Soil depth alters the consequences of species diversity for productivity in an experimental karst herbaceous community

Aims The mechanism by which species diversity drives productivity in different ecosystems is 23 controversial, possibly due to the confounding effects of key environmental variables. Karst ecosystems 24 are fragile and are at great risk of species loss. In these ecosystems, soil depth is a key driver of 25 community diversity and productivity. However, the influence of soil depth on the relationship between 26 species diversity and productivity in karst ecosystems remains unclear. 27 Methods We established artificial karst herbaceous communities with different soil depths and species 28 richness levels and determined how two biodiversity effects—complementarity effect (effect of positive 29 interactions among species) and selection effect (effect due to dominance of productive species)— 30 contributed to productivity. 31 Results Soil depth, species diversity, and different species combinations were significant predictors of 32 productivity. Species diversity significantly positively affected productivity at all three soil depths, with 33 the effect size of species diversity being the greatest in medium-depth soil. Net diversity effects were 34 greater than 0 in all multi-species communities, indicating that complementarity and selection effects 35 both positively influenced productivity. However, complementarity effect had a greater contribution to 36 productivity than selection effect in all multi-species communities. Furthermore, the contribution of 37 complementarity effect increased with increasing soil depth, while that of selection effect decreased. 38 Conclusions Soil depth influences the relationship between species diversity and productivity in karst 39 herbaceous communities. Furthermore, complementarity effect is the major mechanism by which 40 diversity increases community productivity, particularly in shallow soil. Therefore, environmental 41 variables should be seriously considered when studying the relationship between species diversity and function in other ecosystems.


54
The mechanism behind the effect of species diversity on productivity has also been widely studied, 55 and the main mechanisms can be divided into two categories: complementarity effects and selection 56 effects (Hector 1999; Gillman and Wright 2006; Forrester 2014). The former suggests that plant 57 communities consisting of multiple species, which have different resource requirements and utilization 58 patterns, have a higher overall ability to utilize limited resources, thus having greater productivity. In 59 contrast, the latter, also named sampling effects, suggests that communities assembled with high-60 diversity plant mixtures have a higher chance of containing and becoming dominated by the species that 61 achieves the highest productivity when grown alone, thus providing increasingly integrated productivity rather than separately (Loreau et al. 2002, Dai 2007. However, there is still a significant research gap in 65 understanding the relationship between species diversity and productivity and its mechanism in karst 66 plant communities. Recently, the species diversity of karst regions have been threatened by human 67 activities, especially in a socio-economic context where the demand for resources is growing. Hence, it 68 is increasingly imperative to determine the effect of species diversity on productivity in karst regions.

69
It has been reported that abiotic factors may affect the impact of species diversity on productivity 70 (Bärlocher and Corkum 2003, Zhang and Zhang 2003). For example, Tilman (1995) found that 71 community productivity was negatively related with species diversity during normal rainfall years, but  Karst is formed by chemical (dissolution of bedrock, such as limestone, dolomite, and gypsum) and 81 physical processes (water erosion and disaggregation), and occupies more than 10% of the terrestrial 82 surface (Legrand 1973). Karst landscapes are characterized by patchy soil and bare rock due to slow rates 83 of bedrock weathering, and it takes about 300-800 years to create 1 cm of residual soil (Zhu 1997, Nie 84 et al. 2011, Liu et al. 2018). In addition, the rate of soil formation is low and the permeability is high due resources are the key factor limiting community species diversity and productivity in karst ecosystems.

87
Previous studies have found that soil depth can impact species diversity and community productivity in 88 karst regions (Liu et al. 2020). However, how soil depth influences the mechanisms by which plant 89 diversity affects community productivity in karst regions is still unknown. Therefore, we established 90 artificial karst herbaceous communities with different soil depths and species richness levels, calculated 91 the net biodiversity effect on productivity, and then partitioned this effect into the complementarity effect 92 and selection effect. The main objectives were to examine: 1) the effects of soil depth, species diversity, 93 and different species combinations on karst herbaceous community productivity; 2) the relationship 94 between species diversity and productivity and the extent to which this relationship is affected by karst

104
The plant seeds and soil were collected from Zhongliangshan Haishi Park in Chongqing, a typical karst   (Table 2). To prevent random 117 interference, all treatments were replicated three times following a randomized complete block design.

182
The fixed effects, including soil depth, species diversity, and their interaction, all imposed a 183 significant influence on karst herbaceous community productivity (p < 0.05) ( Table 3). The plant 184 community productivities of different species combinations in deep soil were significantly higher than 185 those in shallow soil ( Figure 1). Additionally, the random effects and species combination also 186 significantly affected community productivity, but community abundance had no significant effect on 187 community productivity (Table 3). Productivity differed significantly among monoculture communities,

188
with the monoculture of Xanthium sibiricum being the highest while the monoculture of Arthraxon 189 hispidus was the lowest (Figure 1). It is worth mentioning that the highest productivity among the multi-190 species communities occurred when species diversity was greatest, referring to the mixture of X.

199
The net biodiversity effects (species diversity ≥ 2) on productivity were always greater than 0 in all

227
The species combinations (or species characteristics) also had a considerable impact on the 228 productivity of the experimental herbaceous communities (Table 3)  In this study, we also found that the impact of species diversity 241 on community productivity varied with soil depth (Figures 1b, 2, and 3). This could be because deeper

269
The contribution of the complementarity and selection effects to productivity can be affected by      The relationship between plant species richness and community productivity in (a) shallow soil depth, (b) medium soil depth, and (c) deep soil depth. Statistical signi cance of the regression models is indicated by * at p < 0.05 and ** at p < 0.01. Different lowercase letters indicate signi cant differences among species diversity at p < 0.05. Grey shading represents 95% credible intervals.

Figure 3
The effect size of species diversity on plant community productivity in different soil depths. Data are mean ± SE. S: shallow soil depth; M: medium soil depth; D: deep soil depth.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.