The ecology of catastrophic events: Understanding abrupt spatial transitions in susceptibility of grasslands and croplands to multi-year drought

Background/Question/Methods

Much of the central grasslands region (CGR) of North America experienced a multi-year extreme drought in the 1930s that resulted in broad-scale plant mortality, massive dust storms, and losses of soil and nutrients. All grassland types in the CGR were affected, but to varying degrees. The southern mixedgrass prairie had the most severe impacts on vegetation and soil erosion; these grasslands were also well-studied by early eminent ecologists (e.g., Weaver, Albertson). In some areas, the transition between severely impacted and less impacted sites occurred over short distances, yet our understanding of the drivers of these abrupt spatial transitions remains limited. For example, native prairie sites in southeastern Nebraska suffered high plant mortality while sites on similar soils located < 100 km to the east in southwestern Iowa only had reductions in cover.  Our goal was to synthesize historic data and maps from the 1920s to 1940s along this transition zone in order to determine the climatic, soil, and land use drivers of these abrupt changes in vegetation. We analyzed data through time from the literature for native grasslands and from the USDA Census of Agriculture for cropland to compare land use effects on vegetation change.

Results/Conclusions

Our results show that the transition from reductions in native grass cover in Iowa to increase in grass mortality in Nebraska in the 1930s drought was related to small increases in temperature and subtle decreases in precipitation along this spatial gradient. Although soil texture was similar, long-term plant available water decreased from east to west corresponding to an increase in grass mortality and increase in susceptibility to drought.  Reductions in corn yield in the 1930s also decreased along this east to west gradient in plant available water with lower yields for more years in Nebraska compared with Iowa. These abrupt transitions in vegetation response would have resulted in an increase in ecosystem susceptibility to multi-year drought, and an increase in soil erosion by wind and water through time in the grasslands and croplands where grass mortality occurred. These cumulative effects of multi-year drought challenge our ability to predict ecosystem responses to catastrophic events without long-term data on these types of events.