Skip to main content
ARS Home » Southeast Area » Oxford, Mississippi » National Sedimentation Laboratory » Watershed Physical Processes Research » Research » Publications at this Location » Publication #309875
Title: Quantifying uncertainty of measuring gully morphological evolution with close-range digital photogrammetry

Author
item GESCH, KARL - Iowa State University
item Wells, Robert - Rob
item CRUSE, RICHARD - Iowa State University
item MOMM, HENRIQUE - Middle Tennessee State University
item Dabney, Seth

Submitted to: Soil Science Society of America Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/11/2014
Publication Date: 3/13/2015
Publication URL: http://handle.nal.usda.gov/10113/60589
Citation: Gesch, K.R., Wells, R.R., Cruse, R.M., Momm, H.G., Dabney, S.M. 2015. Quantifying uncertainty of measuring gully morphological evolution with close-range digital photogrammetry. Soil Science Society of America Journal. 79(2):650-659.

Interpretive Summary: This manuscript describes the uncertainty associated with photogrammetry. An experiment was devised to replicate the photogrammetry technique used to monitor ephemeral gully erosion. Images were collected of an asphalt surface in a similar fashion to the field condition then images were collected from around the target area, approximately every 45 degrees (i.e. walk around). Point clouds were generated from all pair combinations and analyzed to determine the quality of the elevation data. The analysis showed that the photogrammetric technique used to quantify ephemeral gully erosion resulted in high vertical accuracy and very small uncertainty, suggesting that this method is highly effective and acceptable means to accurately detect small changes within the landscape.

Technical Abstract: Measurement of geomorphic change may be of interest to researchers and practitioners in a variety of fields including geology, geomorphology, hydrology, engineering, and soil science. Landscapes are often represented by digital elevation models. Surface models generated of the same landscape over a time interval can be compared to estimate geomorphic evolution. Any such morphological estimate of change in a landform should include a range of probable values based on the quality of the digital elevation models that represent the surface of interest. This study sought to determine the uncertainty associated with detecting changes in reaches of ephemeral gullies with close-range digital photogrammetry. An experimental surface was constructed, surveyed, and photographed. Using an established procedure, the photographs were used as input to photogrammetry software to generate point clouds. The point clouds were then analyzed to determine the quality of elevation data generated by the photogrammetric technique. For individual point clouds the 2s relative vertical precision was determined to equal 0.916 mm and the 2s absolute (geo-referenced) vertical accuracy was computed as 8.26 cm, and the 95% confidence range (2s uncertainty) of detecting elevation change from one point cloud to another was determined to be ± 1.29 mm. This value could be applied to any volumetrically derived estimate of geomorphic change as an uncertainty range. The high vertical accuracy and small uncertainty in elevation change determined in this study suggest that close-range digital photogrammetry is an effective and acceptable method to accurately detect small changes in ephemeral gullies or other geomorphic features of interest.