This study assesses the surface energy balance system (SEBS) model for estimating farm-scale actual evapotranspiration (ETa) of medicinal herbs in three diverse climatic regions of southern Iran: Pasargad (semi-dry, semi-cold), Arjan (semi-wet, semi-cold) and Darab (dry, warm). The research aimed to establish ground truth ETa using the soil water balance (SWB) approach and to calibrate and validate the SEBS model utilizing Landsat 8 imagery. SEBS was calibrated against FAO Penman-Monteith (FAO P-M) potential evapotranspiration (ETo) data from nearby water bodies, improving the root mean square error (RMSE) to 0.081-0.088 mm/day and R-2 values to 0.90-0.96 across sites. Validation against field-observed ETa, derived from the SWB (considering irrigation, effective rainfall and return flow, with negligible soil moisture change), revealed an R-2 of 0.81 and an RMSE of 246.0 m(3)/ha (approximately 5% of average SEBS ETa) after one outlier was removed. Although SEBS generally underestimates the observed ETa by approximately 10%, it effectively demonstrates that plant ETa is governed by water needs rather than excessive application. The findings confirm SEBS's capability for large-scale water management, highlighting current overirrigation and the model's utility for optimizing water use and acknowledging a challenge in applying the model to small farm plots owing to image resolution.