Effects of regional species pool, climate and forest structure on tree layer species richness in natural evergreen broad-leaved forests: A regional-scale analysis from Jiangxi, China

The ongoing loss of species diversity highlights necessitates elucidating the drivers of species richness to inform biodiversity conservation and sustainable forest management strategies. This study leveraged data from 1739 vegetation plots from natural evergreen broad-leaved forests from the 2019 forest management inventory in Jiangxi Province, China. A boosted regression trees model was employed to assess regional effects on species richness, followed by the construction of a structural equation model (SEM) to elucidate the direct and indirect pathways of key drivers on tree layer species richness. The results revealed a tree layer species richness range of 3–17, with higher richness values predominantly in the northern, western, and southern regions. Species richness declined significantly with increasing latitude (P < 0.05). About 40 % of the variation in tree layer species richness can be accounted for by the regional species pool (SP) and the selected regional environmental factors. The SEM revealed that SP, stand basal area per unit area (BA) and precipitation seasonality (PS) directly and positively affected tree layer species richness, while precipitation of driest quarter (PDQ) negatively affected richness directly. Tree layer species richness was affected indirectly by the anomaly in annual mean temperature from the Last Glacial Maximum to the present (Anomaly_RAMT) and regional annual mean temperature (RAMT) through SP, and by PDQ and precipitation of warmest quarter (PWAQ) through BA. Hence, conserving high-richness areas can help maintain elevated levels of biodiversity. Incorporating regional and local environmental factors, along with the regional species pool, significantly improves the comprehension of the mechanisms that affect species richness.