United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
Ada OK 74820
Research and Development
EPA-600/S2-83-072 Nov. 1983
Project Summary
Irrigation Return Flow Water
Quality Monitoring, Modeling
and Variability in the Middle
Rio Grande Valley, New Mexico
L.W. Gelhar, P.J. Wierenga, K.R. Rehfeldt,
C.J. Duffy, M.J. Simonett, T.-C. Yeh, and W.R. Strong
A 250-acre (100 hectare) irrigated
farm in the middle Rio Grande valley at
San Acacia, New Mexico was intensively
monitored for the five-year period from
1977 through 1981. The quantity and
quality of the applied irrigation water
and the drainflow leaving the farm were
observed, and 39 observation wells
were installed to monitor ground-water
levels and quality. The data indicate that
there has not been a statistically
significant change in the total dissolved
solids concentration in the shallow
ground water underlying the site during
the monitoring period. A significant
increase in nitrate concentration in the
shallow ground water and the drains
was observed as portions of the farm
were converted from alfalfa to corn.
The observations also included exten-
sive systematic measurements of the
temporal and spatial variability of
chemical and physical parameters at
the site. The spatial observations,
which included measurements of infil-
tration rate, grain size distribution,
electrical conductivity and chloride
concentration, showed substantial
variability, with correlation scales on
the order of 10m in the horizontal
direction and less than 1m in the
vertical. The temporal data show major
weekly variations in composition of the
irrigation water, ground water, and
drainflow.
The data from deep wells demon-
strated the layered structure of the
aquifer underlying the site and pointed
to the possibility that upwelling regional
ground-water discharge was producing
a major influx of relatively high salinity
water beneath the site.
The data collected at the San Acacia
site were used to test two computer-
based models which simulate the flow
and water quality behavior. A two-cell
lumped-parameter model, which em-
phasisesthe dynamic nature of the
water and mass balances, adequately
simulated the average monthly chloride
concentration of the drain water.
A modified version of the U.S. Bureau
of Reclamation (USBR) hydrosalinity
model, which incorporates the layered
structure of the aquifer and the contribu-
tion from high salinity regional inflow,
was used to simulate a 30-year period
under steady state average flow condi-
tions, and the four-year monitoring
period under transient flow conditions.
The results of the 30-year simulation
are in reasonable agreement with the
monitored water quality of the drain
flow at the end of the period. However,
the modified model was not able to
simulate the observed seasonal variation
of salinity emissions from the site
during the monitoring period. The
chemical reactions included in the
USBR model did not have a significant
effect on the model predictions for the
San Acacia site.
This Project Summary was developed
by EPA's Robert S. Kerr Environmental
Research Laboratory, Ada, OK, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
-------
Introduction
Pollution from irrigated agriculture has
presented a difficult problem in the
national water quality strategy. Irrigated
agriculture occupies large land areas,
and as a result its impact on waterquality
is extensive, but at the same time rather
diffuse. Furthermore, control of pollution
from irrigated agriculture requires the
simultaneous cooperation of many differ-
ent land users, irrigation districts and
state and federal institutions. Such a
process necessitates careful planning
and organization. Although it is generally
recognized that irrigated agriculture
contributes to degradation of water
quality, the extent of such pollution varies
from location to location and is frequently
ill-defined. For example, in somewestern
states suspended and settleable solids
are the major causes for degradation of
waterquality, while in the arid southwest
increases in the salinity of streams by
salt-loading and salt concentrating
mechanisms are the main reasons for
water quality degradation. In some areas
water quality is degraded by nutrients
from fertilizer discharged in soil water,
and in other areas irrigation return flow
has increased levels of phosphates or
nitrates originating from natural deposits.
Thus in order to define the impacts of
irrigated agriculture on water quality it
appears necessary to perform field
monitoring studies under various environ-
mental conditions and at various locations.
The general objectives of this project
are: (1) to monitor irrigation return flow
quality on a 250-acre (100 hectare) test
site in the Middle Rio Grande Valley at
San Acacia, New Mexico, (2) to use data
obtained at this site to test the Bureau of
Reclamation-EPA irrigation return flow
model, and (3) to evaluate the effects of
spatial and temporal variability of flow
and water quality. The following detailed
subobjectives will be required:
1 To evaluate the effects of irrigation on
water quality in the Middle Rio
Grande Valley of New Mexico through
intensive monitoring of a farm opera-
tion at San Acacia, New Mexico.
2. To evaluate the predictive capabilities
of the Bureau of Reclamation hydro-
salinity model applied to the San
Acacia site, and to consider modifica-
tions necessary to improve this
model.
3. To evaluate the spatial and temporal
variability of flow and water quality
inputs and parameters at the San
Acacia site and apply stochastic
analyses which incorporate data on
the variability of chemical and flow
parameters at the San Acacia site.
Monitoring and Variability
Monitoring information provided an
extensive data base on the irrigated
hydrologic system at the San Acacia site
over four complete irrigation seasons.
