Predictors of soil functions and aboveground-belowground interactions
Plant functional groups (PFGs) emerged as the dominant predictors of soil functions, explaining approximately 81% of the variance. Plant species richness and PFGs collectively accounted for the majority of the explained variance across four of the five soil functions we examined. However, microbial diversity primarily explained variance in PHOS activity in pre-drought measurements (Figure 2a). These findings were robustly supported by our statistical models (Table 1), underscoring the predominant role of PFGs, particularly N-fixing plant group, in driving all soil functions, both independently and in synergy with microbial diversity, and independent from drought disturbance. Importantly, it is essential to note that the contribution of PFGs extends beyond functional group diversity in non-monoculture contexts. Conversely, microbial diversity exerted a comparatively low influence on soil functions associated with nitrogen cycling during pre-drought sampling. However, its importance increased notably during the drought period, where it shared equal explained variance with PFGs (Figure 2b). These results were further validated by regression models, which identified interactions between PFGs and microbial diversity as the best-fit models with the lowest AIC values (Table 1). While drought had a negligible direct effect, accounting for less than 1% of the variance, it had a significant impact on microbially driven PHOS activity. Following a month of recovery from the drought, we observed the soil functions displaying signs of recuperation. Approximately 78.9% of the variance in these recovered soil functions was attributed to PFGs, with plant species richness and microbial diversity contributing to approximately 12.9%, and 8.1% respectively (Figure 2c).
Relationship between soil functions and plant species richness across microbial dilution gradient
We examined the intricate relationships between microbial and plant diversity by assessing correlations between plant species richness and soil functions within individual soil dilutions. Additionally, we explored associations between microbial communities and soil functions at various levels of plant richness. Notably, we observed a negative correlation between plant species richness and ammonium concentrations in D0 and D2 (ρ = -0.32, p < 0.05), which were statistically insignificant in D6 (ρ = -0.16, p > 0.05) (Figure 3). Conversely, plant species richness exhibited a consistent negative relationship with both ammonium and nitrate levels across all soil diversity levels (Figure 4i). The impact of drought on nitrogen cycling processes was particularly pronounced in D6. This resulted in a shift in the relationship between plant richness and nitrogen levels, with higher microbial diversity in D0 and D2 stabilizing the effect (Figure 4(ii)). Importantly, after the recovery period, the relationships largely returned to their pre-drought negative associations (Figure 4(iii)). Plant species richness showed a significant positive correlation with phosphate only in soils with higher microbial diversity (D0 and D2). However, this relationship was absent in D6 soils (R2 =0.02, p =0.24) (Figure 4c(i)). While plant species richness had no discernible effect on PHOS activity during the plant establishment stage (ρ = -0.04, p > 0.05; Figure 3i, and Figure 4e(i)), the drought phase resulted in a significant positive correlation across all soil diversity levels (Figure 4e(ii)). In case of N mineralization, increasing plant species richness had a positive influence (ρ = 0.35, p < 0.05; Figure 3i, and Figure 4d(i)) during pre-drought conditions. However, the drought period led to a decoupling of this relationship, which was further influenced by the soil biodiversity level (Figure 4d(ii)). Given that higher plant species richness likely corresponds with increased nutrient demand, a negative correlation (ρ = -0.37, p < 0.05) with nitrate levels was observed (Figure 3i). Overall, the presence of plants significantly influenced the relationships between microbial diversity and soil functions during the plant establishment stage (pre-drought). Components of the nutrient pools were notably affected with increasing plant species richness (Figure 3i and 4).
Plant functional groups as the main predictor of soil functions during drought and recovery
Individual PFGs played a significant role in shaping the intricate relationships between microbial diversity and soil functions (Figure 3 and Table 1). To gain a more nuanced understanding of their contributions, we employed multiple linear regression models to dissect the unique roles of each PFG in multitrophic plant-microbial interactions (Table 2, and Figure 5). During the early flowering stage (pre-drought), all PFGs and their combinations exhibited similar impacts on soil functions (Figure 3). However, when drought was induced, extractable NH4, NO3, and N mineralization became significantly influenced by PFGs (Table 1). This influence was primarily driven by the presence of N-fixing plants, both in isolation (p < 0.05) and in combination with other species (p < 0.05) (Figure 5). Notably, NH4 exhibited a negative correlation with N-fixing plants, whereas NO3 and N mineralization displayed a positive relationship with N-fixing plants (Table 2 and Figure 5). Following the recovery from drought, many of these effects of PFGs on soil functions also rebounded. The relationships between soil extractable phosphate and PHOS activity with microbial richness remained largely unaffected by PFGs, with the exception of the presence of C4 plants during the recovery phase (Table 2 and Figure 5).
DISCUSSION
This study represents a substantial advance in our understanding of the intricate relationships among microbial and plant diversity, particularly of plant functional groups (PFGs) and their impacts on essential soil functions under water-limiting conditions. Our results are supported by previous research, emphasizing the potential adverse effects of microbial and plant diversity loss on ecosystem functions , while advancing the discipline by revealing that the composition of plant communities, particularly their functional groups, could be a dominant predictor of soil functions that could also affect microbial response to environmental stressors. Crucially, our findings contribute further evidence to the growing body of research that highlights the sensitivity of biotic interactions between plant and microbial diversity to biodiversity loss . Of note, our results underline that the composition of plant communities, with a special focus on functional groups, significantly impacts the rate and resilience of key soil functions . This highlights the imperative need for holistic evaluations that encompass both above- and belowground biodiversity loss, coupled with explicit consideration of functional groups, in the development of effective management and conservation policies.
