Georeferenced Video Mapping to Classify Streambank Erosion Susceptibility
Boat-mounted, georeferenced videos were used to develop a method to rapidly document streambank conditions and assess erosion susceptibility over long stream segments. Traditional streambank survey methods are often spatially limited, time-consuming, and expensive. A novel Streambank Video Mapping S...
| Format: | Text |
|---|---|
| Language: | unknown |
| Published: |
BYU ScholarsArchive
2019
|
| Subjects: | |
| Online Access: | https://scholarsarchive.byu.edu/josh/vol15/iss2/3 https://scholarsarchive.byu.edu/context/josh/article/1121/viewcontent/Georeferenced_Video_Mapping_to_Classify_Streambank_Erosion_Susceptibility.pdf |
| Summary: | Boat-mounted, georeferenced videos were used to develop a method to rapidly document streambank conditions and assess erosion susceptibility over long stream segments. Traditional streambank survey methods are often spatially limited, time-consuming, and expensive. A novel Streambank Video Mapping System (SVMS) comprised of a GPS receiver, two streambank-facing video cameras, and thalweg depth sensor gathered continuous georeferenced data on both streambanks in a single pass. A modification of the commonly applied Bank Erosion Hazard Index was used to classify streambank conditions. The modified index, named the Bank Erosion Susceptibility Index (BESI), was derived from a combination of the variables including bank angle, bank height, bankfull height, channel depth, bank surface protection, and riparian diversity. These variables were combined into a four-parameter model to estimate bank erosion susceptibility. Using the SVMS, two stream segments in Tennessee were surveyed. In 1.33 hours, 7.7 km of the New River and in 1.08 hours, 7.6 km of Beaver Creek were surveyed for an average collection rate of 6.3 km/h. Maps of streambank erosion susceptibility were developed and the streambank health of the two stream reaches were compared. For each streambank, variables were visually classified from the georeferenced video by five independent observers. To assess the accuracy of the SVMS approach site field measurements were physically collected and compared to the video assessments with average errors less than 5 percent. |
|---|