of  water  from  the  Colorado
River  to  the  Southern  California  project
(MWD). Figure 1  exemplifies the major  river
basins in  the  Western  United  States,  which
show the overlapping character of physical and
administrative  units  as  exemplified  in  State
boundaries. When  we think  of  problems  of
   Figure 1: River Basins of the Western States
water management,  we have to keep in mind
the existence of a myriad  of systems and sub-
systems, each only relatively  autonomous, and
open since they have a wide variety of linkages
on  different  levels.  In other words, a  specific
irrigation company is usually part of a federa-
tion of  irrigation systems within a subsystem of
a given  basin and part of larger inter-basin ex-
changes.
  The   previous  general  discussion  has  at-
tempted to show the overall trends affecting
water use in the West. This part of the country
will also continue to have continuous (although
not highly increasing) water demands  for ir-
rigated  agriculture.  As Table 2 indicates,  the
western region  of the United States will experi-
ence moderate increase  in agricultural  irriga-
tion between  1980  to  2020.  Various  other
studies  have  shown  similar trends in the slow
rate of increase in irrigation  water use. A most
interesting estimate of water use and projected
requirements by region is  that included  in  the
composite Table 3, based on projections of the
Water Council. As contrasted to  other  regions
of the nation, most of the  western regions pre-
sent us  with stationary trends of projected ir-
rigated land use.
  Irrigation, just as other of man's activities,
can have detrimental  effects on  the environ-
ment.  As indicated before,  although it has long
                                          TABLE 2

                Projected estimates of agricultural irrigation in the Western Regions
                       of the United States, 1980-2020 (thousands of acres)
                           Region
      1980
2000
2020
Souris — Red — Rainy
Missouri Basin
Arkansas — White — Red
Texas Gulf
Rio Grande
Upper Colorado
Lower Colorado
Great Basin
Columbia — North Pacific
California
TOTAL — WESTERN REGIONS
MAINLAND UNITED STATES
90
8,050
5,600
6,510
2,050
1,900
1,820
2,340
7,350
9,050
44,760
49,990
230
8,950
6,400
7,350
2,180
2,150
2,190
2,510
7,810
9,600
49,370
56,910
250
9,600
6,690
7,770
2,200
2,250
2,400
2,570
8,490
11,540
53,760
62,890

-------
292    MANAGING IRRIGATED AGRICULTURE

                                          TABLE 3

Regional projections of population and irrigated land in the conterminous United States, 1960-2020
                                   Population
                                         Thousands of Acres
         Region
 1960   1980   2000   2020
1960   1965   1980   2000   2020
North Atlantic
South Atlantic-Gulf
Great Lakes
Ohio
Tennessee
Upper Mississippi
Lower Mississippi
Souris-Red-Rainy
Missouri
Arkansas-White-Red
Texas Gulf
Rio Grande
Upper Colorado
Lower Colorado
Great Basin
Columbia-North Pacific
California
43,896
19,727
25,474
18,793
2,979
11,759
4,619
652
7,845
7,122
8,109
1,604
317
480
970
5,359
15,584
56,693
29,099
33,171
23,498
4,118
15,180
5,871
791
10,337
8,972
12,491
2,649
454
3,038
1,790
7,581
28,167
74,993
42,602
43,293
30,742
5,643
20,004
7,815
1,023
14,260
11,952
18,230
4,173
700
4,768
2,822
10,463
43,317
101,726
61,438
57,640
41,241
7,785
26,766
10,587
1,368
20,079
16,055
25,901
6,063
1,025
7,194
4,285
14,444
64,003
240
850
100
35
15
80
700
10
6,600
3,100
5,100
1,950
1,370
1,520
1,700
5,450
8,420
310
1,500
140
55
20
140
900
15
7,400
3,800
5,500
2,000
1,440
1,660
1,860
6,250
8,850
380
1,800
230
90
30
210
2,100
90
8,050
5,600
5,500
2,050
1,800
1,750
1,950
7,700
10,150
550
2,750
350
180
40
390
3,050
240
9,000
6,400
5,500
2,050
2,000
1,800
2,000
9,500
10,750
700
3,750
470
260
50
550
4,150
250
9,600
6,850
5,500
2,050
2,000
1,800
2,000
11,200
11,100
                Total.
176,289243,900336,800467,600     37,240 41,840 49,480 56,550  62,280
been  recognized  that  the  quality  of water
draining from irrigated areas  has consistently
been degraded, it is only recently that attention
has been given to the possibility of controlling
or alleviating  the degraded quality of our water
resources.  This interest has been accentuated
not only  by  recent general concern with en-
vironmental mismanagement,  but also  by in-
creasing federal  legislation  addressing  itself
to the establishment of a national  policy for the
prevention, control,  and  abatement of water
pollution.
  The water  quality problems associated  with
irrigation return flow are of particular concern
because irrigated  agriculture, especially in the
West, is the largest consumer of public water
supplies, and  it is likely to be so in the foresee-
able future. As a matter of fact in federal state-
ments  questions  have been raised  regarding
the need for  the  various  forms of federal as-
sistance in relation to the outlook for national
food  and  fiber  requirements.  A number  of
authorities, private as well  as  federal,  have
recently raised serious  questions as to the per-
manence  of  economically  feasible  irrigation
                      agriculture in the West in  view of the  rapidly
                      increasing demands on this fixed resource.
                         Perhaps we need to  elaborate a little bit on
                      the problems of water quality  associated with
                      irrigation in the Western United States. One of
                      the things we need to bring  forward is an essen-
                      tial question and a fundamental reassessment
                      which begins to emerge concerning the need for
                      future water  projects  in the  seventeen states
                      west of the 100th meridian. Indeed, the constant
                      hunt  for  water  is  coming under  increased
                      criticism, especially for all these  big projects
                      designed to bring more of it to arid lands. A
                      question that has been repeatedly  raised runs
                      as follows: "Should the government pay farm-
                      ers not  to till the soil in States with high rainfall
                      while it  subsidizes  farm irrigation  in  States
                      with low rainfall?"
                         Thus, it appears that there are two inter-
                      related  problems: first, the question of water
                      supply which is associated also with the present
                      wasteful use  of limited water; and,  second,  the
                      economic impact of increased natural, as well as
                      man-made pollution. According to EPA, BLM,
                      and CRBC estimates, man in his works is al-

