search

epri-dev / SCICHEM

This repository utilizes Git LFS to serve large files - please follow the README-Download-Instructions.txt file in order to access these files. This is a version of the SCIPUFF puff model with chemistry. It allows for the simulation of power plant plumes as a series of puffs that are transported in the atmosphere while undergoing chemical transformation.
17 stars 4 forks source link

Issues with SCICHEM Simulations Using Daily WRF input files: Identical Results on Consecutive Days #8

Open stwcn1 opened 1 month ago

stwcn1 commented 1 month ago

Hello, I ran a SCICHEM simulation for the period from July 1 to July 5, using WRF output data for meteorological files. The WRF output data are daily simulation results. However, there are several unreasonable issues with the simulation results:

  1. In the simulation from July 1 to July 5, the simulation results for July 2 seem to be duplicated on July 3 and July 4, as shown in Figure 1. (a-c: July 2,d-f: July 3,g-i : July 4)The spatial distribution and maximum values of ozone concentration are identical, which is not reasonable.

2.I conducted another simulation for the period from July 3 to July 4. The results of this simulation differ from those of the July 1 to July 5 simulation, as shown in Figure 2.(a-c: first simulation ,d-f: second simulation) The spatial distribution and maximum values of ozone concentration are inconsistent.

Why is this happening? Should I perform daily SCICHEM simulations to obtain more reasonable results given such daily WRF output meteorological data? Thank you in advance!

c93eede90a9f98fbf30a697b10efa7e 47959e0a17c982ea7e09f294932abbe
DougHenn commented 1 month ago

My guess is that the meteorology is not being updated past July 2. First check the project log file (.log) for lines beginning with "Reading met file" to see that the WRF file is being read as the calculation proceeds. The following lines (beginning with "t =") will give the time in hours from the start of the calculation. If the meteorology is being updated, then I would plot the WRF wind fields to see if they vary past July 2, especially comparing the July 3 or 4 fields from the first file with the individual day file fields. I don't know if you have tools to plot WRF NetCDF files. I use in-house software but there are public tools available. As an alternative, you can save the meteorology from the SCICHEM calculation, which the GUI can then plot. In the project .msc file you would have to set

LOUT_MET = T, TOUT_MET = 3600. Set to WRF update interval in seconds LOUT_3D = T, LOUT_2D = T, LFORMAT = F F produces binary output. T produces human-readable files which are much larger and slower to read

I would be happy to look at the project log file, if you'd like.

Regards, Doug

stwcn1 commented 1 month ago

Thank you very much for your reply. I checked the log files and found that the meteorological data read was not according to the multiple WRF files I specified in the LST file, but only one of the files was read. How should I modify the LST file? Here are my log file and LST file. log.txt lst.txt

DougHenn commented 1 month ago

Well, I am perplexed. The list of WRF files looks correct. But the log file indicates that only wrfout_d02_2018-08-01_12_00_00 is being read, meaning the entire month of July has been skipped over. It also shows that the first file in the list (wrfout_d02_2018-06-30_12_00_00) was read to set up the terrain field, so SCICHEM was clearly reading the list file correctly. Perhaps it's misinterpreting the dates and times in the WRF files, although I have not seen that before. However, it seems like wrfout_d02_2018-08-01_12_00_00 has July dates on it (although I'm deducing this from the log file output), which woudl be odd.

I assume there are multiple times on the WRF files. If so, what is they expected interval?

Not sure what to suggest. If the WRF files aren't too big, maybe you can send them to me. Actually, I would only need the first three or so, and maybe the last (2018-08-01). And I would need the project input files .inp, .scn and *.msc. (I already have the WRF list file.)

If you can't send them, I would suggest specifying a single file at a time and seeing what time(s) SCICHEM thinks it's reading by looking in the log file.

stwcn1 commented 1 month ago

The dates specified in the WRF files range from 12 AM of the previous day to 6 PM of the following day, totaling 30 hours. There is overlapping time between consecutive WRF files. Could this be the source of the issue? Additionally, I would like to ask you whether there is a significant difference between the mcw file obtained through MMIF conversion and directly inputting the Wrfout file?

DougHenn commented 1 month ago

Possibly. The code is supposed to handle that situation, but I wrote that a few years ago and it probably hasn't been tested much since then (and perhaps there were other changes that could have affected how that is handled). I don't know if you can get WRF files without overlaps, but it would be good to test that. And, if you send me a few WRF files, I can test it directly.

Regarding the translation of WRF files by MMIF, I recommend using the WRF directly if possible. One issue, I believe, is that MMIF averages the velocity components, which are located on a staggered grid, onto common (cell center) points, thus reducing the accuracy somewhat. The effect on the dispersion calculation is usually small but can be noticeable in regions with strong velocity gradients such as over significant terrain or land cover changes.

stwcn1 commented 1 month ago

Hello, how can I send you the wrfout data? The file size limit on GitHub is 25MB.

