The speed of change in climate and land cover is associated with the speed of biodiversity changes in avian assemblages of the United States

PLOS One | https://doi.org/10.1371/journal.pone.0330153 August 29, 2025 5 / 20 as a control variable in the models as differences in biodiversity are associated with altitudinal gradients [59]. We built GAMMs for each buffer size (12.5 km, 25 km, and 50 km). For all models, we used a significance level of α = 0.05 to determine the significance of predictor variables. Model diagnostics were performed through visual inspection of residual plots (“mgcv” R package, [57]). Most of the residuals’ Q-Q plots showed little or no deviations from the normality assumption. Exceptions were some GAMMs for the species replacement of Great Plains, and the species loss of Eastern Temperate Forests (depending on the resolution of the buffer used). In general, the residual versus predicted plots did not show any pattern. Exceptions were some GAMMs that showed heteroskedasticity in the residual versus predicted plots for the species richness, species replacement and/ or species loss (depending on the resolution of the buffer used). In these instances, there was some variability in the data that was not captured by the fitted GAMMs. In the main text, we mainly present the scenario with the intermediate buffer resolution (25 km). Differences between models across buffer sizes are described in the Supporting Information and S1-S7 Tables in S1 File. We assessed the relative importance of climate change and LULC change per ecoregion by comparing the number of climatic variables to the number of LULC variables that had a significant association with the biodiversity metrics analyzed

in the GAMMs. Since the total number of climate predictor variables included in the GAMMs was lower than the total Fig 1. Methodological approach followed in the study. Step I.A: Four bird biodiversity metrics (response variables) were assessed across five ecore- gions of the United States: one for species richness, and three for assemblage dissimilarity (total assemblage dissimilarity: Sorensen index, Species replacement: Simpson Index, and Species loss: Nestedness index). Step I.B: Predictor variables. We analyzed the relationship between the rate of change of each of the four biodiversity indices in time (Step II.A) to the rate of change of environmental variables (Step II.B) with generalized additive models (Step III). In the equations in Step I.A: a = # species common to both sites, b = # species that occur in the first site but not in the second and c = # species that occur in the second site but not in the first (Baselga, 2010). https://doi.org/10.1371/journal.pone.0330153.g001