-------
                                                       SOCIOLOGICAL CONSIDERATIONS
                                          293
ready significantly  increasing the salt  load  in
the Colorado River's natural salinity. As a gen-
eral rule, in salinity concentrations above 500
mg/1, the value of water begins to diminish not
only because of increased costs in water soften-
ing, corrosion, etc., but also because of the need
for greater amounts of leaching water  and the
damages incurred from diminished  crop  yields
or the inability to grow certain high-value crops.
it  is now estimated that the bill for this salt
reaches  $16 million a year and it is expected  to
reach $28 million by 1980 and  $51 million  by
the year 2010, unless the salt loads are  reduced
substantially. At  the same time, it should  be
noticed  that the major efforts concerning return
flow quality problems are directed at control  of
the source, rather than treatment and reclama-
tion of degraded water.
   The natural problems of salinity are accentu-
ated by  the larger trends  of growth described
above. So as man  has been and will be taking
water out of the river and its tributaries, as well
as  darning the streams and polluting  existing
water supplies, water used  for  irrigation will
not only be picking the salts in the land, but
will be  increasing the salt  levels because  of the
above conditions. The vast natural evaporation,
municipal use,  and the malpractices in the use
of water based  on the convenience of the irriga-
tor and  protection of his  water right, are not
only diminishing water supplies and increasing
natural  water  salinity, but  they are also ex-
celerating problems of man-created pollution in
the water systems of the  West. Many federal
studies  have been pointing out that in  order
to maintain  salinity levels  (700  parts  per mil-
lion of dissolved  solids  is increasingly used  as
a  possible standard),  water use will  probably
have to  be curtailed and steps taken to remove
the salts.
   Given the natural problems of water salinity,
the increasing  demands  for  water  and  the
parallel  trends  of population growth, urbaniza-
tion, and industrialization and increasing water
quality requirements, we may have also increas-
ing conflicts in water use.  On the one  hand,
water quality  is assuming greater  importance
due to  the  pressures  of  population  growth,
municipal expansion and competition among a
wide variety of  uses  of  this limited  resource.
On the  other hand, with each water use there
are also associated quality considerations per-
taining to both the water extracted from and
that returned to the source. A simple diagram
(Figure 2)  summarizes in  a graphic way  the
problematic situation giving rise  to a  water
management  system,  where a  configuration
of  structures and procedures transforms  the
water resources into water-related products and
services, and where  the  desired outputs have
both use,  as well as quantity and  quality limita-
tions.
  From  this  simple diagram describing the
essential parameters  of  a water management
system, we  would  like to  generate a discussion
on the social aspects involved in water quality
management projects.  Many other  individuals
have been addressing themselves to the natural
and  technical dimensions of the  problem  at
hand. It has been  contended, however, that we
have been approaching the need for water qual-
ity  management   with  little  preparation  and
knowledge  of  either policy considerations  or
technology for improving the quality of irriga-
tion return flow. Thus, it becomes apparent that
in order to  deal  effectively  with problems  of
water quality management,  consideration also
has to be given to the following major  ques-
tions:

  1. What  should  be  the future  of  water
development projects for new irrigation in the
Western United States?
  2. What are the larger considerations for  an
understanding  for agriculture in  a changing
society?
  3. How do we  account for the  relationship
between man, technology, society, and environ-
ment.
  4. How can we make some tentative recom-
mendations  concerning  the social  aspects  of
water  quality  management especially  in  the
Western United States?
  The  last  point brings again  forward  the
central thesis that  in order to speak of the social
environment of the Western United States, one
should have to consider a complex combination
of  normative  resources,  community environ-
ments,  cultural traditions,  water management
systems,  sources  of  social conflict, and their
interconnection with limited physical resources.

-------
294
MANAGING IRRIGATED AGRICULTURE
 Water Resources
   Problems
      or
     Needs
                               Water Resource
                              Management System
                     Inputs
                                 Facilities
                                 Personnel
                                 Procedures
                                         Outputs
                                                             Products-,
                                                            Services-l
                                                                     Uses and Goals
                                                                     ( Municipal, Irrigation
                                                                      Recreation, etc)
                                                                    Characterstics and
                                                                       Limitations
                                                                    ( Quantity, Quality,
                                                                       Time, Place)
         Figure 2: A Simplified Version of Water Resources Problems and of Water Management
      Towards A Systematic Viewpoint of
    Man-Made And Natural Environments
  Sociologists  traditionally  have  always  in-
cluded  the environment in  their  theoretical
frameworks,  but the interconnection  between
physical and social environments has not always
been  clearly stated.  In order to  be  able to
examine any kind of a physical  system, it  be-
comes imperative to provide  a much broader
view  of natural resources, and a more careful
analysis and  delineation of  individual  and
aggregate levels involved in the presentation of
a system. This is particularly true in irrigated
agriculture which has  to be  examined within
a complex socio-technical  framework  that is
operating  with  varying degrees  of success in
relationship to  an  encompassing  natural  en-
vironment.
  Increasingly  in current years, many of  the
basic decisions  related to water development
are made not only with technical  considerations
in mind, but at the  same time increasingly call
for a better  knowledge of the political,  legal,
and social economic institutions  which control
the supply and allocation of water. Yet, in any
water development project, we have a number
of unknown  processes  providing us with com-
plex interactions between physical, economic,
                                         political and social factors. This complex whole
                                         does not permit accurate prediction and mea-
                                         surement of the consequences of any proposed
                                         water development action, or even a  very  ac-
                                         curate argumentation either in favor or against
                                         large-scale  water  policies.  The above simply
                                         indicate that the limited water supply, the in-
                                         creasing  population,  the problems  of water
                                         quality, and the multiplicity of uses call for new
                                         integrated forms of coordination and under-
                                         standing  between  various disciplines  affecting
                                         and being affected by a given water  develop-
                                         ment project.
                                           Thus, it has become apparent that the prob-
                                         lems of  water quality are  very much inter-
                                         connected in an  intricate pattern with not only
                                         problems of overall supply,  but are part of a
                                         general  integrated policy  of  regional  and
                                         national welfare. A policy of improved water
                                         management, in terms of both quantity  and
                                         quality, requires a coordinated effort towards an
                                         efficient and effective maximization  of limited
                                         natural  resources.  Therefore,  the  problem of
                                         water quality management is not only one in-
                                         volving the very  careful considerations of physi-
                                         cal potentialities, but also a clear and thorough
                                         understanding of  problems of  political, eco-
                                         nomic, legal, and social feasibility. As  a matter

