Advances and Research Gaps in Ecological Niche Modeling of Amazonian Wetland Plants: A Systematic Review

Ecological niche modeling (ENM) is widely used to predict species distributions and support biodiversity conservation under environmental change, yet its application to Amazonian wetland plants has not been systematically synthesized. We conducted a systematic review following the PRISMA 2020 guidelines to evaluate methodological approaches, environmental predictors, model performance, and research gaps. Literature searches were performed in Web of Science, Scopus, and Consensus (as a complementary AI-assisted academic search platform), yielding 1789 records, of which 48 met the eligibility criteria. Correlative models predominated, with MaxEnt, Random Forest, and ensemble frameworks being the most frequently applied algorithms. Across studies, integrating hydrological and edaphic predictors consistently improved model performance and ecological realism compared with climate-only approaches. Future climate projections indicated greater vulnerability for habitat-specialist species, whereas western Amazonia and the Andean foothills were repeatedly identified as potential climatic refugia. Major limitations included geographically biased occurrence records, limited high-resolution environmental datasets, and the underrepresentation of several Amazonian wetland ecosystems. Overall, the evidence indicates that reliable ENMs for Amazonian wetlands require integrating climatic, hydrological, and edaphic drivers rather than relying solely on macroclimate. Future research should prioritize geographically representative sampling, improved environmental datasets, transparent workflows, and process-informed modeling to strengthen ecological forecasting and conservation planning under climate and land-use change.