Hi there,
I need to run CMAQ version 5.3 for the year 2005 and downloaded the emissions modeling platform (EMF) version 4.3 from the EPA website. However, EMF version 4.3 supports the CB05 chemical mechanism. I am wondering if there is any tool to convert CB05 speciation to CB06 speciation. I tried using the GSPRO and GSREF files from the EMF for 2014 but SMOKE returns a cross referencing error with the 2014 files.
Thanks,
Rajesh
Hi, we think that using a more recent GSPRO/GSREF like 2011, 2014, or 2016 platform CB6 profiles would be a viable approach, although SCCs do change over time. Normally we leave the older SCCs in the cross references. Are there specific sectors that you are having trouble with?
Hi Alison,
Thanks a lot for your response. I was trying it just for one sector “other” to see if replacing the GSPRO/GSREF profiles in EMF 2005 with those from EMF 2014 would work. I did several other tests today to see if some combination of SMOKE versions would work. I used SMOKE 2.7 with the EMF 2005 and everything works great for the CB05 mechanism. All different tests are described below.
Test 1:
When I used the GSPRO and GSREF for CB06, spcmat gave the following error:
“TMP_POL_FILE” opened as OLD:READ-ONLY
File name “/glade/scratch/rkumar/NEI_2005/tier3_prop/2005cr_05b/intermed/othar/area_othar_dat/PMC.ncf”
File type GRDDED3
Execution ID “???”
Grid name " "
Dimensions: 36826 rows, 1 cols, 1 lays, 7 vbles
NetCDF ID: 11 opened as READONLY
Time-independent data.
Closing file TMP_POL_FILE
Reading source data from inventory file…
Generating unique lists from inventory data…
Reading speciation cross-reference file…
ERROR: Overflow prevented while parsing line PARSLINE
First 200 characters of line contents are:
“10100300”,“75014”,“SO2”,“! lignite coal”
*** ERROR ABORT in subroutine PARSLINE
Formatting problem.
Test 2:
Since spcmat from SMOKE 2.7 had problem reading the GSPRO and GSREF files, I used spcmat from SMOKE 4.5 but then I get the following errors:
ERROR: No speciation cross-reference available (and no default) for:
Region: 208024
SCC: 2102004000 POL: VOC
ERROR: No speciation cross-reference available (and no default) for:
Region: 208024
SCC: 2102005000 POL: VOC
ERROR: No speciation cross-reference available (and no default) for:
Region: 208024
SCC: 2102006000 POL: VOC
*** ERROR ABORT in subroutine ASGNSPRO
Problem assigning speciation profiles to sources
Based on these two test, I used SMKINVEN, SPCMAT, GRDMAT, and SMKREPORT from SMOKE 4.5 when I switched to GSPRO and GSREF to EMF 2014. Then all the components worked well but I got error in the “temporal” program as listed under Test 3 and Test 4.
Test 3:
I used the “temporal” program for SMOKE 4.5 with the default files from EMF 2005, I get the following error:
ERROR: Bad xref in "TREF" at line 17 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 18 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 19 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 20 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 21 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 22 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 23 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 24 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 25 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 26 for field 9 (profile-ID)
ERROR: Bad xref in "TREF" at line 27 for field 9 (profile-ID)
Maximum number of errors exceeded
*** ERROR ABORT in subroutine PROCTPRO
ERROR: No TXREFs matching inventory found
If we use the “temporal” program from SMOKE 2.7 (used for EMF 2005), I get the following error:
—>> WARNING in subroutine READSET
Requested variable "IFIP " not available in file set "AREA "
*** ERROR ABORT in subroutine RDINVCHR
Error reading variable IFIP from INVENTORY file
Test 4:
Finally, I changed the ATREF and ATPRO to EMF 2014 and used “temporal” program from SMOKE 4.5 but this yielded the following error:
ERROR: No month-of-year profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000
ERROR: No day-of-month nor day-of-week for source 1 : Region: 000000110000 SCC: 00000000002000000000
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day MONDAY
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day TUESDAY
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day WEDNESDAY
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day THURSDAY
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day FRIDAY
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day SATURDAY
ERROR: No hour-of-day nor Met based profile profile found for source 1 : Region: 000000110000 SCC: 00000000002000000000 day SUNDAY
*** ERROR ABORT in subroutine ASGNTPRO
Problem assigning temporal profiles to sources
This error repeated for multiple sources.
Another important thing I found was that the intermediate netcdf file produced by SMKINVEN versions 2.7 and 4.5 are different. The ROW dimension is significantly reduced with version 4.5. Here are portions of the headers of those files.
SMKINVEN 2.7 file header
netcdf area_othar {
dimensions:
TSTEP = 1 ;
DATE-TIME = 2 ;
LAY = 1 ;
VAR = 6 ;
ROW = 79744 ;
COL = 1 ;
SMKINVEN 4.5 file header
netcdf area_othar {
dimensions:
TSTEP = 1 ;
DATE-TIME = 2 ;
LAY = 1 ;
VAR = 4 ;
ROW = 2841 ;
COL = 1 ;
Rajesh, Yes, I think when you go all the way back to the 2005 platform there are differences in the SMOKE features that are used and in the file formats. I would suggest that you start with some of the more recent platforms – like 2011 - and recent SMOKE versions – then you will find more consistency. I believe temporal profile xrefs is one of the file formats that changed along the way. I may be able to provide you with some ancillary data that will work with the more recent SMOKE versions, but I can’t do it this week for various reasons… We aren’t currently supporting those very old versions of SMOKE you are mentioning.
Alison,
Thanks a lot for your prompt response. Yes, I understand that it is difficult to maintain the consistency between different versions and especially when the decadal scales are involved. I’ll try the more recent versions and keep you posted on how it goes.
Best regards,
Rajesh