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

Changing WRF domain in the middle of a simulation

J

janac4

Member
Hello,

I have a running WRF with ndown which includes a 5 km domain followed by a 1 km domain (of 660x330 cells) up to T+24 h forecast. I need a T+72 h forecast of a subregion of my 1 km domain, and would like to use the first 24 hours from my big 1 km domain and continue the run in a smaller 1 km domain, is it possible?
After my first ndown simulation, I was thinking about executing WPS with the remaining hours (until T+72h) and repeat the ndown process with the new 1 km domain, but I don't know if the model is gonna work.
Also I'm not sure if I should take into account new spin-up hours or not, and I should determine a way to merge data (specifically precipitation) in the 24-25 hours of the simulation.
What would be the best approach to achieve this kind of results?

Thank you a lot.
 
Please see my answers below:

"I have a running WRF with ndown which includes a 5 km domain followed by a 1 km domain (of 660x330 cells) up to T+24 h forecast. I need a T+72 h forecast of a subregion of my 1 km domain, and would like to use the first 24 hours from my big 1 km domain and continue the run in a smaller 1 km domain, is it possible?"

I don't think so. WRF needs lateral forcing data to run. In this case, you have no foricng data.

"Also I'm not sure if I should take into account new spin-up hours or not, and I should determine a way to merge data (specifically precipitation) in the 24-25 hours of the simulation."

WRF is under constraint of initial and lateral forcing. I don't think spin-up time is a big issue.
 
Thanks for your answer,
What would you think would be the best workflow to achieve my main idea? Having a bigger domain for the first 24 hours and after that simulating up to 72 hours with a smaller 1 km domain. In all the cases the 1 km domain is done with ndown process with the same 5 km parent domain.
 
I'm thinking about this again and I would like to simplify the question.
If I have a complete WRF run of a 5km domain for T+0h up to T+100h, is there possible (and how could I do it) to achieve d02 files of smaller domain just form T+24 to T+48h? Input data should come from 5km wrfouts (I'm thinking about using ndown).
I'm not sure this is possible since both domains would start at different times, but maybe there's a way I'm not familiar with.
Thanks a lot.
 
You must log in or register to reply here.
Top