Modifying SEBAL ET Algorithm to account for advection by using daily averages of weather data

The use of Remote Sensing (RS) in crop evapotranspiration (ET) estimation is aimed at improving agricultural water management. The Surface Energy Balance Algorithm for Land (SEBAL) is one of several methods that have been developed for this purpose. This has been a preferred model as it requires minimal climate data. However, it has a noted downside of underestimating ET under advective conditions. This is primarily due to the use of evaporative fraction (EF) to extrapolate instantaneous ET to daily values, with the assumption that EF is constant throughout the day. A modified SEBAL model was used in this study, which requires daily averages of weather data to estimate advection which is then introduced into the 24-hour ET sub-model of SEBAL. The study was carried out in southeastern Colorado, a semi-arid area where afternoon advection is a common feature. ET estimated using the original and modified SEBAL was compared to the lysimeter-measured ET. Results showed that the modified SEBAL algorithm performed better in estimating daily ET in overall, but especially on days when there was advection. On non-advective days, the original SEBAL was more accurate. It is therefore recommended that the modified SEBAL be used only on advective days, and guidelines to help identify such days were proposed.