Error running ptfire-rx on Feb 29, 1980

SMOKE
1

Hello Everyone.

I am working with 2019 NEI emissions re-dated to 1980 and am having problems running ptfire-rx with the script crashing when it get to Feb 29, 1980. Looking at the log files it appears ELEVPOINT fails for this date with elevpoint_ptfire-rx…log file writing the classical “ACRESBURNED” error (see below). I am also not able to generate the smkmerge_ptfire-rx…log file so the script is stopping when processing ELEVPOINT. Comparing the elevpoint_ptfire-rx…log file for Feb 29 to Feb 28 the error appears to be in making stack groups but I am not sure how to correct this so the stack groups are successfully made. I have attached the log files I think relevant.

I have modified the scripts so SMOKE knows 1980 is a leap year.

Any and all suggestions would be greatly appreciated.

Cheers,

David

p.s. I am having a similar problem with ptagfire as well.

image
image1234×827 30.2 KB

elevpoint_ptfire-rx_feb_2024_assessment_19800229_ADMF36.log.txt (40.0 KB)

smkinven_ptfire-rx_2024_assessment.log.txt (30.4 KB)

smkinven_ptfire-rx_feb_daily_2024_assessment.log.txt (81.4 KB)

temporal_ptfire-rx_feb_2024_assessment_19800201.log.txt (942.9 KB)

timelog_ptfire-rx_2024_assessment_ADMF36_daily.txt (20.8 KB)

timelog_ptfire-rx_2024_assessment_ADMF36_onetime.txt (525 Bytes)

2

Elevpoint is trying to read Feb 29th from the PDAY file, but there were not daily inventory emissions on that date because the base year (2019) was not a leap-year. One option is to duplicate a fire within your domain at the daily inventory level and change the duplicate’s date column to Feb 29th. Next rerun ptfire-rx for February starting from the daily smkinven step. This should also solve a similar issue that you are likely to see when you get to Mar 1.

3

Hi James,

I will give this a try. Thanks.

For my understanding…these sources are irregular, so in principle there could be many days when no fires are occurring in the modelling domain. If this is the case, why wouldn’t Elevpoint simply create a zero member stack group? Thanks.

Cheers,

David

4

David,

In more recent versions of SMOKE there was an attempt to add a feature to write a fake stack group (ELEV_WRITE_FAKE_SRC) so this wouldn’t be an issue. According to the elevpoint log file this was turned on in your run but unfortunately here the issue persisted. It isn’t clear if the behavior is related to the leap-day or a general issue in the implementation.