⤺ Search TRRP online library of documents and data
ID: 69
[4.16 MB] SubstratePolygons.lpk
Alvarez, J., D. H. Goodman, K. De Juilio, and A. Martin. 2015. Substrate Map. Data Package for the Trinity River Restoration Program (TRRP). Hoopa Tribal Fisheries, U.S. Fish and Wildlife Service, and Yurok Tribal Fisheries, Arcata, California. Available: https://www.trrp.net/library/data?id=69.
Description This is a series of TINs and polygons representing substrate particle size as it is distributed along the Trinity River corridor from Lewiston Dam to the North fork of the Trinity River. The lateral extent of the map is focused on the 2000 cfs channel as predicted by a two-dimensional hydraulic model. Particle size is represented by the D84 in mm. The map was generated by snorkeling along the river and occularly estimating the D84 rock in a patch. That rock was then measured and a GPS point taken, attributed with the measurement. These points were as dense as was necessary to accurately depict changes in the faces. In homogenous zones the points are more spread out than in heterogeneous areas. Patches of sand, bedrock, or other distinctly isolated substrate size were delineated with polygons. These should be overlaid on the TIN to create the true representation of the particle size map. The substrate map was generated from 22,436 D84 points. The average density of GPS points taken is 35.2/100 meters of river length. Forty-six validations of the D84 estimates were conducted by revisiting GPS locations and conducting a pebble count. The validation was spread over a range of estimated values to evaluate bias across particle size. The measured D84 was compared to the estimated and residual values were plotted. The median residual was 3.5 mm, the mean was 4.4 mm and the variance was 4.09. Accuracy was better with smaller particle sizes. There was no bias observed. Use Considerations The data was collected to reach the 2000 cfs line as predicted by a hydraulic model. If the model output seemed incorrect, a best guess of the actual boundary was selected. When the TIN was created the interpolation area was far greater than the points. The TIN was trimmed to a more reasonable extent that would be useful for creating roughness inputs for a 2D model. The data was not trimmed all the way down to the exterior point boundary so if being used for fine resolution analysis then the safest path would be to trim this product with the 2D model output that guided its creation. Another option is to use the ArcGIS tool “Delineate TIN Area” select a distance smaller 200 feet between nodes (which is the distance currently used to shrink the interpolation area) and remove interior and perimeter areas without points. This will remove islands. Many of the islands were too tall to fall within the scope of this project however an effort was made to create the islands close to reality with the data on along the 2000cfs margin. Therefore the resolution is much less than the in channel map and the bigger the island is the more likely it is not accurately represented.
First Posted: 0000-00-00 00:00:00
Post Updated: 2024-01-22 17:50:03