Development of a model for hydro-salinity simulation in irrigated river basins

Abstract

Agricultural irrigation has led to river salinization in many basins. When irrigation water filtrates through the soil, salt dissolves in the water, and the saline water enters the groundwater and returns to the river. In this study a model is developed to simulate hydro-chemical processes in soil and spatial/temporal distribution of irrigation return flows so that river salinity can be predicted. Major chemical reactions in the root zone soil are simulated including precipitation, dissolution of soil minerals and association and dissociation of major ion pair species. To determine the quantity and quality of the deep percolated water, soil water composition is predicted at five soil depths: surface, 1/4, 1/2, 3/4 and bottom of the root zone. The spatial and temporal distribution of groundwater return flows to the river system are expressed as response functions multiplied by the volume of the stress applied on the basin. The response functions are based on two-dimensional porous media flow equations. A probabilistic approach is developed to simulate travel time of groundwater flow for irrigated fields that have subsurface drainage devices. All related hydrological components are integrated into the computation of river water quantity and quality. These components include: groundwater return flows, irrigation tail water, tributary inflows, river diversions, phreatophyte consumption, river channel losses and river depletion due to pumping. The river system is divided into reaches, and the river quantity and quality can be simulated from upstream reaches to downstream reaches sequentially. The study area selected for implementing this hydro-salinity simulation is the Arkansas River Basin in Colorado. The simulation period is from January 1986 to December 2001, a total of 192 months. The first 60 months (1986 to 1990) are used to calibrate the model. The other 132 months (1991 to 2001) are used to validate the calibrated model. A GIS program, ArcView 3.2, is applied to generate a groundwater table so that the response functions for groundwater return flows in each canal service area can be developed. The simulation results show that irrigation return flows, including surface and groundwater return flows, significantly increase river quantity in reaches 12, 13, 14, 21, 22, 23 and 24. Meanwhile, groundwater return flow is a prime driver of river salinity. Both river quantity and river salinity show significant seasonal fluctuation. While the results presented in this dissertation attempt to represent the actual conditions in the Arkansas River Basin no external funding was provided for this study and therefore in some instances the data used is a best estimate of the actual conditions.