Agricultural water management in a humid region: sensitivity to climate, soil and crop parameters

Abstract

A sensitivity analysis of irrigation water requirements at the regional scale was conducted for the humid southeastern United States. The GIS-based water resources and agricultural permitting and planning system (GWRAPPS), a regional scale, GIS-based, crop water requirement model, was used to simulate the effect of climate, soil, and crop parameters on crop irrigation requirements. The effects of reference evapotranspiration (ET o) methods, available soil water holding capacities (ASWHC), crop coefficients ( K c), and crop root zone depths ( z) were quantified for 203 ferneries and 152 potato farms. The irrigation demand exhibited a positive relationship with K c and z, a negative relationship with ASWHC, and seasonal variations depending on the choice of ET o methods. The average irrigation demand was most sensitive to the choice of K c with a 10% shift in K c values resulting in approximately 15% change in irrigation requirements. Most ET o methods performed reasonably well in estimating annual irrigation requirements as compared to the FAO-56 PM method. However, large differences in monthly irrigation estimates were observed due to the effect of the seasonal variability exhibited by the methods. Our results suggested that the selection of ET o method is more critical when modeling irrigation requirements at a shorter temporal scale (daily or monthly) as necessary for many applications, such as daily irrigation scheduling, than at a longer temporal scale (seasonal or annual). The irrigation requirements were more sensitive to z when the resultant timing of irrigation coincided with rainfall events. When compared with the overall average of the irrigation requirements differences, the site-to-site variability was low for K c values and high for the other variables. In particular, soil properties had considerable average regional differences and variability among sites. Thus, the extrapolation of site-specific sensitivity studies may not be appropriate for the determination of regional responses crop water demand.