REPORT MARCH

661 2025

Recommended practices for methane

emissions detection and quantification
technologies – upstream
Acknowledgements
This Report was prepared by Carbon Limits AS under the supervision
of the IOGP Low Carbon Operational Efficiency Committee. IOGP, OGCI,
Ipieca and Energy Institute are grateful to their Member Companies for
their participation and feedback.

Front cover photography used with permission courtesy of

© Mr.PK/Shutterstock and © Xmentoys/Shutterstock

About
This Report provides oil and gas operators with a framework and
guidelines to help select and deploy methane emissions detection and
quantification technologies that are tailored to their sites and objectives
in the upstream oil and gas industry. It is accompanied by an online
technology filtering tool, detailed technology data sheets covering over
fifty technologies, and decision trees to guide deployment.

Feedback

IOGP welcomes feedback on our reports: publications@iogp.org

Disclaimer

Whilst every effort has been made to ensure the accuracy of the information contained in this publication and any associated material, neither IOGP nor any
of its Members past present or future warrants its accuracy or will, regardless of its or their negligence, assume liability for any foreseeable or unforeseeable
use made thereof, which liability is hereby excluded. Consequently, such use is at the recipient’s own risk on the basis that any use by the recipient
constitutes agreement to the terms of this disclaimer. The recipient is obliged to inform any subsequent recipient of such terms.

Please note that this publication and any associated material is provided for informational purposes and adoption of any of its recommendations is at the
discretion of the user. Except as explicitly stated otherwise, this publication must not be considered as a substitute for government policies or decisions or
reference to the relevant legislation relating to information contained in it.

Where the publication contains a statement that it is to be used as an industry standard, IOGP and its Members past, present, and future expressly
disclaim all liability in respect of all claims, losses or damages arising from the use or application of the information contained in this publication in any
industrial application.

Any reference to third party names is for appropriate acknowledgement of their ownership and does not constitute a sponsorship or endorsement.

Copyright notice

This publication and the contents of these pages are © International Association of Oil & Gas Producers™ (“IOGP™”). Permission is given to reproduce, quote
from, republish, redistribute, and recommunicate this publication in whole or in part for purposes without independent economic value [2] and provided that
(i) International Association of Oil & Gas Producers is always acknowledged as the source, together with the words “© International Association of Oil & Gas
Producers™ (“IOGP™”), and (ii) any acknowledgments of other sources as may appear in this publication, in respect of such material, are included unaltered.
All other rights are reserved.

These Terms and Conditions shall be governed by and construed in accordance with the laws of England and Wales. Disputes arising here from shall be
exclusively subject to the jurisdiction of the courts of England and Wales.
 REPORT MARCH
661 2025

Recommended practices for

methane emissions detection
and quantification technologies
– upstream

Revision history

VERSION DATE AMENDMENTS

1.0 September 2023 First release

Update of the Decision Trees, some technologies data sheets, and addition
2.0 March 2025
of 6 new technology data sheets.
Recommended practices for methane emissions detection and quantification technologies – upstream

Contents

Introduction 7

1. Criteria for methane technology selection presented in the online database and
technology data sheets 9
1.1 Operator preferences 10
1.1.1 Access to site (tool filter) 11
1.1.2 Business model (Tool filter) 11
1.1.3 Sampling frequency during operation 11
1.1.4 Deployment method (tool filter) 11
1.1.5 Visual/non-visual product (tool filter) 12
1.1.6 Sensor classification and types 12
1.1.7 Operating Regions 13
1.1.8 Operational Since 13
1.2 Area characteristics 13
1.2.1 Offshore applicability (tool filter) 14
1.2.2 Access to offshore installation required 14
1.2.3 Daylight (tool filter) 15
1.2.4 Readings near bodies of water (tool filter) 15
1.2.5 Cloud cover (tool filter) 15
1.2.6 Snow cover (tool filter) 15
1.2.7 Precipitation (tool filter) 16
1.2.8 Wind 16
1.3 Aim of deployment 16
1.3.1 Capacity to monitor multiple sites per deployment (tool filter) 17
1.3.2 Detection at site level (tool filter) 17
1.3.3 Detection at equipment level (tool filter) 17
1.3.4 Detection at component level (tool filter) 18
1.3.5 Quantification at basin level (tool filter) 18
1.3.6 Quantification at site level (tool filter) 18
1.3.7 Quantification at equipment level (tool filter) 19
1.3.8 Quantification at component level (tool filter) 19
1.4 Technology characteristics 20
1.4.1 Detection threshold (tool filter) 20
1.4.2 Quantification at detection threshold level (tool filter) 21
1.4.3 Frequency of technology deployment (tool filter) 21
1.4.4 Quantification uncertainty 22
1.5 Technology validation 22
1.5.1 Validation of detection threshold/quantification threshold (tool filter) 24
1.5.2 Quantification performance and uncertainty (tool filter) 24
1.5.3 Validation of false positives (tool filter) 24

