Plant diversity loss has limited effects on below-ground biomass and traits but alters community short-term root production in a species-rich grassland

Abstract

1 INTRODUCTION

The interplay between biodiversity and ecosystem functioning is a central and long-standing question in ecology. According to biodiversity–ecosystem functioning (BEF) experiments, frequently conducted on grassland communities, there is a general consensus about the positive association between biodiversity and productivity. The complementarity effect is one of the proposed explanations for the enhanced ecosystem functioning in diverse communities (Cardinale et al., 2007; Loreau & Hector, 2001; Tilman et al., 2001). This effect stems from the fact that, in more diverse communities, variations in species strategies are expected to optimize resource use, resulting in increased productivity. Although complementarity effects can arise from a wide range of mechanisms, in plants, they have often been linked with horizontal (von Felten & Schmid, 2008) or vertical niche differentiation (Li et al., 2018), especially below-ground, where the distribution of roots is expected to exhibit more variability in diverse plant communities (Brassard et al., 2013). The other proposed mechanism explaining the experimental BEF positive relationship is the selection effects, wherein the likelihood of a high-functioning species being present in an assemblage increases with the number of species (Cardinale et al., 2011).

In the last decades, there has been a debate on whether results from BEF experiments can provide insights into the effects of biodiversity loss on ecosystem functioning in natural communities (Doherty & Zedler, 2018; Lepš, 2004; Wardle, 2016). Most BEF experiments rely on random selections of species from a species pool, which often does not represent realistic communities on which species loss can be simulated. Although such experimental designs offer controlled simulations that can provide important mechanistic information (Eisenhauer et al., 2016), they can be misleading in some aspects as natural species loss does not behave as a random process (Schläpfer et al., 2005; Schmid et al., 2002). Indeed, the probability of a species to be lost from a community depends on multiple factors, but primarily on its relative abundance (Genung et al., 2020; Smith & Knapp, 2003; Spehn et al., 2005). Furthermore, most BEF experiments are based on evenly sown communities, where all species are initially sown with equal abundance, which fails to represent the environmental filters and biotic factors that govern assembly processes (Diaz et al., 2003). On the other hand, removal experiments in natural communities provide a promising alternative approach. Following a realistic scenario of species loss, it has been shown that a few dominant species could compensate for the removal of rare and subordinate species and mitigate the effect of diversity loss on productivity (Lisner et al., 2023; Sasaki et al., 2017; Sasaki & Lauenroth, 2011; Smith & Knapp, 2003), further questioning the transferability of traditional BEF experiments to real-world communities.

Although most BEF experiments to date have focused on above-ground productivity (Cardinale et al., 2007), some studies have explored other ecosystem functions (Spehn et al., 2005; van der Plas, 2019), including below-ground productivity. However, the latter have mainly followed the same traditional experimental set-up with artificial communities (Mueller et al., 2013; Oram et al., 2018; Ravenek et al., 2014) consistently showing a positive relationship between species richness and root biomass. In natural communities, some observational studies found this relationship as well (Mommer et al., 2010; Zeng et al., 2020) but others contradict the idea of any effect of species richness on root biomass (Archambault et al., 2019; Meinen et al., 2009).

Whether the results obtained on the above-ground biomass would transfer to the below-ground biomass remains an open question since high species complementarity could be expected below-ground, due to the more symmetrical competition for resources (Lamb et al., 2009). Moreover, species above-ground biomass may not be correlated with below-ground biomass due to large variations in biomass allocation between species and years (Ottaviani et al., 2020). To address various knowledge gaps in BEF theory related to below-ground plant characteristics, our study examines the impact of a realistic plant diversity loss on below-ground functions. Employing a removal experiment on a semi-natural grassland, a gradient of species richness was established based on the observed dominance structure. This approach allows for the simulation of a more realistic scenario of species loss, integrating environmental and biotic factors inherent in natural communities. We compared above-ground and below-ground biomass along the diversity gradient after 3 and 6 years of species removal. Root characteristics were also determined in different soil layers. Finally, ingrowth cores were used to measure the spring (3 months) and annual root production in all plots. Comparing our results with the outcomes of classical BEF experiments using artificial communities will help us to test the reliability of previous work to assess the consequences of species loss on natural ecosystems. Our study aims to answer the following questions:

1. Does short-term root production and long-term standing root biomass decrease with species loss?
2. Is there any impact on the vertical distribution of roots? And on their traits?

We hypothesize that species loss may lead to a decrease in root biomass and alterations in fine root vertical distribution due to the loss of complementarity effects and reduced interspecific interactions. Regarding root traits, measured at the level of the community, we also expect a shift towards trait values associated with a lower acquisition rate and, more importantly, a lower trait variability with the loss of subordinate and rare species. Furthermore, we anticipate that plots with higher diversity will colonize more rapidly the available soil space and have higher root production in ingrowth cores, especially at short term.