To confirm my answer from yesterday:
Yes, Monthly_onroad_inv_respeciate_12US1_EQUATES_2017.csh is the only script to run for the onroad-as-nonpoint approach. There is no RPD/RPP/etc to worry about, and there is also no separate California/Texas sector with this method.
Yes, assign EMISINV_A to MYR_2011_SMOKE_MOVES_MOVES3_AQstyle_05oct2021_nf_v2.csv and comment out EMISINV_B because this 2011 file has both gas and diesel in it (unlike some of our other packages).
Running the Monthly_onroad_inv_respeciate_12US1_EQUATES_2017.csh script I now get several errors to do with gridding surrogates
ERROR: Gridding surrogate code 244 is not in surrogates file, but was assigned to source:
Region: 056045 SCC: 2209210540
*** ERROR ABORT in subroutine ASGNSURG
Problem assigning gridding surrogates to sources
The log file (with several lines of the same error for different surrogate codes and regions removed, as the whole file is too large to attach):
grdmat_onroad_inv_respec_dec_2011ek_saprc_v6_11g_12CA2.txt (83.9 KB)
The surrogate codes that are resulting in errors are 205, 242, 244, 258, 259, 304, 306, 307 and 506.
I’m using the mgref and srgdesc files for the 2017 platform:
mgref_onroad_MOVES3_28aug2020_v0.txt (13.8 KB)
srgdesc_CONUS12_2017NEI_17dec2019_29apr2020_v1.txt (5.1 KB)
As you noted, the package we pointed you to uses some newer surrogates.
You can download the 12km surrogates from the 2017 or 2016v2 platform and it should work:
Are the surrogates in the ‘USA_100_NOFILL.txt’ file?
The files from the 2017 platform (https://gaftp.epa.gov/Air/emismod/2017/spatial_surrogates/surrogates_CONUS12_2017NEI_23aug2021.zip ) are what I’m using.
The 100 surrogate is one of them – that is population.
There are quite a few others.
It is specifically the USA_100_NOFILL.txt file that is being called in Monthly_onroad_inv_respeciate_12US1_EQUATES_2017.csh
Monthly_onroad_inv_respeciate_12US1_EQUATES_2017.csh (7.1 KB)
Are the surrogate files called anywhere else?
I am still getting the same errors when using the surrogates from https://gaftp.epa.gov/Air/emismod/2016/v2/spatial_surrogates/CONUS12_surrogates_04mar2021_contents.txt or https://gaftp.epa.gov/Air/emismod/2017/spatial_surrogates/surrogates_CONUS12_2017NEI_23aug2021_contents.txt
The log file is:
grdmat_onroad_inv_respec_dec_2011ek_saprc_v6_11g_12CA2.txt (48.6 KB)
We think that the SRGDESC file you are using is corrupted (we reviewed the original post).
The file only has the Canada (CA) and Mexico (MX) surrogates, not the regular US surrogates listed, so SMOKE is not finding the US surrogates it needs.
You can use this one instead:
srgdesc_CONUS12_2017NEI_17dec2019_29apr2020_v1.txt from this package: https://gaftp.epa.gov/air/emismod/2016/v2/ancillary_data/2016v2_gridding_ge_dat_16sep2021.zip
The SRGDESC file points to all the specific surrogate files, such as the ones your Grdmat is showing as missing.
We are wondering if the srgdesc you are using was downloaded directly from our packages or if you have edited it somehow?
If the file as downloaded was incorrect we would like to correct it.
I did not edit any of the files. I downloaded it from https://gaftp.epa.gov/Air/emismod/2017/ancillary_data/ge_dat_for_2017gb_gridding_29jun2020_contents.txt with the other files needed for the 2017 platform.
The SRGDESC files in this package look correct. There is one for US and one for Can & Mex. Is there any chance you are running with the Can/Mex file instead of the US one somehow?
Please double check that the surrogates themselves are located in the correct place (/home/lfk/SMOKE/2011_NEI/ge_dat/gridding/surrogates/CONUS12_2017NEI_04mar2021/).
Grdmat is saying it can’t find any of the surrogates it is looking for, so either the surrogates aren’t where Grdmat is looking for them, or they are missing from the SRGDESC file that is being assigned.
Redownloading the SRGDESC file worked, I’m not sure why the one I was using wasn’t for the US.
The onroad sector is now completing, but when trying to merge all the files I’m again getting the error
ERROR: Variable “VOC_INV” in file 13 has units “g/s”
that are inconsistent with a previous file that had units “moles/s” for this variable
*** ERROR ABORT in subroutine MRGGRID
Inconsistent units for common variables among the files!
(log file:
mrggrid_all_2011ek_saprc_v6_11g_20111201_12CA2_saprc.txt (11.9 KB))
The units for VOC_INV in the onroad files are still in g/s, but are in moles/s for all the other units. How do I set the units to be moles/s? I can’t find any reference to this in the run scripts or in the SMOKE guide.
Glad to hear that you made progress! Thank you for sharing your log.
We would have expected VOC_INV to have units of g/s in all sectors and are unsure how it ended up with moles/s units in certain sectors and not others. According to the Mrggrid log, VOC_INV has g/s units in np_oilgas, ptnonipm, pt_oilgas, and onroad_inv, and moles/s units in all other sectors.
GSPRO differences from one sector to the next could cause this. Have you used different GSPROs for rwc versus np_oilgas, for example? If so, perhaps those can be made consistent.
However, since VOC_INV isn’t used by CMAQ and is only carried through SMOKE for summary purposes, the best option could be to remove VOC_INV from the premerged emissions altogether, or at least from the for sectors that are causing the SMOKE error. One way to do that is m3tools (m3xtract), but that can be tedious if you have lots of files. We think the NCO utility can also do this and that might be easier to work with if they have experience with it.
OR this may be the easiest way to do it: Remove all references to “VOC_INV” from the GSPRO and reprocess np_oilgas, ptnonipm, pt_oilgas, and onroad_inv. The new emissions should be identical to the previous ones except VOC_INV should be missing and then the merge process should hopefully work.
Good luck!
The np_oilgas, ptnonipm and pt_oilgas sectors were using gspro_cmaq_saprc07t_08oct2021.txt (gspro_cmaq_saprc07t_08oct2021.txt (3.7 MB)) while the other sectors were using gspro_SAPRC07TB_16apr2014.txt (gspro_SAPRC07TB_16apr2014.txt (2.0 MB)).
I changed all sectors to use gspro_cmaq_saprc07t_08oct2021.txt and re-ran and now everything is merging.
Thanks for all the help getting SMOKE running with saprc speciation.
This is great news – glad that it is working…