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

Home Browse Search

:.W1. 500 E! 400 u 300 U Vi 200 c d 100 0 Et 0 0000 0000 Year Arkansas River at La Junta (m /s) Fort Lyon Canal (m /s) Figure 3. Total annual discharge of the Arkansas River at La Junta and the Fort Lyon Canal, 1971 -94. allow for diversion of the river at a maximum discharge or 26 m /s (Dash 1994). Average daily diversions of about 9 m /s are a function of the quantity of flow in the river and the administration of water rights in the basin. Diversions from the canal to the irri- gated lands are made through numerous small headgates. Streamflow is not adequate to meet crop demands in many years, especially in spring and late summer. Therefore, ground water pumped from the alluvial aquifer serves to meet irrigation needs during periods of lower streamflow. Irrigation from ground water sources is derived from as many as 92 large capacity irrigation wells that are approximately evenly spaced over the length of the study area, primarily north of the river. During the 24 year study period, the number of irrigation pumps that were used in any one month varied. The alluvium underlying the study area consists of deposits of Holocene and Pleistocene clay, sand, silt, and gravel, and is under- lain by relatively impermeable Upper Cretaceous limestone and shale. Average thickness of the alluvium in the study area is about 10 in (Person and Konikow 1986), with an average saturated thick- ness of about 6 km (Konikow and Person 1985). The estimated effective porosity of the aquifer is 0.2, and the transmissivity aver- ages 9.3 X 10 -3 m /s (Konikow and Person 1985). The aquifer is hydraulically connected to the Arkansas River, which is a par- tially penetrating stream. Ground water flow generally is toward the northeast, parallel to the river. In the eastern part of the study area, however, a component of the flow trends southeasterly from the canal to the river. Streamflow recovery, downstream from the canal headgate, is supplied mainly from irrigation return flows. Description of Model and Data Requirements The two - dimensional, distributed parameter flow and solute transport model (Konikow and Bredehoeft 1978), commonly referred to as the Method of Characteristics (MOC) model, was applied to the study area in several earlier studies (Konikow and Bredehoeft 1974a, 1974b; Person and Konikow 1986). The model used by Person and Konikow (1986), which was calibrated for 1971- 82 conditions, was adopted for use in this study; their boundary con- ditions and initial aquifer characteristics were used. The model and data requirements will be briefly described here. For a more complete description, refer to Konikow and Bredehoeft (1978). The model approximates solutions to the coupled, partial dif- ferential equations governing fluid flow and solute transport. The distribution of the heads in the aquifer is calculated by solving the equation of fluid flow by an alternating direction implicit procedure. Once the distribution of heads in the aquifer is calculated, Darcy's law is used to calculate the flow velocities. The solute transport equa- tion is solved by the method of characteristics to estimate salinity. These equations are solved on a monthly time step for each of the applicable grid cells in the 20 cell by 44 cell rectangular model grid. Each grid cell measures 201 in by 403 m. Cells outside the aquifer boundaries were treated as no -flow boundaries. Boundary cells on the southwestern and northeastern ends of the model grid were treated as constant flow boundaries based on estimated ground water underflow into and out of the study area. Data required for the head and flow calculations include pre- cipitation, streamflow, canal flow, irrigation from surface water sources, ground water withdrawals (municipal and irrigation), crop and phreatophyte evapotranspiration, ground water underflow, and recharge. Precipitation and other climatological data were col- lected at the La Junta airport, which is located about 1.6 km north of the study area. Precipitation was applied evenly over the entire stuady area at a constant rate throughout each month. Streamflow data were collected on a monthly basis near the upstream end of the study area at the La Junta gauging station by the Colorado Division of Water Resources. Surface water diversion data for the Fort Lyon Canal were collected by the Fort Lyon Canal Co. at a Parshall flume located about 1.6 km downstream from the diversion point. Of the total diversions to the canal, 5.5% was delivered for appli- cation to the irrigated land in the study area. The amount of applied surface water was estimated on the basis of the proportion of canal shares in the study area relative to the number of shares in the entire canal system (Gronning Engineering Co. 1993). All irrigated land in the study area was assumed to receive equivalent surface water irrigation application rates. Surface water application rates varied with streamflow. During 1971 -82, when streamflow was relatively small, average surface water applications were 0.6 m/yr (Figure 4). During 1983 -94, streamflow typically was larger, and average sur- face water applications increased to 0.8 m/yr (Figure 4). a -0 1 a d 0.5 0 F Year Surface Water ® Ground Water Figure 4. Estimated annual irrigation application in the study area, 1971 -94. 78 — m %n n 0, .-. m v o m 00 00 00 00 0o a a