Page 24 - A bird’s-eye view of recreation - Rogier Pouwels

A bird's-eye view of recreation
In recreation studies GPS devices are considered to be promising for gathering information on visitor densities and visitor behaviour (Beeco and Brown 2013). They provide accurate data on distribution, speed of movement and time spent at specific locations (D'Antonio et al. 2010, Beeco and Brown 2013). In recent years monitoring with GPS devices has also been used in combination with graph theory to evaluate the use of path structure (Taczanowska et al. 2014, 2017), in combination with recreation suitability mapping (Beeco et al. 2014), in combination with Public Participation GIS (Korpilo et al. 2017) and for spatial analyses of movement patterns (Van Marwijk and Pitt 2008, Renso et al. 2012). However, most studies using GPS devices for monitoring have focused on their utility for visual analyses and to find hotspots (Beeco et al. 2013). Few studies use monitoring information to understand what drives visitor densities in nature areas (Beeco et al. 2014). The exceptions are studies by Meijles et al. (2014) and Zhai et al. (2018). However, although both studies provide managers with information about which features determine visitor densities, this information might still lack relevance to managers. Managers not only need to know which features drive visitor densities, but also how visitor densities depend on these features, what the type of response curve is (Monz et al. 2013). This information would enable them to link potential management interventions, such as changing the features that drive visitor densities, to recognized values such as social and ecological thresholds.
In this study we aim to develop tools and rules of thumb that managers can use in decision-making processes with stakeholders to generate support for potential management interventions when visitor densities exceed social or ecological thresholds. For this support managers need to know how their interventions will lead to a change in visitor densities. We use monitoring data from GPS devices gathered in the New Forest (UK) to develop a random forest model (Breiman 2001) to identify which landscape and environmental features account for the spatial variation in visitor densities in the area. This model is then used as a tool to estimate visitor densities for the whole area. To illustrate its possible applications we use it to assess the impact of potential interventions on the population size of Nightjar (Caprimulgus europaeus), one of the protected species in the New Forest and sensitive to disturbance (Langston et al. 2007). As developing this type of tools needs much data and specialized expertise we also derived rules of thumb that managers can use to estimate the impact of management actions on visitor densities.
22