Water balance irrigation scheduling: comparing crop curve accuracies and determining the frequency of corrections to soil moisture estimates
1997
Steele, D.D. | Scherer, T.F. | Prunty, L.D. | Stegman, E.C.
Water balance methods are commonly used to schedule irrigations and use evapotranspiration (ET) functions and crop curves to estimate crop ET (ETC). Because the methods may over or underestimate ETC, field-measured values of available soil moisture content (SMC) are often used to correct or adjust estimates of soil moisture during the growing season. This study was conducted to (1) compare the accuracies of four crop curves, based on the Jensen-Haise reference ET method, for corn in the northern Great Plains, and (2) determine the need for and frequency of in-season corrections to SMC estimates. The comparisons were based on differences between estimated and measured SMC for the 1990, 1991, 1992, and 1994 seasons using nonweighing lysimeters near Oakes, North Dakota. The SMC data were compared to estimates using bias and absolute errors, r2, Friedman Rank Sums, and sign distributions. The SMC estimates were corrected to measured values at three frequencies: start of season only, approximately monthly, and approximately semimonthly. The crop curve based on days past planting was generally the most accurate, followed by the crop curve based on cumulative growing degree days. All of the methods tended to overestimate ETC. Selection of a correction frequency is more important than selection of a particular independent variable--days or weeks past emergence, days past planting, or cumulative growing degree days since planting--for a crop curve. For the crop curves, soil types, and climatic conditions of this study, none of the crop curves should be used without in-season SMC corrections on at least a monthly frequency, and semi-monthly corrections are preferred. The methods employed in this study can be transferred to other sites, climates, and crops.
显示更多 [+] 显示较少 [-]