An operator’s methane, counted in CO2 terms, could rise 12% on paper in 2025 filings with no change at any site. The US Environmental Protection Agency (EPA), the federal environmental regulator, revised those conversion values with effect from 1 January 2025 [1]. The values feed its Greenhouse Gas Reporting Program (GHGRP), under which about 8,000 facilities, including large emission sources, are required to report each year [2]. Analysts comparing operators, tracking a trend or pricing an acquisition read those totals, so the conversion behind them matters as much as the emissions.
A greenhouse gas (GHG) is a gas that warms the Earth by absorbing energy and slowing its escape to space [3]. Three of them, methane, carbon dioxide and nitrous oxide, account for almost all of an oil and gas operation’s reported total. Reporting adds the three together as carbon dioxide equivalent (CO2e), which requires a conversion factor per gas [1]. That factor is the global warming potential (GWP), the energy a tonne of gas absorbs over a set period relative to a tonne of CO2 [3].
The GWP values are published by the Intergovernmental Panel on Climate Change (IPCC) in its periodic assessment reports [4]. The fourth, fifth and sixth of those reports are known as AR4, AR5 and AR6, and each revised the values [4]. The factor is not a constant, and the value a programme requires changes the reported total [4]. The Greenhouse Gas Protocol’s 100-year methane values from those three reports run from 25 to 30, a fifth apart, without a molecule changing.
Colorado is the worked example because both emissions studies cited here were done in its Denver-Julesburg Basin, though its composition figures are statewide. Readers in Texas, New Mexico, Oklahoma, Wyoming, Pennsylvania, West Virginia and Ohio get a table linking each agency’s emissions inventory or reporting page. This post covers, in order, which gases an operation emits, what CO2e means, and which GWP basis a programme requires. It closes on where reported numbers come from and where they break, for operators’ environmental staff and for analysts comparing filed totals.
Quick Answer: Oil and gas operations emit mainly methane, carbon dioxide and nitrous oxide. Reported totals under EPA’s programme are summed as CO2 equivalent using 100-year global warming potentials [1]. Methane’s value is 25 under AR4, 28 under AR5 and 27.0 under AR6 for non-fossil sources, and 30 or 29.8 for fossil sources [4]. The EPA reporting programme moved to the revised values effective 1 January 2025 [1].
Which Greenhouse Gases Oil and Gas Operations Emit
Oil and gas operations report three greenhouse gases: methane, carbon dioxide and nitrous oxide; combustion emits all three, venting and leaks add methane and CO2. Methane (CH4), the main component of natural gas, arrives from venting and leaks, and as unburned methane from flares and engines. Carbon dioxide (CO2), the reference gas against which all global warming potentials are calculated, comes from combustion and from the gas itself. Nitrous oxide (N2O), a by-product of burning fuel, comes from engine and heater exhaust.
EPA’s oil and gas reporting counts these three, not everything an operation releases [5]. The same streams carry ethane and heavier hydrocarbons, which the median Colorado produced-gas analysis puts at about 12.3 mole % ethane alone [6]. Benzene, toluene, ethylbenzene and xylene travel with them and are regulated as hazardous air pollutants under a separate rule [7].
Those are counted by air-quality programmes rather than converted to CO2 equivalent.
Nitrous oxide is by far the smallest of the three: 0.2 of the 322 million tonnes CO2e the sector reported for 2023 [5]. It forms in combustion exhaust, from engines, heaters and flares, rather than coming from the produced stream. It stays in the reporting tables because Subpart W requires it, not because its factor of 265 to 298 makes it large.
| Gas | Where it comes from in upstream operations | Scales with |
|---|---|---|
| Methane | Venting, fugitive leaks, incomplete combustion | Equipment counts and operating time |
| Carbon dioxide | Combustion exhaust, and CO2 already in the produced gas | Fuel burned and gas composition |
| Nitrous oxide | Combustion exhaust | Fuel burned |
Carbon Dioxide Is Already in the Gas
Produced gas arrives at the surface carrying carbon dioxide before anything is burned [6]. Checking a basin’s produced-gas CO2 means pulling its filed gas analyses, which TetraSoft Atlas, our subscription platform, already holds [6]. Taking the latest produced-gas analysis of each Colorado well, more than 4,000 wells report CO2, at a median of 2.19 mole % [6]. Nearly 61% of those analyses carry 2% or more [6].
The fifth percentile sits at 0.04 mole % and the ninety-fifth at 3.02 mole % [6].
That matters for two calculations at once. Subpart W vent estimates count both gases from their mole fractions, though at Colorado’s median composition vented CO2 is about 0.3% of the vent’s CO2e. A combustion estimate has to avoid counting the same carbon dioxide twice.
The practical test is which stream the carbon dioxide left in. Carbon dioxide vented with the gas belongs to the vent estimate. Carbon dioxide leaving a stack belongs to the combustion estimate, whether it was formed there or arrived in the fuel.
Why Methane Dominates Vented and Leaked Emissions
Methane dominates upstream vented and leaked emissions for two reasons that compound, starting with what the gas is made of. Methane is the main component of natural gas, so leaked or vented gas is mostly methane by volume. The same latest analyses give more than 4,000 Colorado wells a median methane content of 72.9 mole % [6].
The second reason is the conversion factor, which counts that mass at roughly thirty times its own weight. A typical Colorado produced gas, about 73% methane by volume, is about 55% methane by mass yet still converts at over ten times its weight.
The consequence is that each tonne of methane error moves the CO2e total 28 times as far as a tonne of CO2 error. In reported totals, carbon dioxide still leads: 2023 onshore production filings carried 63.9 million tonnes CO2e of CO2 against 27.5 of methane [5]. EPA’s profile uses AR4’s 25 for methane, and at the current 28 methane would be about 30.8, still behind.
Combustion equipment reported 44.5 million tonnes, so an equal 10% error moves the total more on combustion, though methane inventory errors can far exceed 10%. An intensity metric divides that CO2 equivalent total by production, so both the factor and the conversion carry into it. Colorado’s intensity programme is built on exactly that ratio, and is covered in Colorado Regulation 7: GHG Intensity Verification Factor.
What Carbon Dioxide Equivalent Actually Means
CO2 equivalent is a unit of account rather than a measurement. A mass of methane becomes a mass of CO2 equivalent by multiplying it by a global warming potential over a chosen time horizon [4]. EPA’s reporting programme uses the 100-year horizon [1]. That horizon is itself a choice.
Methane’s effect is concentrated early because it leaves the atmosphere quickly. A 100-year figure averages a strong early effect across a long tail. Reporting rules name one horizon so that submissions are comparable; EPA’s is 100 years, while New York’s statewide emission limits use 20-year values.
The factor is revised as the science is reassessed; methane’s 100-year value rose 12% from AR4 to AR5 [4]. The natural gas unit converter converts a methane or nitrous oxide mass to CO2e on either EPA’s current values or the AR4 ones.
| 100-year GWP | AR4 | AR5 | AR6 |
|---|---|---|---|
| Methane, non-fossil | 25 | 28 | 27.0 |
| Methane, fossil | not published | 30 | 29.8 |
| Nitrous oxide | 298 | 265 | 273 |
AR4 and AR5 as tabulated by the Greenhouse Gas Protocol; AR6 from IPCC Working Group I, Chapter 7, Table 7.15. [8]
Fossil and Non-Fossil Methane Are Different Rows
The fossil row is not a rounding variant of the non-fossil one [4]. It adds the radiative forcing of the carbon dioxide formed when that methane oxidises in the atmosphere [4]. Under Greenhouse Gas Protocol guidance, methane from fossil fuel fugitive sources such as oil and gas systems takes the fossil row [4].
The same guidance puts methane in combustion exhaust on the non-fossil row instead [4]. Using the fossil row there would count the same carbon twice. EPA’s reporting rule makes no such split: its Table A-1 lists a single methane value, 28 [9].
AR4 published no separate fossil value [4]. That absence is a fact about the report rather than a gap to fill with the non-fossil figure.
The same tonne of methane is worth between 25 and 30 tonnes of CO2 equivalent depending on which cell of the GWP table applies. [4]
Which Basis a Reporting Programme Requires
The basis is chosen by the programme, not by the reporter. EPA amended the Greenhouse Gas Reporting Rule’s general provisions to reflect revised global warming potentials [1].
