A Statistical Comparison of Sidewalk Slopes Derived from Multi-Resolution Digital Elevation Models in Support of Accessibility



Rodgers, Robin E

Journal Title

Journal ISSN

Volume Title



Sidewalk slope is a major factor taken into consideration when visually or mobility impaired pedestrians select a route to their destination. In an effort to improve campus accessibility, a testbed pedestrian routing environment, called the George Mason University (GMU) Geocrowdsourcing Testbed, has been developed by researchers in the Department of Geography and GeoInformation Science (GGS) at GMU in Fairfax, Virginia. In order to determine how to best incorporate slope as an attribute in the testbed’s sidewalk network, this research considers the effect that using Digital Elevation Models (DEMs) of varying spatial resolutions has on the accuracy of slope calculations. Lower resolution DEMs are often free and more easily acquired than higher resolution DEMs, such as those derived from Light Detection And Ranging (LiDAR). Therefore, if a lower resolution DEM sufficiently captures variations in slope, it may be deemed acceptable to use for this type of application, when a high resolution DEM is not available. To test this proposition, slope values for the sidewalk network were derived from four DEMs with the following resolutions: 1/3 arc-second (about ten meters); five meters; 1/9 arc-second (about three meters); and one meter (m). The slope values from the lower resolution DEMs were statistically analyzed and compared to the high resolution 1 m DEM and to each other to detect significant differences. The results conveyed that the differences between the slope values derived from the lower resolution DEMs and the slope values calculated from the 1 m DEM were statistically significant. Consequently, it was concluded that, for the incorporation of slope in the GMU Geocrowdsourcing Testbed, a high resolution (1 m or higher) DEM needs to be used for slope calculations, to provide the most accurate results for visually and mobility impaired pedestrian routing.



Accessible routing, DEM, LiDAR, Sidewalk network, Slope, Spatial resolution