Effects of increased temperature on plant communities depend on landscape location and precipitation

Abstract

Global climate change is affecting and will continue to affect ecosystems worldwide. Specifically, temperature and precipitation are both expected to shift globally, and their separate and interactive effects will likely affect ecosystems differentially depending on current temperature, precipitation regimes, and other biotic and environmental factors. It is not currently understood how the effects of increasing temperature on plant communities may depend on either precipitation or where communities lie on soil moisture gradients. Such knowledge would play a crucial role in increasing our predictive ability for future effects of climate change in different systems. To this end, we conducted a multi-factor global change experiment at two locations, differing in temperature, moisture, aspect, and plant community composition, on the same slope in the northern Mongolian steppe. The natural differences in temperature and moisture between locations served as a point of comparison for the experimental manipulations of temperature and precipitation. We conducted two separate experiments, one examining the effect of climate manipulation via open-top chambers (OTCs) across the two different slope locations, the other a factorial OTC by watering experiment at one of the two locations. By combining these experiments, we were able to assess how OTCs impact plant productivity and diversity across a natural and manipulated range of soil moisture. We found that warming effects were context dependent, with the greatest negative impacts of warming on diversity in the warmer, drier upper slope location and in the unwatered plots. Our study is an important step in understanding how global change will affect ecosystems across multiple scales and locations.

Keywords: biodiversity; context dependency; global change experiment; open‐top chambers; precipitation; primary productivity.

Figure 1

Effects of experimental treatments on abiotic properties. The left panel illustrates the effects of open‐top chamber (OTC) treatment (OTC vs. Control) and slope location (lower vs. upper slope) on (a) average soil temperature, (c) average soil moisture, and (e) total plant available nitrogen (${\text{NO}}_3^{-}+{\text{NO}}_4^{+}$) across the 2012 growing season. The right panel shows the effect of climate manipulation (OTC vs. control) and precipitation (control vs. added water) on (b) average soil temperature, (d) average soil moisture, and (f) total plant available nitrogen (${\text{NO}}_3^{-}+{\text{NO}}_4^{+}$). Error bars are standard error of the mean. Note that the upper slope location in the left graphs (OTC × slope experiment) is the same plots as the unwatered treatment in the right panel (OTC × water experiment). They are aligned for comparison, but due to the experiment design are analyzed as distinct models

Figure 2

The effect of open‐top chambers (OTCs) (OTC vs. Control) and slope location (lower vs. upper slope; left panel) and OTCs and watering treatment (right panel) on productivity based on biomass (a,b) and percent cover (c,d) and diversity (e ^H′) calculated based on biomass data (e,f) and total percent cover data (g,h). Means shown are unadjusted averages for each treatment. Error bars are standard error of the mean. Note that the Upper slope location in the left graphs (OTC × slope experiment) is the same plots as the unwatered treatment in the right panel (OTC × water experiment). They are aligned for comparison, but due to the experiment design are analyzed as distinct models