/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.
------- |