search

NCSU-CHAZ / Sunny-Day-Flooding

Student research related to the Sunny Day Flooding Project
MIT License
0 stars 0 forks source link

Update Cape Fear River bathymetries? #18

Open ththelen opened 2 years ago

ththelen commented 2 years ago

Hi @caseydietrich, I reviewed the NC9 bathymetries along the Cape Fear River and the representation of the river bottom elevations appears to be somewhat artificial as far south as Carolina Beach. Circle in blue in Fig 1 is a stretch of constant -13.5 m bathymetry in the Cape Fear just east of Carolina Beach. Unless they dredge just this specific portion of the river to deeper depths than the surrounding areas (which would be strange), this would not seem to be an accurate representation of the river channel. Up further north past Wilmington, there are additional areas of constant riverbed bathymetry mixed in with stretches of riverbed with varying bathymetry. See the two locations with constant -8 m riverbed depths circled in blue in Fig 2 as an example of this.

Image Fig 1. Cape Fear River east of Carolina Beach as represented by the NC9 mesh.

Image Fig 2. Cape Fear River north of Wilmington but south of USGS gauge as represented by the NC9 mesh.

I am currently under the impression that we will have to redefine mesh bathymetries for the entire length of the Cape Fear up to the USGS gauge. To that end, I am currently downloading 10GB of CoNED data that covers the Cape Fear from Bald Head to the USGS gauge. My biggest concern with redefining all the riverbed bathymetries is that in the locations where the river channel is only two vertices wide up near the USGS gauge, we might alias the river channel if the CoNED riverbed does not line up exactly with the location of the mesh vertices. Let me know if you think that this is the correct approach to move forward.

ththelen commented 2 years ago

Also, there is no discernable tidal signal at the Cape Fear River gauge we have been using to calculate our ADCIRC flux BC

Image

ththelen commented 2 years ago

Good morning @caseydietrich, could you let me know what you think about this proposed path forward so I can start working on mesh edits and get some runs going?

https://github.com/NCSU-CHAZ/Sunny-Day-Flooding/issues/18#issuecomment-1279119827

caseydietrich commented 2 years ago

@ththelen Sorry for the delay in my reply. This is a bummer about the poor representation of the river bathymetries in the NC9 mesh. I'm not sure of the best way to proceed, but we want to be careful to manage the impact to your workload. It can be a heavy lift to develop a good model for a coastal river.

The big question is: how much do we care about the river? I think it is important, because we want to understand its contributions to the total water levels at Carolina Beach. But you and Katherine can see better for the big picture. Do we want to spend a lot of time on fixing the river, or should we explore other work-arounds? It might be worthwhile to have a quick meeting to discuss this.

If we do want to proceed with mesh edits, then there are two or three immediate questions:

  1. Based on the existing mesh and bathymetry, how far upstream do the tides propagate? You could use synthetic stations at regular intervals to investigate this. This will be imperfect (because the bathymetry is wrong), but it will give us an idea of the influence of the tides. It would be great to cut everything upland/upstream from where the tide signal disappears, and put the river boundary as close to the coast as possible.

  2. Do the CONED datasets have a good/realistic bathymetry? This seems like a big uncertainty to me. Often these DEMs have fake values in the rivers, due to the limitations of LiDAR etc. This could be where the bad mesh bathymetry comes from, it may have been interpolated from a bad DEM. In which case, we are limited in how we can respond.

2A. If the CONED bathymetry is good, then is it possible to extract a thalweg from the river? I assume this is possible with a GIS tool, and I'm hoping you know how (or can find out) how to do it. This would allow us to control the layout of the mesh around the river, and we would avoid the aliasing you mention.

ththelen commented 2 years ago

Hi @caseydietrich, based on your description of the issue, I would probably not be able to get around to re-working the Cape Fear River until next semester. In terms of outstanding work, I still have to remesh the tidal inlet and progress my 3Di model a bit before I go to the Netherlands for training.

Long term, reworking the river would push back the simulation of May 2022 events in ADCIRC and setting up a working 3Di model with pipe flow. I am not sure if it makes sense to spend a lot of time reworking a river that based at least on preliminary results may not contribute significantly to flooding.

I suspect @anardek will also have thoughts on this issue. If we would like to meet to discuss this issue, this Thursday during our 1:15 - 2:00 pm meeting slot may be the best option. I have to move next week's check in meeting anyways since I will be out of town for a conference. If you are both available, I could bump next week's modeling check in to this week. I will send this schedule change out in Google Calendar and you can let me know by responding to the invite or messaging on Github if that works for you.