Aprendizado Profundo para Coleta de Evidências Locais de Mudança Climática e Clusterizaçã de Eventos Extremos

The frequency, duration, and intensity of various extreme climate events have increased as the climate system warms. For example, climate change leads to more evaporation, which can exacerbate droughts and increase the frequency
of heavy rain and snowfall. These extreme climate events often result in extreme conditions or impacts, whether by crossing a critical threshold in a social, ecological, or physical system or by co-occurring with other events. These climate changes are evident in Earth Observation (EO) data. In this context, a growing demand is expected for resilient models that can adapt to the effects of global warming. In this project, we introduced a brief overview of deep learning approaches in the literature for climate data, including the exploration of numerical, hybrid models and those using machine learning for weather prediction, as well as models that employ self-supervised learning techniques, and remark some applications of the models studied.