-------
of  fact,  from  the sociological point of view
part  of  this  problem  involves  also  a delin-
eation  of  the  organizational  capability   of
present water  management systems  for new
alternatives and  new organizational forms, an
understanding and utilization of a social climate
of  receptivity  towards  new cultural  practices
and towards  changing  social  and  economic
circumstances.
  As it has been repeatedly emphasized before,
because of large  national trends and new  or
expanded demands while the overall supply and
quality of water are vital in any future planning
of resource utilization, equally important from
the sociological point of view are the following
major considerations:
  1. Organizational   innovations   applied   to
increased efficiency in the distribution of water.
  2. New   cultural   practices  and   attitudes
towards innovative forms of water use and an
understanding of the demands for efficient use.
  3. An understanding  of  the broader  com-
munity culture and  of the  institutional struc-
tures involved  in obtaining  water supplies,  the
forms  of distribution, and the meeting of larger
society goals.
  4. The recognition, at the same time, that as
pressures  for  existing supplies increase,  there
is also stronger concern for proper social control
against water quality deterioration.
  To  be  able  to develop  a more systematic
framework  for a discussion  of  social' aspects
of water quality  management, we  have to  see
an  irrigation system  within its larger context.
Figure  3  tries  to exemplify  the  relationship
between  nature,  society, and  man,  and  the
central role  of  technological developments.
This diagram simply indicates the close inter-
connection  of  the major components of  the
proposed  system which  operate  within  a bio-
sphere. Thus,  each  one  of these component
parts—man, society, nature, and technology—
affect, and  in  their  turn are affected by each
other,  and a change  in any of these factors will
have  consequences  and  alternatives  for  the
others.  In the  case  of  water quality  manage-
ment,   technical  breakthroughs   in   salinity
control will affect its component  part of  the
biosphere,  and so will also structural changes
in  society  (e.g.  ubanization),  new  cultural
practices, ecological shifts, etc.
      SOCIOLOGICAL CONSIDERATIONS     295

                    Biosphere
 Figure 3: Relationship Man, Society, Nature, and
 Technology

  Such  general  remarks and  impressionistic
observations  help   us   introduce  a  systems
analysis  point  of view.  Water management is
perceived as  a  system  operating  in  a given
environment where inputs (physical and social)
processed through the "organization"  result in
outputs or goals  established for the functioning
of the system.  The systems  approach  makes it
possible  to identify not only component parts,
but also various conditions which enable  the
parts  of an  irrigation system to function "ef-
fectively" in relation to each other  and  "ef-
ficiently" towards the achievement  of desired
goals.
  The most important element is to describe the
kinds of environments in which a given water
management system operates. To start with,
there  are two major environments in which sys-
tems or  subsystems operate.  First, the external
environment which is  both  natural and man-
made, and  second,  the internal  environment
which encompasses  all  subsystems operating
primarily inside the  boundaries of the external
environment. Simply, a  particular system im-
plies a collection of people,  devices, and proce-
dures intended to perform  some function.  A
systems model is a working model of a social
model which is capable of achieving a  goal and
involves  the systematic exploration,  analysis
and evaluation of all the possible consequences

-------
296
MANAGING IRRIGATED AGRICULTURE
of proposed alternatives to an on-going system.
More specifically, an irrigation system may be
designed in such a way as to be able to include
a variety of environments  and inputs and be
defined as the application of water by human
agency to assist the growth  of crops and grass.
Figure 4 is a simplified version of an irrigation
system  designed  to  achieve  maximum agri-
cultural productivity.
  Figure 4 indicates  that four major environ-
ments,  i.e.,  the  physical  environment with
particular ecological  constraints, the existing
socio-demographic conditions  and  spatial  ar-
rangements,  the economic  potentialities,  and
the normative milieu (which includes  not only
norms,  values and institutions, but also legal
constraints and  cultural resources) provide  the
necessary inputs for the operating  of  a system
or organization. In addition, there are additional
inputs into an irrigation subsystem, arising from
constraints and/or  facilitators of  the  larger
system  or  external environment, such as  the
state of technology and technical innovations,
the larger  political network and the linkages
with other administrative units,  and  the con-
straints  from  larger resource  policies  affecting
                                          the operation of the irrigation subsystem. Per-
                                          haps, from the sociological point of view, the
                                          most important inputs are  those  patterns of
                                          normative behavior which influence water  use.
                                          The most crystallized part  of the normative
                                          environment, i.e. the legal norms and the legal
                                          interpretation of institutionalized arrangements
                                          are among the crucial elements for the success-
                                          ful operation of an irrigation system.
                                            The inputs from a variety of environments
                                          are  processed  through   structures  and  pro-
                                          cedures  which  attempt  to  maximize desired
                                          goals.  This  organizational  system varying in
                                          size, scope,  integration,  and  complexity from
                                          region to  region, and from basin to  basin in-
                                          cludes  both physical facilities  developed  for
                                          meeting the  need  of increased productivity,
                                          etc., and  the  intangible  aspects of  organiza-
                                          tional infrastructure, such as rules of operation,
                                          pattern of leadership and  command, efforts for
                                          control, integration,  information, and  com-
                                          munication and ways of interacting with other
                                          organizational environments.
                                            The  desired  goals  or  objectives  in an  ir-
                                          rigation system  denote  the output  part  and
                                          revolve around  a variety of goods and services


Constraints / Needs Feedbac
DnPuts] (More C

(Population,
Distribution)
/
Economic
(Capital Resource +—f
Allocation)
\.
\
Physical
'*-* (Natural Resources, —
Ecological Limitations)
/
Normative
(Values, Institutions,
Laws)
k and Changes in Inputs
apitol. Population •
New Values, etc.)
Processing Mechanisms
[Thruput, " System"]
1 Institutionalized Organization
i
-*\ Physical Infrastructures
1 1
1 Rules of Operation
I
I
i i
Fee

Desired Goals
[Output]
— T

Objectives:
-Increased Prod
	 » -Enhancement o

-Individual, Coll




activity
f Land
t
ective Betterment

dback to the Organization
(Participation, Morale)

                                 Additional Constraints from External Environment
                                 -Technology
                                 -Political Network and Administrative Apparatus
                                 -State, Regional, Federal Policies
                     Figure 4:  A Simplified Version of a Local Irrigation System

