Scale-dependent evidence for the abundant-center hypothesis and underlying mechanisms

Whether species are more abundant near the center of their distribution has been intensively debated. Previous studies have largely focused on local sampling units, often overlooking the inherent scale-dependence of ecological patterns. Here, we examine how sampling grain influences the relationship between species abundance and distance to a species' distributional centroid, using inventory data of 212 tree species from the USDA Forest Inventory and Analysis program (FIA). We found scale-dependent support for the abundant-center hypothesis. At the fine spatial grain of individual FIA plots, positive and negative abundance-distance relationships were roughly balanced, with the mean regression slope approaching zero. This pattern reversed at coarser grains (> 0.5°), where the odds of detecting a negative trend more than doubled and the mean regression slope became significantly negative. Additionally, we identified niche overlap as a key factor constraining species abundance at smaller grains, with its impact diminishing at larger scales. Our results underscore the necessity of incorporating biotic interactions in fine-grain species abundance and distribution modeling. Conversely, traditional climate envelope models based on abiotic variables may effectively capture abundance patterns at coarser grains. This scale-explicit perspective helps reconcile previous inconsistent findings and advances our understanding of how ecological processes structure species distributions across scales.