Fuzzy Math: Why Pennsylvania’s Oil and Gas Pollution Numbers Don’t Add Up

Authors: Nadia Steinzor & Raphael Breit

As discussed in a previous blog post, in 2019, Pennsylvania adopted emission reduction goals as part of its Climate Action Plan, committing the state to greenhouse gas (GHG) reduction goals of 26% by 2025, and 80% by 2050, compared to 2005 levels.

For reductions related to the oil and gas industry, Pennsylvania  measures its progress through the Department of Environmental Protection (DEP)’s Marcellus emissions inventory, which tracks operator self-reported GHG emissions estimates.

Accurate emissions estimates are crucial for assessing Pennsylvania’s progress toward its goals. Unfortunately, as Earthworks has written about before, DEP–just like other state and federal agencies–is likely undercounting its emissions. A recent analysis indicates that oil and gas methane emissions in Pennsylvania are more than 15 times higher than what operators report to the state.


EPA and DEP Numbers Don’t Match

The emissions data picture gets fuzzy fast. Earthworks wanted to get a better sense of the scope of oil and gas pollution in relation to climate goals, so we analyzed the emissions that several well known facilities submitted to DEP and the US Environmental Protection Agency (EPA)’s Greenhouse Gas Reporting Program (GHGRP). We found that operators often didn’t report their emissions, and when they did they reported completely different estimates to DEP and EPA–even for the same location in the same year. 

This trend was consistent across major facilities and dates back years. But there were several possible reasons why the numbers might not match up that we needed to investigate before drawing conclusions.


Different Reporting Requirements

One possible reason for reporting discrepancies is that operators are not required to report the same things to each agency.

Operators calculate emissions released from different sources of pollution at a well site or facility (e.g., stacks, tanks, valves, dehydrators). These are based on the engineering specifications of particular equipment, resulting in calculations known as “emissions factors.” Operators then calculate emissions based on the number of hours the specific source was in use, the quantity of fuel burned, the molecular composition of the gas that was emitted, temperature, and other considerations.

Critically, there is no universal methodology. Each agency is free to use its own emission factors. State and federal policies define which pollution sources must be included and which are exempt. Operators sometimes mix and match emission factors for the same equipment. But as Earthworks dug deeper, we discovered that there was even more to the story.


Different and Confusing Volume Formats

Most agencies and organizations express emissions in metric tons of carbon dioxide equivalent (CO2e) per year. But we discovered that DEP reports its numbers in imperial (US, or “short”) tons and does not convert different pollutants into CO2e. So DEP reports 1 ton of methane as 1 ton of methane, but EPA presents 1 ton of methane as its value of CO2e.

To smooth out this difference, Earthworks tried to convert DEPs numbers. First we converted DEPs numbers into metric tons and then we attempted to convert them to CO2e. CO2e numbers are calculated using global warming potentials (GWPs) – multipliers which estimate the impact of GHG pollutants compared to carbon dioxide over a set time period. We attempted this conversion using three different GWPs:

  • 25, which is what EPA uses in its inventory. This number reflects the 100-year GWP for methane in the IPCC’s Fourth Assessment Report (2007).
  • 34, which is the 100-year GWP for methane in the IPCC’s Fifth Assessment Report (2014).
  • 86, which is the 20-year GWP for methane in the IPCC’s Fifth Assessment Report (2014).


Something is Missing
Unfortunately, despite concerted effort, none of these calculations resulted in a match between the data from DEP and EPA inventories. Nor could we discern a clear pattern to explain the discrepancies. There are any number of reasons why. DEP isn’t clear about which GWP it uses. Some components and pollution sources which may be included in one estimate may be excluded from the other. 

The bottom line is that operators are allowed to report completely different emissions numbers to different agencies for the same time period. Without a full picture of where these numbers come from and what they represent, there is no way to know if any of the reported numbers are accurate or not. 

This makes it very difficult to  trust the emissions numbers from either agency as representative of reality. As a result, it is currently impossible to get an accurate understanding of the scale of oil and gas emissions at these facilities or across Pennsylvania–and in turn to know how far off the mark the state is in meeting its climate goals, and what it must do to get on track.


Example Table – Bluestone Gas Plant’s Reported Emissions

Bluestone Gas Plant DEP Inventory DEP Methane Data Converted to CO2e EPA GHG Inventory
Methane IPCC AR4 100 yr GWP (x25) IPCC AR5 100 yr GWP (x34) IPCC AR5 20 yr GWP (x86) Methane – CO2e
2014 59 1,327 1,805 4,565 4,049
2015 59 1,328 1,806 4,567 1,799
2016 25 568 773 1,955 2,962
2017 35 792 1,077 2,724 5,164
2018 32 730 993 2,512 5,790


Houston Gas Plant DEP Inventory Methane Converted to CO2e EPA GHG Inventory
Methane IPCC AR4 100 yr GWP (x25) IPCC AR5 100 yr GWP (x34) IPCC AR5 20 yr GWP (x86) Methane – CO2e
2014 45 1,015 1,380 3,491 8,624
2015 40 905 1,231 3,113 2,378
2016 75 1,699 2,310 5,844 3,455
2017 99 2,243 3,050 7,715 6,866
2018 227 5,153 7,009 17,728 6,939


Trillith Compressor Station DEP Inventory Methane Converted to CO2e EPA GHG Inventory
Methane IPCC AR4 100 yr GWP (x25) IPCC AR5 100 yr GWP (x34) IPCC AR5 20 yr GWP (x86) Methane – CO2e
2014 12 281 382 966 N/A
2015 38 852 1,159 2,930 13
2016 20 456 620 1,569 N/A
2017 29 666 906 2,291 N/A
2018 10 233 317 801 N/A