Modelling of recreational trails in mountainous areas: An analysis of sensitivity to slope data resolution

Abstract

This paper investigates how the spatial resolution of slope data affects the modelling of recreational trails in mountainous areas. It measures the impact of spatial resolution on a) the estimation of cost functions that link walking speeds and slope, and b) the modelling of moving times and mapping of routes using these functions. Cost functions that build on a mountain hike GPS record and slope data derived from a Digital Elevation Model (DEM) at varying resolutions are estimated and their ability to predict accurate moving times is evaluated. The cost functions are supplemented with recently published cost functions to map the quickest routes between the hike start and end points with least-cost-path analyses. The results indicate that spatial resolution has a critical influence in the modelling: the similarity between routes modelled at a same spatial resolution with distinct cost functions considerably surpasses that of routes simulated at varying resolutions with a same cost function. Furthermore, employing finer resolutions enhances the similarity between routes mapped with different cost functions, while also improving the accuracy of predicted moving times. These findings provide evidence that DEM resolution should receive prime attention in the modelling of trails with slope-dependent cost functions.