Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

MPAS from WRF background

J

janac4

Member
I am familiar with the WRF model for atmospheric forecasting, but I would like to learn how to use MPAS. So far, I have been able to run the example cases, but I would like to apply it to my WRF study domain.
In WRF, I use two domains (5 km and 1 km, one-way nesting) centered in Panama, with the higher-resolution domain having a grid size of 600x300 cells. The input data is from the 25 km GFS.
If there is no predefined MPAS grid that reaches a 1 km resolution for my specific study area, can I generate one myself? I have noticed that existing grids go down to 3 km resolution, but even those have a larger extent than I need, and I want to limit the simulation to a smaller area.
How should I approach this issue? Is it possible or cost-effective to use the MPAS model to replicate the WRF domain? What advantages would this bring?
 
Please see my answers below:

(1) If there is no predefined MPAS grid that reaches a 1 km resolution for my specific study area, can I generate one myself? I have noticed that existing grids go down to 3 km resolution, but even those have a larger extent than I need, and I want to limit the simulation to a smaller area.

The mesh generation is kind of hard and we basically won't recommend uses to produce mesh themselves. Hopefully we can generate a variable-resolution mesh with the highest resolution that can reach 1-km. However, Note that this will take time and the high-resolution mesh may not cover exactly the area you are interested. We have to consider the general need of the broad community....

(2) Regional MPAS is an ideal system that can replace WRF. The advantages of using regional MPAS can be found in the document here
 
Ming Chen said:
Please see my answers below:

(1) If there is no predefined MPAS grid that reaches a 1 km resolution for my specific study area, can I generate one myself? I have noticed that existing grids go down to 3 km resolution, but even those have a larger extent than I need, and I want to limit the simulation to a smaller area.

The mesh generation is kind of hard and we basically won't recommend uses to produce mesh themselves. Hopefully we can generate a variable-resolution mesh with the highest resolution that can reach 1-km. However, Note that this will take time and the high-resolution mesh may not cover exactly the area you are interested. We have to consider the general need of the broad community....

(2) Regional MPAS is an ideal system that can replace WRF. The advantages of using regional MPAS can be found in the document here
Click to expand...

Thanks for your answer,
Even though it is not recommended,
is it possible to have access to the code or receive guidance on the process? From a computational standpoint, I understand it may be quite costly, but given the limited scope, the expense should not be as significant as global meshes.
 
You must log in or register to reply here.
Top