Data from: Factors determining forest diversity and biomass on a tropical volcano, Mt. Rinjani, Lombok, Indonesia

Tropical volcanoes are an important but understudied ecosystem, and the relationships between plant species diversity and compositional change and elevation may differ from mountains created by uplift, because of their younger and more homogeneous soils. We sampled vegetation over an altitudinal gradient on Mt. Rinjani, Lombok, Indonesia. We modeled alpha- (plot) and beta- (among plot) diversity (Fisher’s alpha), compositional change, and biomass against elevation and selected covariates. We also examined community phylogenetic structure across the elevational gradient. We recorded 902 trees and shrubs among 92 species, and 67 species of ground-cover plants. For understorey, subcanopy and canopy plants, an increase in elevation was associated with a decline in alpha-diversity, whereas data for ground-cover plants suggested a hump-shaped pattern. Elevation was consistently the most important factor in determining alpha-diversity for all components. The alpha-diversity of ground-cover vegetation was also negatively correlated with leaf area index, which suggests low light conditions in the understorey may limit diversity at lower elevations. Beta-diversity increased with elevation for ground-cover plants and declined at higher elevations for other components of the vegetation. However, statistical power was low and we could not resolve the relative importance to beta-diversity of different factors. Multivariate GLMs of variation in community composition among plots explained 67.05%, 27.63%, 18.24%, and 19.80% of the variation (deviance) for ground-cover, understorey, subcanopy and canopy plants, respectively, and demonstrated that elevation was a consistently important factor in determining community composition. Above-ground biomass showed no significant pattern with elevation and was also not significantly associated with alpha-diversity. At lower elevations communities had a random phylogenetic structure, but from 1600 m communities were phylogenetically clustered. This suggests a greater role of environmental filtering at higher elevations, and thus provides a possible explanation for the observed decline in diversity with elevation.

热带火山是一类重要却研究不足的生态系统，其植物物种多样性、群落组成变化与海拔的关联模式可能不同于构造抬升形成的山地，原因在于火山土壤更为年轻且均一性更强。我们在印度尼西亚龙目岛的林贾尼火山沿海拔梯度开展了植被调查，以海拔及选定的协变量为解释变量，分别构建了α（样地内）多样性、β（样地间）多样性（以费希尔α指数（Fisher’s alpha）表征）、群落组成变化及地上生物量的统计模型；同时分析了沿该海拔梯度的群落系统发育结构（community phylogenetic structure）。本次调查共记录到92种木本植物（含乔木和灌木）共计902株，以及67种地表覆盖植物。针对林下层、亚冠层与冠层植物，海拔升高与α多样性下降呈显著负相关；而地表覆盖植物的α多样性则呈现驼峰型分布模式。海拔始终是决定所有植被组分α多样性的最关键驱动因子。地表覆盖植被的α多样性还与叶面积指数（leaf area index）呈显著负相关，这表明林下低光照环境可能会限制低海拔区域的物种多样性。地表覆盖植物的β多样性随海拔升高而上升，而其余植被组分的β多样性则在高海拔区域出现下降。但本研究的统计效力较低，无法明确各因素对β多样性的相对影响程度。针对样地间群落组成变异的多变量广义线性模型（Multivariate Generalized Linear Models，GLMs）分别解释了地表覆盖层、林下层、亚冠层和冠层植物67.05%、27.63%、18.24%及19.80%的偏差变异（deviance），结果证实海拔是决定群落组成的一贯重要因子。地上生物量未呈现随海拔变化的显著分布模式，且与α多样性也无显著关联。低海拔区域的群落系统发育结构呈随机分布模式，但在海拔1600米以上的区域，群落则呈现系统发育聚集模式。这表明环境过滤作用在高海拔区域发挥了更强的调控作用，也为海拔升高导致物种多样性下降的观测结果提供了潜在的科学解释。