Stimulated Effects of Irrigation on Salinity in the Arkansas River Valley in CO

Home Browse Search

damage annually. Miles estimated that 81,000 hectares within the valley were irrigated with class C4 water, which is the U. S. Salinity Laboratory's highest rating of salinity hazard for crops. The objective of this study was to estimate how changes in his- torical irrigation practices could affect the quantity and quality of ground water and surface water in the Arkansas River valley. Mathematical modeling represents a powerful tool to help evalu- ate the potential impact of changes in irrigation management prac- tices on the quality and quantity of surface and ground water. In the Murray Basin in Australia, for example, ground water flow mod- eling has been used to determine the effectiveness of alternative grown water management schemes to prevent stream an land s - ..aa � o : vx.. � , ° R M T ENT j P R O W E R 3� • isation associated with irrigation (Simpson and Herczeg 1991; _ '09' OTERO 1 ( f Ghassemi et al. 1989. Within the Roswell and Carlsbad Basins in a. -. -. New Mexico, analytical and three - dimensional ground water flow 0 1 :o 30 40 50WILE5 modeling are currently being used to assess the impact of reduced ground water pumping on changes in base flow to the Pecos River Figure 1. Location of study area. in order to satisfy an interstate water compact between New Mexico and Texas (Barroll 1996). A two - dimensional flow and solute trans- river and irrigation canal flow, ground water withdrawals, and sur- port model_(Konikow and Bredehoeft 1978) was applied to a 17.7 face water quality. The present study incorporated this variability km reach of the Arkansas River to examine the relations between by using monthly data of these conditions to estimate the potential hyd r o logic s tresses (such as irrigation, pumping, and streamflow); effects of changes in irrigation practices. ground water levels in the alluvial aquifer; and salinity of ground water and surface water. Rather than using this model in a strictly Description of Study Area predictive sense, the hydrologic and chemical consequences of The study area is located in the alluvial valley of the Arkansas various modeled irrigation management scenarios were compared River, near La Junta in southeastern Colorado (Figure 1), and is to historic conditions. about 17.7 km long and 2.4 km wide. This area was chosen because This model has been applied to this site previously (Konikow its hydrologic framework and water use patterns are representative and Bredehoeft 1974a, 1974b; Person and Konikow 1986) and of much of the Arkansas River valley downstream from Pueblo, was originally developed for this area in the Konikow and because of the availability of detailed hydrologic data (hydrology, Bredehoeft (1974a) study. Konikow and Bredehoeft (1974a) mod- geology, water use, and water quality), and because of the availability eled the period from 1971 -72 to evaluate how ground water and river of a previously calibrated model (Konikow and Bredehoeft salinity changes in response to varying hydrologic stresses. Konikow 1974a).The climate is semi -arid; mean annual precipitation is about and Bredehoeft (1974b) extended the simulation period to 1975 by 27.9 cm/yr; and mean annual temperature is 12 ° C. The primary crops recycling the first year of input data. They assumed that the hydro- in the study area are alfalfa, corn, sorghum, and wheat. About logic stresses measured in the 1971 -72 study period were repre- 1993 hectares (about 27 %) of land in the study area are flood irri- sentative of long -term averages. The simulation was extended to gated with surface water and ground water. For the purposes of mod - 1982 by Person and Konikow (1986); however, the model was mod- eling, the irrigated acreage in the study area was subdivided into ified by incorporating an unsaturated zone solute transport function three sub -areas. This subdivision is based on land use patterns, irri- and by allowing the quantity of the applied surface water to vary gation application rates, and irrigation canal administration proce- based on actual streamflow measurements during 1972 -82. These dures. Sub -areas 1, 2, and 3 compose about 20, 33, and 47% of the studies emphasized the difficulties in using models to make pre- total irrigated area in the study area, respectively (igure 2). dictions of future hydrologic and water quality conditions sub- Surface water is diverted for irrigation from the Arkansas stantially beyond the model calibration period. The present study River by the Fort Lyon Canal at its headgate, which is located uses a 24 year study period as the basis for model calibration, a sub- about 1.6 km west of the upstream end of the study area. The stantially longer period than previous studies. Natural and anthro- unlined canal follows the northern boundary of the study area pogenic effects throughout the 24 year study period produced a high (Figure 2) and generally diverts a substantial portion of the stream - degree of variability in many hydrologic conditions, including flow from the river (Figure 3). Direct -flow water rights for the canal Figure 2. Map of study area, sub -areas of irrigated acreage, and aquifer boundaries. 77