Using Photogrammetry to Document Stream Topography.

Bank erosion occurs in many human impacted streams. Stream topography measurements over time are necessary to quantify stream bank erosion. The goals of this project are to evaluate topographic survey methods for costs, effort, and accuracy; to develop a stream geometry monitoring method based on photogrammetry; and to verify as-built restoration with designed stream geometry.

            Photogrammetry is a technique to create 3D surface models from individual photo images. The accuracy of photogrammetry depends on image quality, the number of images, image angle, and vegetation. Lab and field experiments were conducted to examine these parameters. Volume models created from a 23.0cm x 20.5cm x 16.5cm box were evaluated. Treatments included two cameras, camera angles, and vegetation density. A volume model was considered accurate when it matched at least 99% of the box volume. Models with no vegetation were accurate with a minimum of 21 images from both the Samsung Galaxy S4 and DSLR.

            When less than 21 images were taken, a trend of lower accuracy among models with and without vegetation was visible. Results indicate that model error is increased and accuracy diminished with fewer images.

            Field studies with two truncated soil pyramids (2m x 2m, and 3m x 3m, respectively) planted with German millet were evaluated for volume accuracy. 40 images were taken of each pyramid to ensure proper overlap amongst images with the phone and DSLR. Image sets were taken weekly from week zero (no vegetation) to week four (100% vegetation cover). Results indicate at week zero both pyramids had 1.5 percent error or greater with both cameras.

            Stream restoration projects pre- and post-construction will be measured using photogrammetry to further evaluate the impact of vegetation on photogrammetric model accuracy. Ground control points will be installed to determine stream geometry change (i.e. erosion or deposition).