Climate change is an anthropic phenomenon that leads to changes in the atmosphere and causes an increase in average temperature and the probability of extreme weather events on the planet. This is the biggest environmental concern today and understanding its impacts is essential. Climate change causes changes in the physiology and metabolism of organisms, altering biodiversity, dynamics of populations and communities, structures of trophic networks and functioning of natural ecosystems. However, there is an urge to incorporate community approaches, trophic relationships and ecosystem processes in the assessment of the impacts of climate change. This is extremely important in tropical freshwater ecosystems, where there is a lack of studies on this topic. One way to use such approaches is the study of microbial communities, due to their great diversity and varied and rapid responses to environmental changes. Microorganisms are directly and indirectly impacted by climate change in different ways. The microbiota is also able to respond to disturbances caused by climate change, enabling its use as bioindicators and biomarkers. In order to utilize microorganisms as bioindicators, it is first necessary to characterize these communities through sequencing the 16S rRNA gene (phylogenetic marker for bacteria). The use of microorganisms as bioindicators of climate change for tropical freshwater ecosystems allows the development of a community index for replication in different environments. Therefore, this current project aims to investigate how warming affects the phylogenetic diversity of microbial communities in tropical freshwater ecosystems, for the selection of bioindicators of climate change and the construction of health indices for these environments. For the purpose of this work, three different experimental models will be used: (1) artificial mesocosms, (2) geothermal streams and (3) natural microcosms (bromeliads). Sequencing of the 16S rRNA gene combined with bioinformatics and statistical analysis (diversity indices, LEfSe, IndVal, etc.) will provide a deep characterization of the studied environments. Afterwards, the use of machine learning (ML) will enable the assignment of bioindicators of climate change, for establishing the community index and predictive model. The index and predictive model will allow replication for different environments and, consequently, a better understanding of the impacts of climate change for freshwater continental tropical ecosystems.