Biodiversity serves as the fundamental underpinning for ecosystem functions and services. As a result of human-induced global change, there is a growing awareness of the substantial alterations in terrestrial aboveground biodiversity, particularly within alpine regions. However, it remains uncertain whether belowground biodiversity will exhibit similar responses, both in terms of magnitude and manner, to anthropogenic global changes as aboveground biodiversity.

Here, we conducted a meta-analysis to assess the impacts of warming, nutrient addition, and grazing on plant and soil microbial biodiversity in alpine grasslands on the Qinghai-Tibetan Plateau, which are known to be climate-sensitive and vulnerable. The analysis included 819 experimental observations from 152 studies, focusing on species richness, Shannon diversity, and Pielou’s evenness.

We found that plant biodiversity exhibited greater sensitivity to climate warming and anthropogenic activities compared to soil microbial biodiversity. Specifically, plant richness and Shannon diversity were reduced by warming and nutrient addition, while plant evenness was increased by grazing. However, only microbial richness was increased by grazing and microbial evenness was increased by warming slightly.

The responses of biodiversity to climate warming and anthropogenic activities were modulated by multiple factors. Specifically, the negative effects of warming on plant biodiversity were more pronounced in long-term experiments under warmer or drier environmental conditions. The negative effects of nitrogen addition on biodiversity were enhanced by the intensity and duration of nitrogen treatment. Appropriate intensity and frequency of grazing were beneficial to sustaining plant biodiversity. Soil microbial biodiversity was weakly regulated, where bacterial Shannon diversity was more sensitive to nutrient addition, while fungal species richness was sensitive to grazing.

Synthesis: Our findings reveal a mismatch between aboveground plant and belowground microbial biodiversity in response to climate warming and anthropogenic activities in alpine grasslands, with plant biodiversity being more sensitive. In the context of future global change, plant biodiversity may be at greater risk than soil microbial biodiversity. In addition, biodiversity responses of different experimental and environmental conditions should be distinguished, and more attention is needed on biodiversity conservation in alpine steppe, or areas with warmer and drier environmental conditions, high-intensity fertilization or heavy grazing. 

We searched the literature for the responses of biodiversity to warming, nutrient addition, and grazing from 1900 to 2022 in the Web of Science and China National Knowledge Infrastructure. The search string is the following key-word combinations:(((tibet*) AND (grassland OR meadow OR steppe) AND (diversity OR richness) AND (plant OR micro* OR bacteria* OR fung*)) AND ((warming OR increasing temperature OR elevated temperature OR climate change) OR (fertiliz* OR nutrient addition OR nitrogen addition OR phosphorus addition) OR (graz* OR livestock grazing))).

We screened studies using the following criteria: (i) based on field manipulative experiments containing control and treatment groups; (ii) contained at least one of the following biodiversity indices: richness (species richness for plants and operational taxonomic unit for soil microbes), Shannon–Wiener diversity index, and Pielou’s evenness index; (iii) included only natural grassland ecosystems, excluding farmland and wetland ecosystems; (iv) included only bacterial and fungal community diversity, excluding special functional groups except arbuscular mycorrhizal fungi (AMF). We extracted the mean, standard deviation, standard error, and sample size from the tables or extracted these numerical values from digitized graphs using Engauge Digitizer (v.10.8). After the selection process (See PRISMA in Fig.S1), a total of 152 published articles and 819 observations were included in this meta-analysis.

We also collected data defining experimental conditions (the duration, intensity, and method of treatment), grassland types (alpine meadow and alpine steppe), environmental conditions (mean annual temperature (MAT), mean annual precipitation (MAP), and altitude) and taxa (bacteria and fungi), from the original study or papers cited in that study. The WorldClim database (www.worldclim.com) was used for supplementation according to the longitude and latitude of study sites if MAT and MAP were not reported.