Comparative Analysis of Evapotranspiration in Sugarcane Fields Using MODIS Satellite Data and Ground-Based Weather Stations

The objective of this study is to explore the feasibility of using satellite imagery to monitor evapotranspiration (ET). Spatial evapotranspiration estimates may differ from crop evapotranspiration, either being overestimated or underestimated. The approach of this assessment involves using satellite imagery products to monitor evapotranspiration and comparing the results with data from weather stations. MOD16A2GF is a satellite-based evapotranspiration product, was selected for comparison with weather station data. The study found that MOD16A2GF can be used to assess evapotranspiration effectively during Thailand's dry season, as the trend of evapotranspiration aligns with weather station data. However, during the rainy season, MOD16A2GF estimates were lower than those from the weather stations. Over the entire year, MOD16A2GF underestimated reference evapotranspiration by 3.90 mm/8 days in 2020 and 8.21 mm/8 days in 2021. Similarly, crop evapotranspiration was underestimated by 6.72 mm/8 days in 2020 and 11.17 mm/8 days in 2021. The correlation between ET from satellite data and ET from weather station shows that 87% of the variability in one dataset can be explained by the other, indicating a strong relationship. A prediction model using the Random Forest algorithm achieved the highest R2 value of 0.76 and an MSE value of 132.47. Therefore, this study concludes that ET values obtained from MOD16A2GF products, using PET multiplied by the crop coefficient, can effectively monitor evapotranspiration in the study area.