Suitability of terrestrial laser scanner derived surface roughness for predicting rill erosion in rangeland ecosystems

Authors: EITEL, JAN - University Of Idaho; Williams, Christopher - Jason; VIERLING, LEE - University Of Idaho; AL-HAMDAN, OSAMA - University Of Idaho; Pierson Jr, Frederick

Submitted to: Trans American Geophysical Union
Publication Type: Abstract Only
Publication Acceptance Date: 10/16/2010
Publication Date: 12/13/2010
Citation: Eitel, J.U.H., Williams, C.J., Vierling, L.A., Al-Hamdan, O.Z., and Pierson, F.B. 2010. Suitability of Terrestrial Laser Scanner Derived Surface Roughness For Predicting Rill Erosion in Rangeland Ecosystems. Presented at 2010 Fall Meeting, American Geophysical Union, San Francisco, California, 13-17 December 2010, Abstract B41C-0316.

Interpretive Summary: Surface roughness is known to effect overland flow velocity and soil erosion, but its field quantification at appropriately fine spatial scales has proven elusive. Field portable terrestrial laser scanners (TLS) potentially provide a new rapid, objective, and repeatable approach to accurately measure surface roughness across a range of spatial scales. Surface roughness can be derived from surface topography mapped at the sub-cm level using field portable TLS instruments. We tested the suitability of TLS derived surface roughness for predicting rangeland rill erosion. Rill erosion was simulated and measured in the field by releasing concentrated flow on 8.5 m2 plots that were randomly distributed across rangeland sites in a southeastern Oregon and southwestern Idaho, USA. Local surface roughness (locRMSH) of field plots was calculated as the standard deviation of TLS mapped surface heights within moving windows varying in size from 30x30 to 90x90 mm. The mean locRMSH of the rilled area and entire plot were negatively correlated (r2> 0.71, RMSE < 95.97 g min-1, and r2 > 0.74, RMSE < 90.07 g min-1, respectively) with measured rill erosion. The strength of the locRMSH - rill erosion relationship and regression model parameters were affected by the sliding window size, emphasizing the scale dependence of the locRMSH - rill erosion relationship. Removing hillslope angle trend from locRMSH decreased the strength of the rill erosion - locRMSH relationship (from r2 < 0.83 to < 0.26). Our results suggest that TLS represents a useful tool to measure surface roughness related to overland flow erosion processes.

Technical Abstract: Surface roughness is known to effect overland flow velocity and soil erosion, but its field quantification at appropriately fine spatial scales has proven elusive. Field portable terrestrial laser scanners (TLS) potentially provide a new rapid, objective, and repeatable approach to accurately measure surface roughness across a range of spatial scales. Surface roughness can be derived from surface topography mapped at the sub-cm level using field portable TLS instruments. We tested the suitability of TLS derived surface roughness for predicting rangeland rill erosion. Rill erosion was simulated and measured in the field by releasing concentrated flow on 8.5 m2 plots that were randomly distributed across rangeland sites in a southeastern Oregon and southwestern Idaho, USA. Local surface roughness (locRMSH) of field plots was calculated as the standard deviation of TLS mapped surface heights within moving windows varying in size from 30x30 to 90x90 mm. The mean locRMSH of the rilled area and entire plot were negatively correlated (r2> 0.71, RMSE < 95.97 g min-1, and r2 > 0.74, RMSE < 90.07 g min-1, respectively) with measured rill erosion. The strength of the locRMSH - rill erosion relationship and regression model parameters were affected by the sliding window size, emphasizing the scale dependence of the locRMSH - rill erosion relationship. Removing hillslope angle trend from locRMSH decreased the strength of the rill erosion - locRMSH relationship (from r2 < 0.83 to < 0.26). Our results suggest that TLS represents a useful tool to measure surface roughness related to overland flow erosion processes.