Satellite-Based Assessment of Crop Coefficient for Sugarcane in Maui, Hawaii.

Water availability is one of the limiting factors for sustainable production of biofuel crops. A common method for determining crop water requirement is to multiply daily potential evapotranspiration (ET_o) calculated from meteorological parameters by a crop coefficient (K_c) to obtain actual crop evapotranspiration (ET_c). Generic K_c values are available for many crop types but not for sugarcane in Maui, Hawaii, which grows on a relatively unstudied biennial cycle.  In this study, an algorithm is being developed to estimate sugarcane K_c at development stage using normalized difference vegetation index (NDVI) obtained from Landsat 7 satellite images. Crop canopy cover was measured with a handheld multispectral camera from two sugarcane fields at the Hawaiian Commercial & Sugar Company (HC&S) plantation during the Landsat 7 satellite overpass days. An Eddy Covariance (EC) tower system was set up within each of these two fields and gathered EC flux at a 30-minute interval. Reference evapotranspiration was calculated from the network of automated weather stations at HC&S plantation using a modified Penman equation. Crop canopy cover was highly correlated with satellite NDVI values (r²=0.96, RMSE=0.04). A linear relationship between NDVI and measured K_c was obtained. Satellite -based ET_c maps of HC&S plantation were developed using the NDVI-based K_c values and reference ET from HC&S weather station network. The satellite-based ET_c was compared and validated with field measurements of ET using Eddy Covariance tower. A series of satellite-based ETc maps were developed to indicate the water demand of sugarcane plants at HC&S plantation.  These results validate the use of satellite imagery as a tool for estimation of ET of sugarcane plants in Maui, Hawaii.