The role of niche overlap, environmental heterogeneity, landscape roughness and productivity in shaping species abundance distributions along the Amazon–Andes gradient

Abstract Aim Statistical and ecological mechanisms shape species abundance distributions (SADs). A lack of correlation between ecological gradients and SAD shape would suggest that SADs are caused by purely statistical reasons. We evaluated the variation in the shape of SADs for communities in landscapes of differing variable connectivity, environmental heterogeneity, species niches overlap and productivity. Location Rainforests in the Madidi region (Bolivia). Methods We compiled biological and environmental information on 65 sites (a site being a group of two to six 0.1‐ha plots where woody plants of a diameter at breast height ≥ 2.5 cm were inventoried). We built unveiled (complete) SADs for each site and fitted Gambin models to those SADs. The Gambin α parameter served as a metric of SAD shape. Low α values characterize logseries‐like SADs, while high α values characterize lognormal‐like SADs. For each site, we estimated landscape roughness, environmental heterogeneity, species niche overlap and productivity. These variables were related to SAD shape by means of variation partitioning. Results SADs changed from logseries‐like to lognormal‐like along the elevational gradient. Many of our predictor variables were correlated: 40.4% of the variation in SAD shape could not be attributed to specific factors. However, 50.62% of the variation in the SAD shape could be assigned to individual predictor matrices: 28.4% was explained exclusively by niche overlap, 15.41% exclusively by environmental heterogeneity, 5.20% exclusively by landscape roughness and 1.6% exclusively by productivity. Main conclusions Ecological processes related to the topographical/environmental complexities that vary across the elevational gradient are correlated with the SAD shape. Purely statistical mechanisms are apparently not sufficient to explain the changes in SAD shape. The most important factor is the mean overlap of the niches of the species of an assemblage: avoiding competition with co‐occurring species could be the most important mechanism driving species relative success at the ≤100 km 2 scale.