One of the intriguing aspects we observed is the dynamic nature of the effects of PFGs and diversity on soil functions over time and in relation to various microbial processes. At the early pre-planting stage, we detected a negative impact of soil microbial diversity loss on soil nitrogen cycling and inorganic nitrogen concentrations. The decrease in soil bacterial diversity was associated with a significant reduction in nitrate accumulation in low-diversity soil (D6), suggesting a direct negative relationship between the rate of nitrification and microbial diversity . The loss of the nitrifier bacterial community, responsible for converting ammonium into nitrate, coincided with declining microbial diversity . This effect was likely amplified in the absence of plants at the pre-planting stage. Similarly, we observed low mineralization rates due to the higher microbial consumption of ammonium and excess accumulation of nitrate, where no competition occurred between different microbial communities and no plant nutrient demand was present .
During the early flowering stage (pre-drought), the impact of plants on soil functions, particularly the N cycle, was predominant, explaining approximately 98% of the variance. This effect was not surprising, as it was primarily attributed to the presence of N2-fixing plants, known for their influence on N pool dynamics . More novel however was identifying the fundamental role of PFGs in elucidating how ecosystem functions respond to climate change . For example, PFGs emerged as the primary factor accounting for the majority of the effects of plant species richness on soil functions during drought. In our statistical analyses, it is important to acknowledge the potential confounding effect between PFGs and plant species richness, where the presence of both variables in the model may make it challenging to disentangle their individual contributions to observed outcomes.
Specifically, our findings align with the concept that the influence of plant diversity on ecosystem functions is intricately linked to the presence of greater plant functional diversity . This is exemplified by a positive correlation we observed between the rate of N mineralization and microbial richness in the presence of legumes during drought conditions. Legumes increase soil N availability by N2fixation with microbial symbionts, and the mineralization of N-rich legume litter . Therefore, presence of legumes increases N mineralization rates, but soil biodiversity loss may lead to decrease in legume performance and hence soil functions . For example, a recent study reports the detrimental effect of soil biodiversity loss, which decreased the plant yield under drought conditions . However, a different pattern emerged in the case of the P-cycle. Plant species richness had minimal impact on phosphate across all microbial diversity levels. Still, it did strengthen the relationship of PHOS enzyme activity with plant species richness and showed a significant positive correlation with microbial richness. Notably, plants themselves play a role in P cycling as plant roots exude phosphatases as well as carboxylic acids that can solubilize PO4 . In particular, N2-fixers have a high P demand, possibly explaining why an increase in PHOS activity was observed, without an increase in P. This finding is consistent with a previous study which observed that biodiversity strengthens plant and soil feedback and hence increases P cycling . These findings significantly advance our understanding of how environmental stressors alter the relative contributions of microbial communities to soil functions . Moreover, global change drivers, including warming, drought, land use intensification, and other anthropogenic activities, pose threats to both plant and soil biodiversity . In line with our hypothesis, the results indicate that the strength and direction of plant diversity’s effects on soil functions are influenced by microbial diversity during drought.
Specifically, under drought conditions, microbial diversity become one of the most important drivers of function. The likely reduction in plant metabolic and growth rates during drought, combined with the microbial ability to sustain these functions under stress, can elucidate this shift in relative contribution . Microbes engage in dynamic interactions with plants and enhance their resilience during changing environmental stress. As an example, microbes are known to release extracellular polymeric substances into the soil, enhancing water retention capacity . While a previous study reported no significant impact of microbial diversity loss on primary production , our findings in this study indicate a positive influence of microbial diversity in reinforcing the effects of plant diversity on soil functions. These effects are relative and context-dependent. Therefore, these varying responses could be attributed to differences in resource availability and other abiotic conditions , indicating a context-dependent relationship between plant-soil biodiversity and functions under climate change. This is further supported by the finding that fungal richness may have been indirectly influencing plant composition importance in predicting aboveground functions (.
Most importantly, our results suggest that PFGs play a pivotal role in driving context dependency in the BEF relationships supporting and helping explain prior observations. For example, our results are consistent with the few studies that examine plant functional group identity and richness independently of species richness . Our findings also imply that the effects of biodiversity on ecosystem processes exhibit variation based on the types and traits of plants involved both within and among PFGs . This is consistent with evidence that increasing the number of PFGs in grassland plots increased primary productivity , especially if the PFGs differed in their resource use strategies, such as nitrogen fixation or rooting depth and even when holding species richness constant . Similarly, a meta-analysis by showed that the impact of biodiversity on ecosystem functioning was stronger when PFGs were defined based on traits related to resource acquisition and allocation, rather than taxonomic affiliation. These reports exemplify how PFGs can modulate the BEF relationship by influencing the complementarity and redundancy of species within ecosystems.
In conclusion, this study employed an innovative experimental design to provide mechanistic insights into the relative significance of plant and microbial diversity, PFGs, and their interplay in shaping essential soil functions amid environmental stressors. The results clearly illustrate that while plant diversity can enhance soil functions, increasing plant species richness cannot fully compensate for the loss of microbial diversity. Moreover, the decline in microbial diversity directly influences the magnitude of plant diversity impact on soil functions. The significance of soil microbial diversity was particularly pronounced in the under and environmental stress, where PFGs mediate microbial responses to drought. Hence, both plant species richness and PFGs play pivotal roles in shaping soil functions . These findings underscore the vital need to conserve both microbial and plant diversity, paying close attention to PFGs, in order to maintain the stability of multiple soil functions in a changing world and advances our understanding of the intricate interactions between above- and belowground biodiversity and their combined effects on soil functions. It provides a new line of future investigation to unravel the specific mechanisms governing these interactions and elucidate their broader implications for sustainable agriculture and ecosystem management. Such knowledge is essential for the development of effective strategies to preserve both above- and belowground biodiversity, ensuring the long-term health and productivity of our ecosystems.