Positive species richness effect on AGB manifested through niche complementarity
Recent studies have provided evidence for the need to explore beyond species richness, how stand biomass and carbon relates to functional trait and structure-based diversity (Conti & Díaz, 2013; Dimobe, Kuyah, Dabré, Ouédraogo, & Thiombiano, 2019; Finegan et al., 2015; Hao, Zhang, Zhao, & von Gadow, 2018; Lin et al., 2016; Prado-junior et al., 2016; Wang et al., 2011; Zhang & Chen, 2015). Others researchers also documented the importance of phylogenetic diversity (Lasky et al., 2014; Satdichanh et al., 2019; Wasof et al., 2018). In this study, we compared the relative importance (and mediation) of functional trait, structure and dominance metric in explaining species richness effects on biomass stock. Interestingly, none of the single-trait functional diversity (FDvar) and dominance (CWM) metrics explained AGB variation in our stands. Contrary to some previous studies (Conti & Díaz, 2013; Finegan et al., 2015; Fotis et al., 2018), our study showed that FDvar and CWM of wood density and plant maximum height in these mixed species stands did not influence AGB. Lin et al. (2016) showed that biomass carbon responds most strongly to CWM values of wood density and maximum tree height in subtropical evergreen broad-leaved forest in China. Similarly, Wasof et al. (2018) showed that biomass of the forest understorey was mainly related to CWM of plant traits (leaf area and plant height) in temperate deciduous forests in Northern France. Some of our previous studies also showed that CWM of traits correlate strongly with biomass and carbon stock (S. Mensah, du Toit, et al., 2018; S. Mensah, Veldtman, Assogbadjo, et al., 2016; Sylvanus Mensah, Salako, Glèlè Kakaï, & Sinsin, 2020). Although not anticipated, the conflictual result is not surprising and might be due the complexity of these stands ecosystem structures, perhaps because of the low functional trait values (low level of functional diversity) of the dominant species.
Nevertheless, our analyses showed that among the multi-trait functional diversity indices, functional evenness had a significant and positive relationship with biomass, partly corroborating reports of positive relationship between multi-trait functional diversity metrics and AGB in forest stands (Dimobe et al., 2019; Hao et al., 2018; S. Mensah, Veldtman, Assogbadjo, et al., 2016; Rawat, Arunachalam, Arunachalam, Alatalo, & Pandey, 2019). Our finding is also in line with the general expectation that evenness should positively correlate with biomass production (Kirwan et al., 2007; Nijs & Roy, 2000; Wilsey & Potvin, 2000). Functional evenness reflects the evenness of species contribution to AGB (in this case) within the stand (Mason et al., 2005). As such, our result suggests that there is homogeneity in the distribution of the relative density (i.e., lower variance in the abundance of different species) across the multivariate trait space, reducing interspecific competition (Tilman, 1982). Because high values of functional evenness indicate effective resource utilization and development of productive communities (Kelemen et al., 2017) through niche complementarity and facilitative effects (Polley, Wilsey, & Derner, 2003; Polley, Wilsey, & Tischler, 2007), the positive functional evenness and AGB relationship (Figure 2) as well as the significant mediation role of functional evenness (Figure 4), as observed in this study are supportive of the niche complementarity, as a mechanism driving positive species richness effect on AGB in our mixed species stands.
Furthermore, the positive effects of structural diversity metrics on AGB stress the importance of niche complementarity hypothesis, as the main mechanism operating in these stands. Particularly, we found that CV DBH and CV Npb were the most important predictors of AGB, after functional evenness. These results indicate that structural diversity promotes AGB, as also pointed out in previous studies (Wang et al., 2011; Yan, Zhang, Wang, Zhao, & Gadow, 2015). The structural diversity metrics as computed here (i.e. CV DBH, CV Ht and CV Npb) reflect the amount of both intra and interspecific vertical and horizontal tree size and crown variation within the plot, (Seidel et al., 2019; Wang et al., 2011), and thus are somewhat indicative of resources capture and use by species (Yachi & Loreau, 2007). For example, greater intra and interspecific vertical and horizontal tree size and crown variations would translate into forest vertical stratification and crown complementarity, which allow for greater light infiltration and promote complementary use of light by trees in the subcanopy, canopy and above canopy layer, leading to higher performance at stand level, as previously shown in multi-storey Afromontane natural forest in South Africa (S. Mensah, du Toit, et al., 2018). We thus argue that the positive effects of both CV DBH and CV Npb on AGB result from resource-use efficiency and complementarity due to high structural (tree size and crown) differentiation, supporting facilitation and niche differentiation or complementarity, as also shown in spruce-dominated forest stands in Canada (Wang et al., 2011).