A mixed-modes approach for estimating hiking on trails through diverse forest landscapes: the case of the Appalachian Trail

Abstract

Many hiking trails traverse the forests and public lands across North America. It has therefore become important for federal management to gain an understanding of total use on these trails. However, there has never been a formal attempt to estimate hiking on these long, backcountry trails. This paper presents an approach that utilizes two survey instruments (exit-site tallies and a trail-user questionnaire) and two primary estimation components (standard and augmented sites) to estimate hikers over a spatial and temporal span. For illustrative purposes, the methodology is applied to a 175 km segment of the Appalachian Trail from 1 June through 14 August 2007. Two alternative estimation methodologies are presented and compared. The model-based approach may be preferred to the design-based approach when sample size is small because it smoothes erratic strata estimates and yields smaller standard errors. However, the design-based approach relaxes an assumption and is more appropriate as sample size increases. In our survey of the Appalachian Trail, there was a 5.6% difference between the visitation estimates based on these two approaches, and such stability reinforces confidence in the methodology.