The bobcat (Lynx rufus) is native to much of the United States, including South Dakota where it is an important furbearer, is harvested through managed seasons in areas west of the Missouri River, and is closed to harvest east of the Missouri River. Bobcat ecology has been well documented throughout most of the US, but information on the ecology of this species is lacking in the Northern Great Plains of South Dakota. We addressed the following objectives in 3 regions of South Dakota (East River, West River, and the Black Hills): (1) identify habitat selection by bobcats, (2) determine survival rates and cause of mortality of bobcats, (3) estimate potential population size, and (4) identify food habits of bobcats. From July 2006 to January 2010 we radio collared 10 bobcats in the Badlands (4 female, 6 male), 16 bobcats in the Black Hills (8 female, 8 male), and 15 bobcats in Bon Homme County (7 female, 8 male) with GPS or VHF collars. Home range size averaged 26.7 km2 (SE = 7.3, n = 4) for females and 72.1 km2 (SE = 17.7, n = 5) for males in the Badlands; 18.8 km2 (SE = 4.3, n = 4) for females, and 47.9 km2 (SE = 7.4, n = 3) for males in the Black Hills; and 16.0 km2 (SE = 16.0, n = 3) for females, and 64.7 km2 (SE = 12.63, n = 4) for males in Bon Homme County. Home ranges varied by sex (P = 0.008). During the summers of 2008 and 2009 we analyzed microhabitat characteristics at bobcat locations at the Badlands (n = 176) and Black Hills (n = 173) study sites. Microhabitat characteristics at Badlands relocations of bobcats differed (P < 0.001) from Black Hills relocations. At the Badlands site, logistic regression identified bobcat relocations being closer to drainage features, having taller small shrub vegetation, and with greater amounts of medium visual cover. At the Black Hills site, logistic regression identified bobcat relocations as being associated with steeper slopes, being closer to drainage features, having greater amounts of grass, shrub, and bare ground cover types, and having greater amounts of medium visual cover. Annual survival was 0.43 (95% CI = 0.17–0.73) for 10 bobcats (6m, 4f) at the Badlands site; 0.76 (95% CI = 0.39–0.94) for 16 bobcats (8m, 8f) at the Black Hills site; and 0.49 (95% CI; 0.22–0.77) for 15 bobcats (8m, 7f) at the Bon Homme site. Survival for bobcats occupying harvested areas was 0.59 (95% CI; 0.22–0.77). Human caused mortality was the most common source (n = 6) of mortality, followed by unknown causes (n = 5). Survival did not vary between study sites (P = 0.428) or between harvested and unharvested populations (P = 0.609) and combined survival for all sites was 0.54 (95% CI = 0.35–0.716). Using a GIS, we analyzed landscape scale habitat features using bobcat locations from the Badlands (n = 700), Black Hills (n = 700), and Bon Homme County (n = 800) sites. Data from the Badlands and Bon Homme County were combined to create a prairie data set, whereas the Black Hills data set represented the forested region of the state. Logistic regression of the prairie data set identified bobcats selecting for areas with greater amounts of high cover grassland, wooded vegetation, and terrain ruggedness. In the Black Hills, logistic regression identified bobcats selecting for areas with greater amounts of terrain ruggedness.