Fractal dimension of linear network segments: experiments on hiking trailsPrah, Klemen (Avtor) Shortridge, Ashton (Avtor) fractal dimensionbox-countinghiking trailsGISSloveniaComplex terrestrial linear networks like roads, rivers, and hiking trails have different degrees of spatial variation in different environments and across different geographic scales. Fractal dimension is a longstanding conceptual approach to characterizing such spatial variation over multiple scales, and a number of algorithms have been developed to calculate fractal dimension of a linear feature. In this paper, we develop, implement, and test a variant of the widely published box-counting algorithm that uses rasterization to make the process computationally efficient to run on datasets containing hundreds or thousands of high-spatial resolution linear features. This variant separately calculates fractal dimension in the horizontal and vertical dimensions. We then apply this variant on a dataset of over 300 randomly sampled segments of Slovenia's 10,000 km hiking trail network. Segments range from trails in gently varying terrain to routes in alpine regions with extremely steep and rough topography. Two methodological approaches were applied and compared: a raster-based R workflow and a vector-based ArcGIS Pro/Python workflow, the latter applied to a smaller subset of the dataset. Two main datasets were used: vector data representing hiking trails and a lidar derived digital terrain model (DTM) with a 1 m horizontal resolution, both covering the entirety of Slovenia. The two approaches were further tested and compared using five synthetic curves. Results show that, while like other box counting implementations, ours suffer from some inaccuracy, fractal dimensions in the horizontal and vertical dimensions provide useful insight into the variation in hiking trails in a wide range of conditions and offer promise for new approaches to classify trail difficulty.20262026-07-14 15:01:47Drugo98869sl