Impacts of grazing, fire, and precipitation variability on woody plant cover expansion in semi-arid grasslands of southeastern Arizona

Northern Chihuahuan semi-arid grasslands are highly managed systems supporting rich biodiversity and many endemic species as well as providing a valuable economic resource for cattle-ranching livelihoods, with 90% of grasslands open to grazing. Chihuahuan grasslands share many characteristics with other managed grazing systems, which occupy 25% of the global land surface and are the most extensive form of land use. These grasslands are experiencing land-cover modification from woody plant cover expansion, leading to diminished biodiversity and grazing capacity. Ongoing research indicates that grazing, fire suppression, and precipitation variability are the primary drivers causing increased woody plant cover in Chihuahuan grasslands; however, there is debate concerning the dominant driver. While it is understood that historical land use and climate variation have facilitated initial woody encroachment in the region, the current relative influence of the three drivers remains unclear. This research explores how grazing, fire/suppression, and seasonal precipitation variability influence woody plant cover in the semi-arid grasslands of southeastern Arizona and identifies the dominant driver behind observed changes. This research used the Landsat Thematic Mapper record from 1984 to 2008 to map changes in woody plant cover and identify spatial patterns and temporal trends of woody plant cover expansion. Spectral mixture analysis (SMA) was used to quantify the percent of woody plant cover in each pixel; trend analysis was used to track per-pixel changes over the time-series. Trend analysis was further refined by segmenting trends around fire events to accommodate abrupt and non-monotonic effects of fire on woody plant cover. The overall trend in the region shows increasing woody plant cover with most values ranging between 5-20% over the 25-year period and significant spatial variability in expansion amounts across the region. The Random Forests decision tree approach was used to rank the influence of grazing, fire, and precipitation variability and determine the dominant driver of woody plant cover change in the region. In addition to the three primary independent variables, initial woody plant cover, topography, soils, and drainage variables were also added. The Random Forests ranking results clearly identify one driver as having a substantial influence over woody plant cover dynamics relative to the other drivers. In addition, they show the drivers of change in this region extend beyond just grazing, fire, and precipitation, indicating that legacy effects and other variables are also important drivers of change in the region. Results will enable resource managers to better understand the dynamics of woody plant cover expansion within the region and allow scientists to develop new models and applications to answer both natural and societal land management questions and support decision-making processes.