COMMUNITY STRUCTURE OF SHORTGRASS-PRAIRIE RODENTS: COMPETITION OR RISK OF INTRAGUILD PREDATION?

Abstract

Local assemblages of rodents in northern shortgrass prairie may reflect either predatory or competitive effects of northern grasshopper mice (Onychomys leucogaster) on other species such as deer mice (Peromyscus maniculatus). In an effort to distinguish these two types of effects, during the summer of 1994 I compared abundances, patterns of microhabitat use, and diets of P. maniculatus on four areas of shrub-dominated prairie in north-central Colorado to these characteristics on four similar sites where 0. leucogaster was removed. The abundance of P. maniculatus decreased during the study on both control and removal sites, but the decline was greater on controls, where numbers of 0. leucogaster increased. Declines in the abundance of P. maniculatus on individual study plots were correlated with abundances and recolonization rates of 0. leucogaster. Only 6% of P. maniculatus present on controls during pre-removal trapping were captured 7 wk later, compared to 32% of those initially present on removal sites. Furthermore, changes in the number of P. maniculatus were inversely related to the amount of shrub cover on sites, suggesting that shrubs may have provided refuge from 0. leucogaster, which rarely uses these microhabitats. Analyses of microhabitat use support this assertion; individual P. maniculatus increased their use of shrubs on control sites when 0. leucogaster was abundant, but there were no shifts in microhabitat use on removal sites. The proportions of arthropods in the diets of the two species were similar prior to Onychomys removals, and although P. maniculatus consumed fewer arthropods as the experiment progressed, diets did not differ between control and removal sites. In addition, Ord's kangaroo rats (Dipodomys ordii) increased slightly in number following removals, and western harvest mice (Reithrodon- tomys megalotis) colonized two removal sites. These results suggest that predatory or aggressive interference by 0. leucogaster, rather than exploitative competition, was re- sponsible for the observed changes in abundance and microhabitat use. Risk of aggression or predatory interference therefore may influence the local abundance and distribution of P. maniculatus and other small rodents on areas of shortgrass prairie where habitat char- acteristics permit coexistence.