Estuaries are among the most dynamic and ecologically complex environments on Earth, acting as critical transition zones between terrestrial and marine systems. These environments host benthic macrofaunal communities that are essential for nutrient cycling and ecosystem resilience, yet their high spatial and temporal variability often challenges ecological predictability. Traditionally, the inherent physical diversity across different estuarine types—such as bays, rivers, and lagoons—has been thought to preclude the development of generalizable ecological models, often limiting studies to site-specific descriptions. This study investigates whether a universal set of environmental and climatic predictors can reliably forecast macrofaunal community structures across geomorphologically distinct estuaries. Here, we show that a hybrid machine learning framework, integrating unsupervised Self-Organizing Maps with supervised Random Forests, achieves approximately 86% predictive accuracy on a decadal scale across multiple estuarine types. Our results identify salinity, sediment grain size, primary productivity, and wind regimes as universal determinants that transcend local geomorphology, directly challenging the assumption that estuarine physical diversity prevents generalizable predictions. By resolving the non-linear mechanisms through which physical forcing shapes benthic assemblages, this framework moves from descriptive ecology to robust quantitative forecasting. These findings demonstrate that estuarine ecosystems follow identifiable ecological trajectories, providing an important tool for proactive management and restoration in the face of global climate change. • The hybrid machine learning demonstrated strong predictive power to assess shifts in macrofaunal communities. • Random forest analysis revealed spatial factors ranking as the most influential predictor of macrofaunal communities. • Salinity gradients, sediment granulometry, chlorophyll- a , and tendency of coastal winds emerged among the top drivers of temporal and spatial macrobenthic variability. • Opportunistic taxa, such as H. australis and M. schubarti , respond predictably to environmental drivers. • The model supports proactive management and habitat restoration, enabling resilience of estuarine ecosystems under future climate change scenarios.
Francisco et al. (Fri,) studied this question.