Scheduling cotton irrigations using remotely-sensed basal crop coefficients and FAO-56

Abstract

Techniques to more accurately quantify crop evapotranspiration (ET C) are needed for determining crop water needs and appropriate irrigation scheduling. In this study, remotely sensed observations of the normalized difference vegetation index (NDVI) were used to estimate cotton basal crop coefficients (Kcb), which were then applied within the dual crop coefficient procedures of the Food and Agricultural Organization (FAO), Paper 56 (FAO-56) to calculate daily ETC. An experiment in central Arizona during 2003 compared irrigation scheduling using a remotely sensed Kcb technique (NDVI treatment) with the FAO-56 Kcb curve (FAO treatment). The FAO curve was locally developed for optimum crop conditions and standard cotton density. Final lint yield means were not significantly different between the two irrigation methods, which included sub-treatments of two levels of nitrogen and three plant densities. However, NDVI attained higher yields under low N input, whereas FAO generally had higher yields under high N. The ETC estimated using the NDVI-KCb method was in closer agreement with measured cumulative ETC than the FAO Kcb. For high N treatments, the mean absolute differences between measured and estimated cumulative ETC during the growing season for typical, dense, and sparse populations (10, 20, and 5 plants m-2, respectively) were 4, 17, and 4 mm, respectively, for NDVI, whereas they were10, 32, and 13 mm, respectively, for FAO. Although additional research is needed for improving our remote sensing technique, it potentially offers an improvement over the FAO Kcb curve for quantifying actual ETC.