Drivers of temporal beta diversity and ecological resilience of plant communities across the United States

Israel Temitope Borokini , Alyssa M Walters, Michael D France, Aidan Peers, Monica Lohr, Sophie Kusek

Long-term monitoring surveys offer more reliable assessments of biodiversity responses to climate change. In this study, we evaluated community shifts in 1105 1 m2 plots across 20 NEON sites, representing each of the 20 eco-climatic domains across the U.S. and Puerto Rico, using long-term plant surveys (5-10 years). We computed alpha, spatial and temporal beta, and gamma diversity for the plots in both the earliest surveys (T1) and those conducted in 2023 (T2). We used permutational analysis of variance (PERMANOVA) and permutational multivariate analysis of dispersion (PERMDISP) to test for statistical significance in compositional variance between the two surveys, and investigated the effects of T1 alpha diversity, elevation, soil, and climatic variables on temporal beta diversity using generalized linear mixed effect models (GLMM). We also checked if temporal beta diversity was significantly structured along geographical gradients and among different vegetation classes. The results showed increases in alpha and gamma diversity, comprising mostly native species; therefore, temporal beta diversity was high for most of the NEON sites, driven mainly by turnover and species gains between T1 and T2 periods. We observed latitudinal and longitudinal but not elevational, gradients in temporal beta diversity; herbaceous vegetation classes experienced the lowest temporal community changes, while forest ecosystems and pasture hay showed the highest shifts. PERMANOVA and PERMDISP tests showed significant community shifts in 16 NEON sites that is explained by compositional variance within or between survey years. GLMM showed significant effects of three soil (soil pH, silt content, fine organic carbon) and climatic (monthly Tmax standard deviation, monthly Tmean standard deviation, and AET coefficient of variation) variables on temporal beta diversity, while T1 alpha diversity has no effect. Given the land use history in many of the NEON sites, observed temporal beta diversity may indicate post-disturbance succession or climate-induced reshuffling of regional species pool.