-------
                                                        SOCIOLOGICAL CONSIDERATIONS
                                                      297
for  individuals,  communities,   and  regions.
Recently, in addition to such obvious objectives
as  increased  productivity,  wealth,  land  en-
hancement, security, etc., increased  mention is
made of such qualitative goals as enhancement
of quality of life, esthetic satisfaction, regional
balance and others, which provide us with in-
finitely  more  difficult  problems of evaluating
the  performance  and  effectiveness of  an ir-
rigation system.
  System or organization and inputs from di-
versified environment are linked with the out-
put of an irrigation system, through two feed-
back loops. One,  provides  feedback  to  the
"organization"    or  processing  mechanisms
through  increased participation of  individuals
organizational  morale and direct  interaction
between,  e.g. users and/or  officials and  mem-
bers  of  the organizational units.  The  other
feedback  loop  is  understood in terms of addi-
tional inputs through  increased  or achieved
objectives, such  as e.g. more capital invested,
increased  population,  and changed attitudes
or norms vis-a-vis any of the four environments
(and other external constraints) which provided
the  initial  input.  Ideally,  feedback processes
could serve as corrective  mechanism, monitor-
              other Sources
              Precipitation, Recycled
              Cloud-Seeding, Desolino
              tion, Non-Irrigated
              Surf ace Runoff, etc.
Water
Control
            ing devices, and additional inputs which could
            increase the  organizational  effectiveness  and,
            thus the achievement of desired and expanding
            objectives.
              While Figure 4 shows the overall structure of
            an irrigation system and the more or less static
            consideration in terms of the constraints of the
            natural and socio-cultural environments, Figure
            5  attempts to provide an overview of the dy-
            namic aspects in the operation of a given irriga-
            tion  system.  An irrigation  system's functions
            can be broken  into  the following major areas,
            each  of which requires a vast array of not only
            organizational  structures,  but  also  complex
            rules  and  procedures  and  critical considera-
            tions of decision-making and policy guidelines:
              1.  Water supply and water source considera-
            tions, including new  or potential  sources  of
            supply.
              2.  Water control  aspects and characteristics
            of diversion, storage, reservoirs, and wells, and
            the assorted institutional forms of regulation.
              3.  Water distribution systems, the means of
            transmission and patterns of water flow.
              4.  Water utilization,  the system of irrigation
            and crop operations, (including crop and leach-
Water Distribution
 (Canals, Laterals,
   Distributions)
                  Figure 5:  Irrigation Water System Functions and Dynamic Processes

-------
 298     MANAGING IRRIGATED AGRICULTURE
 ing  requirements), as well as cultural practices
 and scheduling programs.
   5. Water reclamation and aspects of drain-
 age, including  field outlets,  release of poor
 water, and irrigation return flow.
   Both figures  4 and 5  try  to emphasize the
 multiplicity of levels of analysis and  the multi-
 plicity  of functions in an irrigation system.
 Most important, at each level and for each sub-
 system  component part and function, problems
 of institutional order arise, difficulties  of or-
 ganizational arrangements, and  need for spe-
 cific understanding  of the  normative rules in-
 volved at each stage or phase of a dynamically
 operating irrigation system.

            Water Quality Control:
          Preliminary Considerations
   The  general  argument  and  preoccupation
 with the systems approach is part of an  effort
 of  integrating  physical  and non-physical di-
 mensions in irrigation  systems. Such an inte-
 gration implies  that any change in either phys-
 ical  or social  dimensions  of  our system will
 have impacts on each other. If we  perceive,
 then, that  the  sociological examination of a
 water management system  or  an  irrigation
 system  involves a careful delineation  of inputs,
 organization, and output, we may be able then
 to examine the  problem of water quality man-
 agement as one  affecting all parts of the system.
 If the problem under consideration  is, let us
 say,  increased salinity  of a  river as a  result
 of both natural and  man-made  pollutions it
 will be easier to examine its effects, and con-
 sequences as a result of three types of changes
 in an irrigation system:

   1. Changes  in  the  external  environment
 which  lead  to  changes  in  inputs,  such as
 changes in the total number of people, changes
 in the economy, changes in technology, etc.
   2. Changes in organizational structures and
 procedures  because of changes in size, capa-
 cities, and technology, different roles or organi-
 zational power, etc.
   3. Changes in  output  or  what   has  been
described as  "goal alterations",  which result
from goal displacement, new targets of society,
and  new expressed  or latent policies of federal,
state, or local authorities.
   It has been  repeatedly stated  that the prac-
tice of agriculture  may have detrimental  ef-
fects on the environment. Presently, the major
irrigation return flow quality problem  areas in
the West are the San Joaquin Valley, the Colo-
rado  River Basin, and  the Rio  Grande River
Basin. In addition to these areas, many other
areas in such States as Nevada, California, and
Utah have serious problems and are affected by
irrigation return flow.
   The research concerning such areas  and the
suggestions made  on potential   solutions  for
controlling irrigation return flow, seem to em-
hasize  that a basic attack in minimizing quality
problems is the improvement of water manage-
ment  practices on  the  affected  crop lands.
Physical suggestions  to  the  solution  of this
problem and improved  practices that can  be
used on the farm include a diversified array of
technological solutions. At the same time, there
is wide recognition  that there are  various in-
stitutional methods which can be used to control
irrigation return flow quality.
   If we assume,  for the  time being, that there
are solutions to the technical problems imposed
by the return flow hydrologic cycle, we need to
see from the social point of view what is the so-
called  institutional problem in improving water
quality  management.   Essentially,  the  key
problem seems to be a simple one, that is, how
to minimize water passing below the root zone
or how to  account for deep percolation losses.
This key problem, and the fact that more salts
are dissolved in the water, is not equally pre-
valent  in every geographical area in the West.
Not only every  area  is  differentially  affected
from quality problems arising from irrigation
return  flows, but also the historical establish-
ment of an irrigation project has  something to
do with the overall problem of salinity. In other
words, certain  areas of the West  not only had
less salt to start with, but over time their salts
were washed out. The areas of concern today
are then those areas  of high natural salts whose
problems  of high salinity are accentuated by
irrigation practices which allow the additional
solution of salts in the deep percolating water.
It seems that a  dilemma appears in all these
areas  of concern, that is, that although some
water  is needed  to  maintain  agricultural pro-
ductivity (leaching requirements)  people quite