4
Recommended practices for methane emissions detection and quantification technologies – upstream

1.6 Deployment aspects 25

1.6.1 Time considerations for technology deployment 25
1.6.2 Ease of deployment 25
1.6.3 Training 25
1.7 Other factors when selecting technology 26
1.7.1 Presentation of output and results 26
1.7.2 Safety, regulation, and social responsibility 26
1.7.3 Scalability of technologies 26
1.7.4 Third-party deployment/service providers 27

2. Deployment – decision trees 28

2.1 Tree 0 - General Tree 28
2.1.1 If the objective is to reduce emissions 28
2.1.2 If the objective is to report and reduce emissions 30
2.2 Tree 1 - Screening of components and sites 30
2.2.1 Do we need site screening? 32
2.2.2 Source screening 32
2.2.3 Continuous improvement: update or improve existing source level inventory continuously 33
2.3 Tree 2 - Source level quantification 34
2.4 Tree 3 – Measurement based emissions quantification – site/group of equipment 37
2.4.1 What is the aim of the site level quantification? 37
2.4.2 Other constraints when selecting site level quantification technology 40
2.5 Tree 4 - Reconciliation for a single site 40
2.5.1 Information required 40
2.5.2 Source level quantification 42
2.5.3 Comparison between source level inventory at the time of the measurement and site level,
measurement-based quantification 43
2.5.4 Additional considerations 45
2.6 Tree 5 – Reconciliation for a group of sites 46
2.6.1 Required information 48
2.6.2 Compare site level and source level estimates 48
2.6.3 Root causes of discrepancies in reconciliation 49
2.7 Tree 6 – Reconciliation to produce a single Measurement Informed Inventory (MII) 51
2.7.1 Step 1: Define Scope and Identify Emission Sources 51
2.7.2 Step 2: Categorize and Stratify Emission Sources 51
2.7.3 Step 3: Establish Initial Inventory and Expected Emissions Distribution (EED) 52
2.7.4 Step 4: Develop Sampling and Measurement Strategies 52
2.7.5 Step 5: Deploy Technologies and Collect Data 53
2.7.6 Step 6: Analyse Data and Evaluate Quality 53
2.7.7 Step 7: Choose Reconciliation Pathway 54
2.7.8 Step 8: Reconcile Inventories and Estimate Measurement Informed Inventory (MII) 54
2.7.9 Step 9: Evaluate Objectives 57
2.7.10 Step 10: Develop Report 57

5
Recommended practices for methane emissions detection and quantification technologies – upstream

2.8 Elements of frequency – site level measurement-based quantification 57

2.8.1 History of successful reconciliation 58
2.8.2 Potential super emitters 58
2.8.3 Understanding emissions variability 58
2.8.4 Combining different technologies 59
2.8.5 Large-scale reconciliation 60
2.8.6 Overview of factors influencing frequency of reconciliation 60

3. Technology combinations 61
3.1 Example 1 – Framework for combination of numerous technologies 61
3.2 Example 2 – Framework for combination of numerous technologies 62
3.3 Example 3 – Simulations of technology combinations 62
3.4 Example 4 - Aerial measurement combined with OGI and permanent sensors 63
3.5 Example 5 – Aerial measurements combined with permanent sensors 63
3.6 Example 6 – Site level quantification combined with OGI 64

