| Abstract: | An inadequate streamflow supply in Saginaw Bay, Michigan, affects the irrigation development in the region. Linear programming (LP) models are developed in this paper to help irrigation planners determine optimal irrigation scenarios. To incorporate the spatial variations of the soil and climate into the optimization models, soil associations are used as the basic analysis unit for estimating crop yields, irrigation requirements, and expected economic returns. The yields and irrigation requirements of corn, drybeans, and soybeans are simulated by the CERES-MAIZE, BEANGRO, and SOYGRO models at each of the twelve soil associations in the study area over the period of 1951–1980. The yield and irrigation requirement of sugarbeets are simulated by the YIELD model for all the soils in the study area over the period of 1956–1987. The crop prices, simulated crop yields and irrigation requirements, and variable costs are used to compute the expected gross margins and other coefficients in the LP models. The optimization results provide irrigation scenarios that are linked to specific soil associations, which are more useful to decision makers for identifying and targeting individual soils for irrigation development than are the spatially lumped optimal solutions. This study demonstrates that through integration of GIS and simulation models, spatial variations of the climate, soil, and water resources can be better represented in the optimization models to produce more realistic irrigation scenarios. |
