Introduction
Many grassland ecosystems are subjected to livestock grazing, and these grazing systems provide up to a third of global food consumption and support the livelihoods of more than one billion people in the world(Suttie et al. 2005; Kemp et al. 2013). Increasing demand for livestock products has driven a global increase in grazing (Tilman et al. 2011; Kemp et al. 2013; Fetzel et al. 2017), which is threatening grassland biodiversity and ecosystem functioning (Schönbach et al. 2011; Eldridge et al. 2016). Yet, understanding how grazing impacts biodiversity, ecosystem functioning, and their relationship is still a major challenge in the context of climate change at broader spatial and temporal scales.
Biodiversity at multiple trophic levels (i.e., plants, animals, or microbes)and dimensions (i.e., taxonomic and functional diversity) respond differently to grazing due to their differential biological response mechanism(Fischer et al.2019; Wang & Tang 2019; Filazzola et al. 2020), which makes it difficult to estimate the overall grazing effects on biodiversity and its relationship with ecosystem functioning. Therefore, recent studies have acknowledged using multi-diversity by integrating different dimensions of diversity across multiple trophic levels simultaneously to reflect the overall effects of grazing on total local biodiversity(Allan et al. 2014; Wang et al. 2019). In addition, biodiversity becomes more important in maintaining multiple ecosystem functions (hereafter ecosystem multifunctionality, EMF) (Byrnes et al. 2013; Lefcheck et al. 2015; Gamfeldt & Roger 2017) than individual functions of productivity (Ma & Chen 2016; Zhang et al. 2018), stability (Loreau & Mazancourt 2013; Hautier et al. 2015), carbon storage (Yang et al. 2019), or nutrient availabilities(Komarov et al. 2018; Wang et al. 2020) based on a common positive biodiversity–ecosystem functioning (BEF) relationship in natural grasslands (Tilmanet al. 2001; Zhang et al.2018). Furthermore, evidence is also mounting that the positive BEF relationship can be facilitated by intermediate environmental stress (Baert et al. 2018; Guo et al. 2019). However, how grazing affects multi-diversity, EMF, and their relationships in managed ecosystems still remain poorly understood and a systematic assessment at the global scale is lacking.
Grazing effects on biodiversity and ecosystem functioning are highly dependent on intensity quantified by stocking rate or density (Schönbach et al. 2011; Herrero-Jáuregui & Oesterheld 2018), and grazing duration (Porensky et al.2017). For example, previous studies suggested that species diversity and aboveground net primary productivity (ANPP) is maximized at moderate grazing intensity due to reduced competitive exclusion and increased compensatory growth in the plant community compared with non-grazing condition(McNaughton 1983; Milchunas et al. 1988), whereas high-intensity grazing with long duration decreases plant diversity and community productivity by reducing abundance of annuals and several weedy species in the plant community (Porensky et al. 2017; Zhang et al. 2018), as well as soil water-holding capacity and nutrient availability (Zhang et al. 2017; Sitters et al. 2020). Moreover, the magnitude and direction of grazing intensity effects on biodiversity vary across the wide range of ecological contexts(Olff & Ritchie 1998; Gao & Carmel 2020). In relatively humid and productive grasslands, moderate grazing may increase plant species diversity through reducing the dominance of palatable species and improving the availability of limited resources (i.e., light and nutrients) to rare species colonization (Olff & Ritchie 1998; Gao & Carmel 2020). In relatively arid and low productive grasslands, intermediate grazing may reduce species diversity by increasing dominance of grazing-tolerant species and aggravating resources stress (i.e., water and nutrients) to rare palatable species (Herrero‐Jáuregui & Oesterheld 2018; Zhang et al.2018). Collectively, these disparities of grazing effect not only depend on grazing intensity, but also are driven by environmental gradients.
In addition, different domestic herbivores (e.g., cattle and sheep) may have differential impacts on biodiversity due to their different grazing behavior (i.e., distinct diet of selectivity) (Grant et al. 1985; Dumont et al. 2011; Bremm et al. 2012; Tóth et al. 2016). For example, compared to cattle, sheep are more likely to reduce taxonomic and functional diversity by reducing the amount of forbs at light or intermediate grazing intensities, but differences in selectivity between sheep and cattle may decrease with increasing grazing intensity(Tóth et al. 2016). Overgrazing and the use of inappropriate livestock may lead to grassland degradation and desertification (Tóthet al. 2016; Gao & Carmel 2020). Therefore, understanding how grazing intensity interacts with livestock type in changing grassland multi-diversity and EMF are essential for determining sustainable grazing management strategies.
Here, we conducted a global meta-analysis from 138 grazing intensity studies to evaluate grazing effects on the multiple biodiversity and ecosystem functioning in grassland ecosystems worldwide. Together, 16 biodiversity metrics across 5 groups (i.e., plant species diversity, functional group diversity, functional diversity, insect and soil microbial diversity) and 12 individual ecosystem functions (i.e., above- and belowground biomass, temporal stability of plant community, soil nutrients and moisture, net ecosystem productivity, ecosystem respiration and gross ecosystem productivity) were aggregated to estimate the multi-diversity and EMF, respectively, by weighted-averaging the natural log-transformed response ratio (lnRR) of each variable. In particular, we examine how the relationship between multi-diversity and EMF changes with intensifying grazing disturbance by incorporating the regulation of livestock types and grazing duration across a wide range of the aridity index.