4. Other recommendations and overarching elements 65

4.1 Understanding uncertainty 65
4.2 Data management and security 66
4.3 Internal practices and processes independent of the provider 67
4.4 Lack of independent standards for comparing technologies 67
4.5 Interpretation of test results 67
4.5.1 Site layout 67
4.5.2 Probability of Detection (PoD) 68

5. Conclusion 69

Glossary 70

List of Acronyms 72

Appendix A: Methodology and data sources 73

A.1 Literature review 73
A.2 Interviews with technology providers 74
A.3 Interviews with service providers and operators 74

Appendix B: List of technologies assessed 75

Appendix C: Selected peer-reviewed articles 77

6
Recommended practices for methane emissions detection and quantification technologies – upstream

Introduction

Many upstream oil and gas operators are aiming to implement and/or improve methane emissions
detection and quantification at their sites1, including in response to requirements of regulators,
company practices, identification of mitigation opportunities and reporting initiatives. This document
and its accompanying online tool and set of technology data sheets2 provide oil and gas operators
with guidelines for selecting and deploying methane emissions detection and quantification
technologies tailored to the situation at their sites, with the aim of improving upstream methane
management and emissions reporting.

Technologies for detecting and quantifying methane emissions have improved significantly and
continue to evolve. Following such improvements, reporting standards have also evolved, requiring
robust detection and quantification. The selection of appropriate technologies to meet these needs
depends on several factors.

The technology filtering tool that accompanies this report guides the operator by asking questions
related to purpose, location, prevailing weather conditions, as well as details on the detection
threshold, frequency, and uncertainty required. The technology data sheets provide more nuances
to the assessment and the technology filtering tool is a simplification of a complex assessment.
Operators are always invited to check the technology data sheets and to contact technology providers.
Next, a set of decision trees in the second part of this Report provide guidance in deployment.

At the heart of the technology filtering tool is a set of technology data sheets for over fifty
technologies that are searchable according to the factors mentioned above. The independent
consultancy, Carbon Limits, developed the technology data sheets based on multiple sources,
including peer-reviewed academic literature, public datasets, and interviews with operators, service
providers, and technology providers. Sources for all information in the technology data sheets are
identified. Further information on methodology and data sources is provided in Appendix A. A list of
reviewed academic papers is provided in Appendix C.

This document and its accompanying technology filtering tool and technology data sheets do not
recommend one technology or approach over another. They have been developed to provide a
framework of detailed technology characteristics so that operators can make informed decisions
on selecting and deploying the technology (or combinations of technologies) best suited to their
specific circumstances, taking into account the objectives of technology deployment.

Section 1 of this Report provides an overview of the criteria by which the technology filtering tool
helps users select the technology.

Section 2 of the Report provides guidance for deployment, based on decision trees for different
activities, including quantification at source level, quantification at site level, reconciliation for a
single site, and reconciliation for a group of sites and/or a single site with multiple measurements
over time. By answering a series of questions, an operator can obtain guidance best adapted to their
unique situation.

1 In the body of the report, sites are synonymous with “facilities”: see Glossary for more information.
2 https://www.iogp.org/workstreams/environment/environment/methane-emissions-detection-and-quantification/methane-detection-
and-quantification-technology-filtering-tool/tool/

7
Recommended practices for methane emissions detection and quantification technologies – upstream

The importance of combining technologies was highlighted by many interviewees. Recognizing

that there is neither a universal technology nor a universal combination of technologies,
Section 3 provides recent examples of operators’ experience, highlighting the benefits of certain
combinations.

Section 4 covers several recommendations which emerged from interviews and discussions.

In a fast-evolving methane measurement and reporting space, new information is always available.
Version 1.0 of this Report was finalized in January 2023. Version 2.0 was finalized in December 2024
and reflects the best knowledge available at the time.

