RESUMO
Rainfall use efficiency (RUE) is crucial for understanding the changes in grassland productivity due to variations in future rainfall patterns. Recently, numerous studies have been conducted on the relationship between RUE and the amount of rainfall, but there has been little research on the influence of rainfall distribution and the interactive effect of rainfall amounts and distribution on RUE. Here, a simulated rainfall experiment was conducted to evaluate the impacts of rainfall amount (average rainfall amount (R0), 334 mm; decreased (R-) and increased (R+) rainfall amounts, 233 mm and 434 mm, respectively) and dry intervals (comprising 6-day, 9-day, 12-day, 15-day, 18-day and 21-day intervals between rainfall) on productivity and RUE in Leymus chinensis (Trin.) Tzvel., a dominant grass of the Eastern Eurasian Steppe. Our results showed that (1) for biomass production and RUE, moderate extension of dry intervals was conducive to enhancing total biomass production and RUE. The peak values of total biomass and RUE appeared during the 15-day interval for R-, and the 18-day interval for R0 and R+. (2) For biomass allocation, extension of dry intervals decreased the stem to leaf ratio (S/L) and the root to shoot ratio (R/S). (3) Further, the S/L ratio was significantly negatively correlated with RUE. These results suggest that variations in rainfall patterns can alter the RUE by changing the S/L ratio, and finally influence biomass production in L. chinensis. These findings have important implications for understanding and predicting the effect of future climate change on productivity in semi-arid grassland.