inundation-mapping icon indicating copy to clipboard operation
inundation-mapping copied to clipboard

[8pt] Evaluate backpooling module and update if needed (inclu. Guam overflooding)

Open EmilyDeardorff opened this issue 5 months ago • 4 comments

There have been a lot of updates to FIM since the mitigate branch outlet backpools module was completed in February 2024. These changes, particularly the new floodplain adjustment processing, could have impacted the performance of the branch outlet backpools mitigation. Evaluate the current performance of the backpools processing and update, if needed, the code or coefficients accordingly.

Sites to test: An important part of this task will be finding, if possible, sites where the backpools are malfunctioning as well as testing the known backpooling sites well in order to test the updated functionality. We can use the site(s) that were used in the original branch outlet backpool issue to benchmark the changes (HUC 13080002 branch 6077000088, although the branch ID probably has changed).

Guam overflooding: One potential site where backpooling could be causing problems is Guam site ugug8. In this site, there is over-inundation going up a different segment of the stream (shown below, the over-inundation is circled). Evaluate whether this over-inundation could be mitigated by updates to the backpooling module.

Image

EmilyDeardorff avatar Oct 29 '25 20:10 EmilyDeardorff

The field submitted this site (DAWM2 HUC 02070008) that you could use as a possible testing ground.

Image

CarsonPruitt-NOAA avatar Oct 30 '25 13:10 CarsonPruitt-NOAA

Here's another on the Connecticut River (RKGV1 HUC 01080106)

Image

CarsonPruitt-NOAA avatar Oct 30 '25 13:10 CarsonPruitt-NOAA

It looks like this is being caused by the floodplain adjustment. Compare v4.8.16.0 (blue with black catchment boundaries) with the floodplain adjustment turned off (red with red catchment boundaries).

Image

Here's another on the Connecticut River (RKGV1 HUC 01080106)

Image

mluck avatar Nov 12 '25 18:11 mluck

It looks like the backpool mitigation might not be able to help with this case (01080106). Most of the erroneous inundation (blue) is being caused by a pixel catchment (red) that is not at the outlet of the tributary.

Image

mluck avatar Nov 13 '25 16:11 mluck