Caro-COOPS.org Discussion Board

lianxie · Joined: 25 Feb 2025 Posts: 17

Hi Madilyn,

The modeling is ready to go (albeit using the old elevation and bathymetry data), but I don't know the status of new bathymetry/elevation as well historical data collection (waterline, surge time series, streamflow, etc.). As soon as I find out about these
data from Jeremy, we can produce the images and animations.

--- Lian

dwhite · Joined: 25 Feb 2025 Posts: 3

Lian and Madilyn,

In regard to the merged data sets. We have the data from NGDC this is the 90 meter resolution data. The generation of the 30 meter resolution merged data is an ongoing effort. There is currently only one (Tampa Bay) that has been developed and it is available to the public (http://chartmaker.ncd.noaa.gov/bathytopo/welcome.html). I have been looking into this process and it is not an easy exercise. This is compounded by the fact that the NED data for Charleston and much of the South Carolina coast appear to have the noise problem or the stripping observed in DEMs (please see examples that I sent out a couple weeks ago (for more info see: http://gisdata.usgs.net/NED/About.asp)). We can have new 30 meter data developed from the ASTER sensor (http://edcdaac.usgs.gov/aster/ast14dem.html). It is more accurate and it would not have the stripping problem, but again this is going to increase development time.

The Tampa Bay effort required the development of the VDATUM tool (http://chartmaker.ncd.noaa.gov/bathytopo/vdatum.htm). This tool does not exist for the east coast yet so that is another problem. It is difficult for me to provide a completion date at this time because I do not have a good idea yet of all of the problems that we will encounter. I am not optimistic that we can do this within a couple months. I have just started the process of locating the people who worked on the Tampa Bay project to get a better idea of the time and effort needed for this and the status of devleoping the VDATUM tool for the east coast.

Dave

fletcher · Joined: 25 Feb 2025 Posts: 2

Dave,

I good example of the needs that become identified only after you start working on a problem. Are we sure no one is working on a 30 meter resolution for the east coast? As soon as you have all the information you can get, let's look at this, in conjunction with the NCSU folks, and see where we're at and what the options are. The need for a retrospective analysis is still there. But what I'm hearing is that we cannot do that until we have the 30 m data. Does that apply to SC only?

lianxie · Joined: 25 Feb 2025 Posts: 17

Hi Madilyn,
On the modeling side, we can do a model simulation (hindcast) for Hugo with or without the 30m data. Technically, we can run the model with our existing 100 m (grid size) data from NOAA or any other elevation/bathymetry data. In that case, we will need to clearly point out the limitations of the results due to elevation data error. The current NOAA elevation data we use is incremented at 1m interval. If this is the best data we have for the Charleston area, we'll go ahead to produce the results with this data set. I was hoping that a more accurate bathymetry and land elevation data set can be created for the Charleston region before we conduct the surge and inundation simulation for this area.
I suggest that if it will take a long time to create a good high resolution elevation and bathymetry data set for SC, and if there is no other (better) elevation/bathymetry data for SC (than what we have), we just use what we have here and run the Hugo case and calibrate the model for Charleston using whatever flood/surge records we can find. Can some one at SC collect Hugo flood/surge information (for example, actually extend of flooding, surge height at all available gauge stations, and even water markers on some historical/symbolic structures)? We need to have some ground truth to calibrate the model for the Charleston area. Data for other storms that affected SC will also be useful. Thanks!
--- Lian

dwhite · Joined: 25 Feb 2025 Posts: 3

Lian,
I am tracking down a USGS report that cataloged flood surge (high-water marks) along the SC coast at approx 315 locatoins. "U S Geological Survey Open File Report 90-386, Storm-tide elevations produced by Hurricane Hugo along the South Carolina Coast, September 21-22, 1989". By Eirk Schuck-Kolben. Our library only has this report in microfiche, but I can photocopy it. The data would need to be entered by hand.
Dave

lianxie · Joined: 25 Feb 2025 Posts: 17

Hi Dave,
The high water mark (surge+tide+wave) data will be useful. It wouldn't be a big problem to enter the data by hand since, I guess, the data file will be fairly small. In addition to the high water mark data, time series of tidal gauge data that show the time history of water level at selected locations will also be useful. More importantly, direct observations of flood extend (such as estimates from aerial photographs) will be very useful for inundation validation. Informal records (from technical or news reports, interviews, etc) about actual water/dry land lines (boundaries of flood) in the region are also useful. Thanks!
--- Lian

lianxie · Joined: 25 Feb 2025 Posts: 17

Hi Dave,

It depends on the distribution of the 315 points. Unless some points are clustered very close to each other, I'd like to see all of the data. If there are points very close to each other (say within a 100m x100 m box), and if the geographic condition within the box is fairly uniform, we only need to pick a representative point from that box. However, if the geographic conditon is non-uniform,I would like to see all of the data even if they are close to each other. The reason is that, since the model (say at 100m grid size) will not be able to resolve the differences of the geographic conditons (coastal shape, elevation, etc.) within a 100x100 m grid box, the model data for that region (box) should be compared to the average height of all of the observational data points within that model grid box, not to the specific value of a geographic point.

--- Lian