search

USEPA / flowsa

Library that attributes resource use, waste, emissions, and loss to economic sectors
MIT License
23 stars 19 forks source link

Refac ghg national mecsallocationfix #321

Closed catherinebirney closed 1 year ago

catherinebirney commented 1 year ago

Update attribution of EIA MECS using BEA. Resolves attribution errors.

WesIngwersen commented 1 year ago

running 2016 locally from e964ef3

bl-young commented 1 year ago

I'm just waiting for 2020 to finish then I can post all to sharepoint.

WesIngwersen commented 1 year ago

There are a number of balance check issues in my local run of 2016. I confirm i updated my local BEA Make and Use FBAs before the run. All the FBAs for the GHGI tables referred to below were created on 2023 -01-26 so reflect the new GHGI

petroleum in EPA_GHGI_T_3_22b (1.53% less than FBA)

coal_manufacturing in EPA_GHGI_T_A_9 (4.69% less) natural_gas_manufacturing in EPA_GHGI_T_A_9 (0.03% less than FBA)

coal_manufacturing in EPA_GHGI_T_3_8 (0.23% more than FBA)) ng_manufacturing in EPA_GHGI_T_3_8 (0.03% less than FBA) ng_manufacturing in EPA_GHGI_T_3_9 (0.03% less than FBA)

bl-young commented 1 year ago

At a glance I was not able to identify where those data loss are coming from. May need @catherinebirney to review the EIA MECS/BEA functions. I have a hard time following those. I will say that one possible thing to explore is whether difference in NAICS years could be causing a small issue see #212

WesIngwersen commented 1 year ago

Basically petroleum is handled just as it was before for stationary combustion, where as there were no errors in the petroleum activity set before