Systemic Safety Screening in San Juan National Forest: A Practical Approach with Limited Crash Data

As a national forest, the San Juan National Forest (SJNF) falls under the jurisdiction of the United States Forest Service (USFS), an agency of the United States Department of Agriculture. SJNF covers 1.8 million acres in southwest Colorado and contains three regions: Mancos-Dolores, Columbine, and Pagosa. The Office of Federal Lands Highway (FLH), in partnership with the Federal Highway Administration (FHWA) Office of Safety, developed a forest road safety plan (FRSP) to assess policies, identify relevant risk factors, and recommend key countermeasures to reduce roadway departure crashes in the forest. However, crash data in SJNF are limited; only 38 crash records on SJNF roads were available, provided by the Colorado Department of Transportation (CDOT), for the years between 2010–2018. Furthermore, the FLH project team could only generally locate 28 of these records (i.e., not to a specific location, but at the intersection of two roads or at a rounded distance offset from an intersection). As a result, the project team needed a practical method for assessing risk on the forest’s 2,498 mi of roadway to deploy effective low-cost countermeasures. 104 TR Circular E-C283: 13th International Conference on Low-Volume Roads Al-Kaisy and Huda published a framework for screening low-volume roads in Montana that does not necessarily require crash data (although the presence of a crash within the last 5 years is a risk factor). Segment-level risk factors noted in this research include the following: Road width, Horizontal curve radius, Vertical grade, Driveway density, Side-slope steepness, Fixed objects on roadside, Unpaved road, Poor pavement condition, Posted speed limit, and Traffic volume. Although these risk factors are substantiated in the existing literature (2,3), many of these risk factors do not necessarily apply to SJNF or are difficult data to obtain. Almost all roads in SJNF have a historic traffic count of less than 500 vehicles per day, and roads are typically only wide enough to accommodate a single direction of traffic. Furthermore, only 17 of the forest’s roughly 2,500 mi are paved. This limited the potential risk factors present (as well as applicable countermeasures) for roads in the forest. However, the project team was able to identify readily available data that could support risk-factor screening in SJNF (e.g., horizontal curvature, vertical grade, traffic volume, and exposure). This extended abstract documents a practical approach to systemic screening using readily available data for most low-volume road contexts in the United States.


  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; Tables;
  • Pagination: pp 103-110
  • Monograph Title: 13th International Conference on Low-Volume Roads
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 01893303
  • Record Type: Publication
  • Files: TRIS, TRB, ATRI
  • Created Date: Sep 15 2023 1:04PM