-------
                                                       SOCIOLOGICAL CONSIDERATIONS     299
often over-irrigate and, therefore, permit more
deep  percolation  losses.  Essentially, then, the
problem of water quality management is  one
of  exceeding  a satisfactory  level of  leaching
requirements.
  The question of how these leaching require-
ments are to be met is answered by proper ap-
plication efficiency which involves  a two-fold
consideration,  i.e., leaching requirements  and
crop  requirements. To be able then  to have
proper  application efficiency we  need essen-
tially  two major things.
  1. Find the exact technical specifications for
application  efficiency. This  is  a technological
problem to which increased concern  and re-
search is addressing itself presently.
  2. Answer the  question of  how  to imple-
ment  technology, which in turn is based upon
three  further considerations: a) the implementa-
tion of  technology in  accordance with existing
water rights; b) implementation of technology
through organizational innovations, i.e.,  with
improvement in size of the units that are cap-
able presently, and increasingly  so in the future,
to meet both the economic and organizational
demands  of  expanded   efficiency;  c)  imple-
mentation  of technology in accordance  with
changing  attitudes of people,  i.e.,  not  only
changing  attitudes  in relation  to  irrigation
practices,  but  also  changing  attitudes  with
general  patterns of water use in a given region.
  It  becomes  apparent,  therefore,  that  the
western salinity problem  is very closely tied to
the major considerations of population, urbani-
zation,  industrialization,  and ecological trends
mentioned  at  the  beginning of the paper. The
belated  concern with  increasd  salinity in  the
Colorado  River has  come about as a result of
the  fact that as we take  more  water from the
basin, we leave less water for application ef-
ficiency. The   less  the  water,  and  with  the
volume  of- salts remaining  more or  less  the
same, any withdrawal from a given river basin
is bound  to raise the concentrations of salts.
Thus, the agricultural problems have  become
the focus not  only  because  of outright agri-
cultural damages, but also because the problem
has become a larger one involving a more com-
plex equation  of a multiplicity of water uses.
If salinity criteria  are  established, their imple-
mentation would  conflict with existing water
rights,  interstate  compacts,  even with  inter-
national  agreements  concerning the  division
and allocation of any given basin's water re-
sources.   In  addition, long-established  water
rights with no consideration of improvement in
the  salinity  level,  may  preclude any future
development  of  irrigated  land  and  impose
restrictions  which  are  totally  unaccepted  in
present practices. All in all, however,  inde-
pendent  of the establishment of any salinity
criteria,  it  is again  concerted  public policy,
comprehensive planning and larger social con-
siderations that will ultimately  determine new
water  uses, rather than  piecemeal restrictions
or efforts for post hoc enforcement in both the
quality and quantity aspects  of  irrigated agri-
culture.
  As it has been  emphasized above,  a crucial
problem in any water management system are
the types  of incentives and  the structure of the
organizational  arrangement that may permit
irrigators  to control irrigation return flow. In-
centives for efficient management usually come
in the form of economic incentives, either nega-
tive or positive.  At the same time, the larger
law regulations contribute substantially to both
the creation and the solution of irrigation return
flow  quality  problems. Indeed,  in many in-
stances, there  is no incentive to  conserve  water
in most of the irrigated valleys in the West. A
key problem  is that  most  irrigators  feel that
they must use  their full water right because they
are afraid of losing  any portion  of the unused
right. Despite repeated observations and find-
ings that such  attitudes of  excessive use  of
water  right   frequently  contribute  to  local
drainage  problems,  the  practice persists  be-
cause it is rooted in deep seated  personal fears
as to water use  and on the  notion that the ex-
ercise  of  the  right means the preservation  of
the right. A paradox then seems to emerge, i.e.,
that efficient  farmers who  through improved
technological practices save  water, are not able
at the same time to  use the  conserved  or  saved
water to irrigate additional land or to supple-
ment their water  supply for lands  having an
inadequate water right. A farmer who  has used
inefficient and many times  flooding techniques
has a built-in advantage and, thus, any incentive
from the legal  point of view is diminished by the
realities of the persistent attitudes in the use of

-------
300    MANAGING IRRIGATED AGRICULTURE
present water rights. This  simply  compounds
any effort for improving his water management
practices.
  It should  be  emphasized that although the
law per  se may be satisfactory, the difficulties
arise mostly  from its interpretation. Provision,
therefore, should be made in the law, as well as
in everyday  practice, of  building  incentives,
both economic and legal that  would make pos-
sible water saved by a given farmer to be sold or
rented to a farmer having an  inadequate water
supply. Such an attempt points also to the fact
that  water  supply  should  be redistributed
within an irrigated valley in an efficient way to
meet total water agricultural demands  rather
than be based on capricious  individual farmer
requirements. The distribution of existing water
supplies has to be met not only in tighter regula-
tory procedures, but also in  a more efficient
organizational  structure that  would permit
better  monitoring  and  budgeting of  available
water.
  At the same time, safeguards need to be built
in any interpretation of the effort towards water
saving, so that not only plans  for a comprehen-
sive water resources development  for a  given
basin can be made, but also  inordinately high
prices could be avoided.
  Improving  irrigation  management  implies
not only technological  innovations and better
interpretation of water law, but  also  other
organizational improvements. Such organiza-
tional   improvements  include,  for  example,
better irrigation scheduling which is very closely
associated with  the present negative aspects of
western  water laws. When we combine better
information  regarding  soil moisture  manage-
ment in the  root zone  of the crops and  more
sersitive  equations  concerning   application
efficiency criteria,  the  farmer may still wish
to use his full water right. It is assumed, at this
point,  that various positive incentives and re-
sulting changing attitudes will allow saving of
water, and transfer of saved water to other uses
of  a comprehensive water resource  develop-
ment plan of an irrigation basin.
  A major problem in the overall management
efforts is the lack  of single management units
in most of the irrigated valleys of the West.
Most of the valleys in the western United States
are characterized by the existence  of quite a
number  of fragmented irrigation companies
and  districts,  with each company responsible
for water delivery to  only a part of the valley.
In most  cases,  separate  institutions  exist to
handle the water drainage system.  In  order to
develop not only effective irrigation return flow
quality control programs, but also to be able to
improve the overall efficiency of a water system
in a  given irrigated valley, it is imperative that
the entire  irrigated system  should be coordi-
nated  on  an  integrated  basis.  We  need to
move from private,  single company  oriented
management units to  what we may describe as
primarily  valley-wide  alternatives.  This  brings
forward a very difficult, but challenging  water
management problem, that is, an  attempt to-
wards  consolidating  separate small irrigation
companies into   single entities  which  would
have advantages  of size,  economies   of  scale,
and  potential for comprehensive  valley-wide
development. Water quality degradation will be
reduced- if attempts are made to face the  prob-
lem not only as an integrated approach of vari-
ous units within  a valley,  but  also as a  much
larger  effort to  integrate the particular  basin
with other surrounding systems. We are speak-
ing, therefore, here not only for comprehensive
systems on a small scale within a given valley,
but for comprehensive larger  entities encom-
passing a  number of  basins,  a collection of
systems, and hopefully the larger region of na-
tural water flow.  In other words, we are  speak-
ing for  inter-valley coordination,  as well as
inter-basin and inter-region coordination.
   Many  other suggestions  have been offered
concerning the solution of problems of irrigation
return  flow quality control. Essentially, most of
them can  be  summarized  around  a cluster of
individual and community attitudes which seem
to have consequences for the quality of a given
water  system  supply.  There is agreement that
there is  a host  of social practices, traditional
ways of irrigating and using the water, as well
as modern practices which directly degrade the
quality of a  given water  supply system. The
immediate solution and the attack against such
practices  can  be  easily perceived  as  a  simple
problem  of social control. This  means that
perhaps  one of the immediate attacks to  this
problem  is to require that  anyone degrading
the  quality  of  water should   pay   the cost

