Hi all,
I got multi-temporal high-resolution DEMs and would like to study river longitudinal profile evolution. I plot these profiles and their general patterns are similar but there are some large bias (i.e., elevation offsets) among these profiles. These large bias are ~5-10 m, which are caused by DEM generation rather than actual physical process. Any ideas to eliminate these bias and put all the profiles in uniform elevation reference? Thanks!
You might review some of my materials and comments to related questions on stream ordering, etc. Most of my LiDAR involvement in last decade, but have been using topographic contour dems or maps for decades. Basically in many landscapes, you have burn in streams to remove barriers. A recent paper and powerpoint by Andy Maceyka (GIS) and I may be helpful in describing our efforts to improve stream coverage in coastal plain.
You might review some of my materials and comments to related questions on stream ordering, etc. Most of my LiDAR involvement in last decade, but have been using topographic contour dems or maps for decades. Basically in many landscapes, you have burn in streams to remove barriers. A recent paper and powerpoint by Andy Maceyka (GIS) and I may be helpful in describing our efforts to improve stream coverage in coastal plain.
Dear Kang,
Please be more specific: are these data LiDAR data? What type of processing have been applied? Or do you have data generated by stereophotogrammetric methods? Do you have the original data (unprocessed data, e.g., the original point cloud)? What is your point density? Do you extract the drainage network yourself? What is the typical relief, say in 100 m? (If it is low, it is a problem tackled by William in their abstract, too.) Is it a karst area or the surface covered at least partly by carbonate rocks?
Without knowing the answer to these questions, I assume that you see the problems of the DTM generation from LiDAR data: some points coming from the vegetation are considered as terrain points this may cause that type of problems you described.
If you can answer the questions listed above, it is possible to figure out how the (presumably) erroneous points are generated, and what you can do against them.
Kind regards,
Balázs
- Badges
- Science topic
- Similar topics
- Geoscience
- Geoinformatics