This guide walks through the tools you need, the monitoring methods available, how to interpret what your sensors tell you, the mistakes that get operators fined, and the regulatory frameworks tying it all together.
Key Takeaways
- Ground-based, continuous, airborne, and satellite methods differ in accuracy, scale, and cost.
- EPA's OOOOb pathway lets approved continuous systems replace routine audio, visual, olfactory (AVO)/optical gas imaging (OGI) inspections.
- Separating normal emissions from true fugitive leaks means catching real risk, not chasing false alarms.
- Poor sensor placement and infrequent inspections cause most missed leaks.
- Continuous, AI-validated monitoring complements quarterly inspections, cutting costs while strengthening regulatory defensibility.
What You Need to Monitor Methane Emissions at Oil & Gas Sites
Picking the right sensing technology and site configuration determines whether you catch real emissions events or generate noise that regulators won't accept as evidence. Detection thresholds, sightlines, and power availability all factor into the decision.
Tools and Technologies Required
Operators typically draw from five categories of equipment:
- Handheld or fixed Optical Gas Imaging (OGI) cameras: infrared devices that visualize invisible methane plumes
- Continuous emissions monitoring systems: combining video, acoustic, and gas sensors for round-the-clock coverage
- Airborne sensors: drone- or aircraft-mounted spectrometers for wide-area screening
- Satellite-based methane sensors: for basin-level or facility-level detection of large releases
- EPA Method 21 handheld analyzers: for component-level leak confirmation using a 500 ppm threshold
Most operators now layer these technologies rather than choosing one. Continuous ground sensors handle daily coverage, while periodic aerial or satellite passes screen for anything the ground network might miss at scale.
Preconditions and Setup
Selecting the right equipment mix is only the first step. Before deploying any continuous system, confirm it meets EPA's technical definition. Under 40 CFR 60.5398b, automated Periodic Screening requires a qualifying sensor system capable of measuring a mass emission rate and detecting releases during each quarterly screening cycle.
Beyond the regulatory threshold, a handful of site conditions determine whether a deployment actually works:
- Sightlines: sensors need unobstructed views of tanks, wellheads, and compressors, with wind direction and speed accounted for
- Power and connectivity: continuous systems need reliable onsite power plus either data transmission or onsite edge computing for low-connectivity wellsites
- Calibration window: AI-based platforms typically need 1-2 days of initial site learning to distinguish normal operating signatures from genuine anomalies
Well Checked Systems' Zensory.ai™ platform, for example, runs a roughly two-day AI Site Learning cycle per location before it begins issuing validated alerts. That baseline period matters because it's what allows the system to tell a known process emission apart from an actual fugitive leak later on.
Methods to Monitor Methane Emissions in Oil & Gas Operations
The right method depends on required accuracy, coverage area, budget, and whether you're pursuing a traditional LDAR pathway or an EPA-approved alternative-technology route.
Method 1: Optical Gas Imaging (OGI) Camera Inspection
Handheld or tripod-mounted infrared cameras let technicians see methane plumes that are invisible to the naked eye. This has been the standard periodic LDAR survey method for years.
What you need: an MWIR OGI camera, calibrated bump-test gas, and an inspection route checklist.
Steps:
- Conduct an Audible, Visual, Olfactory (AVO) walk-around before imaging
- Scan components (valves, flanges, hatches, connectors), moving downwind to upwind
- Document any visible plume, log its location, and schedule repair per regulatory timelines
OGI excels at pinpointing exact leak locations for repair crews. Its weakness is timing: surveys run quarterly or semiannually, so any leak that starts and stops between visits simply goes unrecorded.
Method 2: Continuous Multi-Sensor AI Monitoring
Fixed, always-on systems layer multiple sensor types with AI to detect and validate methane events around the clock rather than during scheduled visits. This is the category EPA's alternative-technology pathway was designed to accommodate.
