Environmental change affects metacommunity structure including diversity patterns among local communities directly through influences on abiotic factors and dispersal, and indirectly via interactions among co-occurring species that can be complex. One critical question is how interaction complexity affects metacommunity structure and whether responses are predictable in real-world systems. Here, we use a global dataset of freshwater lake fish species occurrences on three continents to demonstrate that environmental processes are the main drivers of species’ distribution and diversity among the three major metacommunity factors: environment, space, and species association, which suggests that anthropogenic changes of abiotic factors will heavily influence the structure of metacommunities. At the same time, the species association process quantified from co-occurrence patterns had no negligible impacts on specific species and lakes indicating that the ongoing changes are modulated not only by the direct impacts of changing abiotic factors but also indirectly, through cascading effects of species interactions. Our global analysis indicates that even under the current high rate of environmental change, impacts on metacommunity structure can be predicted by an identifiable set of underlying processes.