DougHenn commented 1 month ago

How big are the files?

stwcn1 commented 1 month ago

One day's worth of wrfout data is approximately 500MB.

DougHenn commented 1 month ago

If you give me your email address, someone in my company can send you a link to upload

stwcn1 commented 1 month ago

I'm sorry for the delayed response. Here is my email address(ztwa1218@gmail.com). Please send me the link to upload the file.

DougHenn commented 1 month ago

My turn to apologize but I was away this week and so didn't follow-up on setting up the link until today. Hopefully you'll get an email today.

From: stwcn1 @.> Sent: Sunday, May 19, 2024 8:56 PM To: epri-dev/SCICHEM @.> Cc: DougHenn @.>; Comment @.> Subject: [EXTERNAL] Re: [epri-dev/SCICHEM] Issues with SCICHEM Simulations Using Daily WRF input files: Identical Results on Consecutive Days (Issue #8)

I'm sorry for the delayed response. Here is my email @.**@.>). Please send me the link to upload the file.

— Reply to this email directly, view it on GitHub [github.com]https://urldefense.com/v3/__https://github.com/epri-dev/SCICHEM/issues/8*issuecomment-2119489373__;Iw!!NFAdMAnI0yk!FtSQ-bU1UIaxHBIGLzhLDQGEUD_clsZ9Xw244CoP4nDEuJDUOoWPs-oEnNjDc33ePdjgRJQyi3tGVXvcfZUphShGIC2j$, or unsubscribe [github.com]https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AOG3X4AUKPX2TAAN7OYTL6DZDFC3ZAVCNFSM6AAAAABHZJZYGKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMJZGQ4DSMZXGM__;!!NFAdMAnI0yk!FtSQ-bU1UIaxHBIGLzhLDQGEUD_clsZ9Xw244CoP4nDEuJDUOoWPs-oEnNjDc33ePdjgRJQyi3tGVXvcfZUphfJtKlNi$. You are receiving this because you commented.Message ID: @.***>

'NOTICE: This email message and all attachments transmitted with it may contain privileged and confidential information, and information that is protected by, and proprietary to, Parsons Corporation, and is intended solely for the use of the addressee for the specific purpose set forth in this communication. If the reader of this message is not the intended recipient, you are hereby notified that any reading, dissemination, distribution, copying, or other use of this message or its attachments is strictly prohibited, and you should delete this message and all copies and backups thereof. The recipient may not further distribute or use any of the information contained herein without the express written authorization of the sender. If you have received this message in error, or if you have any questions regarding the use of the proprietary information contained therein, please contact the sender of this message immediately, and the sender will provide you with further instructions.'

DougHenn commented 1 month ago

You should have gotten the link (via email).

Please send the first two WRF files and the last. Also, the project input files (.inp, .scn, .msc and list file)

From: Henn, Douglas [US-US] @.> Sent: Friday, May 24, 2024 11:19 AM To: epri-dev/SCICHEM @.>; epri-dev/SCICHEM @.> Cc: DougHenn @.>; Comment @.***> Subject: Re: [EXTERNAL] Re: [epri-dev/SCICHEM] Issues with SCICHEM Simulations Using Daily WRF input files: Identical Results on Consecutive Days (Issue #8)

My turn to apologize but I was away this week and so didn't follow-up on setting up the link until today. Hopefully you'll get an email today.

From: stwcn1 @.> Sent: Sunday, May 19, 2024 8:56 PM To: epri-dev/SCICHEM @.> Cc: DougHenn @.>; Comment @.> Subject: [EXTERNAL] Re: [epri-dev/SCICHEM] Issues with SCICHEM Simulations Using Daily WRF input files: Identical Results on Consecutive Days (Issue #8)

I'm sorry for the delayed response. Here is my email @.**@.>). Please send me the link to upload the file.

— Reply to this email directly, view it on GitHub [github.com]https://urldefense.com/v3/__https://github.com/epri-dev/SCICHEM/issues/8*issuecomment-2119489373__;Iw!!NFAdMAnI0yk!FtSQ-bU1UIaxHBIGLzhLDQGEUD_clsZ9Xw244CoP4nDEuJDUOoWPs-oEnNjDc33ePdjgRJQyi3tGVXvcfZUphShGIC2j$, or unsubscribe [github.com]https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AOG3X4AUKPX2TAAN7OYTL6DZDFC3ZAVCNFSM6AAAAABHZJZYGKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMJZGQ4DSMZXGM__;!!NFAdMAnI0yk!FtSQ-bU1UIaxHBIGLzhLDQGEUD_clsZ9Xw244CoP4nDEuJDUOoWPs-oEnNjDc33ePdjgRJQyi3tGVXvcfZUphfJtKlNi$. You are receiving this because you commented.Message ID: @.***>

'NOTICE: This email message and all attachments transmitted with it may contain privileged and confidential information, and information that is protected by, and proprietary to, Parsons Corporation, and is intended solely for the use of the addressee for the specific purpose set forth in this communication. If the reader of this message is not the intended recipient, you are hereby notified that any reading, dissemination, distribution, copying, or other use of this message or its attachments is strictly prohibited, and you should delete this message and all copies and backups thereof. The recipient may not further distribute or use any of the information contained herein without the express written authorization of the sender. If you have received this message in error, or if you have any questions regarding the use of the proprietary information contained therein, please contact the sender of this message immediately, and the sender will provide you with further instructions.'

stwcn1 commented 1 month ago

Hello, I have already uploaded the WRF files and the project input files. Please check them.

