Detecting vegetation cover change on the summit of Cadillac Mountain using multi-temporal remote sensing datasets: 1979,2001, and 2007

摘要

This study examines the efficacy of management strategies implemented in 2000 to reduce visitor-induced vegetation impact and enhance vegetation recovery at the summit loop trail on Cadillac Mountain at Acadia National Park, Maine. Using single-spectral high-resolution remote sensing datasets captured in 1979, 2001, and 2007, pre-classification change detection analysis techniques were applied to measure fractional vegetation cover changes between the time periods. This popular sub-alpine summit with low-lying vegetation and attractive granite outcrop-pings experiences dispersed visitor use away from the designated trail, so three pre-defined spatial scales (small, 0-30 m; medium, 0-60 m; and large, 0-90 m) were examined in the vicinity of the summit loop trail with visitor use (experimental site) and a site chosen nearby in a relatively pris- tine undisturbed area (control site) with similar spatial scales. Results reveal significant changes in terms of rates of vegetation impact between 1979 and 2001 extending out to 90 m from the summit loop trail with no management at the site. No significant differences were detected among three spatial zones (inner, 0-30 m; middle, 30-60 m; and outer, 60-90 m) at the experimental site, but all were significantly higher rates of impact compared to similar spatial scales at the control site (all p < 0.001). In contrast, significant changes in rates of recovery between 2001 and 2007 were observed in the medium and large spatial scales at the experimental site under management as compared to the control site (all p < 0.05). Also during this later period a higher rate of recovery was observed in the outer zone as compared to the inner zone at the experimental site (p < 0.05). The overall study results suggest a trend in the desired direction for the site and visitor management strategies designed to reduce vegetation impact and enhance vegetation recovery at the summit loop trail of Cadillac Mountain since 2000. However, the vegetation recovery has been rather minimal and did not reach the level of cover observed during the 1979 time period. In addition, the advantages and some limitations of using remote sensing technologies are discussed in detecting vegetation change in this setting and potential application to other recreation settings.