Scenarios
In 2017 the topographic survey validation process was scoped to contain checks for within-season issues and across year review of projections and inconsistencies of files were put on hold until further review of the program. We strongly recommend a programmatic review of projections and reoccupation consistency prior to use of Geomorphic Change Detection metric and data use.
Here we outline potential sources and outcomes of reoccupation issues that may be a result of 2011-2017 survey procedures, equipment, processing workflow, transformations, and data merging.
| |
Unprojected FDS/Folder Exists |
GCD? |
Notes |
| New Visit |
Yes |
N/A |
Normal |
| Visit Reoccupied (missing 0-1 bm’s) |
No |
Yes |
Normal |
| Crew Error |
Yes |
Maybe* |
Crew used toolbar incorrectly |
| Out of Tolerance |
Yes |
Maybe* |
Not occupied in field, in office used manual affine/rotation/elevation to best fit survey. Use previous hinge if at all possible. |
| Missing 2 benchmarks |
Yes |
Maybe* |
in office used manual affine/rotation/elevation to best fit survey. Use previous hinge if at all possible. |
| Missing 3 Benchmarks |
Yes |
No |
Control Network Reset. Occasionally use manual best fit to align with orientation of previous surveys, but elevation will not be the same. |
| LiDAR |
“Transformation” FDS may exist (in custom folder for topo projects?) |
Y |
Previous surveys were re-transformed if control points were captured for LiDAR control network. |
| RTK |
No |
Yes |
It would be good to have a Point Quality or uncertainty for the benchmarks for these visits. |
An expert’s assessment would be required to determine if GCD results are representative in these cases. Higher uncertainty should be used in these cases.
Ideal Assessment Procedure
Capture the Timeline of the Control Network for a site.
- Instances of Transformation Tool use (unexpected “Unprojected” instance)
- Continuation of Benchmarks in each survey, verified by Benchmarks captured in Aux.
- Assess the spatial agreement (delta xy) of Benchmarks between surveys, with attention to SP vs SS.
- Evaluate Elevation Agreement between benchmarks (delta Z), especially the designated Hinge point and OC points. (Though I don’t think we are able to reliably identify the hinge point used).
- A Retransformation (say due to lidar) could retroactively reset the control network, however, all benchmarks/control network should still be in spatial agreement.
- The goal is to identify
- A coherent timeline that shows each visit falls in place spatially, as represented by the benchmarks.
- Inconsistencies that could indicate visits could be misaligned between years.
- Valid reasons for miss-occupied visits exist, and these are important for explaining erroneous values in GCD
This procedure would be a useful feature to add to the Site Properties tool, and ideally could be used build the new Control Network file for re-occupying a site.
In-Survey Drift
- High Backsight Check Error values
- Multiple Backsight checks or station setups between Side Shots.
- Some uncertainty can be accounted for in point quality calculation.
- Does this propagate through each station setup? I don’t think so.
- Could affect benchmarks that were collected at different station setups.
- 2011 - 2012 surveys could be more prone to drift (or unknown drift status)
- Green crews did not recognize when drift was occuring drift
- No procedures or little training to recognize and deal with drift.
- Initial visits to sites (no existing context)
- Nikon/Instrument differences/issues
- Proprietary software did not carry over backsight information to toolbar.
- Check Reoccupation values, backsight checks for store points?
- Strange or overlapping features within survey (LW on wrong side of TB, etc)
- GCD Results could show drift over the longitudinal survey area.
HOME