DougHenn commented 1 month ago

I have downloaded test.zip but it only contains the input files and one WRF file, which is not one of the files listed in the .lis file. It's "wrfout_d01_2018-04-09_12_00_00", not "wrfout_d02_2018-04-09_12_00_00" or the other two files. I'm not sure if there's a problem on our side. Can you check what files you uploaded?

I also forgot to tell you to include the .IMC file as well as the ambient background file (maybe cz.amb). Please send them to me (attachments should be fine).

In the meantime, I'm running with different chemistry and background files as well as the single WRF file. That seems to be running fine but doesn't really address your issues.

From: stwcn1 @.> Sent: Sunday, May 26, 2024 9:21 PM To: epri-dev/SCICHEM @.> Cc: DougHenn @.>; Comment @.> Subject: [EXTERNAL] Re: [epri-dev/SCICHEM] Issues with SCICHEM Simulations Using Daily WRF input files: Identical Results on Consecutive Days (Issue #8)

Hello, I have already uploaded the WRF files and the project input files. Please check them.

— Reply to this email directly, view it on GitHub [github.com]https://urldefense.com/v3/__https://github.com/epri-dev/SCICHEM/issues/8*issuecomment-2132484303__;Iw!!NFAdMAnI0yk!GhOgI4KtlQoyLWJ5L9zGLKI4UehDyaMrTWueHLkTdFMN0XFhvQteq9v47dy1XjDoBqaO9VxEO5Mg4dlg2WX80ll1hieD$, or unsubscribe [github.com]https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AOG3X4DBLEHXJ4LYIU3B6FTZEKDB7AVCNFSM6AAAAABHZJZYGKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZSGQ4DIMZQGM__;!!NFAdMAnI0yk!GhOgI4KtlQoyLWJ5L9zGLKI4UehDyaMrTWueHLkTdFMN0XFhvQteq9v47dy1XjDoBqaO9VxEO5Mg4dlg2WX80j2VYYQo$. You are receiving this because you commented.Message ID: @.***>

'NOTICE: This email message and all attachments transmitted with it may contain privileged and confidential information, and information that is protected by, and proprietary to, Parsons Corporation, and is intended solely for the use of the addressee for the specific purpose set forth in this communication. If the reader of this message is not the intended recipient, you are hereby notified that any reading, dissemination, distribution, copying, or other use of this message or its attachments is strictly prohibited, and you should delete this message and all copies and backups thereof. The recipient may not further distribute or use any of the information contained herein without the express written authorization of the sender. If you have received this message in error, or if you have any questions regarding the use of the proprietary information contained therein, please contact the sender of this message immediately, and the sender will provide you with further instructions.'

stwcn1 commented 1 month ago

I apologize for the incorrect upload of the WRF files. I tested the wrfout_d01 files and found the same issue: the terrain seems to be read from the first file in the lst file, and the meteorological data does not seem to correctly recognize the time in the wrfout files. Additionally, I tested a single wrfout file that contains all the simulation times and found that this issue does not occur. Perhaps I should consider simulating on a daily basis or using wrfout files that contain all the simulation times. Additionally, here are the IMC and AMB files I used for my simulation. imc.zip

DougHenn commented 1 month ago

Using WRF files with all of the simulation times may be a good workaround for you, particularly if they aren't too large. However, I do want to find out if SCICHEM is reading times incorrectly (or there is some other problem). Are you able to provide a few files with single output times? (d01 grids would be okay if they show the same behavior.) I can have my colleague send another link if so.

stwcn1 commented 1 month ago

Hello, I did not understand which files you want me to provide when you say "few files with single output times." Could you please explain that again?

In my recent testing, I simulated two days, from April 9th at 00:00 to April 11th at 00:00. The meteorological data used were the d01 WRF files that I sent to you. When I checked the log file, even though three WRF files are specified in the lis file: wrfout_d01_2018-04-09_12_00_00, wrfout_d01_2018-04-10_12_00_00, wrfout_d01_2018-04-11_12_00_00,it seems that only the data from wrfout_d01_2018-04-09_12_00_00 was read. I have sent all these files to you in the previous link. You can modify the simulation time in the inp file and test it. Here is my log file. log.txt