Mesquite control increases grass density and reduces soil loss in southern Arizona

Abstract

In 1974 we selected 8 pairs of gully headcuts on the Santa Rita Experimental Range. Mesquite (Prosopis velutina Woot.) on the watershed of 1 headcut of each pair was killed with diesel oil. Densities of perennial grasses and shrubs, shrub cover, surface erosion, headcut advance, and gully depth were recorded at 3-year intervals, 1974-1986. Four of the watershed pairs were in pastures grazed yearlong: 4 were in Santa Rita rotations. Each grazing schedule included 2 watershed pairs that were about 200 m higher in elevation than the other 2. In 1974, before mesquite was killed, perennial grass densities were low with little difference between assigned conditions. By 1977 perennial grass density was greater where mesquite was killed than on untreated watersheds and was greater at upper elevations. Lehmann lovegrass (Eragrostis Lehmanniana Nees.) made up almost all of the grass density increase at the upper elevation. Native perennial grasses, which were largely replaced by Lehmann lovegrass at the upper elevation, accounted for almost all of the gain at the lower elevation. Lower density (P less than or equal to 0.05) on low elevation rotation grazed watersheds in 1986 was the result of summer drouth in 1985 and 1986 that coincided with March-October grazing in 1986. Soil loss (mm) during each 3-year period was lower at headcut-soil surface grids where mesquite was dead. Advances in headcuts and changes in gully depth showed similar trends. On the 4 pairs of watersheds that were equipped to measure runoff, there was more total runoff per millimeter of rainfall where mesquite was alive than where mesquite was killed.