Comparison of Underwater Video Mapping and Pebble Count Measurements in Determining Stream Channel Substrate

GPS videography has been used to capture substrate data with post-processing efforts being made to characterize streambed particle distributions. Advantages of this method include less field time, minimal streambed disturbance, convenience of post-field processing, and digitally stored data which can be used with GIS when geo-referenced. The question being asked is ‘what are the drawbacks of this method and how accurate are the data.’

Current video mapping techniques were used to capture stream channel substrate footage in three distinct float patterns. All data were geo-referenced by embedding geographical coordinates to each frame during recording. Post-processing of the video data included a comparison of three methods used to "retrieve" samples from the video: time interval, distance interval, and visual estimation. These data were compared to the data gathered by three distinct pebble count transects: thalweg, zig-zag, and proportional. The accuracy and efficiency of both methods - video and field – were tested by comparing them to a frame method used as a control. Data were compared by particle size, diameter size class, and percent distribution of substrate.

The proportional pebble count data proved to be significantly different at α=0.15 from the frame data for particle size. No one method of pebble counting showed itself to be more accurate than the others. Significant differences at α=0.15 were found between video data and frame data for all three retrieval methods. Visually estimating data proved to be significantly different for measurements of cobble while video distance and time interval data showed significant differences in small gravel. Particle measurements from video data proved to be more time consuming and less accurate than current pebble count techniques.

Although visual estimation of video data shows less precision than pebble counting methods, it maintains the video-mapping benefit of reduced field time by allowing quick analysis of data in the laboratory. When performing large scale substrate classifications, the use of an underwater video mapping system can provide: 1) reconnaissance data of locations for more accurate substrate analysis and 2) a quick method of detecting substantial changes to the streambed over long distances due to land use effects or natural phenomena.