search

clawpack / geoclaw

Version of Clawpack for geophysical waves and flows
http://www.clawpack.org/geoclaw
BSD 3-Clause "New" or "Revised" License
75 stars 86 forks source link

Overestimation "eta" for Harvey #620

Open sourabhmet opened 2 weeks ago

sourabhmet commented 2 weeks ago

Hi,

I am using scripts from "https://github.com/mandli/surge-examples/tree/master/harvey_2017" for harvey,2017 surge height. the model overestimating the "eta" at taxes coast.

Could you please help me to understand which parameters or coefficients I can modify to reduce surge height.

Thanks in advance

mandli commented 2 weeks ago

There are a lot of parameters that can be changed. Can you provide more detail as to how/where things are being over estimated?

sourabhmet commented 2 weeks ago

Here, I used NCEI 90 meter resolution bathymetry datasets and track file I used same as you suggested in the examples.

Only change, I added the fgmax and fgout from the example and saving the data into netcdf form. After that I am extracting "eta" for 1000 locations in Texas, Then for few locations I am getting values is about 12.6958 (m) (near to landfall point). I am assuming it's in meter because the model provides the output in meters. However, in that location observed around 8feet surge height.

Apart from that I have not made any change in the setrun.py

mandli commented 2 weeks ago

Thanks @sourabhmet. Can you provide the location of one of the points that you are seeing the 12 m surge? Also, can you provide a plot of the surface height that shows the location and one for the entire domain?

sourabhmet commented 1 week ago

Hi Mandli,

Sorry but I cannot show the map caused by some limitations. However, I extracted the elevation at this location is 13 meter.

Regards

Sourabh Shrivastava, PhD

On Wed, Jun 19, 2024 at 23:38 Kyle Mandli @.***> wrote:

Thanks @sourabhmet https://github.com/sourabhmet. Can you provide the location of one of the points that you are seeing the 12 m surge? Also, can you provide a plot of the surface height that shows the location and one for the entire domain?

— Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/620#issuecomment-2179262806, or unsubscribe https://github.com/notifications/unsubscribe-auth/AME6JU4PPJOXZIDFWWKVAJLZIHCJRAVCNFSM6AAAAABJQHDDAKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNZZGI3DEOBQGY . You are receiving this because you were mentioned.Message ID: @.***>

mandli commented 1 week ago

Well that does make it difficult. From the Harvey example there's no where that I can see that produces more than 2 meters of surge (and it's more like 1 meter at the highest) so I am not entirely sure what you have found. Things that I can suggest to look into:

  1. Make sure that the storm does not approach the boundaries of the domain that go over water. This can cause issues with how the water column is forced.
  2. Double check to make sure you are extracting the surface value of the water and not either dry topography or the depth at a wet point. This can be done by looking at $h$ and checking to see if it is zero (the first case) or the depth rather than the surface. These are all related of course through $h = \eta - b$ and $h$ and $\eta$ are output at each gauge.
  3. Overall just make sure that the there isn't anything else going awry, such as a blow-up, which can happen for a number of reasons.
sourabhmet commented 1 week ago

Hi Mandli,

  1. I rechecked the storm is not approaching the boundary of the domain.
  2. I also checked h, and its max value is 13.98 (for same location).
  3. What elese could be the reason ?

One more thing , I am using AFTC track data directly, Do I have to make any change in the data such as its wind speed unit is knots, should I change it to m/s?

Please help

mjberger commented 1 week ago

Can I suggest you send your setup - input files, etc. to us so we can try to directly see what you are doing?

If they are small you could email them to me at @. @.>. If they are too large you will have to upload them somewhere and we can download them.

— Marsha

On Jun 25, 2024, at 3:46 AM, sourabhmet @.***> wrote:

Hi Mandli,

I rechecked the storm is not approaching the boundary of the domain. I also checked h, and its max value is 13.98 (for same location). What elese could be the reason ? Please help — Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/620#issuecomment-2188210715, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAGUGC47CFIBNV7BZ33KEZ3ZJEN5TAVCNFSM6AAAAABJQHDDAKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOBYGIYTANZRGU. You are receiving this because you are subscribed to this thread.

sourabhmet commented 1 week ago

Please share your email. for Datadownload I will share you link and corresponding script to convert Raster to Ascii

mjberger commented 1 week ago

I already sent my email: @. @.>

I will need all your input files, including whatever has changed in the ike directory you are basing it on.

— Marsha

On Jun 25, 2024, at 9:36 AM, sourabhmet @.***> wrote:

Please share your email. for Datadownload I will share you link and corresponding script to convert Raster to Ascii

— Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/620#issuecomment-2188988783, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAGUGC2FBHBGLJNWXNE7KU3ZJFW65AVCNFSM6AAAAABJQHDDAKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOBYHE4DQNZYGM. You are receiving this because you commented.

sourabhmet commented 1 week ago

Marsha,

I cannot see you email, its showing me *@*** *@****>, something like this :-)

mjberger commented 1 week ago

this i fine too - send it to any email you have for me like this one you used.

— Marsha

On Jun 25, 2024, at 9:43 AM, sourabhmet @.***> wrote:

Marsha,

I cannot see you email, its showing me @** @***>, something like this :-)

— Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/620#issuecomment-2189004350, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAGUGCZLFAIYJREQWETU4SLZJFXZZAVCNFSM6AAAAABJQHDDAKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOBZGAYDIMZVGA. You are receiving this because you commented.