ABSTRACT
Summary Production of grasses and forbs was measured for 4 years, beginning in 1976, following aerial application of 2,4,5-T to 3,634 ha of mesquite (Prosopis glandulosa Torr.) dunelands on the Jornada Experimental Range in south central New Mexico. Plant production was also measured on a 3,318-ha control area. The mesquite dunelands are representative of large areas in the southwestern U.S.A. where mesquite has invaded former desert grasslands. Objectives of the study were to determine forage increases attributable to reduced mesquite competition and to evaluate the relationships between production and precipitation. Both sprayed and control areas were grazed by cattle. Temporary exclosures were used to exclude cattle from three sampling sites on the sprayed and control areas each year. The sampling sites were all on sandy ranges with relatively deep soils (>50 cm to caliche layer). A dune-centered sampling system permitted the determination of differences in production on dune and interdune areas. The mesquite dunes, occupying 28% of the land surface, produced more annual forbs and fewer perennial grasses than the interdune areas. Stem kill of mesquite averaged 56% at sampling sites on the sprayed area. End-of-season harvests were used to compare treatments. Perennial grass production was 7-, 8-, and 4-fold greater on the sprayed than on the control area in the first 3 years following treatment, respectively. Maximum perennial grass production of 642 kg/ha occurred in the first season following treatment. In the 4th year the control area received 49 mm more precipitation than did the sprayed area, and production of perennial grass 438was nearly equal on the two treatments. Mesa dropseed (Sporobolus flexuosus [Thurb.] Rydb.) contributed 66% -92% of the perennial grass production. Production of annual plants varied widely among years but was greatest in the 4th season when precipitation patterns favored annual plant growth, particularly on the control area where annual forb production was 543 kg/ha. Production of broom snakeweed (Xanthocephalum sarothrae [Pursh] Shinners) cycled from high to low and back to high on both treatments. Mesquite control is an effective tool for improving forage production on arid rangelands, but the success of control efforts depends not only upon the degree of brush kill but also upon the posttreatment precipitation patterns.
