Mitchel P. McClaran

Questions How does desert grassland vegetation respond to long‐term grazing removal? Is grass response the result of differences in the number or the size of plants? Does the response differ over time and in relation to precipitation patterns? Location Santa Rita Experimental Range, southwestern United States. Methods Four times between 2011 and 2020, we measured the cover of woody plants and native and non‐native perennial grasses, and the density, size, and biomass of individual perennial grasses on 40 permanent transects inside and outside 10 long‐term (88–104 years old) livestock exclosures (0.1–4.0 ha) occurring on the same ecological site. We used linear mixed models to compare vegetation variables in grazed vs ungrazed transects through time and calculated the cumulative frequency distributions of grass plant diameters. Results The cover of woody plants did not differ by grazing treatment. Instead, the exclosures had a greater cover, density, and biomass of native grasses and cover and biomass of the most abundant native grass Arizona cottontop ( Digitaria californica ). Moreover, ungrazed populations of natives and Arizona cottontop showed a plant size structure skewed to larger sizes. Non‐native grasses showed no differences between grazing treatments. Patterns of inter‐annual precipitation influenced woody and grass plant abundance, but not their response to livestock removal. Conclusions Long‐term grazing removal in desert grasslands affected native grass abundance, but not that of non‐native grasses and woody plants. Response of native grasses to livestock removal was characterized more by plant size rather than the number of plants, and, importantly, the population size structure skewed to smaller plants in grazed areas suggests that grazing limits plant vigor and longevity. Absence of a non‐native grass response likely reflects lower palatability and greater grazing resistance of non‐natives. Absence of woody plant response is due to their low palatability and the permeability of exclosures to seed dispersal.