These observations provided a unique
data base for comparative testing of the
USBR-EPA irrigation return flow model.
The temporal variability observations
emphasized that water quality parameters
of the irrigation water, the ground water
and the drainflow are subject to significant
short-term (weekly) variations which can
easily mask any long-term trends. These
observations, demonstrate that the irrigated
hydrologic and water quality system is
highly dynamic and point to the difficulty
of coming to any definitive conclusions
about water quality trends in such
systems from only occasional (annual or
semi-annual) observations.
Similarly the spatial variability observa-
tions demonstrated that the flow and
water quality parameters at the San
Acacia site are highly variable in both the
horizontal and vertical direction. Generally
the correlation scales in the horizontal
(on the order of 10m) are significantly
larger than the scales in the vertical (a
meter or less). The correlation scale of
water quality parameters tends to be
larger than that of the infiltration rate.
These observations also demonstrated
the difficulty of realistically characterizing
the properties of such an irrigated system
from a small number (2 or 3) of observation
wells or soil samples.
Modeling
Two modeling approaches were used
in this study to determine the characteris-
tics of the middle drain flow system at the
San Acacia site. The first approach
employed multiple-celled lumped-param-
eter models, and the second consisted of
a profile finite element flow model
coupled with the U.S. Bureau of Reclama-
tion hydrosalinity model. The results of
the modeling studies indicated that the
mixing process producing observed drain
concentrations involves a complex tran-
sient relationship between the irrigation
recharge and a poor quality regional
inflow.
The application of these models at the
San Acacia site demonstrated the dynamic
structure of the irrigation return flow
system. The observations of flow and
water quality showed a high degree of
temporal and spatial variability. A simple
lumped-parameter water and mass
balance model adequately reproduced
the dynamic effects, but the USBR-EPA A
model failed to represent the observed ™
temporal system. We do not recommend
modification of the USBR-EPA model to
represent unsteady conditions with
regional inflow. This would require
complete revision of the modeling
approach and would greatly increase data
requirements. The need to simulate
dynamic water quality changes in irrigated
systems would, in our opinion, be better
served by improving the lumped-parameter
model to include the effect of chemical
reactions. The data requirements of the
lumped-parameter model are less severe,
parameter estimation is systematic, and
the simple structure is easily modified to
reflect site specific conditions. In any
case, we recommend that a systematic
lumped-parameter water and mass
balance model be used initially to
evaluate overall hydrologic conditions
and provide regional management-
oriented predictions.
Conclusions
Intensive monitoring of the irrigation
system established that the San Acacia
site, which originally appeared to be
hydrologically isolated from the surround-
ing area, was, in fact, significantly
influenced by the surrounding regional ^
ground-water flow system. This is espe- fl
cially true in relation to salinity, where
over 50 percent of the total dissolved
solids emission in drains originates from
sources outside the farm. The models
were modified to account forthis external
contribution.
Analysis of total dissolved solids data
obtained from monitoring shallow wells
underlying the irrigated area of the San
Acacia site shows that there has not been
a statistically significant change in the
salinity of the shallow ground water ovei
the duration 6f the project. A significant
increase in nitrate concentration in the
shallow ground water and the drains was
observed as portions of the farm were
converted from alfalfa to corn.
Extensive systematic observations of
chemical and physical parameters of the
subsurface flow system at the site show
that these parameters are highly variable
in time and space. The correlation scale of
spatial variability is generally observed to
be an order of magnitude larger in the
horizontal direction than in the vertical.
A two-cell lumped-parameter model,
which emphasizes the dynamic nature of
the water and mass balances, was found
to adequately simulate the average
monthly chloride concentration in the
middle drain, despite the complexities of ^
the real flow system. V
-------
A modified version of the Bureau of
Reclamation model adequately simulated
steady-state average water quality at the
end of a 30-year period but the modified
model was not able to simulate the
observed seasonal variation of salinity
emissions from the site during the
monitoring period. The chemical reactions
included in the Bureau of Reclamation
model did not have a significant effect on
the predictions at the San Acacia site.
In this and other applications of
lumped-parameter models, it has been
shown that these simple models adequate-
ly simulate the behavior of conservative
solutes. It is recommended that the
lumped-parameter models be generalized
to include the effect of chemical reactions.
L W. Gelhar is presently with Massachusetts Institute of Technology, Cambridge.
MA 02139; P. J. Wierenga is with New Mexico State University, Las Cruces, NM
88003; K. R. Rehfeldt, C. J. Duffy, M. J. Simonett. T. -C. Yeh, and W. R. Strong
are with the New Mexico Institute of Mining and Technology, Socorro, NM
87801.
James P. Law. Jr., is the EPA Project Officer (see below).
The complete report, entitled "Irrigation Return Flow Water Quality Monitoring.
Modeling and Variability in the Middle Rio Grande Valley, New Mexico, "(Order
No. PB83-261 719; Cost: $37.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
P.O.Box 1198
Ada, OK 74820
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
------- |