The accurate estimation of precipitation and its spatial variability within a watershed is crucial for reliable discharge simulations. The study is the first detailed analysis of the potential usage of the Cellular Automata technique to merge different rainfall data inputs to hydrological models. This work shows an improvement in the performance of hydrological simulations when satellite and rain gauge data are merged.

In this study, we investigate the combination of surface soil moisture information with process-related features, namely, evaporation efficiency, soil water index and normalized difference vegetation index, using artificial neural networks to predict root-zone soil moisture. The joint use of process-related features yielded more accurate predictions in the case of arid and semiarid conditions. However, they have no to little added value in temperate to tropical conditions.

Soil moisture and evapotranspiration (ET) are important to flood forecasting. An energy balance scheme based on the representative temperature (RET) was developed and coupled to the original mass balance scheme of the Xin'anjiang model. Validation against both runoff and land surface temperature confirmed the accuracy and applicability of the improved model (XAJ-EB). RET serves as a new constraint to the model and can be used for model calibration and validation.

This study presents the sensitivity of an oceanic model of the Adriatic Sea to the horizontal resolution and to the meteorological forcing. The model is run with two different configurations and with two horizontal grids at 1 and 2 km resolution. To study the influence of the meteorological forcing, the two storms have been reproduced by running ROMS in ensemble mode. Possible optimizations of the model set-up are deduced by the comparison of the different run outputs.