In data_cleaning.clean_cems, when we impute missing CO2 values, we use energy source codes from the power sector data crosswalk, and if missing, we calculate plant-month weighted average emission factors based on EIA-923 data. However, when calculating CH4 and N2O values, we had previously not validated that there were no missing energy source codes, so some of the resulting CH4 and N2O data will be missing.
We need to fix this either by:
Implementing a similar two-step process that we use for CO2
Using a different source of data for assigning energy source codes to CEMS data (currently our approach for assigning energy source codes to EIA-923 data is not consistent with how we do it for CEMS data).
I'd like to try and patch this before our initial public release.
In
data_cleaning.clean_cems
, when we impute missing CO2 values, we use energy source codes from the power sector data crosswalk, and if missing, we calculate plant-month weighted average emission factors based on EIA-923 data. However, when calculating CH4 and N2O values, we had previously not validated that there were no missing energy source codes, so some of the resulting CH4 and N2O data will be missing.We need to fix this either by:
I'd like to try and patch this before our initial public release.