What you need: a multi-sensor edge-AI monitoring unit, onsite power, and data logging capability.
Well Checked Systems' Zensory.ai™ platform pairs three sensor types in one unit: high-resolution video, acoustic anomaly AI, and Long-Wave Infrared (LWIR) OGI, covering video, acoustic, and infrared sensing simultaneously. The LWIR approach delivers day/night detection at roughly one-third the cost of traditional mid-wave IR cameras, which matters when you're deploying across dozens or hundreds of sites.
Steps:
- Install sensor units with clear sightlines to key emission points across the site
- Allow the AI learning cycle to establish the site's normal operating signature
- Monitor the acknowledge-dispatch-mitigate alert workflow, acting only on validated anomalies rather than raw sensor feeds
Once a fugitive event is confirmed (not a false alarm), the system's Zentinal Core™ layer alerts the team, and its quantification layer, Zentinal IQ™, calculates volume, duration, and rate. That acknowledge-dispatch-mitigate cycle is built to close within 24 hours, a window operators cite as key to supporting a documented, timely response on validated events.
The trade-off: continuous systems need upfront installation and a brief per-site calibration period. In exchange, they deliver a regulatory-defensible record and filter out the false alarms that plague raw sensor feeds.
Method 3: Airborne and Satellite Remote Sensing
Aircraft-, drone-, or satellite-mounted spectrometers measure methane concentrations over large areas. These tools are built for finding "super-emitter" events and validating facility-level reporting, not for catching small component leaks.
What you need: a contracted aerial survey provider or access to satellite methane datasets, plus wind and meteorological data for plume modeling.
Steps:
- Schedule a survey flight or satellite pass over the target basin or facility
- Combine concentration data with wind models to estimate emission rate and source
- Cross-reference flagged locations against site records to dispatch repair crews
Detection thresholds vary sharply by platform, and that gap matters more than most operators realize:
| Platform | Published detection threshold |
|---|---|
| Broad-area public satellite (Sentinel-5P) | 500-8,800 kg/hr per pixel per overpass |
| Broad-area imaging satellite | ~500 kg/hr for point sources |
| Aircraft-based spectrometer | ~200 kg/hr for point sources |
| Point-source imaging satellite | ~180-240 kg/hr typical |
Sources: peer-reviewed satellite detection-limit studies
Freely available satellite data catches the big releases. It routinely misses the smaller, chronic leaks that ground-based continuous monitoring is built to catch.
How to Interpret Methane Monitoring Results
Misreading a result carries real cost either direction. Miss a reportable leak and you're exposed to regulatory and safety risk. Over-react to normal operations and you've wasted a dispatch. Accurate interpretation is what drives the correct next action.
- Normal/Acceptable: Readings reflect expected process emissions, such as pneumatic controller venting or routine tank breathing, within the learned baseline. Log it and take no action.
- Minor Issues: Small deviations above baseline stay below the regulatory detection or reporting threshold. Flag for the next scheduled inspection rather than an immediate dispatch.
- Out-of-Spec: A validated reading exceeds the site's normal signature or the regulatory threshold, indicating a true fugitive emission. This triggers the acknowledge-dispatch-mitigate response, dispatching a technician within a defined window (commonly 24 hours) to repair the source.
Logging the duration and volume of every validated event matters for two separate reasons. First, it drives repair-prioritization decisions: a small leak running for three weeks may have lost more product than a large leak caught in an hour.
Second, it produces the auditable record regulators and ESG frameworks now expect, rather than a rough estimate reconstructed after the fact.
Regulatory Frameworks and Compliance Considerations
Three overlapping frameworks now shape how operators must monitor and report methane, and they don't all ask for the same thing.
EPA 40 CFR Part 60 Subpart OOOOb lets approved continuous or periodic-screening systems reduce traditional AVO/OGI inspection frequency, but only subject to EPA Administrator approval of the specific method.
