I am running SMOKE v4.8.1 for the 2019 emission platform. When I run ‘Monthly_onroad_ca_adj_RPD_daily_12US1_2019ge_cb6_19k.csh’, it showed the following error. I don’t know why the Grid Cell temperature is a large negative value. How to fix it? All the data are downloaded from the CMAS google drive.
*** ERROR ABORT in subroutine MOVESMRG
ERROR: Grid cell temperature -459.67 out of range of minimum emission factor data 0.00
Date and time 0:00:00 June 1, 2019 (2019152:000000)
What meteorological data are you using for this run (especially the METCRO2D files)? It is possible some data got corrupted at some stage of the data transfer. Can you provide the Movesmrg log for the day that failed?
Attached is the log file.
movesmrg_RPD_onroad_ca_adj_jun_2019ge_cb6_19k_20190601_12US1_cmaq_cb6ae7.txt (162.6 KB)
I used the EQUATES 2019 MCIP files. To test your hypothesis, I switched to another set of MCIP files from our own WRF simulations, but got the same error. So it looks like it not caused by the MCIP input file.
Hao
We have not seen this issue when we processed this year of data. Have you considered trying a newer version of SMOKE?
Perhaps there is a memory allocation issue causing the value to be zero that is not properly handled.
Also, are you using a natively compiled version of SMOKE or one you downloaded?
I am using the SMOKE v4.8.1 contained in the zip file from the UNCI Google Drive. All other sectors such as RPD/RPH/RPHO/RPP/RPS/RPV are finished for CONUS and CA, exceipt this one.
I can try new version of SMOKE. Any suggestion on the specific version for the 2019ge platform?
I tried SMOKEv4.9 precompiled executbile, and got the same results.
Is there any option to overide this warining like 2016 NEI platform: Mobile sector error: grid cell temperature out of range?
One more thing. the Value of -459.67 is absolutel zero in Fahrenheit. Is it possible there is a conversion error if MCIP to SMOKE? MCIP has the temperature in K, may 0 for missing value?
We agree with this being absolute 0 and did not find this issue when we ran SMOKE with the same data, so are wondering if there could be a data download error, or an error in the compiled version of SMOKE running on your computer. If you scan the data with a netCDF data inspection tool, do you see 0 in there for the temperature data?
I don’t think it is caused by data corruption, because I have this error for all days in June 2019.
Second, I tried another MCIP dataset using WRF simulations conducted in our group, but found the similar errors.
Lastly, the same MCIP inputs are successufly used to conduct 12US1 RPD/H/HO/P/V/S runs and California RPH/H0/P/V/S runs. The only problem happened in ca_adj_RPD.
Is it possible the eftables of California are corrupted during data upload or transfer?
Hello again. We recall that there were some segmentation faults cropping up prior to SMOKE v5.1.
Can you try using the latest version of SMOKE (5.2.1) to see if you still have the error?
If the error still occurs, we could re-examine the CA emisison factors.
I tried to use SMOKE v5.1, and it stopped at the same location without showing the -459.67 error. The command line shows the following error. Seems like the temperature still has problem.
forrtl: severe (408): fort: (3): Subscript #2 of the array MTMP_INVT has value 0 which is less than the lower bound of 1
Image PC Routine Line Source
movesmrg 000000000041ADEA movesmrg 1123 movesmrg.f
movesmrg 00000000004080BD Unknown Unknown Unknown
libc-2.28.so 0000145F54C327E5 __libc_start_main Unknown Unknown
movesmrg 0000000000407FDE Unknown Unknown Unknown
6747.576u 15.814s 2:09:49.39 86.8% 0+0k 389600+199064io 0pf+0w
now checking log file /data/chemstg99/haohe/2019v1/2019ge_cb6_19k/intermed/onroad_ca_adj/RPD/logs/movesmrg_RPD_onroad_ca_adj_jun_2019ge_cb6_19k_20190601_12US1_cmaq_cb6ae7.log
ERROR: detected in movesmrg
ERROR: Running smk_run for part 4 for 20190601
Can you try setting MTMP_OUTPUT_YN = N and let us know if that helps
Set MTMP_OUTPUT_YN = N works and solve the problem! I searched online but didn’t find the explanation about this option. Can you explain what function it controls? Thanks!
When MTMP_OUTPUT_YN = Y, Temporal will output additional intermediate files which support the generation of emissions reports via Smkreport such as hourly reports. It must require additional data that triggered this error. We are so glad to hear that the problem is fixed! Hopefully tracking it down has had some useful side effects. Using a newer version of SMOKE is a good idea for a variety of reasons because it includes various bug fixes and feature additions as well.