That rule published on 25 April 2024 and took effect on 1 January 2025 [1]. It moved Table A-1, the rule’s list of each gas’s GWP, to the AR5 100-year values [1]. For certain gases that AR5 does not list, it took AR6 values instead [1].
Methane rose from 25 to 28 and nitrous oxide fell from 298 to 265 [9]. The methane figure it adopted is the AR5 row without the fossil adjustment, so a GHGRP filing uses 28 rather than 30 [1].
Programmes move at different times, so two filings in the same year can use different factors. Two totals compare directly when they use the same conversion basis; otherwise one must be recomputed from mass.
Subpart W is the GHGRP section for oil and gas [10]. It covers ten industry segments, starting with onshore production, offshore production, processing, transmission compression and underground natural gas storage [10]. The rest are LNG storage, LNG import and export, natural gas distribution, gathering and boosting, and transmission pipelines [10].
EPA proposed in September 2025 to end most GHGRP reporting and suspend the remaining Subpart W segments until reporting year 2034 [11]. That proposal is not final, and the reporting-year 2025 deadline now falls on 10 March 2027 [12]. What it currently requires is covered in Subpart W Proposed Rule: Is GHGRP Reporting Still Required?.
Get regulatory updates and emissions data insights in your inbox. Subscribe to the TetraSoft newsletter.
Where the Numbers Come From, and Where They Break
Most of a reported inventory is calculated rather than measured at the source: combustion from fuel records, much of the rest from factors. For some sources, equipment counts are multiplied by emission factors; others use measurement or engineering calculations, and the results are summed. The conversion to CO2 equivalent is then applied on top [1].
Two independent choices therefore sit between an operation and its reported number. Either can move the total without anything changing in the field.
Readers in seven other producing states find their agency’s emissions inventory or reporting page in the table below, plus EPA’s national facility data. Most of these inventories track ozone-forming and toxic pollutants; Colorado collects oil and gas GHG reports annually, and EPA publishes GHGRP facility totals.
Colorado is one of eight states TetraSoft Atlas covers, and we plan to expand Atlas to every US state [13]. The eight produced 68% of US crude oil, including lease condensate, and 83% of US marketed natural gas in 2025 [14].
| State | Agency | Record to use |
|---|---|---|
| Texas | Commission on Environmental Quality | Point Source Emissions Inventory |
| New Mexico | Environment Department, Air Quality Bureau | Emission Inventory |
| Oklahoma | Department of Environmental Quality, Air Quality Division | Emissions Inventory |
| Wyoming | Department of Environmental Quality, Air Quality Division | Emissions Inventories |
| Pennsylvania | Department of Environmental Protection | Unconventional Natural Gas Emission Inventory, which also reports methane |
| West Virginia | Department of Environmental Protection, Division of Air Quality | Emissions Inventory |
| Ohio | Environmental Protection Agency | Fee emission reports, the yearly report large permitted sources file |
| Nationwide | US EPA | Facility-level GHG emissions data, GHGRP totals for facilities at or above 25,000 t CO2e a year; an onshore production facility is one operator’s wells in a basin |
Six of the seven state rows hold air-pollutant inventories, Pennsylvania’s also counting methane; the federal row holds the GHGRP CO2e totals.
An Inventory Is Mostly Arithmetic on Factors
A factor describes an average piece of equipment under normal operation. Applied to a count, it produces an expected mass rather than an observed one. The result is a plausible modelled total, not a direct measurement of what the equipment emitted.
How independent observation compares against inventories built this way is covered in Why Satellites Show About 2x More Methane Than Inventories.
Where a Measurement-Informed Inventory Fits
The gap sits in the mass, not in the conversion. A global warming potential can be looked up and checked against the report that published it. The mass it multiplies can be looked up in EPA’s facility data, as the figure each operator calculated and reported.
TetraSoft MAES, the Mechanistic Air Emissions Simulator, is an emissions model developed at Colorado State University (CSU) [15]. The University of Texas at Austin is a co-developer [16]. TetraSoft provides commercial and academic access to MAES through a special license with CSU [17].
MAES’s mechanistic models estimate a site’s emissions from the fluid flow through its equipment and each unit’s state [16]. It generates an expected emission range rather than a single value [15].
