Abstract
Resource availability and heterogeneity are recognized as two essential environmental aspects to determine species diversity and community abundance. However, how resource availability and heterogeneity determine species diversity and community abundance in highly heterogeneous and most fragile karst landscapes is largely unknown. We examined the effects of resource availability and heterogeneity on plant community composition and quantified their relative contribution by variation partitioning. Then, a structural equation model (SEM) was used to further disentangle the multiple direct and indirect effects of resource availability on plant community composition. Species diversity was significantly influenced by the resource availability in shrubland and woodland but not by the heterogeneity in woodland. Abundance was significantly affected by both resource availability and heterogeneity, whereas variation partitioning results showed that resource availability explained the majority of the variance in abundance, and the contribution of resource heterogeneity was marginal. These results indicated that resource availability plays a more important role in determining karst plant community composition than resource heterogeneity. Our SEMs further found that the multiple direct and indirect processes of resource availability in determining karst species diversity and abundance were different in different vegetation types. Resource availability and heterogeneity both played a certain role in determining karst plant community composition, while the importance of resource availability far exceeded resource heterogeneity. We propose that steering community restoration and reconstruction should be highly dependent on resource availability, and multiple direct and indirect pathways of resource availability for structuring karst plant communities need to be taken into account.
Keywords community abundance, karst shrubland and woodland, resource availability, resource heterogeneity, species diversity
Introduction
Understanding the mechanisms that maintain vegetation is an essential ecological goal for the sustainable development of fragile ecosystem. Two general ecologically-based hypotheses related to resources have been developed to explain plant community composition at the local scale. The resource heterogeneity hypothesis suggests that species diversity is a function of heterogeneity in resources because of species specialization in heterogenetic resource gradients (Ricklefs 1977; Huston 1979; Tews et al. 2004; Scott and Baer 2019). Resource heterogeneity generally increases niche diversity and provides opportunities for speciation events, allowing for increasing species coexistence (Rosenzweig 1995, Silvertown 2004; Do Carmo 2016; Feeser 2018). But this is not true in small spatial patches, where there is a nonsignificant or even negative heterogeneity-diversity relationship (Wijesinghe et al. 2005; Tamme et al. 2010; Gazol et al. 2013). In contrast, the resource availability hypothesis does not necessarily rely on assumptions about species specializations and proposes that the average level of limiting resources should govern species coexistence (Wright 1983; Stevens and Carson 2002; Bakker et al. 2003; Wassen et al. 2005; Kumar et al. 2018). In general, under a low resource availability, the species diversity and abundance of a plant community are lower because only a few species possibly survive and grow under such harsh environments (Désilets and Houle 2005; Comita et al. 2007).
Recent evidence has indicated that the relative importance of resource availability and heterogeneity in influencing species composition is different in different plant communities. For instance, Bartels and Chen (2010) demonstrated that resource availability drives species diversity in both young and mature stands of forest ecosystems, whereas resource heterogeneity dominates in old-growth stands. Shirima et al. (2016) suggested that the mean soil nutrient availability explains considerable variations in tree species richness in moist forests, while vertical soil nutrient heterogeneity is a predictor of tree species richness in miombo woodlands. In fact, species diversity and abundance are scarcely the results of a single factor and direct process, with respect to either resource availability or resource heterogeneity (Whittaker et al. 2001). Soil depth varies with topographic position, which influences species diversity (Baer et al. 2005). Soil acidity has an indirect effect on plant community composition via its effects on the availability of key mineral nutrients (Chen et al. 2004; Zhu et al. 2013). Soil nitrogen increases with stand development due to increased nitrogen fixation with stand age (Hume et al. 2016). Therefore, there is a suite of direct and indirect environmental factors that simultaneously influence the species diversity and abundance of karst plant communities, but their direct and indirect relationships remain poorly understood.
Karst is an important component of global terrestrial ecosystems with the most fragile ecological environments (Huang and Cai 2007; Ford and Williams 2007). Fragmentation of karst landscapes is much higher than other terrestrial ecosystems, where a shallow and discontinuous soil layer with exposed rocks forms various habitats, such as rocky fissures, rocky gullies, soil faces, rocky faces and rocky pores (Zhang et al. 2007; Nie et al. 2011; Chen et al. 2016). The broken terrain dramatically increases the spatial variability of other environmental variables, notably the physical and chemical properties of the soil (Zhong et al. 2014; Toure et al. 2015). Additionally, previous researches have demonstrated that the spatial distribution of soil resource varied obviously in karst region (Tateno and Takeda 2003; Zhang et al. 2006). Therefore, the high edaphic heterogeneity provides numerous ecological niches for plant diversification and speciation (Wang et al. 2014), species diversity and abundance of karst plant community are expected to be determined by resource heterogeneity. Nevertheless, the soil resource shortage also is a remarkable feature of karst landscapes owing to slow soil formation, a shallow soil layer and severe soil erosion, so the species diversity and abundance of karst plant community are likely to be strongly influenced by the availability of one or more limiting resource. As consequence, karst landscapes not only represent limited soil resource availability but also exhibit significant soil resource heterogeneity, the effect of resource availability and heterogeneity on karst species diversity must be taken into consideration simultaneously. In prior studies of karst ecosystem, the fact that some resource availability (e.g. Soil depth, P availability, K availability) limits species diversity and abundance has been widely demonstrated (Crowther 1982; Zhang et al. 2013; Toure et al. 2015; Liu et al. 2020). However, few studies have examined the effects of small-scale resource heterogeneity on karst plant community species composition; and no study has examined the role of resource availability and heterogeneity in karst plant species diversity and abundance.
Here, we chose six sites typical in karst environment in southwestern China to quantify and compare the effects of resource availability and heterogeneity on species diversity and abundance. We then further disentangled the direct and indirect effects of resource availability on karst plant community composition by SEM. We addressed the following questions: (1) How do resource availability and heterogeneity affect both species diversity and abundance in the highly heterogeneous and most fragile karst mountains? (2) What is the relative importance of resource availability and heterogeneity for shaping karst plant community species composition? (3) How is karst plant community species composition driven by multiple direct and indirect pathways of resource availability?
Materials and Methods