US Geological Survey - Southwest Biological Science Center Moab, United States
Climate change, grazing history, and the spread of invasive plants are commonly noted stressors that contribute to ongoing dryland degradation and can even result in the establishment of alternative steady states. Previous research suggests that intense historical grazing practices often severely alter the structure and function of semiarid ecosystems due to soil and biological soil crust (biocrust) degradation, decreases in perennial grass cover, and increases in non-native species abundance and distribution. Additionally, recent research across the Colorado Plateau suggests that climate change (i.e. increasing average air temperatures, increasing aridity) can also negatively impact soils, biocrust, and vegetation, therefore drawing into question the ongoing resilience of historically grazed semiarid grassland ecosystems. Here we present findings from a 20+ year long-term study in Canyonlands National Park, Utah, investigating variations in biocrust and vegetation cover across a historical grazing gradient that has created three distinct alternative steady states: a degraded annualized-bare system (previously heavily grazed), a grass-bare system (previous grazed only in winter), and a high biocrust and grass system (never grazed). Additionally, we combine our plot-level data with remotely sensed datasets (LandCART) and long-term climate data to estimate the total size, impacts, and potential trajectories of the three distinct dryland alternative states.
Results from this study show variations in ecosystem response to ongoing warming across the three alternative steady states. Within the annualized-bare system, there has been a notable decline in C3 grasses after 2001, both for native bunchgrasses and the non-native annual grass Bromus tectorum. Conversely, in the grass-bare system and the high biocrust system, native perennial grass cover for both C3 and C4 grasses has remained consistent, with observed interannual variation due to annual variations in ambient air temperature and precipitation patterns. Late-successional biocrust components (lichens and mosses) are the most abundant in the high biocrust system, yet that ecosystem has observed recent declines in heat-sensitive biocrust species bringing biocrust cover closer to that of the grass-bare system. The grass-bare system consistently has the highest cover of non-native plant species including both Bromus tectorum and Salsola tragus (Russian thistle). Future work will continue to build on this analysis by comparing the plot-level data across our gradient to larger landscape patterns to estimate the total landscape area in the three alternative steady states. Additionally, we will aim to map the alternative states through time to see if ecosystem types are shrinking or expanding across our study location.