Large-scale assessment of potential habitat to restore elk to New York State

Recent public interest in restoring elk (Cervus elaphus) to eastern states has created the need to assess habitat suitability over large landscapes. Our objectives were to identify the areas in New York, a landscape of 126,000 km(2), with the highest potential to support elk and to determine whether restoration to these areas was biologically feasible. We assessed habitat suitability for elk in 2 stages. Stage 1 used low-resolution data to eliminate approximately 98,800 km(2) (78%) of the state's land area that was predominantly agricultural, urban, near a major roadway, or did not historically support native elk. Stage 2 used a habitat suitability model to evaluate remaining areas. We used food, cover, and habitat diversity relationships to evaluate suitability based on 7 land-cover classes from satellite imagery classified by the New York Gap Analysis Project. We used a road-density relationship to explore the influence of human access and potential exploitation (i.e., hunting or poaching) on habitat suitability. We evaluated accuracy and resolution of the satellite imagery to determine whether it was appropriate to use in our habitat model and determined suitabilities at 4 scales (16.0 km(2), 38.5 km(2), 63.0 km(2), and 102 km(2)), representing potential annual home-range sizes for individual elk and elk herds. Evaluations of suitability were sensitive to scare of analysis; however, we were able to identify areas that had high suitability independent of scale. No areas in New York were identified as optimal. If human exploitation of elk is not substantial, a large amount of the habitat in peripheral Adirondack and Catskill regions is highly suitable for restoration (>4,100 km(2) across both regions). If human exploitation is substantial, portions of these 2 regions are moderately suitable, but little high suitability habitat exists (3 km(2)).