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

No Difference in wrfvar_output

bigveech

New member
Hi everyone,

I'm running into an issue while trying to assimilate data using WRFDA, and I was hoping to get some guidance.

I'm using observation data from this dataset to ensure everything is set up properly before attempting to assimilate my own data. Specifically, I’m working with data from 2023-10-16 at 12Z. The input data is generated by WRF's `real.exe` using GDAS data in my region.

The obs file I’m using is in Little_R format (OBS:2023101612.txt). After running `da_wrfvar.exe`, I noticed that the `wrfvar_output` is identical to the `wrfinput_d01` file that I provided as input. I used a diff tool to check the files, but there’s no difference between them. Additionally, I’ve run the forecast using `wrf.exe` and checked values at various points, but again, there’s no change.

I even modified the observation data slightly, thinking that maybe the original obs file was too close to the actual data, but this still produced no difference.

Has anyone run into this issue before or have any insight on what could be going wrong? I would expect at least some small difference, even if the observations were close to the model's input.

Thanks in advance!
 

Attachments

  • Figure_1.png
    Figure_1.png
    49 KB · Views: 2
Hi everyone,

I'm running into an issue while trying to assimilate data using WRFDA, and I was hoping to get some guidance.

I'm using observation data from this dataset to ensure everything is set up properly before attempting to assimilate my own data. Specifically, I’m working with data from 2023-10-16 at 12Z. The input data is generated by WRF's `real.exe` using GDAS data in my region.

The obs file I’m using is in Little_R format (OBS:2023101612.txt). After running `da_wrfvar.exe`, I noticed that the `wrfvar_output` is identical to the `wrfinput_d01` file that I provided as input. I used a diff tool to check the files, but there’s no difference between them. Additionally, I’ve run the forecast using `wrf.exe` and checked values at various points, but again, there’s no change.

I even modified the observation data slightly, thinking that maybe the original obs file was too close to the actual data, but this still produced no difference.

Has anyone run into this issue before or have any insight on what could be going wrong? I would expect at least some small difference, even if the observations were close to the model's input.

Thanks in advance!
Hello,

Please include your namelist files and your rsl.out.0000 and rsl.error.0000 file. Without those we cannot help. From the sounds of it, it sounds like no observations are being assimilated but I can't tell without looking into those files.

Thank you,
JeremyB
 
I actually got it to work. I think I was comparing it to an old file that was identical to the input 🤦. I was able to get some data assimilated but I'm not getting the results that I was expecting.

Are you experienced in data assimilation? I'd love to pick your brain if you're up for it.
 
No problem! Any chance you can point to anyone on the forum that's got a lot of DA experience?
Hello,

Unfortunately I don't have anyone I can point you to. However feel free to ask question in this forum. There are lots of people that can help.
 
Top