-------
                                                       SOCIOLOGICAL CONSIDERATIONS
                                          301
of treating this water. However, this approach
treats the  problem of  the  degradation of the
quality of water only from a symptoms point of
view  rather  than addressing itself to the es-
sential cause. It is not only difficult  to assess
penalties, but also to  provide a very complex
and expanded system of policing and mechan-
isms  of  law  enforcement.  The  attack to the
problem  would be to try to reach the roots or the
causes of the water quality degradation, that is,
the kinds of social practices  and the types of
attitudes which would improve water use before
degradation  of water  takes  place. From the
social point of view, in changing practices, and
provided that there are also technological  solu-
tions to both natural and man-made pollution of
water, we should be concerned with the follow-
ing major areas:
  1.  Priority of use (and the interpretation of
     legal doctrine).
  2.  Geographic area (and  the increasing scope
     of planning).
  3.  Population affected.
  4.  Political units involved.
  5.  Disciplinary  scope (and  the attempt to-
     wards a multidisciplinary synthesis).
  As it was  pointed out above to bring about
changes  in the organizational behavior of all
types of  units involved in water management, as
well   as   effective response  from  individual
irrigators,  three  major  categories  of policy
decisions and social action  must be made:  first,
strong incentives for efficient or new uses  (eco-
nomic benefits, redefinition of the doctrine of
beneficial use,  etc.); second, structural changes
(such  as  new  organizational  arrangements,
creation  of inter- and  intro-state agencies, ap-
pelate bodies,  water brokerage—either private
or public, etc.); and, third, "regulatory counter-
incentives"   (such  as  stricter  enforcement,
pricing   policies,  etc.).  More  than  anything
else,  however",  all the above changes or at-
tempts for modification must be  guided  by a
pervasive spirit of social consciousness and a
new  world outlook of individuals and collectiv-
ities  away from their  small  closed  system of
their  particular community, to the larger and
much more complex regional scene.
  To be  able to give answers to the above major
areas of  concern, we need to develop an assess-
ment methodology which would  enable us to
identify in a  systematic  way potential causes
and corresponding effects, a description of their
characteristics, and  possible  consequences in
the overall water system. The following diagram
of Figure 6 attempts to show the steps involved
in a systematic  methodology  utilizing a multi-
disciplinary model of assessment and of alterna-
tive plans of evaluation.
  In trying to determine effective water man-
agement, we need to keep in mind that alterna-
tives offered  should be  evaluated  under three
different conditions  of "effectiveness." Tradi-
tionally, the  most widely used  term has been
that  of efficiency which attempted to relate in
simple,   economic   benefit-cost  analysis  the
relationship   between  resources  (input)   and
proposed  goals  or  attempted targets (output).
An efficient system has been an easy one, that
with minimal cost,  and  that cost has  always
been understood in terms of dollar values.
  The term effectiveness  has been  used mostly
in terms of organizational performance or the
meeting or purely organizational goals, that is,
the relationship between  a given organization
(thruput)  and perceived goals (output). It  has
been  also used in the context of the  overall
measure of achievement  for a system, derived
from its subsystem's performance or related to
its interactions with other systems.
  And  last,  but not least, the term efficacy is
used, which attempts today to incorporate the
meeting of social goals and a  much more com-
prehensive relationship between input, thruput,
and  output. Efficacy, in other words, attempts
to move beyond purely economic considerations,
or criteria of organizational effectiveness,  and
tries to  answer the question of how  a parti-
cular system  can efficiently,  effectively,  and
guided  by principles of social awareness, meet
goals of a given society. The term of efficacy
brings forward  an increasing awareness of a
whole  number  of  intangible benefits to  be
accrued from a given water system that can not
be directly measured by  existing economic or
other quantitative criteria. Qualitative criteria
and  the consideration of social goals transcend-
ing purely utilitarian criteria provide us with the
very difficult task of trying to strike a balance
of fulfilling water use  goals in expedient, tech-
nologically and economically feasible ways, to

-------
302    MANAGING IRRIGATED AGRICULTURE
larger questions of social policies and attempts
of environmental balance. This implies that any
water management system, as well as any at-
tempt  for comprehensive  water use develop-
ment would  be also dependent on subjective
models  which are  much more difficult to con-
struct, yet they contain long-range policies for
a social use of natural resources.
  These  kinds of  considerations  probably
provide the  answer to a  point raised  earlier.
As it will  be recalled, many questions have been
expressed as to the  advisability of continuous
water use development  in  the arid  West under
the adverse or fragile ecological conditions of
the territory. It would have been easier to have
water systems responding only to technological
imperatives.  However, the  valleys of the West
and  the  irrigation  systems are not  abstract
simulation models responding to the  whims of
any  experimenter.  They  include  individuals
and  communities that have developed a pattern
of life and whose  welfare and future may even
depend  on inefficient water  systems. Even a
marginal or not particularly efficient agriculture
fulfills the purpose of being a supportive social
system for a  number of individuals and part of
the ongoing life of a number of Americans. It
is not easy to dictate a social policy that would
be based  solely on  criteria of  efficiency  and
effectiveness  without considering at  the same
time the so-called "human factor." And in many
respects the  human factor involves  questions
of inefficiency (and ineffectiveness) because the
social costs of dislocation and disruption may be
so high—when examined under criteria of ef-
ficacy—that they may dictate a policy of con-
tinuing  present practices with little technologi-
cal intervention.
  The  above  discussion,  however,  does  not
mean that there is  very little to be done with the
irrigated valleys of the  West or with any pres-
ently declared inefficient  or ineffective water
management system. It only points out that the
problem of water quality  management control
is a complex one that requires considerations
beyond accepted technical, economic, or even
political constraints. Our  effort for improving
water  quality  management implies,  therefore,
a manyfold attack and  a series of efforts aimed
at improving project irrigation efficiency and
effectiveness, under the larger rubric of efficacy
and the achievement of larger social goals.
  An  artificial  distinction  between  technical
and   non-technical  propositions  concerning
water quality control may serve as the basis for
summary and tentative conclusions.
  A)  Technical Recommendations
     1. Water measurement.
  Research has  shown that where water mea-
surement is not practiced  throughout a  given
water system, irrigation  efficiencies are  gener-
ally very low.
     2. Modification of delivery schedule.
  It  has become apparent  that the  modifica-
tion  of present delivery  schedules,  especially
where the continuous flow method is found to
be inefficient, are necessary conditions for an
efficient system. It should be recognized, how-
ever, that the major  obstacle to overcome such
a difficulty are  the  long-established local cus-
toms of water delivery and use.
     3. Treatment.
  Concern  is expressed and recommendations
have made that  studies are  needed to evaluate
methods of reduction of salt  or nitrate  in various
soils and that they are apparently technical sol-
utions to such an attempt.
     4. Water charge schedule.
  Studies  again have shown that when  excess
water charges have  been levied, water use  has
remained reasonably low without reduction in
crop yields. The water charge also should  be
used for all irrigation districts, not only against
those using excessive water, thus  making  the
fanner cognizant of the need for good irrigation
practices.
     5. Modernization of project facilities.
  Many of the irrigation valleys are character-
ized  by nearly  obsolete or worn out   struc-
tures since a number of the projects constructed
date back to the early 1900's. This is particularly
true of turnouts, canal linings, and other struc-
tures. A program for replacement and moderni-
zation is needed in order to  increase efficiency.
  B) Non-technical recommendations.
      1. Improvement in operational practices.
  Organizational innovations  need to be intro-
duced in many  irrigation systems  to be able to
increase water deliveries  in an efficient manner.
If we are  aiming towards efficient  water use
and  proper water scheduling, as well as  an ef-
ficient delivery schedule, we must have reason-

