Missing 2018gg_18j SMOKE-MOVES eftables

SMOKE Emissions Inventory
1

I am running the onroad RPD emissions using the 2018v2 platform, but my jobs have been aborted with errors such as:

“Could not open file rateperdistance_smoke_nata_20210219_2018-20210208_45045_7.csv at line 1 of the county factors file list.”

I found that the EFTABLES were not uploaded for the 2018v2 platform (https://gaftp.epa.gov/Air/emismod/2018/v2/2018emissions/moves_eftables/). I also attempted to use EFTABLES from the 2017 and 2019 platforms, but those tables have the prefix “rateperdistance_smoke_aq_cb6_saprc”, which does not match the expected format in the provided scripts.

Could someone kindly assist with updating the EFTABLES for 2018gg_18j?

2

Based on the TSD for the 2018v2 platform posted in the EPA air emissions modeling website for that platform, the onroad emissions are unchanged from those in the original 2018 platform. The MOVES eftables for that platform are here: https://gaftp.epa.gov/Air/emismod/2018/v1/2018emissions/moves_eftables/

Have you tried processing with these 2018 emission factor tables?

3

Hi Alison, I did download eftables from the 2018gc platform.

But after replacing “smoke_nata” with “smoke_aq_cb6_saprc”, I found there was no file named “rateperdistance_smoke_aq_cb6_saprc_20210219_2018xxxx.csv”, only something like “rateperdistance_smoke_aq_cb6_saprc_20210217_2018xxxx.csv” existed. All the files used for the 2018gg platform shown in the corresponding scripts cannot match those in the 2018gc platform.

Is it okay to use these eftables coming from the 2018gc platform instead?

4

Hi Alison, I just created some MRCLIST files and updated the scripts, but I encountered an issue when running the jobs for RPD emissions. The error message I received was as follows:

*** ERROR ABORT in subroutine MOVESMRG
ERORR: Missing speed distribution profile for:
SCC: 00000000002203510372 || FIPS: 000000001049 || DOW: WEEKDAY
Date and time 0:00:00 Jan. 3, 2018 (2018003:000000)

I have downloaded all the EFTABLES from the 2018gc platform, but based on the error, it seems that the file list may be incomplete. I’ve attached my script, the MRCLIST file, and the log file for your reference. Could you check if all the EFTABLES have been uploaded?

Monthly_onroad_RPD_daily_4SE2_2018gg_18j.csh (8.8 KB)
MRCLIST_MOVES3_2018gc_20210217_RPD_23feb2021_v0.txt (51.2 KB)
movesmrg_RPD_onroad_jan_2018gg_18j_20180101_4SE2_cmaq_cb6ae7.log.txt (402.5 KB)

5

2018v2 onroad is the same as 2018gc onroad, except that 2018gc was a HAP-CAP case, while 2018v2 was CAP-focused. For 2018v2, SMOKE-MOVES was not rerun, instead the HAPs were removed from the 2018gc gridded emissions.

The 2018v2 script package references the HAP-CAP EF tables, which should work properly even if you want to run a CAP-focused case (because the MEPROCs and GSPRO included in the package only select the CAP species). You seem to be trying to use the CAP-focused EF tables instead of the HAP-CAP EF tables, which should work, with the benefit being smaller EF table file sizes, but you need to edit the EF table filenames in the MRCLIST files to do so. In addition to the smoke_nata versus smoke_aq_cb_saprc filename difference, there is also a difference in the date stamp in the filenames, 20210219 versus 20210217. The reason why is just an artifact of the moves2smk step; it doesn’t indicate any kind of underlying difference in the factors themselves.

So in short, the “smoke_aq_cb6_saprc_20210217” EF tables are okay to use.

6

Thank you for this information, Alison! I did get the emis_mole_RPD_onroad_*.ncf even though there are some error messages as I mentioned above. Does this mean this kind of error in my case is acceptable?

7

Using the AVGSPD_SCCXREF from 2020 platform should fix that error:

https://gaftp.epa.gov/air/emismod/2020/ancillary_data/onroad_ge_dat_2020ha2_22sep2023.zip

filename: ge_dat/onroad/AVGSPD_SCCXREF_RPD_MOVES3R1_25apr2022_v0.csv

8

Thank you for providing this file, Alison! It did work!

I replaced the AVGSPD_SCCXREF file among all the onroad emission processes. The RPD scripts will spend 2.5 hours (wall-clock time) for a one-day run. So I ran multiple scripts in parallel for different months. However, I have encountered the following errors for some months:

SCRIPT ERROR: timetracker script found multiple existing entries
              for the following primary keys in the file already:
      Sector   = onroad
      Job Name = Annual_area
      Program  = grdmat
      Run Date = inv
      This should not happen, so script does not know how to
      replace the entries for these keys
ERROR: Problem calling timetracker from smk_run script

I’m wondering if this means that different processing streams can’t run on the same node?

9

You can ignore the errors from the timetracker. They don’t impact the functioning of the core SMOKE programs.