2018 ESA Annual Meeting (August 5 -- 10)

COS 135-2 - Evaluating the impact of brush management on ecosystem services in desert grasslands

Friday, August 10, 2018: 8:20 AM
333-334, New Orleans Ernest N. Morial Convention Center
Adam T. Naito1, Steven R. Archer1, Philip Heilman2 and Katharine I. Predick1, (1)School of Natural Resources and the Environment, University of Arizona, Tucson, AZ, (2)USDA-ARS Southwest Watershed Research Center, Tucson, AZ
Background/Question/Methods

Encroachment of unpalatable woody plants into North American grasslands has been an on-going phenomenon since the beginning of the 20th century. Land managers have traditionally used brush management (BM) to limit shrub encroachment, restore lost forage production, and improve groundwater recharge. Within these narrow perspectives, BM is often not an economically feasible grassland management tool. However, numerous other ecosystem services (ES) are potentially impacted by BM, and an accounting for them would provide a more complete assessment of the viability of BM as a conservation practice. Accordingly, we are examining the impacts of BM on a portfolio of ES (herbaceous diversity, forage production, net ecosystem exchange, carbon sequestration, erosion) at the watershed scale in a southeastern Arizona desert grassland encroached by velvet mesquite (Prosopis velutina). We collected pre-treatment data of these ES from September 2015 to May 2016 in each of four watersheds prior to applying an herbicide cocktail consisting of clopyralid, aminopyralid, and triclopyr. The herbicide was then applied on two watersheds in June 2016, while retaining the other two watersheds as controls. Following treatment, we established four permanent transects to evaluate mesquite foliage recovery and continued to track ES responses.

Results/Conclusions

Herbaceous diversity increased in the treated area, but this was primarily the result of an increased proportion of native and non-native annuals. Although herbaceous ANPP increased 25% the growing season following BM, the contribution of non-native perennials (48% increase) far exceeded that of native perennials (5% increase). Precipitation in the subsequent year was below-average and ANPP decreased on both BM and control sites, but more so on the latter than the former (65 vs. 33%). Taken together, BM-induced enhancement of ANPP in wet years may be offset by ANPP reductions in dry years. The herbicide application initially caused a 90% reduction in mesquite foliar cover. Foliar cover on treated plants in the year following BM was reduced relative to untreated mesquite (50% vs. 80%, respectively), but mesquite plant mortality was less than 1%. Thus, a single application of our herbicide cocktail was short-lived and not particularly effective in reducing mesquite canopy cover. Our preliminary results indicate that BM in this desert grassland may shift the community composition in favor non-native perennials and annuals over native perennials, perhaps triggered by periods of precipitation shortfall, with unintended adverse consequences for herbaceous biodiversity and ANPP.