Background/Question/Methods:   Numerous studies have shown that past land use is important in structuring modern ecological communities in terrestrial systems. For example, data across forest, grassland, and desert ecosystems suggest that post-colonial and prehistoric agricultural activities can affect species assemblages, biogeochemical cycling, and ecosystem responses to interacting stressors such as climate change and atmospheric deposition. But are all prehistoric activities equal in their long-term impacts, or are there some which “disappear” from the ecological canvas over time? In collaboration with archaeologists, we are exploring this question across a gradient of human land use intensity in arid desert ecosystems of the southwestern US which supported agroecologically active and well studied populations of humans until 1200-1400 AD.

Results/Conclusions:   In a desert grassland, herbaceous plant communities on prehistoric agricultural terraces used for dryland farming were subtly but significantly more diverse (Shannon-Weaver H’) and even (Pielou’s J) than plant communities on agricultural features that had been intentionally cleared of rocks or on natural terraces without significant evidence of human use. Farmed terraces were also composed of finer textured soils (silt + clay) than other agricultural features, including rock-free areas used for agave production, or natural terraces. Preliminary data, however, suggest that prehistoric activity has not left long-term legacies in soil nutrient pools. Vegetation patterns in this arid ecosystem may be associated primarily with differences in soil properties between landscape features that alter water availability.