-------
                                                       SOCIOLOGICAL CONSIDERATIONS     303
able assurance that water can be obtained when
it is needed or when one is entitled to receive it.
Streamlining of companies innovative adminis-
trative procedures,  trained personnel, and over-
all  organizational  effectiveness  are  parts  of
the guarantee for improvement in operational
practices.
      2. Consolidation of irrigation companies
        and districts.
  From  previous discussion, it has become ap-
parent that it is difficult to accomplish the task
of  an  efficient  water  management  system,
which will be capable of providing an efficient
water  delivery system  as well as minimizing
degradation of the quality of the water supply,
unless there exists the proper  operational unit
which may allow the employment  of  better
qualified  personnel, streamlined  procedures,
and the administrative  facilities  for a  better
management. Consolidation  of irrigation com-
panies and districts becomes a necessary con-
dition for  a better management,  reflected not
only in lower operating costs per unit area, but
in more  efficient regulation and control of the
water supply, avoidance of duplication, and pro-
vision for  increased  capital  availability  for
better irrigation projects. At  the same time, the
coordination and  combination of small irriga-
tion districts may permit exchanges  or regula-
tions  of water between districts, and if practiced
at a larger scale, permit better regulation be-
tween basins  and  interregional arrangements.
The elimination of overlapping personnel and
facilities  are parts  of the guarantee  for a new
operational structure that  will  be able  to
respond to the changing needs of agriculture.

      3. Institutional changes.
  The major  institutional change required in
the West is the  interpretation of the western
water  laws, which will  provide  incentives for
efficient  water  use. This is particularly true for
States having major quality  problems resulting
from irrigation return flow. There,  not only
studies should  be  undertaken  to  delineate the
changes  required  in water law interpretation,
but  also procedures must be  devised so that
such  interpretations could be  incorporated in
water law structures  and  mechanisms  that
could  be  implemented.  The interpretation of
western  law must always be followed by the
kinds of administrative  and procedural  struc-
tures which will permit regulation and control
and  effective  interpretation of  a  limited  re-
source and of the individual water right.
     4. Social Control and regulation.
  The  above  recommendations need  to  be
incorporated into  control measures  which will
make feasible the  improvement of downstream
water quality.  Coordinating committees repre-
senting local, state, and federal interests may be
developed, that would attempt to show only the
particular  benefits to  be gained from a con-
certed  policy.  Larger mechanisms, such as  ap-
pelate  administrative bodies,  could  also help
resolve potential social conflicts. These social
conflicts  may arise for two reasons:  either  be-
cause the scarcity  of water leads to conflicts of
interests  between  individuals  and groups;  or,
because efficient use of water requires organi-
zed,  coordinated, cooperative, even compre-
hensive action  on the  part of all persons or
organizations sharing the resources  of a given
system.
     5. Comprehensive  planning  and  social
        policy.
  It  becomes apparent  that  the problem of
water quality return flow is only part  of a larger
developing  complex resource utilization scene
in the western  United States. Already  frame-
work studies attempt to  provide a  systematic
examination of the  requirements for compre-
hensive planning in various basins. What is also
needed is an understanding  of the emerging
new  trends which will  affect  not  only agri-
cultural practices,  but will determine the ques-
tion  of a valuable environment in the Western
United States.  Agriculture  needs  to be  ex-
amined  in  the  context of major  population
trends, the increasing centrifugal forces of ur-
banization, metropolitanization and  suburbani-
zation. At the  same time,  there  should be
cognizance of the fact that increased  industriali-
zation and the multiplicity of uses of the water
in the  West will also contribute to the growing
problem  of return flow quality control.  This is
also  to be understood in the context of water as
a precious recreational use in the West, whose
increasing role  will  also  affect attempts  for
control and improvement.
     6. Education programs.
  All  the above,  technical and non-technical
proposed solutions for improved water man-

-------
304
MANAGING IRRIGATED AGRICULTURE
              Consideration
               of Regional
              and National
                 Policy
                               Evaluation
                             of Alternative
                                 Plans
                                         Final Evaluation
                                         Recommendations |
                                         Policy Principles
   Figure 6: Research Strategy and Assessment Methodology in the Implementation of Irrigation Systems

