Diminishing precipitation change effects on plants in grasslands from population to ecosystem levels

As one of the serious issues facing the world currently, precipitation changes are causing substantial shifts in earth's biotic systems. Precipitation affects plant growth processes at multiple ecological scales, including population, community and ecosystem level. However, the absence of reconciliation across different scales means that the impact of precipitation changes on plant growth continues to be a significant source of uncertainty in future climate projections. We explored the effects of precipitation change on grassland plant growth at population, community, and ecosystem level, by conducting a meta-analysis of 3625 observations from 168 studies worldwide. Our findings reveal that the effects of precipitation change on grassland plant growth differ significantly across various scales, and they are gradually weakening, from population (e.g., photosynthesis: d = 1.674, p < 0.001), community (e.g., carbon content: d = 1.598, p < 0.001), to ecosystem level (e.g., plant abundance: d = –1.267, p < 0.001). Further, at both the population and community levels, plant growth showed greater sensitivity to increases in precipitation (e.g., photosynthesis: population, d = 1.674, p < 0.001; community, d = 0.717, p = 0.001) compared to decreases (e.g., photosynthesis: population, d = –0.532, p < 0.001; community, d = –0.474, p = 0.45), and more sensitive to decreased precipitation at ecosystem level (e.g., plant abundance: increased, d = 0.152, p = 0.601; decreased, d = –1.267, p < 0.001). Our synthesis highlights that accurately quantifying the varying responses of plant growth necessitates explicit attention to organizational levels and the magnitude of experimental manipulation. Our finding has important implications for reliably modelling terrestrial carbon and water cycles, thus accurately predicting the impact of future climate change.