Influence of soil disturbance, seed pressure, aridity, and burn history on community structure of winter annuals in North American deserts

Background/Question/Methods

Desert ecosystems are currently showing a dramatic increase in the arrival and spread of non-native species. Traditionally, annual plants have patchy distributions in desert scrub ecosystems, being restricted to nutrient rich areas beneath shrubs, and are not found in open areas between shrubs. We hypothesize human intervention through soil disturbance and increased seed pressure, along with changes in aridity and the fire regime will change plant community structure and species associations. We investigated the density of native and non-native annuals in relation to shrubs in the Mojave and Sonoran Deserts both before and after applying environmental treatments. We used both indirect gradient (DCA) and direct gradient (CCA) analyses to explore how community structure is related to soil disturbance, seed pressure, aridity, and burn history.

Results/Conclusions

Prior to treatment in the Mojave Desert, indirect gradient analysis revealed several distinct groupings of species – species mostly likely to be found together than with other species. Direct gradient analysis revealed that species in those groups of species are linked to the microhabitat and shrub side (north or south of shrub) environmental variables. There were no distinct species groups in the Sonoran desert before treatment, but rather a continuum of species along a gradient. Direct gradient analysis reveals that this gradient appears to be strongly linked to microhabitat. In both deserts our focal species, the invasive grass Schismus was found close to the centroid in both deserts, showing no strong association with a given side of the shrubs or microhabitat. The first winter post treatment in the Mojave Desert was a dry year and there were few plants found, and those data were not analyzed. In the Sonoran, we compared CCA analyses from pre treatment with post treatment controls and found similar community structure with regards to microhabitat and transect direction. When we added the environmental variables we found that microhabitat was still one of the strongest influences on community structure, but that abundances of specific species and communities were very strongly influenced by treatments.