-------
                                                      SOCIOLOGICAL CONSIDERATIONS     305
agement, depend very much on strong educa-
tional  programs  and informational  inputs at-
tempting to reach all those responsible for the
operation and use of a given system. Everybody
involved, from the single user, the communities,
the board  of directors, the local, state, and
federal authorities need to incorporate into their
way of thinking the idea of social responsibility
and wise use of every type  of natural resource.
This is part of a larger biospheric commitment
in the use of resources as part of a limited com-
modity which needs to be wisely extracted and
also wisely used.  Training programs  should be
incorporated into all research and development
activities, making it  possible for all  people to
participate  early  in  the formulation of larger
policies and to, thus,  be able to understand the
consequences of  present practices and future
trends.
  It has been found  in many studies  that when
individuals  have the chance to participate early
in the  formulation, administration, and actual
running of  a project they are much more able to
understand  the implications  of certain  actions
and much more able to respond to the challenge
of change.  The  challenge of change indeed is
the  one that  will  ultimately  determine  all
efforts  for  water  quality control. The manage-
ment of a system  is only a small part of a larger
effort  to   develop   comprehensive  policies,
proper administrative mechanisms, and mobil-
ization of  individuals who will  be able to re-
spond  to the continuous challenge of  growth,
to the  natural limitations of the  region, and to
the challenge of continuously improving  the
environment for the common welfare.
  No research  and  no technological solution
will be able to compensate  the damage done to
an area and to our total environment by people
who have very little  appreciation of the larger
picture and by individuals who cannot perceive
their role in the context of an integrated social
system. The  social awareness of problems and
an intelligent citizen approach is  part of a much
larger challenge for both hard and soft science
practitioners who are  increasingly  asked to
answer questions of future survival  in an  en-
vironment  characterized, not by the quest of
continuous growth, but  by  questions  of quality
and the search for a way  of life  and  a com-
munity of people who recognize  natural limita-
tions and the fragility of our planet.
BIBLIOGRAPHICAL COMMENTS
  Generally,  the literature on sociological as-
pects of water management, or on problems of
water quality control is very limited. Rather
than provide any specific literature, we would
like to cite the works most useful in the above
argumentation.
  A number  of general works are important in
providing general information  on  water re-
sources  and  population trends. These  include,
The Nation's Water Resources prepared by the
Water Resources Council in 1968;  Robert M.
Hagan, et. al, Irrigation of Agricultural Lands,
Madison, Wisconsin,  American  Society  of
Agronomy, 1967; Gilber F. White, Strategies of
American  Water  Management,  Ann Arbor,
University  of  Michigan  Press,  1969;   Otto
Ekstein,  Water Resource  Development, Boston,
Harvard  University Press, 1958; various current
reports of the Census Bureau, including avail-
able returns of the 1970 enumeration; and, fin-
ally, for some major theoretical dimensions and
the role  of water in society, A. Wiener's,  The
Role  of Water  in  Development,  New York,
McGraw-Hill, 1972.
  For the social dimensions in the use of natural
resources, there are a number of select articles
and special papers that are of use. Among them,
B. H. Bylund, "The Human Factor and Changes
in  Water Usage Patterns,"  Water  Resources
Research, 2  (1966):  365-369; Arit  K.  Biswas,
"Socio-Economic Considerations in Water Re-
sources   Planning,"  Paper  presented  at  the
National Water Resources  Engineering meet-
ing of ASCE,  Atlanta,  1972, Samuel Baxter,
"Effects  of   Urbanization,"  Water  Resources
Bulletin,  4 (1968): 51-56; the present  author's
"Social  Processes in Water Management  Sys-
tems" in Treatice on Urban Water Systems,
Fort Collins, Colorado, State University, 1971,
pp.  722-739; R. K.  Linsley,  "Some  Socio-
economic Aspects  of Water Development," in
Water  Resources Use and Management,  Mel-
bourne,  Melbourne   University  Press,  1964,
pp. 429-437; Kenneth Wilkinson and L. W. Cole,
Sociological  Factors  in   Watershed Develop-
ment,  Mississippi  State  University,  Social
Science   Research Center,  Preliminary Report
20, 1967; and Evan Vlachos, "Urban  Growth
and Natural Resources," in Urban Demands on
Natural  Resources,   Denver,  University  of

-------
306    MANAGING IRRIGATED AGRICULTURE
Denver Press, 1970, pp. 259-276.  I have  also
greatly benefited  from an  unpublished manu-
script of Garth Jones, "A  Conceptual  Frame-
work for the Evaluation of Irrigation Systems."
And for a perceptive analysis of regional trends,
socio-economic  conditions and   natural   re-
sources,  one should see Ernest  A. Engelbert,
"Regional Planning for Water Development in
the Western United States; The Relevance of
American Experience  for  Other  Nations," in
International Conference on Water for Peace,
Washington, D.C., 1968, Vol. VI, pp. 256-268.
  In  thinking  about  problems  of water  re-
source management, the following works were
informational: Social and Ecological Aspects of
Irrigation and Drainage,  New York,  ASCE,
1970; U. P.  Gibson, "Integrating Water Quality
Management into Total Water Resources Man-
agement," "Critical Reviews in Environmental
Control,  2 (1971): 1-55; and Irving K. Fox and
L. E. Craine, "Organization Arrangements  for
Water Development," Natural  Resources Jour-
nal, 2 (1962): 1^4.
  For a more  general discussion of the  systems
approach in social sciences, the list of works is
lengthy but  of great benefit can be the discus-
sions in Walter Buckley, Sociology and Modem
Systems  Theory,  Englewood  Cliffs,  Prentice-
Hall, 1967; Joseph H. Monane, A Sociology of
Human Systems, New York, Appleton-Century-
Crofts, 1967; and, the descriptive but insightful
little  volume  by C.  West Churchman,  The
Systems  Approach, New  York,  Delta  Books,
1968.
  A very difficult problem in the present ex-
position  has been  the locating  of  pertinent
works  of problems of irrigation  return flow,
salinity, etc. Most beneficial for a first under-
standing of such problems were the works  of
James P. Law, Jr. and Jack L. Witherow (eds.),
Water Quality Management Problems in Arid
Regions, Ada Oklahoma, 1970; the Report  of
Gaylord V.  Skogerboe and James P. Law, Jr.,
submitted in November  1971 to the Office  of
Research and Monitoring Environmental Pro-
tection  Agency,  under  the title,  "Research
Needs for Irrigation Return Flow Quality Con-
trol;" and Lloyd V. Wilcox, "Salinity Caused by
Irrigation,"  Journal  of  the  American  Water
Works Association, 54 (1962): 217-222.
  Finally, in trying to  develop alternatives for
evaluation  and aspects of technology assess-
ment, in addition to a number of general socio-
logical works (such as  on social indicators), a
few  specialized  works  were  helpful  in the
clarification of the argument. These include:
William K. Johnson, "Assessing Social Conse-
quences of Water Deficiencies," Journal of the
Irrigation  and  Drainage  Division, ASCE,  4
(1971): 547-557;  V.  Vemuri  and N.  Vemuri,
"On  the  Systems  Approach in  Hydrology,
"Bulletin of the International Association  of
Scientific Hydrology, 15, 6 (1970):  17-38; and
the unpublished paper of A.  Bruce  Bishop,
"An Approach to Evaluating Environmental,
Social, and Economic  Factors in Water Re-
sources Planning," (mimeographed, no date).

ACKNOWLEDGEMENT
  This paper is  based  on work supported  in
part from funds provided by the United States
Department of the Interior as authorized under
the Water Resources Act of 1964,  Public Law
88-379.  The support  of the  Environmental
Resources Center at Colorado State University
is also gratefully acknowledged.

-------