Aridity drives plant biogeographical sub regions in the Caatinga, the largest tropical dry forest and woodland block in South America

Our aims were to quantify and map the plant sub regions of the the Caatinga, that covers 844,453 km2 and is the largest block of seasonally dry forest in South America. We performed spatial analyses of the largest dataset of woody plant distributions in this region assembled to date (of 2,666 shrub and tree species; 260 localities), compared these distributions with the current phytogeographic regionalizations, and investigated the potential environmental drivers of the floristic patterns in these sub regions. Phytogeographical regions were identified using quantitative analyses of species turnover calculated as Simpson dissimilarity index. We applied an interpolation method to map NMDS axes of compositional variation over the entire extent of the Caatinga, and then classified the compositional dissimilarity according to the number of biogeographical sub regions identified a priori using k-means analysis. We used multinomial logistic regression models to investigate the influence of contemporary climatic productivity, topographic complexity, soil characteristics, climate stability since the last glacial maximum, and the human footprint in explaining the identified sub regions. We identified nine spatially cohesive biogeographical sub regions. Current productivity, as indicated by an aridity index, was the only explanatory variable retained in the best model, explaining nearly half of the floristic variability between sub regions. The highest rates of endemism within the Caatinga were in the Core and Periphery Chapada Diamantina sub regions. Our findings suggest that the topographic complexity, soil variation, and human footprint in the Caatinga act on woody plant distributions at local scales and not as determinants of broad floristic patterns. The lack of effect of climatic stability since the last glacial maximum probably results from the fact that a single measure of climatic stability does not adequately capture the highly dynamic climatic shifts the region suffered during the Pleistocene. There was limited overlap between our results and previous Caatinga classifications.

本研究旨在量化并绘制卡廷加（Caatinga）的植物亚区分布图，该区域面积达844453平方千米，是南美洲面积最大的季节性干旱森林连片分布区。我们整合了该区域迄今规模最大的木本植物分布数据集（涵盖2666种灌木与乔木，覆盖260个采样点），并对其开展空间分析；将所得分布格局与现行植物地理分区方案进行对比，同时探究驱动该区域植物亚区区系格局形成的潜在环境因子。研究基于以辛普森相异指数（Simpson dissimilarity index）计算的物种周转量开展定量分析，以此界定植物地理区域。我们采用插值法对卡廷加全域的群落组成变异的非度量多维标度（NMDS）轴进行可视化制图，随后依据预先通过k-means聚类分析确定的生物地理亚区数量，对群落组成相异性进行分类。我们采用多项逻辑回归模型，探究当代气候生产力、地形复杂度、土壤特性、末次冰盛期以来的气候稳定性以及人类活动足迹对已界定植物亚区的解释效力。本研究最终界定出9个空间上连续完整的生物地理亚区。最优模型中仅保留了以干旱指数表征的当代生产力这一项解释变量，其可解释近一半的亚区间区系变异。卡廷加区域内特有物种丰富度最高的亚区为迪亚曼蒂纳高原核心亚区与边缘亚区。本研究结果表明，卡廷加区域的地形复杂度、土壤异质性与人类活动足迹仅对木本植物的局部分布产生影响，而非塑造大尺度区系格局的决定性因素。末次冰盛期以来的气候稳定性未对区系格局产生显著影响，这一结果可能源于单一的气候稳定性指标无法充分反映该区域在更新世期间经历的剧烈气候波动。本研究结果与既往的卡廷加分区方案仅存在有限的重叠区域。