Abstract: Estimating the effects of extreme drought on the photosynthetic rates (Pn) of dominant plant species is crucial for understanding the mechanisms driving the impacts of extreme drought on ecosystem functioning. Extreme drought may result from either reduced rainfall amount or decreased rainfall frequency, and the impacts of different patterns of extreme drought may vary greatly. In addition, different grasslands likely appear various sensitivity to different extreme drought patterns. However, there have been no reports on the effects of different extreme drought patterns on dominant species Pn in different grassland types. Here, we conducted multi-year extreme drought simulation experiments (reducing each rainfall event by 66% during the growing season, CHR vs. completely excluding rainfall during a shorter portion of the growing season, INT) in two different grasslands (desert grassland vs. typical grassland) from 2014. The Pn of two dominant species in each grassland was measured in July and August 2017. Both CHR and INT significantly decreased dominant species Pn, with INT causing more negative impacts on Pn regardless of grassland types. The response ratios of Pn in desert grassland were generally higher than that of typical grassland, especially for Leymus chinensis in CHR. These results demonstrated that decreased rainfall frequency had a more negative effect on Pn compared with reduced rainfall amount, with grassland types changing the magnitude, but not the direction, of the effects of extreme drought patterns. These findings highlight the importance of considering extreme drought patterns and grassland types in ecosystem management in the face of future extreme droughts.

Key words: climate extremes, desert grassland, typical grassland, extreme drought pattern, seasonal drought