Conceptual illustration of how grazing with wetness fluctuation would affect net ecosystem productivity (A), and geographic (B) and climatic distributions of data (C) included in this meta-analysis.

Effects of different grazing intensities on ecosystem carbon dioxide (CO2) fluxes and biomass in our field experiment.

CK, enclosure; LG, light grazing; MG, moderate grazing; HG, heavy grazing. Data (means ± SE, n = 3) followed by different lowercase letters indicate differences at P < 0.05 between treatments.

Relationships of the responses of gross primary productivity (GPP) and net ecosystem productivity (NEP) with potential regulating factors from 2019 to 2023 under different grazing intensity.

Overall, overall grazing; LG, light grazing; MG, moderate grazing; HG, heavy grazing. The significant regression lines and 95% confidence intervals are shown with lines (solid for significant) and shaded areas, respectively.

Effects of grazing on gross primary productivity, ecosystem respiration, net ecosystem productivity, aboveground biomass, belowground biomass and root shoot ratio, soil moisture and soil temperature, in global grasslands in the field experiment in this study (A) and from the meta-analysis (B).

Circles and error bars represent average parameter estimates and 95% confidence interval (CI). The star (*) indicates significance when the CI did not overlap zero. Sample sizes of observations for each variable are displayed on the left.

Effects of different grazing intensities on ecosystem CO2 fluxes, biomass and soil temperature and moisture in global grasslands from the meta-analysis.

Circles and error bars represent average effects and their 95% confidence interval (CI). The star (*) indicates significant effects, i.e., the 95% CI did not overlap zero. The sample size for each variable is shown on the left. LG, light grazing; MG, moderate grazing; HG, heavy grazing.

Responses of ecosystem CO2 fluxes and biomass to grazing (A) and different grazing intensities (B-C), at wetness index ≤ 30 and wetness index > 30 in global grasslands from the meta-analysis.

Circles and error bars represent average parameter estimates and 95% confidence interval (CI). The sample size for each variable is shown on the left. LG, light grazing; MG, moderate grazing; HG, heavy grazing.

Relationships of the responses of ecosystem CO2 fluxes with potential regulating factors in global grasslands from the meta-analysis.

The marginal (m) and conditional (c) R2 indicate the proportion of variance explained by the fixed effects and by both the fixed and random effects study), respectively. The black solid lines and gray shaded areas in panels are the mean and 95% confidence interval of the slope, respectively. The sizes of points are proportional to their corresponding weights. RR, response ratio; overall, total grazing; LG, light grazing; MG, moderate grazing; HG, heavy grazing.

Schematic summary of the effects of grazing intensity on ecosystem carbon dioxide (CO2) fluxes and biomass in global grasslands in the meta-analysis.

The arrows indicate negative effects. Asterisks (*) indicate significant effects on variables at P < 0.05. GPP, gross primary productivity; ER, ecosystem respiration; NEP, net primary productivity; LG, light grazing; MG, moderate grazing; HG, heavy grazing.