Improving remote sensing based evapotranspiration modelling in a heterogeneous urban environment

Evapotranspiration (ET) is a key component of the hydrologic cycle. Knowledge of ET is important for modelling hydrologic fluxes and improving water resource management. Various remote sensing-based ET modelling studies have been conducted for investigating water demand in agricultural areas. However, ET modelling studies of urban areas are rare due to the challenges inherent in the heterogeneity of urban landscapes. This study proposes an improved surface energy balance algorithm for urban areas (uSEBAL) to make it suitable for estimating ET in urban environments. In the proposed approach, ET is predicted using improved energy budget components considering urban land cover composition and anthropogenic heat flux. The results of the uSEBAL are compared with the traditional method of SEBAL, and a sensitivity analysis is performed to evaluate the impact of uncertainties in ET estimates. The findings of this study indicate that the variability in urban land cover types impacts spatial variability in energy fluxes and ET. The results also show seasonal influence on ET for different land covers, but no significant influence of seasonality is observed on urban impervious areas, which produces the lowest ET nearly zero mm/day. The analysis of variance indicates that the differences between uSEBAL and SEBAL derived ET values for urban impervious areas are statistically significant (p-value < 0.05). The results also show the variability of ET values between models for other land cover types but show small variations in areas of wetland and dense vegetation. An investigation of factors of changes in ET indicates that surface albedo, and solar radiation highly influence ET, and the errors in the estimation of them can result in the highest uncertainty in the estimation of ET. The model performance metrics indicate that the uSEBAL is better than the SEBAL for estimating ET in a heterogeneous urban environment. This study supports the inclusion of anthropogenic heat in the energy budget, and emphasizes the use of land cover maps while estimate ET in urban areas.