Estimation of Rice Evapotranspiration in Paddy Fields Using Remote Sensing and Field Measurements

Water resources are limited in many parts of the world. Due to the fast growing world population, the demand for domestic and industrial water use is increasing tremendously. This results in reduction of water for agricultural use, especially for major rice growing areas which needs huge amounts of water. The study was carried out in the northwest of Selangor, in the Tanjung Karang Rice Irrigation Project Malaysia. The objectives of this study were to estimate the rice evapotranspiration using satellite data and compare it with the field measurements. Eight sets of non-weighing lysimeters (91 cm x 91 cm x 61 cm) with attached casella hook were installed to measure the crop evapotranspiration at five different locations within the 19000 ha rice irrigation scheme. Global positioning system (GPS) was used to locate the lysimeter position. The rice yields in the lysimeters were manually measured for three seasons. An automatic meteorological station was installed inside the field to collect data required for calculations of the crop evapotranspiration using the CROPWAT software. NOAA satellite data was used as data input to correlate the remote sensing data with field evapotranspiration data. For three seasons, the off (dry)season from December to May, the main (wet) season from July to November, the observed ET from the lysimeters ranged from 3.2 to 5.8 mmlday, while ET by calculation using weather parameters ranged from 3.15 to 5.72 mmlday. There was no significant difference between the blocks in the first season of the experiment because of the small area and not much difference in the environmental conditions within the block. Most of the correlation for the second and the third season were significant at 0.01%. The corresponding ET values from satellite data were 4.04 to 6.54 mmlday. Considering ET measured by lysimeter as the most accurate method, ET determined using satellite data overestimates and by computed method underestimates those obtained by lysimeter. ETc by NOAA data were found to overestimate by 8% to 12% with an average of 10%. Using CROPWAT, ETc were found to be underestimated between 7% and 20% with an average The significance findings of this research are that ET can be estimated for paddy areas in Malaysia with reasonable accuracy using satellite data or computed method. The implications are that much time and cost can be saved using these alternative techniques compared to manual data collection from lysimeters. This will result more efficient water management planning in the rice areas. Generally, by knowing the actual ET, the cropping calendar can be prepared at the beginning of the cultivation season by knowing the amounts of water needed throughout the season