Some general questions /storm surge simulations

82 views
Skip to first unread message

Mobina Namvar

unread,
Sep 18, 2024, 2:10:24 AM9/18/24
to claw-users
Hello Kyle,

 I recently reviewed the storm surge simulations on GitHub and have a few questions regarding the results.


For example, in the case of Storm Surge: Zeta 2020:

- Station 1: Bay Waveland Yacht Club, MS (ID: 8748437)
  - Observed: 2.40 meters
  - GeoClaw Prediction: 1.10 meters

- Station 2: Pascagoula NOAA Lab, MS (ID: 8741533)
  - Observed: 2.00 meters
  - GeoClaw Prediction: 1.25 meters

- Station 3: Bayou La Batre Bridge, AL (ID: 8739803)
  - Observed: 1.95 meters
  - GeoClaw Prediction: 0.85 meters

- Station 8: Weeks Bay, Mobile Bay (ID: 8732828)
  - Observed: 1.15 meters
  - GeoClaw Prediction: 0.50 meters

- Station 9: New Canal Station, LA (ID: 8761927)
  - Observed: 1.10 meters
  - GeoClaw Prediction: 0.30 meters

Or in the case of Storm: Michael 2018:

- One station recorded a surge of 1.75 meters (GeoClaw predicted 0.50 meters).
- Another station reported 2.60 meters (GeoClaw predicted 1.25 meters).

In some instances, GeoClaw's predictions align more closely with the observed data. 

Questions:

1. Do you think adjusting the wave tolerance parameter could help reduce the gap between the observed data and GeoClaw simulations?

2. Among the 37 storm surge examples, most recorded storm surges of less than 1 meter or approximately1 meter, with none reaching 4 meters. Are you aware of any instances where storm surges exceeded 4 meters during a storm?

3. I noticed some storm surges reported negative values (e.g., -0.7), indicating a decrease in normal sea level during the storm. Based on the Saffir-Simpson Hurricane Scale, the first category of storm surge starts around 1.2 cm. If fluctuations in sea level are less than the absolute value of 1.2 cm, can we classify them as storm surges, or are they variations in water level?

4. It seems that heavy rainfall during storms can contribute to increased storm surge. However, it appears GeoClaw currently simulates storm surge primarily due to wind forces. Is there any plan for incorporating rainfall effects in future simulations?

Best wishes,

Kyle Mandli

unread,
Sep 25, 2024, 4:25:56 PM9/25/24
to claw-users
Up front I should state that the examples in this repository were primarily done to provide examples setups and were done by students as summer projects so may well need some additional work to get them up to validation standards.  That being said here are a few comments:
  • Refinement can help in some circumstances, especially around gauges.  This can again take the form of changing the wave tolerance parameter or defining regions around areas of interest.  It's also key to make sure that the gauges are where they really are as often the coordinates reported are not accurate enough to put them in the water.
  • Rainfall can account for the discrepancy between observed and simulated depending on the gauge.
  • The storm is represented with a somewhat simplistic field and sometimes can be important.
  • There are other sources of error in all the input as well.
This is all to say that there are a lot of details to understanding why a gauge may not match often requiring looking carefully at each situation.  The systemic under-prediction in the case of zeta might be broadly the storm representation or something similar.  Michael looks like it might be a datum issue but hard to say.

Kyle
--
You received this message because you are subscribed to the Google Groups "claw-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to claw-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/claw-users/99b72e07-b470-4c72-91fc-f2b2178e53dfn%40googlegroups.com.

Mobina Namvar

unread,
Sep 26, 2024, 1:59:41 PM9/26/24
to claw-users

Thank you Kyle

Best wishes
Reply all
Reply to author
Forward
0 new messages