This isn't a blanket waiver: approval is method-specific, so operators need to confirm their platform qualifies under their state's SIP or FIP before scaling back manual inspections.
OGMP 2.0 Levels 4 and 5 require site-level, measurement-based data rather than estimates built from generic emission factors. Level 5 goes further, reconciling that source-level inventory against independent site-level measurements and quantifying uncertainty.
Continuous monitoring output (the volume, duration, and rate data Zentinal IQ™ produces after Core validates an event) aligns directly with this reporting tier because it's measured, not modeled.
SASB Oil & Gas E&P and TCFD frameworks increasingly expect auditable, measurement-based emissions data for investor disclosure, not periodic estimates. That's a shift from "what we think we emitted" to "what we measured," and it puts continuous monitoring records to work twice: once for compliance, once for ESG reporting.
A few practical distinctions worth keeping straight:
- GHGRP (Subpart W) reporting applies at a 25,000 metric ton CO2e/year threshold and doesn't replace OOOOb repair obligations
- OOOOb covers sources built or modified after December 6, 2022
- Approved alternative-screening thresholds are facility-class-specific: a compressor station and a single-wellhead site face different frequency requirements
Common Errors and Safety Best Practices
Common Errors in Methane Monitoring
Even well-intentioned programs fail for predictable reasons:
- Obscured sightlines — sensors or cameras angled so equipment blocks the emission point
- Infrequent intervals — quarterly or semiannual inspections miss intermittent, short-duration leaks that occur between visits
- Poor signal discrimination — failing to distinguish normal process emissions from true fugitive leaks, which causes either alert fatigue or missed real events
Safety and Best Practices
Fixing these errors takes more than better hardware — it requires disciplined field practices:
- Follow site lockout/isolation and PPE requirements whenever technicians are dispatched to inspect or repair a flagged point
- Position continuous sensors for unobstructed, weather-resistant operation, and validate calibration on a regular schedule
- Favor systems that run on onsite edge computing rather than depending on continuous connectivity — critical for remote wellsites where cellular or satellite links drop out
A monitoring system that pauses every time connectivity fails isn't monitoring continuously at all. Platforms built for onsite edge processing, like the Zensory.ai™ system, keep detecting and validating events locally, then sync data once a connection returns.
Conclusion
Accurate, continuous methane monitoring has moved from a nice-to-have to a survival requirement under EPA OOOOb and OGMP 2.0. The best-performing operators correctly separate normal operations from true fugitive emissions, then act on that distinction fast.
The right monitoring approach turns a validated reading into a dispatched technician within hours, not a fine discovered months later during an audit.
Frequently Asked Questions
Is GHG reporting mandatory?
Yes, for many facilities. EPA's Greenhouse Gas Reporting Program and the OOOOb methane rule both carry mandatory reporting obligations. OGMP 2.0 remains voluntary but is increasingly expected by ESG-focused investors.
What are the 7 GHGs?
The seven gases covered under major reporting frameworks are carbon dioxide, methane, nitrous oxide, HFCs, PFCs, SF6, and NF3.
What does 40 CFR 60.5398b require for automated Periodic Screening?
Under 40 CFR 60.5398b, automated Periodic Screening systems must be capable of measuring a methane emission rate and detecting releases across the site during each quarterly screening cycle.
How is continuous monitoring different from traditional LDAR?
Continuous systems provide always-on sensor coverage, while traditional LDAR relies on periodic technician surveys every quarter or half-year. Continuous monitoring catches leaks that occur and resolve between inspection windows.
How often should oil and gas facilities be inspected for methane leaks?
Frequency depends on facility class and the approved alternative technology's detection threshold, ranging from monthly screening at compressor stations to quarterly or semiannual checks at simpler well sites.
What is OGMP 2.0 Level 5 reporting?
Level 5 is the highest reporting tier. It requires source-level, measurement-based data reconciled against independent site-level measurements, which is why continuous multi-sensor monitoring is increasingly used to meet this standard.
