Bird and Small Mammal Populations in Relation to Seral Stage of Mixed-Grass Prairie, For Pierre National Grassland, South Dakota

The Nebraska National Forest uses a multivariate technique to classify seral stages of clay range sites in mixed-grass prairie. Mean canopy cover x frequency of occurrence indices for buffalo grass lliuchloe dactyloides), western wheatgrass (Agropyron smithii), and green needlegrass (Stipa viridula) are used to classify clay range sites into one of four seral stages. Managers lack necessary information to assess how wildlife communities are affected by changes in seral stage. I estimated bird species density, bird species richness, bird species diversity and small mammal abundance at 3 7 sites (7 low seral stage; 5 low-intermediate seral stage; 19 high-intermediate seral stage; 6 high seral stage). Vegetation structure and composition characteristics were also measured at each site. Bird species richness did not differ (P = 0. 12) among seral stages. Density of grasshopper sparrows, bobolinks, dickcissels, and brown-headed cowbirds increased (P \u3c 0.10) from low to high seral stages. Burrowing owls, upland sandpipers, chestnut-collared longspurs, and horned larks decreased (P \u3c 0.10) from low to high seral stages. Western meadowlarks were abundant in all seral stages but densities were greater (P \u3c 0.10) in low and low-intermediate seral stage compared to high seral stage. Red-winged blackbird density did not differ (P \u3e 0 .10) among seral stages. Bird species whose habitat requirements include tall grasses and increased residual grass cover were more abundant in higher seral stages. Lower seral stages were beneficial to birds that prefer short grass and sparse vegetative cover. Multiple regression analyses indicated that sera1 stage was a satisfactory predictor of density for most bird species. Vegetation structure and composition variables were collinear with seral stage regressors. No significant model was generated for red-winged blackbirds (P \u3e 0.10) using seral stage regressors or vegetation variables. Western meadowlarks were modeled (P \u3c 0.01) using only vegetation variables. Small mammal abundances were variable between years and within seral stages. Abundance of grasshopper mice (Onychomys leucogaster) decreased (P \u3c 0.01) as seral stage changed from low to high. Macro- and micro-scale features of the habitat must be investigated to understand small mammal occurrence and abundance. Seral stage seems to be an adequate predictor of small mammal abundance at a course scale, but may not be reflective of small mammal abundance at a particular site and time. A habitat capability (HABCAP) model, based on relative densities of grassland birds in seral stages of clay range sites, was developed using 1996 data and evaluated using 1997 data. Only minor revisions were necessary to predict grassland bird densities from HABCAP model coefficients developed in 1996 suggesting that model coefficients accurately predicted bird species densities. Managers should maintain a mosaic of seral stages within the landscape to provide all seral conditions for grassland wildlife. Further research is needed to develop separate HABCAP model coefficients for feeding and breeding habitats for upland sandpipers and brown-headed cowbirds and to incorporate species not emphasized in this study into the HABCAP model. Changes in management actions which affect the seral stage of the prairie landscape can influence bird and small mammal populations. This research enables managers to predict the effects of management actions that change the ecological stage of clay range sites on bird and small mammal populations