SOOFIE sensors detect methane well under EPA's 5 kg per hour limit

EPA found that SLB's SOOFIE network, which uses an AI-based algorithm to detect methane events, reached 90% probability of detection at about 0.6 kg/hr in controlled tests.

  • Not rated society

    EPA's finding rests on controlled releases, and no emission cut in the field is reported.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Intended benefit.
Evidence

foundstudiedreportedclaimed

An independent regulatory document supports the benefit.

Compared with

The reference limit of 5 kg/hr for detection. Measured.

Without the system
5 kg/hr
With it
About 0.6 kg/hr at 90% probability of detection
The system’s part
Not isolated. EPA's reviewers wrote that a proprietary AI-based algorithm detects emission events and that a hybrid model with machine learning sets emission rates. They did not compare detection with and without the algorithm, and the sensors and the plume model also shape the result.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
SOOFIE (Systematic Observation of Facility Intermittent Emissions) emissions monitoring system
Made by
SLB (application first filed by ChampionX, LLC)
Used by
Oil and gas operators (approved use)
Where
Colorado, United States
Stage
Tested in a laboratory
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Monitoring Processes, Materials, or Surroundings, Estimating the Quantifiable Characteristics of Products, Events, or Information
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Regulatory document, by authors with no tie to the system.

    U.S. Environmental Protection Agency. Approval letter for the SLB SOOFIE emissions monitoring system as Methane Alternative Test Method MATM-013, with the Acceptance Justification memorandum, 23 September 2025 (docket EPA-HQ-OAR-2024-0619).

    SLB demonstrated a solution sensitivity (90% Probability of Detection) below the 5 kg/hr alerting threshold, based on 2024 ADED test conducted at METEC at Colorado State University.
    A 90% POD threshold was identified at approximately 0.6 kg/hr, which is below the reference detection limit of 5 kg/hr.

Details

AI system

Workflow

Each device measured methane with a semiconductor sensor made of metal oxide, along with temperature, pressure and humidity. SLB set the number and placement of at least three devices with an algorithm that uses wind data. An AI-based algorithm classified each reading as inside or outside an emission event.

Follow-up

Conditions for replication

Wind must carry methane to the sensors within the system's operating range. Devices must stand at least 1.5 times the source height from each probable source.

Entry details

Sector

Climate change mitigation / methane emissions monitoring.

Purpose

SLB built SOOFIE to detect, locate and quantify methane emissions at oil and gas facilities with a network of point sensors.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). SOOFIE sensors detect methane well under EPA's 5 kg per hour limit (AIBD-0826). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0826.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions