Date of Graduation


Document Type


Degree Type



Davis College of Agriculture, Natural Resources and Design


Agricultural and Resource Economics

Committee Chair

Chad Pierskalla.


This research investigates the frequency and patterns of trail erosion on purposefully selected natural surface trails in Monongahela National Forest. Trail erosion is considered a non-self-limiting process which degrades trail beds making them unattractive, difficult to travel on, and has the potential to impact local streams with the introduction of silt. Recreation ecology research delineates factors which influence the rate of erosion on a trail into three broad categories: recreational use, environmental conditions, and managerial actions. This study tests the relationship of variables from each of these categories to the local patterns of erosion. Three sampling strategies are used to document the presence of erosion and to characterize recreational use, environmental conditions, and management parameters. Point samples taken at 600-foot intervals along 63 miles of trail provide information on maximum incision, and a problem census of erosion greater than five inches in depth for ten or more feet document the presence and extent of erosion on study trails. Use of over 500 high-quality field reference points collected with a Global Positioning System (GPS) allowed the placement of wheel distances for eroded segments onto trail routes in a Geographic Information System (GIS) for analysis with other secondary GIS data sets. This approach allowed each trail to be segmented into 600-foot linear segments providing a third sampling frame for consideration with the amount of linear erosion per segment as the dependent variable. These equal sized segments provide a manageable landscape scale unit of analysis. Additionally, GIS elevation data enabled the calculation of two independent variables which are new to trail erosion research, Topographic Relative Moisture Index (TRMI) and Estimated Upslope Flow Length (EUFL).;The results of this study indicate trail erosion on the study trails is comparatively lower than other resource areas with occurrences limited to a subset of trails. Regression models show the primary factors influencing erosion rates for the Forest are recreational use, trail grade, low trail alignments, mid-slope trail positions, open canopies, mean annual precipitation, decreased winter temperature, TRIM, and ELF yet the best model only predicted 26% of the occurrences of erosion. Of all the independent variables, average ELF for linear trail segments described the most variance in the incidence of erosion. The significance of this finding is that there is a great need for shortening the distances water is allowed to travel on a trail with either water diversion structure installation and maintenance or trail rerouting. ELF was also influential in the finding that low-gradient trails with limited use displayed disproportionately higher rates of erosion than other trails, a finding not well documented in previous research. As four of the eight significant predictor variables were GIS based, this study shows the usefulness of utilizing GIS in trail erosion research as it provides insights into landscape scale variables that are not easily measured in the field. Furthermore, this approach integrates well with GIS based trail inventory management systems informing maintenance regimens and facilitating long-term trail degradation monitoring.