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

Help with WRFDA

ashish__shaji

New member
Hi,

I am a beginner at running wrfda. I used wrfda online tutorial and run its test case successfully. I have also gone through the user guide, but I was not understanding most of it. Now, I want to run a real case of wrf data assimilation. Could you please help and give me a user guide or tutorial with all the details and stages of running wrfda for running real cases, including making background error files (be.dat), conversion of observation into assimilation supported format, etc.? Can anyone guide me through this?

Thank you in advance,
 
Hello ashish_shaji,

This seems to be a recurring theme (I've had many requests for this). I may put together a video tutorial or document with this. As for making the background error files, that's something I'm still learning how to do. Also, for conversion of the observations into data assimilation, it also depends on what data you are trying to assimilate. I can convert RAWS station data into little r format, but I have no clue how to do sounding data. If I get around to making this document I will let you know. Most of the real cases is similar to how you run WRF. All you do is setup everything the same and get your wrfinput files, then when you have those files, link them to fg, process your observations, then run da_wrfvar.exe. I'm sorry I'm not more helpful with this, but when I get more time I can try to help some more.

Thanks,
Jeremy B
 
Thanks, Jeremy, for your reply. I'm also learning how to do these things. If I understand/ learn them properly, I'll also let you know.

Thanks,
Ashish Shaji
 
I feel this is my case, I want to assimilate GPS ZTD, and the whole of thing about little_r formatting is just confusing. Can I get help on which file to use or format to litte_r and how I will go about it.

Thanks
 
I feel this is my case, I want to assimilate GPS ZTD, and the whole of thing about little_r formatting is just confusing. Can I get help on which file to use or format to litte_r and how I will go about it.

Thanks
Hello,

Unfortunately the best advice I can give is to write a script to do it. I had some help from AI with the code and after some adjusting it worked. I would setup a very small case and use your observations. When you go to run obsproc, put it in a log files (./obsproc.exe >& test.log &) and check the errors in that log file. It took me some time to get my observations perfect.

Thanks,
JeremyB
 
Ok, JeremyB.
I would appreciate such guidance so much. And I will definitely wait for the demonstration.
Thank you very much, Sir
 
Hello,

Unfortunately the best advice I can give is to write a script to do it. I had some help from AI with the code and after some adjusting it worked. I would setup a very small case and use your observations. When you go to run obsproc, put it in a log files (./obsproc.exe >& test.log &) and check the errors in that log file. It took me some time to get my observations perfect.

Thanks,
JeremyB
Dear JeremyB,
Sorry, It took me a little long, I just got access to the data. Find the attached sample of my ZTD Observation Data I planned to assimilate for your perusal and kindly help me with the coding set up.

Thanks and regards..
 
Ok, JeremyB.
I would appreciate such guidance so much. And I will definitely wait for the demonstration.
Thank you very much, Sir
Hello CalmTide,

Just know it will be a while before it will be out. I do work and don't have much time outside of work.

Thank you,
JeremyB
 
Hello CalmTide,

Just know it will be a while before it will be out. I do work and don't have much time outside of work.

Thank you,
JeremyB
Alright JeremyB, I appreciate your tight schedule so, I can always wait for you.
Thank you
 
Hello JeremyB,
Trust this finds you well.
Just a soft reminder regarding our discussion. Kindly find time for me, pls
Time is really against me on this experiment.
Thanks
 
Hello JeremyB,
Trust this finds you well.
Just a soft reminder regarding our discussion. Kindly find time for me, pls
Time is really against me on this experiment.
Thanks
Hello CalmTide,

Unfortunately I've been pretty busy with work. What is it you need exactly? Do you just need help installing WRFDA or are you still looking for a way to convert your data to little r?

Thank you,
JeremyB
 
Hello CalmTide,

Unfortunately I've been pretty busy with work. What is it you need exactly? Do you just need help installing WRFDA or are you still looking for a way to convert your data to little r?

Thank you,
I thought as much. Sorry about that (your workload).
I have installed WRFDA already.
I'm actually looking for a way to convert my data to little_r.

Thank you,
CalmTide
 
I thought as much. Sorry about that (your workload).

I'm still looking for a way to convert my data to little_r.

Thank you,
CalmTide
Hello CalmTide,

Unfortunately that's something I'm unable to help you with. In my experiences, every observation is different in terms of converting it to little r files. I have codes for specific observations, but it's too difficult at the moment to create a universal code for it. I know you're working with GPS ZTD, however I've never worked with that dataset and I don't know how to convert it to little r format. The best advice I can give you is write a python script to do it. Use AI as well, it was super helpful pointing me in the right direction. Also lookup example little r files and how they're formatted. Try running obsproc on them, and fixing the bugs it provides. Unfortunately these files are very particular. It took me a couple months to get it right for my observations and those are very different from yours. I'm sorry I can't be more helpful in this situation. In the event your need any form of observation for your DA, try prepbufr files. That is by far the easiest way to assimilate observations into your simulation.

Thank you,
JeremyB
 
Hello CalmTide,

Unfortunately that's something I'm unable to help you with. In my experiences, every observation is different in terms of converting it to little r files. I have codes for specific observations, but it's too difficult at the moment to create a universal code for it. I know you're working with GPS ZTD, however I've never worked with that dataset and I don't know how to convert it to little r format. The best advice I can give you is write a python script to do it. Use AI as well, it was super helpful pointing me in the right direction. Also lookup example little r files and how they're formatted. Try running obsproc on them, and fixing the bugs it provides. Unfortunately these files are very particular. It took me a couple months to get it right for my observations and those are very different from yours. I'm sorry I can't be more helpful in this situation. In the event your need any form of observation for your DA, try prepbufr files. That is by far the easiest way to assimilate observations into your simulation.

Thank you,
JeremyB
This weakens me but thanks for the responses and advice.
To earn me fast pace understanding, can you please share a copy of your final codes you used in thepast. I can guide myself and feed you back how it goes.
Also, I read from the manual that GNSS ZTD data is a special case and it's simpler than other forms of satellite data. I don't know if you can devote a little time to check it out, you will get to understand the specificity more quickly. I'm really behind schedule

Thank you,
CalmTide
 
Hello CalmTide,

I'm unable to share my codes with you unfortunately (it has to do with NDAs and other things like that and I could get in trouble). If it's a simpler form then that should help. If you're able to extract pressure, date, time, temp, wind, wind direction, elevations, etc. then you should be able to create it. Sorry about that.

Thank you,
JeremyB
 
Hello CalmTide,

I'm unable to share my codes with you unfortunately (it has to do with NDAs and other things like that and I could get in trouble). If it's a simpler form then that should help. If you're able to extract pressure, date, time, temp, wind, wind direction, elevations, etc. then you should be able to create it. Sorry about that.

Thank you,
JeremyB
It's okay. I'll get my hands busy then. I'll always carry you along with my progress about it.

Thanks a lot,
CalmTide
 
Top