Distribuição geográfica e modelagem de nicho ecológico de três espécies de palmeiras da Amazônia

Abstract

This study aimed to model the climatic niches and identify the current and future potential distribution areas of Astrocaryum aculeatum G. Mey, Bactris gasipaes Kunth, and Euterpe precatoria Mart, species of high sociocultural, economic, and ecological importance in the Amazon. Additionally, it sought to assess the vulnerability of these areas under climate change scenarios and the representativeness of ex situ germplasm collections. Current modeling was based on occurrence records obtained from GBIF and nine bioclimatic variables from the WorldClim database, using eight algorithms. Models with AUC > 0.70 were integrated through ensemble modeling with k-fold cross-validation (k = 5), performance-based weighting, and thresholding based on the joint maximization of sensitivity and specificity. Analyses were conducted in R software (version 4.3.3). Future projections were generated using five global climate models from CMIP6 (ACCESS-CM2, EC-Earth3-Veg, INM-CM5-0, MPI-ESM1-2- HR, and MRI-ESM2-0), under SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios, for the periods 2041–2060 and 2061–2080. Model agreement was integrated using a likelihood criterion, with presence confirmed in areas with at least 66% consensus. The results indicate that all species will experience reductions in climatically suitable areas across all scenarios and time periods. The greatest reduction was observed for Bactris gasipaes, with a projected loss of up to 79.6% by 2070 under the most pessimistic scenario. Astrocaryum aculeatum and Euterpe precatoria also showed contraction trends, with reductions of approximately 49% and 63%, respectively. The species are predominantly distributed within the Amazon domain, and no new climatically favorable expansion areas were identified for the future. The climatic gap analysis was based on georeferenced data from Genesys PGR ex situ accessions and natural occurrences. Bioclimatic variables were subjected to hierarchical clustering (Ward.D2) to define climatic clusters, and principal component analysis (PCA) was applied to visualize the environmental distribution of the accessions. The results revealed significant gaps in the climatic representativeness of germplasm banks, highlighting the need to improve passport data and guide collecting efforts toward underrepresented bioclimatic zones to ensure the conservation of adaptive diversity under climate change.