Remotely sensed data of hydrological parameters are mostly provided on a coarse spatial resolution. Therefore, in order for them to be useful in the development and improvement of various hydrological models and applications, the need of a downscaling approach arises. The aim of a downscaling approach consists of the production of estimates through a variety of methods at finer scales by relying on auxiliary finer scale remotely sensed information. Such downscaling methods are being presented within this chapter focusing on the most studied, in terms of downscaling, remotely sensed hydrological parameters, i.e. soil moisture, land surface temperature, and precipitation. The means to improve hydrological modeling, apart from the development of various downscaling approaches, include techniques that pertain to data fusion as well. In the present chapter, two data fusion techniques are being discussed, focusing on the investigation of floods and their impacts. The first refers to the fusion of remotely sensed precipitation data from two different missions in order to evaluate precipitation and streamflow over a river basin and, thus, contribute to flood assessment in terms of flood intensity and frequency. The second technique involves the blend of a flood risk model with a satellite-based flood assessment system for the purpose of contributing through a model to the assessment of social and economic flood impacts.