Landscape and Fine Scale Habitat Associations of the Loggerhead Shrike

-This study was conducted to determine landscape and fine-scale vegetative variables associated with breeding Loggerhead Shrikes (Lanius ludovicianus) on Fort Riley Military Reservation, Kansas. Because Fort Riley is an Army training site, the influences of training disturbance to the vegetation, and range management practices on bird habitat patterns were also investigated. Breeding birds were surveyed in 1995 and 1996 using point counts. Survey plots were identified, a priori, at the landscape scale as either grassland, savannah, or woodland edge according to cover by woody vegetation. In 1996, fine-scale habitat at survey points and at bird use sites was measured and a principal components analysis used to characterize the fine-scale herbaceous vegetation structure. A military disturbance index was developed to quantify the severity of vehicle disturbance to the vegetation at survey and bird use sites. Shrikes were associated with savannah habitat at the landscape scale. Sites used by Loggerhead Shrikes were characterized at the fine-scale by tall, sparse, structurally heterogeneous herbaceous vegetation with high standing dead plant cover and low litter cover. At the fine-scale, tree and shrub density did not differ between sites used and not used by shrikes. Used sites did not differ from survey sites with respect to military training disturbance, hay harvest, or the number of years since a site was last burned. Our results in this study suggest that the shifting mosaic of vegetation on Fort Riley resulting from training and range management practices maintains adequate habitat for breeding shrikes. Received 25 Sept. 1997, accepted 18 June 1998. Declines in numbers of Loggerhead Shrikes (Lanius ludovicianus) have been reported for most areas of the United States and for western Canada (Arbib 1976, 1977, 1978, 1979; Cade and Woods 1997; Kridelbaugh 1981; Morrison 1981; Telfer 1992). In Kansas, the Loggerhead Shrike is designated as a "Species in Need of Conservation" (Kansas Department of Wildlife and Parks 1986). While reasons for the continued decline of Loggerhead Shrike populations remain unclear, loss of breeding and/or wintering habitat is a probable cause (Kridelbaugh 1981, Smith and Kruse 1992). Shrikes in the eastemn U.S. are associated with pastures of relatively short vegetation that offer adequate foraging and perching sites such as utility wires or isolated trees (BohallWood 1987, Brooks and Temple 1990, Smith and Kruse 1992, Telfer 1992, Yosef and Grubb 1993). Management recommendations resulting from these studies have been to provide short grassy habitat interspersed with scattered elevated perches. These recommendations may not be applicable to shrikes breeding in non-agricultural systems. The stucture of herbaceous vegetation in areas of extensive agriculture and grazing is relatively simple and homogenous. As a result, quantification of habitat in the above studies was restricted to percent cover of broad habitat types such as pastures or rowcrops (Bohall-Wood 1987, Brooks and Temple 1990). Native grasslands exhibit greater structural heterogeneity, and shrikes breeding in these areas may reveal different or more complex habitat association patterns (Prescott and Collister 1993, Chavez-Ramirez et al. 1994). Native grassland may provide suitable and adequate perch sites for foraging shrikes in the absence of powerlines, fences, or other artificial perching substrates (Chavez-Ramirez et al. 1994). The structure of herbaceous vegetation within a grassland may be very different depending on how often it is disturbed by burning, grazing, or, in the case of Fort Riley, by military vehicles. Unburned tallgrass prairie is structurally heterogeneous because of the interspersion of standing dead vegetation, forbs, and woody vegetation. Cover and depth of residual litter can be quite high in unburned prairie. Annual fire in the absence of grazing tends to favor dense, homogenous cover by warm season grasses (Gibson 1989), but periodic early spring burns such as those implemented on Fort Riley eliminate litter and in' Kansas Cooperative Fish and Wildlife Research Unit, U.S. Geological Survey, Biological Resources Division, 204 Leasure Hall, Kansas State Univ., Man-