The Response Of Productivity And Its Sensitivity To Changes In Precipitation: A Meta-Analysis Of Field Manipulation Experiments

Questions Changes in precipitation can influence grassland productivity and its sensitivity to precipitation variation, but general patterns remain elusive.Methods Using data from 59 studies of experimental precipitation manipulations in grasslands, we tested the effect of increases and decreases in precipitation on above-ground net primary productivity (ANPP), below-ground net primary productivity (BNPP), forb and grass productivity, and their sensitivity to changes in precipitation.Results Our results demonstrated that: (a) ANPP sensitivity to increases in precipitation was 87.3% greater than to decreases in precipitation, whereas there was no difference in the sensitivity of BNPP to increases and decreases in precipitation; (b) the sensitivity of grasses was 315.5% higher to decreases in precipitation than to increases in precipitation, but there was no difference in the sensitivity of forbs to increases and decreases in precipitation; (c) the sensitivity of ANPP to increases and decreases in precipitation was 620.8% and 91.7% higher than that of BNPP, respectively; (d) the sensitivity of grasses to precipitation decreases was 157% higher than that of forbs, but there was no difference in the sensitivity of grasses and forbs to increases in precipitation; and (e) the sensitivity of ANPP to increases in precipitation and BNPP to both increases and decreases in precipitation reduced with increasing mean annual precipitation (MAP), and the reduced magnitude in BNPP sensitivity was higher to precipitation increases than to precipitation decreases. In contrast, neither grass nor forb sensitivity varied with changes in MAP.Conclusions Our study suggests that grassland productivity and its sensitivity to changes in precipitation not only depend on the magnitude of precipitation but also on the direction of precipitation change. These findings are important for the management of grasslands under future precipitation change scenarios.