Actual evapotranspiration is an important compound of water and energy balance systems. Two models were used, the Surface Energy Balance System (SEBS) combines remote sensing data and meteorological in-situ data to solve the energy balance equation and assess AET. A hydrological balance model, the Backet Botton Hole (BBH) uses Hydrological and rainfall in-situ data. Results indicate an overestimation of AET-SEBS compared to AET-BBH which are consistent with results found by other researchers.

This work is made to help mitigate some of the outcome of floods and enhance water management and monitoring, particularly for agriculture in the Medjerda river basin. A forecasting model is then built to predict monthly precipitation using pre-processing method in order to extract significant components. It can be concluded that climate indicators can add some additional information to enhance monthly precipitation forecasts at longer lead-times.

Previous evaluation of rainfall estimation by satellite algorithms showed an insufficient over Northern Tunisia. That was why we tried the WRF. We selected four heavy rainy days. For each day, 99 combinations of Cu and PBL are simulated. The sensitivity study highlighted the large difference in the estimation by the different schemes. The use of several verification techniques was extremely helpful to choose the best combinations for each event. The ensemble method gave very satisfying results.