Ants as indicators of exposure to environmental stressors in North American desert grasslands

The relative abundance of ant species was measured by pit-fall trapping at 44 sites in southern New Mexico and southeastern Arizona, U.S.A.. Sites wer e selected for study based on documentation of a history of disturbance or protection from disturbance, exposure to varying intensities of livestock g razing, dominance by an exotic species of plant and vegetation change resul ting from disturbance or restoration efforts. Ant community composition, re lative abundances of species, and species richness were the same on disturb ed and undisturbed sites. None of the metrics based on hypothesized respons es of ants to disturbance clearly distinguished between disturbed and undis turbed sites. Ant communities on sites where restoration efforts have resul ted in distinct differences in vegetative cover and composition were simila r to the ant communities on degraded unrehabilitated sites on the same soil type. Ant communities in riparian cottonwood gallery forests in Arizona an d New Mexico were similar but differed from the assemblages in exotic salt cedar and native ash riparian woodlands. Ant species exhibited remarkable r esistance to human-induced disturbances in these rangeland areas. In grassl ands dominated by the South African grass, Eragrostis lehmanniana Nees, lar ge seed harvesting ants, Pogonomyrmex spp., were greatly reduced in abundan ce compared to native grasslands. Other ant metrics were not different in E . lehmanniana grasslands and native grasslands. We conclude that ants canno t be used as indicators of exposure to stress, ecosystem health or of rehab ilitation success on rangeland ecosystems. Ants are also not useful indicat ors of faunal biodiversity in rangeland ecosystems.