Hot springs are tractable model systems in microbial ecology for investigating the interactions of photosynthetic microbial biofilms. This is because they occur across broad geographic scales, possess readily identified major abiotic variables, and are subject to minimal influence from metazoans. Despite this regional scale investigations are lacking, and major questions persist concerning the evolutionary drivers responsible for biofilm turnover at broad geographic scales. Here, we present the largest study to date, incorporating concurrent measurement of biotic and abiotic diversity and rigorous statistical analysis and modelling. We characterized 395 biofilms from neutral-alkaline hot springs spanning a 2,100km latitudinal gradient in Southeast Asia. The data clearly resolved six biogeographic regions with each defined by a core microbiome comprising specific cyanobacteria and other diverse photosynthetic, chemoheterotrophic, and chemoautotrophic taxa. Our findings demonstrated that the most influential abiotic variables (pH, conductivity, carbonate) accounted for relatively little of the observed variation in biofilm communities, and that extensive biotic interactions spanned multiple trophic levels. Importantly, we present quantitative evidence that stochasticity due to ecological drift was the most important evolutionary driver of spatial turnover at a regional scale. These insights establish a pivotal milestone in understanding of this model system, fostering enhanced testing and comparison with more intricate microbial ecosystems.