Resilience similar to the extent as a reference system is an ecosystem attribute ecological restoration aims to achieve that can be evaluated by productivity response to disturbance. One of the most widespread natural disasters in recent US history was the drought of 2012. Categorized as severe to exceptional, the 2012 drought brought costly consequences for row-crop agriculture, the land-use that has replaced much of the tallgrass prairie ecosystem. As such, grassland restorations in the Midwest and Great Plains often occur in formerly cultivated, deep lowland soils. There is little information on the degree to which restored grasslands are resilient to drought and respond comparably to reference grassland. To assess potential resilience of restored grasslands to drought, we evaluated aboveground net primary production (ANPP) data collected before, during, and after 2012 from three different grassland restoration experiments containing: (1) long-term restored prairie compared to never-cultivated prairie; (2) short-term restored prairie established in average precipitation and drought years; and (3) prairie restored with local and non-local ecotypes across a climate gradient. Means were calculated from independent plots in each experiment each year. Year effects (differences from 2012) were analyzed using repeated measures within analyses appropriate for each experimental design.
Results/Conclusions
Restorations sown with local ecotypes or cultivars of native grasses exhibited resilience to the 2012 drought, with the oldest restoration responding similarly to native prairie. In 2012, total ANPP in 15-yr restored prairie (424 g m-2 y-1) was 32% lower than the average total ANPP from three prior years and was representative of ANPP in lowland native prairie (406 g m-2 y-1). Total ANPP increased to 714 and 703 g m-2 y-1 in the restored and native prairie the year following drought, respectively. Newly restored prairie established under average rainfall conditions exhibited no negative impact from the drought and a linear increase in total ANPP from 2010-2014. Prairies restored across a longitudinal gradient showed either: no strong variation in total ANPP among ecotypes or in response to drought; greater resilience of grass ANPP to drought in prairie sown with local grass ecotypes; or no effect of drought on the linear increase in ANPP of all grass ecotypes, out of which the local ecotype increased at the fastest rate. Restored grassland resilience to the 2012 drought was attributed to the predominance of warm-season grasses in each experiment, the cover of which is inversely related to diversity, another common restoration goal.