Current forecasts estimate that the global population will grow from 7.7 billion people to 9.8 billion people by 2050. Furthermore, urban dwellers are expected to reach nearly 68% of the world’s population. If this scenario becomes reality, the global area covered by urban settlements is expected to increase to over 3 million km², which would be equivalent to the entire country of India. A key parameter in determining the impacts of urbanization on the environment is measuring the amount of impervious surface area (ISA). In remote sensing (RS), mapping of ISA by satellite imagery can be performed within a timely and cost-effective manner. In this regard, adopting geographic object-based image analysis (GEOBIA) and Deep Learning (DL) technology may help with the mapping of ISA in remotely sensed images. Increased ISA may result in severe hydrological impacts in urban areas, being some of the clearest impacts increasing surface runoffs and increases in flood frequency and its magnitude in flood-prone areas. Among all natural disasters, floods cause more impact than any other worldwide (over 30%). To make it worse, urban flooding is expected to occur more often and be more severe as Earth experiences extreme hydrometeorological events occasioned by global climate change. Given the foregoing, this Ph.D. proposal aims to combine a forecasting model of urban expansion and urban land use change with a hydrodynamic model to provide forecasts of extreme flooding events in a pilot city. The expected results will provide crucial geo-information by identifying areas with higher flood risk, which should get maximum attention during any such events. Thus, this work is aimed not only at increasing scientific knowledge in this field but also providing a support tool for Civil Defense agents, and decision and policy makers. With such information, they will be able to make decisions in a well-timed manner to contain and mitigate the effects of floods in Brazilian cities.