EGU Stack Parameters for CEMS data

NEI Modeling Platform
1

Hello,

I am part of a project using INMAP (Intervention Model For Air Pollution). The goal is to acquire stack parameters to use with hourly CEMS data for EGUs, and I thought this forum might have some insights.

At first I attempted using NEI data to merge with CEMS and ran into many issues I believe to be corroborated on page 10 of the Technical Support Document for the 2016 North American Emissions Modelling Platform 7.1, but did not see how to access the matched/adjusted CEMS and NEI data.

Next I tried to match stack parameters from EIA Form 860 data (instead of NEI data) and ran into more issues, some similar and some different (e.g. even though the 860-form instructs a new id for each stack, it appears that some ids of “1-3” or “6-10” could represent multiple stacks with the same parameters. Does this seem plausible?). I found a crosswalk for CEMS to EIA, but need EIA boiler ids that do not match 1-1 with stack ids in the EIA stack parameter data.

I would greatly appreciate any suggestions.

Risa

2

Hi, I am not familiar with INMAP but here is the information regarding to processing NEI point EGU sources annual inventory with CEMS dataset.

https://www.cmascenter.org/smoke/documentation/4.8.1/html/ch02s09s19.html

https://www.cmascenter.org/smoke/documentation/4.8.1/html/ch02s08s05.html

3

The flat files used for EPA modeling platforms are already cross-walked to CEMS data as ORIS_FACILITY_ID and ORIS_BOILER_ID columns.

You shouldn’t have to do much more than that. Is there a reason you can’t just use the flat files?

For 2016, the latest files are in the inventories in this zip file:

https://gaftp.epa.gov/Air/emismod/2016/v2/2016emissions/2016fj_point_inventory_15sep2021.zip

4

Thank you for sharing. Currently the project is trying to use actual hourly emissions (and match them to stack parameters) rather than hourly emissions estimated using annual emissions and heat input, but this could prove useful if that changes.

5

Thank you for the response!

  1. Currently the project I am working on is trying to use actual hourly emissions (and match them to stack parameters) rather than hourly emissions estimated using annual emissions and heat input. Otherwise, the flat files could be utilized as is, yes.

  2. Could you direct me to the code/documentation for how that crosswalk between NEI and CEMS was done? Is it the Smkinven program? I found this page where it appears that in the case of multiple boilers per stack, the program considers each boiler to have a different stack, rather than multiple boilers emitting through one stack: “NOTE that this option will internally modify the unit IDs by appending ‘_##’ when there are multiple ORIS/Boiler IDs Continuous Emissions Monitoring (CEM) sources under same unit IDs.”

  3. A project lead at the EIA has confirmed that the CEMS unit ID is a boiler ID, not stack ID, and with the poor crosswalk they have, there are many cases of multiple boilers per stack and multiple stacks per boiler. I was planning on following up to better understand if these instances reflect reality or are more likely to be mistakes.

Do you know if the unit ID in the files you shared the same as the boiler ID, or stack ID?