MAES does not choose a global warming potential and it does not file a report. Santos et al. (2025), led by this post’s author, combined that model with aerial survey measurements and operator inventories in the Denver-Julesburg Basin [18]. What MAES addresses is the mass the conversion factor multiplies.
Frequently Asked Questions
Is carbon dioxide or methane the bigger problem in oil and gas?
Per tonne, methane: EPA’s reporting rule counts each tonne as 28 tonnes of CO2 [9]. In reported totals, carbon dioxide leads: 262 million tonnes CO2e of CO2 for 2023 against 60 of methane at AR4’s 25, about 67 at 28 [5]. For local air quality the answer differs again. Benzene and the other hazardous air pollutants are governed by their own standard, and the volatile organic compounds in the same stream form ground-level ozone [7].
Why does the same emission produce different CO2e numbers?
Because the conversion factor is a published choice that has been revised [4]. Methane’s 100-year value is 25 under AR4 and 30 under the AR5 fossil row [4]. The same measured mass differs by a fifth between those two cells.
Which basis should an operator use?
Whichever the receiving programme names; confirm it rather than assume it. EPA’s reporting rule moved to the AR5 100-year values effective 1 January 2025 [1]. That put methane at 28 and nitrous oxide at 265 [9]. A voluntary or international framework may name a different assessment report.
What is the difference between fossil and non-fossil methane?
The fossil row adds the effect of the carbon dioxide formed when the methane oxidises [4]. Under that guidance, fugitive methane from oil and gas systems takes the fossil row and combustion exhaust the non-fossil one, to avoid counting carbon twice [4]. A GHGRP filing makes no such split; Table A-1 lists one methane value, 28 [9].
Why is there no AR4 fossil value?
Because that assessment did not publish one [4]. The separate fossil row appears from AR5 onward [4]. Filling the gap with the non-fossil figure would present an invented number as a published one.
Does nitrous oxide matter in an upstream inventory?
Nitrous oxide stays small even after conversion: 0.2 of the 322 million tonnes CO2e the sector reported for 2023 [5]. Its 100-year value is 298 under AR4 and 265 under AR5 [4]. It arises from combustion rather than from the produced stream.
Does venting release carbon dioxide as well as methane?
Yes, because vented gas carries its own carbon dioxide along with the methane. Colorado produced-gas analyses, taking the latest for each well, carry a median 2.19 mole % carbon dioxide [6]. Subpart W counts both gases, though at that composition vented CO2 is about 0.3% of the vent’s CO2e.
Can an operator change its reported total without changing operations?
Yes, and that is the uncomfortable part. A revised global warming potential changes the conversion [1]. A revised emission factor changes the mass, and neither requires anything to happen at the site. A year-on-year trend that spans a basis change is measuring both at once.
Interested in building a Measurement-Informed Inventory for your operations? Contact us to learn about our MAES-based estimation services.
Disclosure: Arthur Santos, the author of this post, is lead author of the Santos et al. (2025) study cited here and a co-author of the Mollel et al. (2025) study.
References
- 89 FR 31802, April 25, 2024. Revisions and Confidentiality Determinations for Data Elements Under the Greenhouse Gas Reporting Rule; effective 1 January 2025. https://www.federalregister.gov/documents/2024/04/25/2024-07413/revisions-and-confidentiality-determinations-for-data-elements-under-the-greenhouse-gas-reporting
- EPA, Greenhouse Gas Reporting Program (GHGRP). https://www.epa.gov/ghgreporting
- EPA, Understanding Global Warming Potentials. https://www.epa.gov/ghgemissions/understanding-global-warming-potentials
- Greenhouse Gas Protocol, Global Warming Potential Values, August 2024. https://ghgprotocol.org/sites/default/files/2024-08/Global-Warming-Potential-Values%20%28August%202024%29.pdf
