Resistance and resilience of a grassland ecosystem to varying drought intensities and durations

Background/Question/Methods

Over the next century, global climate models project more frequent and intense short-term extreme droughts (drought as a pulsed event) as well as long-term reductions in mean precipitation (chronic water stress as a press event). Assessing the effects of these different forms of water stress on ecosystem function through observations alone is challenging due to the rarity and unpredictability of such climatic events and the limited availability of long-term ecological data. Therefore other approaches are necessary to examine the ecological effects of climate extremes. Here I will present the results from two studies, a field-based experiment and a complementary modeling simulation, to examine the effects of both pulse- and press-droughts on a grassland ecosystem. In the field-based study, we experimentally imposed extreme drought (66% rainfall reduction) for two years at the Konza Prairie LTER site, then assessed recovery for two years. This allowed us to evaluate the resistance and resilience of ecosystem function (aboveground net primary production, ANPP) to drought and to identify the underlying mechanisms. In the modeling study, we examined long-term effects of both pulse- and press-droughts using the Community Land Model (CLM4.0), calibrated for the Konza Prairie. Precipitation inputs were altered in three ways for 10-year periods: a control treatment receiving average rainfall, a pulse-drought treatment in which 3 of the 10 years experienced a 66% rainfall reduction, and a press-drought in which rainfall was reduced by 12% each year (both drought treatments received equal total decadal rainfall).

 Results/Conclusions

In both field and modeling experiments, the grassland ecosystem lacked resistance to the extreme pulse-drought as evidenced by large reductions in ANPP (-60% field, -53% model), but both showed immediate recovery in the year following drought. In the field experiment, total ANPP completely recovered one year post-drought due to a rapid demographic response by the dominant grass. Thus, both experiments suggest that this grassland has low resistance to extreme drought, but high resilience. While the reduction in annual precipitation was more subtle in the press-drought treatment, the cumulative effects on total ecosystem carbon over the 10-year period were greater than the pulse-drought treatment (-3.1 vs. -1.5%). This is likely due to the quick recovery in function in the years following extreme drought. These results suggest that in this ecosystem, extreme short-term droughts are likely to have dramatic but transient effects on ecosystem function, while chronic reductions in mean precipitation may have greater overall effects on terrestrial carbon balance.