Applying NEI Emissions to other Years

Alexander Frie

unread,

Jul 16, 2021, 9:41:37 AM7/16/21

to wrf-chem-anthro_emiss

Hello All,

I believe there are many users, myself included, that have been slightly confounded about how to translate NEI 2014/2017 emissions to alternate years. Due to the presence of Weekends/Weekdays/Holidays, It's not immediately obvious how easily transfer emissions from year to year without introducing artifacts or losing some of the information provided by NEI. Based on past comments from the developers, I’ve put together a short procedure for adapting NEI to alternate years.

To run EPA anthro_emis for different years, and maintain weekend/holiday information I believe you can use the following steps:

1. Make Slight modifications to the epa.f90 file.

1a. change:

dateHdr(7) = &

(/ 'all ', 'aveday_N', 'aveday_Y', 'mwdss_N ', 'mwdss_Y ', 'week_N ', 'week_Y' /)

To

dateHdr(8) = &

(/ 'date ', 'aveday_N', 'aveday_Y', 'mwdss_N ', 'mwdss_Y ', 'week_N ', 'week_Y' , 'all' /)

1b. change all instances of 7 to 8

1c. rebuild anthro_emis

2. Modify the smk_merge_dates_... file to include conversions for the year you’re interested in running.

2a. Set first column to the dates for the year you want to use, then set the other columns to the dates in the inventory that correspond to each day for each date type. I've attached a spreadsheet which does this for 2021 using NEI 2017.

2b. Add data from 2a to smk_merge_dates and save as csv.

3. Change names (or make copies) of sectorlist files in the anthro emission folder so the year aligns with the years you're interested in.

Hopefully this can be useful to someone in the future! Developers, please let me know if you anticipate these changes causing more issues than solutions.

Cheers,

Alex

smk_merge_year_change.xlsx

Alexander Frie

unread,

Jul 16, 2021, 9:47:27 AM7/16/21

to wrf-chem-anthro_emiss, Alexander Frie

Sorry, found an error in the spreadsheet. This one should be correct.

Alex

smk_merge_year_change_v2.xlsx

elham baranizadeh

unread,

Nov 30, 2021, 5:42:06 PM11/30/21

to wrf-chem-anthro_emiss, frie...@umn.edu

Hi Alex,

Thanks so much for sharing this. this is exactly what I need. I'm going to to give it a try- I may bother you with follow-up questions.

Cheers,

Elham

Alexander Frie

unread,

Nov 30, 2021, 6:10:20 PM11/30/21

to elham baranizadeh, wrf-chem-anthro_emiss

Hello Elham,

Happy to hear this is useful to you! I've attached an updated excel file which includes an additional lookup table that I had neglected to include in the first versions.

Cheers and good luck!

Alex

_________________________

Alexander Frie, PhD (he/him/his)

Research and Fellowship Coordinator

Minnesota Sea Grant

University of Minnesota

smk_merge_year_change_v3.xlsx

elham baranizadeh

unread,

Nov 30, 2021, 6:27:58 PM11/30/21

to wrf-chem-anthro_emiss, frie...@umn.edu

got it, thank you!

Elham

elham baranizadeh

unread,

Dec 1, 2021, 12:44:33 PM12/1/21

to wrf-chem-anthro_emiss, frie...@umn.edu

Hi Alex,

I implemented your method in epa_anthro_emiss, and so far it's going smoothly. I just had some educational questions:

1) what is the difference between the fields with "_Y" suffix vs the ones with "_N" suffix in smk_merge_dates_Year.txt file?

According to this document (http://crcao.org/wp-content/uploads/2020/03/CRC-Project-A-119_Final_Report.12March2020.pdf), I Understand the each field's meaning as follows:

"aveday" = One representative day per month
"mwdss" = Four representative days per month: [Mon, Weekday (Tue-Fri), Sat, Sun]
"week" = Seven representative days per month: one for each day of the week
"all" = No representative days; each day is processed individually

2) do you know how these date types (i.e. aveday, mwdss, week etc; basically the data in your lookup table) are determined for 2017 inventory? any document you could refer me to?

Thanks a lot for your help.

Cheers,

Elham

elham baranizadeh

unread,

Dec 1, 2021, 4:19:55 PM12/1/21

to wrf-chem-anthro_emiss, frie...@umn.edu

Hi Alex,

My epa_antho_emiss stops at the Labor Day holiday. Here is the details:

I am running the tool for the period from start_output_time = '2018-08-25_00:00:00' to stop_output_time = '2018-09-30_00:00:00'. The run stops at the Labor Day holiday (Sep 3, 2018), because the aveday_Y
, mwdss_Y, and week_Y dates equal to the same date i.e. Sep 3rd, 2018, this results in an error because there is no "ptnonipm" emission source/file on this date.

I noticed in your case year 2021, the Labor Day falls on SEp 6th for which there is no ptnonimp emission file either. DO you mind letting me know how you resolved this issue?

Thanks a lot,

Elham

On Tuesday, November 30, 2021 at 3:10:20 PM UTC-8 frie...@umn.edu wrote:

Reply all

Reply to author

Forward