Get reliable science

Semi-arid rangelands in California face multiple threats from interactions of invasive weeds and drought. To assist ranchers in improving rangeland forage and economic productivity, we partnered with local stakeholders to examine the effects of phenologically-timed grazing to control late-season weedy invasive grasses and promote more desirable forage species. In previous work, we found that simulated grazing (clipping) conducted in spring when forage species reach physiological peak most strongly decreased forage biomass and increased weed cover in rangeland plots. In contrast, fall clipping of standing dry grasses increased forage biomass and reduced weed extent, as predicted by our rancher collaborators. However, May clipping just prior to weed seed production reduced weed biomass most. Here we report on the legacy effects of this three-year period of clipping treatments as evident four years after treatments ended (seven yrs after treatment start), during an extended drought. We monitored vegetation within a one-acre field from which livestock but not rodents were excluded. Most strikingly, we found that plots previously clipped in May now contained notably less yellow starthistle (YST, an undesirable invasive forb), but ~2X more weedy grass biomass than the other treatments. This reversal in weedy grass dominance demonstrates the value of longer-term studies in assessing the trajectories of response to disturbance. In this specific case, we suggest that the weedy forbs and grasses share an antagonistic relationship, likely due in part to differences in their seasonal needs for water availability. While May clipping initially reduced weedy grass biomass, it also reduced YST establishment. Subsequent recolonization by weedy grasses thus was likely less constrained by their weedy forb competitors, and weedy grass biomass rebounded to higher levels than elsewhere. The observed legacy effects underscore the importance of anticipating responses to management mediated by competition among weedy species from different functional groups.