Keystone species have large impacts on community and ecosystem properties, and create important ecological interactions with other species. Prairie dogs (Cynomys spp.) and banner-tailed kangaroo rats (Dipodomys spectabilis) are considered keystone species of grassland ecosystems,and create a mosaic of unique habitats on the landscape. These habitats are known to attract a number of animal species, but little is known about how they affect arthropod communities. Our research evaluated the keystone roles of prairie dogs and kangaroo rats on arthropods at the Sevilleta National Wildlife Refuge in central New Mexico, USA. We evaluated the impacts of these rodents on ground-dwelling arthropod and grasshopper communities in areas where prairie dogs and kangaroo rats co-occurred compared to areas where each rodent species occurred alone. Our results demonstrate that prairie dogs and kangaroo rats have keystone-level impacts on these arthropod communities. Their burrow systems provided important habitats for multiple trophic and taxonomic groups of arthropods, and increased overall arthropod abundance and species richness on the landscape. Many arthropods also were attracted to the aboveground habitats around the mounds and across the landscapes where the rodents occurred. Detritivores, predators, ants, grasshoppers, and rare rodent burrow inhabitants showed the strongest responses to prairie dog and kangaroo rat activity. The impacts of prairie dogs and kangaroo rats were unique, and the habitats they created supported different assemblages of arthropods. Where both rodent species occurred together on the landscape, there was greater habitat heterogeneity and increased arthropod diversity. there was great habitat heterogeneity and increased arthropod diversity.
( Mound-scale plots) To evaluate arthropods associated with mound disturbance patches and rodent burrow systems, we established replicate mound-scale plots with paired non-mound control plots, a design similar to that used by Hawkins and Nicoletto (1992), Schooley et al. (2000), and others. The mound and non-mound plots were spatially intermixed within each landscape-scale plot. Samples were collected from 10 kangaroo rat mounds on the Krat plot, 10 prairie dog and 10 kangaroo rat mounds on the Pdog+Krat plot, and on paired non-mounds located 10 m away from sample mounds, in areas with minimal rodent disturbance.
Pitfall traps also were placed next to prairie dog mounds and paired non-mounds on the Pdog+Krat plot and at kangaroo rat mounds and paired non-mounds on the Krat plot. To prevent rodents from disturbing the traps, pitfall traps at mound-scale plots were surrounded by 2.54 cm wire mesh, large enough to allow large arthropods to pass through.
The pitfall traps were made of 10 oz plastic cups inserted into 12 oz cans, buried with the rim of the cup flush with the ground. Traps were filled with non-toxic propylene glycol as the trapping and preservation agent, and ceramic tiles (0.23 m x 0.23 m) were elevated over the traps. Traps were opened for one month, both during spring (May) and fall (September), from spring 2000 through fall 2001.
Arthropod samples were sorted and identified in the laboratory. Only ground-dwelling taxa were considered. The pitfall dataset resulted in 218 species (from both plot and mound samples). Most taxonomic groups were identified to species, but weevils (Curculionidae), velvet ants (Mutillidae), and spider wasps (Pompilidae) were grouped at higher taxonomic levels and not included in. Nymphs of crickets, grasshoppers, and spiders could not be identified to species. Arthropod voucher specimens from this study were deposited in the collection of the Arthropod Division, Museum of Southwestern Biology (MSB) at the University of New Mexico (UNM).
1/09/2009 -YX. The metadata was entered by Yang Xia from metadata supplied by Ana Davidson 1/17/2008 (Yang Xia) - updated and modified metadata to correct format; - Checked data for missing data points and errors
Additional Personnel with Collecting Samples and Data
Julie McIntyre helped with the field data collection.
Additional Study Area Information
Study Area Name: Pino Gate
Study Area Location: The study site was located near the base of the Los Pinos mountains and directly adjacent to the nothern fenceline of the SNWR at Pino Gate.Elevation: 1600 m
Vegetation: Burrograss (Scleropogon brevifolius), sand dropseed (Sporobolus cryptandrus), and black grama (Bouteloua eriopoda) were the dominant vegetation.
Soils: Deep clayey loam soils Geology: On an upper bajada slope, in a broad swale
Climate: Long-term mean annual precipitation is 243 mm, about 60% of which occurs during the summer. Long-term mean monthly temperatures for January and July are 1.5°C and 25.1°C, respectively.
Site history: Historically, prairie dogs were common throughout the area, but were exterminated by the early 1970’s (John Ford, United States Department of Agriculture Wildlife Services, personal communication). Gunnison’s prairie dogs began to re-colonize the study site from adjacent private land in 1998. During our study, the colony occurred within a 5 ha area, near the base of the Los Piños Mountains in an area with deep clayey loam soils. The site has been long inhabited by kangaroo rats, and represents typical northern Chihuahuan Desert grasslandNorth Coordinate:34.406954South Coordinate:34.406954East Coordinate:106.606269West Coordinate:106.606269