- EPA, Greenhouse Gas Reporting Program, Petroleum and Natural Gas Systems sector profile, 2011-2023 data, October 2024. https://www.epa.gov/system/files/documents/2024-10/subpartw_profile_2023.pdf
- TetraSoft Atlas, multi-state wellhead gas composition dataset. Gas chromatography analyses across eight states, sourced from state agency filings and the USGS Natural Gas Compositional Analyses Dataset, verified 2026-11-03. https://tetrasoftco.com/atlas-landing.html#pricing
- 40 CFR Part 63, Subpart HH, National Emission Standards for Hazardous Air Pollutants From Oil and Natural Gas Production Facilities; Subpart HHH, Natural Gas Transmission and Storage Facilities. https://www.ecfr.gov/current/title-40/chapter-I/subchapter-C/part-63/subpart-HH; https://www.ecfr.gov/current/title-40/chapter-I/subchapter-C/part-63/subpart-HHH
- IPCC AR6 Working Group I, Chapter 7, Table 7.15. Intergovernmental Panel on Climate Change, Sixth Assessment Report, Working Group I. https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_Chapter07.pdf
- 40 CFR Part 98, Subpart A, Table A-1, Global Warming Potentials, as amended at 89 FR 31894, April 25, 2024. https://www.ecfr.gov/current/title-40/chapter-I/subchapter-C/part-98/subpart-A/appendix-Table%20A-1%20to%20Subpart%20A%20of%20Part%2098
- EPA, Subpart W: Petroleum and Natural Gas Systems. https://www.epa.gov/ghgreporting/subpart-w-petroleum-and-natural-gas-systems
- 90 FR 44591, September 16, 2025. Reconsideration of the Greenhouse Gas Reporting Program, proposed rule. https://www.federalregister.gov/documents/2025/09/16/2025-17923/reconsideration-of-the-greenhouse-gas-reporting-program
- 91 FR 65112, October 14, 2026. Extending the Reporting Deadline Under the Greenhouse Gas Reporting Rule for Reporting Year 2025 to March 10, 2027, interim final rule. https://www.federalregister.gov/documents/2026/10/14/2026-20982/extending-the-reporting-deadline-under-the-greenhouse-gas-reporting-rule-for-reporting-year-2025-to
- TetraSoft Atlas landing page. https://tetrasoftco.com/atlas-landing.html
- EIA, Crude Oil Production and Natural Gas Marketed Production by state, annual 2025. https://www.eia.gov/dnav/pet/pet_crd_crpdn_adc_mbbl_a.htm; https://www.eia.gov/dnav/ng/ng_prod_sum_a_EPG0_VGM_mmcf_a.htm
- CSU METEC, Measurement-Informed Inventory. https://metec.colostate.edu/measurement-informed-inventory/
- Mollel et al., 2025, ACS ES&T Air, 2, 723-735, DOI 10.1021/acsestair.4c00168. “Using Prototypical Oil and Gas Sites to Model Methane Emissions in Colorado’s Denver-Julesburg Basin Using a Mechanistic Emission Estimation Tool.” https://doi.org/10.1021/acsestair.4c00168
- TetraSoft MAES platform page. https://www.tetrasoftco.com/maes-platform/maes-landing.html
- Santos et al., 2025, ACS ES&T Air, 2, 1598-1611, DOI 10.1021/acsestair.5c00089. “Using Measurement-Informed Inventory to Assess Emissions in the Denver-Julesburg Basin.” https://doi.org/10.1021/acsestair.5c00089
This post is for informational purposes only and does not constitute legal or compliance advice. Consult qualified legal counsel or a compliance professional for guidance specific to your operations and jurisdiction.
Interactive tool
What the CO2e basis does to your number
The same measured mass converts differently under each published basis. This holds the mass fixed and shows every answer at once.
Spread across the published bases
0.7%
- Lowest
- 2,980 t CO2e
- Highest
- 3,000 t CO2e
By basis
| Basis | CH4 GWP | N2O GWP | Total t CO2e |
|---|---|---|---|
| AR4 | not published | 298 | -- |
| AR5 | 30 | 265 | 3,000 |
| AR6 | 29.8 | 273 | 2,980 |
AR4 published no separate fossil methane value. That row reads "not published" rather than repeating the non-fossil figure, because the absence is a fact about the report.
AR4 and AR5 values as tabulated in Greenhouse Gas Protocol, Global Warming Potential Values, August 2024; AR6 from IPCC AR6 Working Group I, Table 7.15. All are 100-year values. Which basis applies to a filing is set by the receiving programme, not by this tool. Nothing leaves your browser.
Comments
0Share your thoughts. All comments are moderated before appearing.
No comments yet. Be the first to share your thoughts!