Colorado Ongoing Basin Emissions (COBE) Measurement-informed inventory results

Jenna Brown1, Michael Moy1, Arthur Santos1, Ethan Rimelman3, Olga Khaliukova2, William Daniels2, Callan Okenberg2, Dorit Hammerling2, Daniel Zimmerle1, Anna Hodshire3

1Energy Institute, Colorado State University, 2Dept. of Applied Mathematics and Statistics, Colorado School of Mines, 3Systems Engineering Dept., Colorado State University

Download Poster PDF

Background – why did we do this study?

In 2024, the Air Pollution Control Division published a verification protocol establishing new requirements for operators reporting methane emissions from upstream oil and gas facilities. Under this protocol, operators must either create their own measurement-informed inventory (MII) or apply the state-developed default factor. This factor accounts for the differences between measured methane emissions and reported methane emissions for the upstream sector. To inform the 2026 default factor, this study, the Colorado Ongoing Basin Emissions (COBE) project, deployed three different aerial measurement technologies to quantify methane emissions from upstream oil and gas facilities across Colorado during 2024.

GOALS of this study:

  • Collect representative methane measurements of upstream oil and gas facilities throughout Colorado
  • Develop MII’s using the aerial emissions data
  • Compare MII’s to operator-reported emissions

Methodology – how did we create a MII for the state of CO?!

Inventory – In the state of Colorado, oil and gas operators report annual greenhouse gas emission estimates to the Oil and Natural Gas Annual Emissions Inventory Reporting (ONGAEIR) database, maintained by the Colorado Department of Public Health and Environment (CDPHE). For this project, the ONGAEIR 2022 dataset was the most recent publicly available inventory.

To model emissions, we use our Mechanistic Air Emissions Simulator (MAES) which is more than just a bottom-up model.

  • Uses Monte Carlo methods
  • Gas composition analysis
  • Inclusion of failure conditions
  • Throughput variation
To create an MII, you need... Box one: - Inventory: -- Site information ---Production data ---Equipment counts/information ---Annual inventory methane estimate --Site configuration --Gas composition Move to box 3 Box 2 Measure Facilities - Measurement methods should report both when emissions are detected and when facilities are scanned with NO emissions detected. Move to box 4 Box 3: MAES Inventory Models 1. Initialize model with inventory inputs: - Model all equipment onsite ('normal emissions') - Mechanistic Model: (physio-chemical processes) - heaters, flares - Traditional model: (EF x AF) — compressors, tanks, component leaks 2. Compare MAES to reported (inventory) emissions (goal:
View a text version of this table at the bottom of the page

MAES MII Results – the cool science stuff!

Average Methane Emissions (mt/year) on the y-axis. All Basins, CJ, Piceance, Other Basins, PS1, PS2, PS4, and PS6 on the x-axis.
Figure 1: MAES MII and MAES inventory model results by basin and prototypical site compared to ONGAEIR values, all broken out by emission sources. These ratios are calculated by dividing the total MAES MII emissions by the total ONGAEIR emissions. Maintenance emissions are not modeled in MAES and are therefore taken from ONGAEIR and added onto MAES results.
CDF of CH4 Emission Rates (on)
Figure 2: Aggregate cumulative distribution function (CDF) of instantaneous methane emission rates across 10,144 simulated facilities using MAES. The black line represents the empirical CDF. Vertical dashed lines mark commonly used detection thresholds, with annotations showing the fraction of emissions below each threshold.
Y-axis: Methane Emissions (kg/h/site) X-axis: All Basins, DJ basin, Piceance Basin, Other Basin, PS1, PS2, PS4, PS6
Figure 3: MAES MII broken out by average methane emission per site divided into the contribution from rates : 0results-5 kg/h, 5-100 kg/h and above 100 kg/h.

But wait.. There is more!

Colorado School of Mines (CSM) was part of the modeling team and developed an independent statistical model to create a MII for the state of Colorado. The CSM team’s model relies solely on rates estimated by measurement technologies, and separate distributions are used to model emissions below the aerial detection thresholds.

RegionMAES RatioAvg. CSM Ratio
State1.292.60
DJ1.663.09
Piceance0.992.44
Others1.291.83
PS21.343.36
PS41.231.24
Y-axis): Methane Emissions (mt/y)

Conclusions and Future work – why these results matter!

Key takeaways:

  • Colorado is often regarded as a leader in methane regulations, yet these measurements suggest that the statewide inventory underestimates emissions by 29%.
  • MII results show that 68% of total emissions are from rates < 5kg/h, contrary to the common finding that large emitters are contributing disproportionately to a region/basin total emissions.
  • Two independent models made different assumptions about incorporating the measurement data and produced varying sets of state-wide emissions totals and ratios.

Future work:

We are currently proposing a continuation of this to help inform the 2027 verification factor. This would allow us to further investigate the differences between CSU and CSM’s results.


Funding and Contact Information​

Funding for COBE was provided by the Colorado Department of Public Health and Environment Agreement #2024*3364

Poster author:

Jenna Brown
Research Associate
CSU Energy Institute
Colorado State University
[email protected]

Project PI:

Anna Hodshire
Assistant Professor
Systems Engineering Department
Colorado State University
[email protected]


References – to see more about this interesting study!​

COBE final report:
Brown, J. A.; Moy, M.; Santos, A.; Rimelman, E.; Okenberg, C.; Daniels, W. S.; Hammerling, D. M.; Zimmerle, D.; Hodshire, A. L. Colorado Ongoing Basin Emissions (COBE) Final Report.

COBE anonymized dataset:
Brown, J. A.; Hodshire, A. Colorado Ongoing Basin Emissions Study (COBE) Anonymized Final Data Set of Emissions Measurements, 2025. https://doi.org/10.5061/dryad.8kprr4z0p.

For other COBE-related posters, see:  

Ethan Rimelman– COBE: Measurement and Source Attribution
Michael Moy – COBE: Estimating Methane Emission Distributions from Aerial Measurements

Equipment at the METEC Site

To create an MII, you need…


Inventory

  • Site information
    • Production data
    • Equipment counts/information
    • Annual inventory methane estimate
  • Site configuration
  • Gas composition

↓


Measure Facilities

  • Measurement methods should report both when emissions are detected and when facilities are scanned with NO emissions detected.

↓


MAES Inventory Models

    1. Initialize model with inventory inputs:
      • Model all equipment onsite (‘normal emissions’)
      • Mechanistic Model: (physio-chemical processes) – heaters, flares
      • Traditional model: (EF x AF) — compressors, tanks, component leaks
    2. Compare MAES to reported (inventory) emissions (goal: <~10%diff), rerun as needed:
      • Identify issues with model or inventory
      • Diagnostic check for facility representation

→


Classify Emissions – in or out of inventory

  • Send emission data to participating operators to request cause analysis
  • Is emission event above MAES facility estimate?
    • Yes – assign failure type and probability
      • Operator cause analysis
      • Aerial imagery

↓

⤷


MAES MII Models

  1. Mechanistic models update inventory model with probability of observing ‘failure’
  2. Traditional models update inventory model with probability of observing ‘failure’ and use aerial distributions.