8
Recommended practices for methane emissions detection and quantification technologies – upstream

1. Criteria for methane technology

selection presented in the online
database and technology data sheets
Purpose and site characteristics both play a critical role in the selection and deployment of
methane emission detection and quantification technology. To help operators understand
which technologies may be most suitable, a technology filtering tool and technology data
sheets were developed and are provided with this Report.

Using the interactive technology filtering tool, the operator answers a set of questions,
selecting preferences for a range of criteria, to assess which technologies would be
suitable for the operator. The operator may answer only parts of the questions depending
on the specific characteristics of the need. The technology filtering tool simplifies a
complex assessment, and operators are invited to refer to the technology data sheets for a
more detailed assessment.

Detailed technology data sheets have been prepared for each technology assessed
under this project. The information used in the technology filtering tool comes from the
technology data sheets, based on the filtering criteria.

The following sources and validation methods were used to develop the technology filtering
tool and technology data sheets.
• Sources
– Information from peer-reviewed paper prepared by an independent party (such
as academia)
– Information from independent third party (such as operator)
– Information from technology provider (including peer-reviewed paper from
technology provider)
– Certification against a requirement (such as optical gas imaging (OGI), US
Environmental Protection Agency (US EPA) Title 40 – Chapter I – Subchapter C –
Part 60 - Subparts OOOOa and OOOOb)
– Carbon Limits assessment

• Validation
– Validated by independent academic researchers
– Validated by fully blind tests performed with a third party (such as, operator,
academia). Fully blinded tests are tests where the technology provider has
no knowledge of controlled releases being performed and are the most
representative of real-world oil and gas sector surveys.
– Validated by partially blind tests performed with a third party (such as, operator,
academia). Partially blinded tests are tests where the technology provider is
aware of controlled releases, but not of the characteristics of the release, such
as the location or the magnitude.
– In-house testing3

3
Technology provider’s in-house testing

9
Recommended practices for methane emissions detection and quantification technologies – upstream

In some jurisdictions, regulation can influence the choices of technologies and

reconciliation methods. For example, the US EPA has a process for approving alternative
technologies for use in its NSPS OOOOb and EG OOOOc regulations4 and specific methods
for integrating emission observations from other large release events into regulatory
reporting.5 Another example is the EU Methane Regulation, which provides requirements
for technology detection capabilities, as well as performing reconciliation approaches.6

The below sections (1.1 through 1.7) present the information and criteria used in the
technology filtering tool and technology data sheets. Categories followed by “Tool Filter”
are used as filters in the database. All criteria mentioned in this section, whether they
function as filters or not, are fully detailed in the technology data sheets to provide a
comprehensive understanding of each technology.

1.1 Operator preferences

Methane emissions detection and quantification technologies can be selected based on
the operator’s preferences and constraints with regards to site access, business model,
deployment method, and the output of the sensor (visual/non-visual). The sections below
detail each of these filters.

Depending on the filter questions, the operator can choose one or more options. For single
option filter questions, the default is “All”. In this case, technologies applying to all option
types will be displayed in the final technology table. For multiple option filter questions,
the user can tick or untick the boxes depending on the characteristics to be included
or excluded, narrowing the technology choices that will be displayed by the technology
filtering tool.

Figure 1: Operator preferences

4
https://www.epa.gov/emc/oil-and-gas-alternative-test-methods
5
United States Environmental Protection Agency, 2024, https://www.govinfo.gov/content/pkg/FR-2024-05-14/pdf/2024-08988.pdf
6
Regulation (EU) 2024/1787 of the European Parliament and of the Council, 2024

10
Recommended practices for methane emissions detection and quantification technologies – upstream

1.1.1 Access to site (tool filter)

Site access may be required for deployment or installation of technologies. Hot work
permits may be required for installation or deployment, e.g., a permanently installed
sensor on a fixture that requires placement and setup. Some deployments do not require
access to the site.

The relevant question for this issue in the technology filtering tool asks whether to consider
technologies that would require site access for deployment. The possible answers are:
• All - Both “Yes” and “No” options will be displayed.
• No - site access is not required.
• Yes - site access is required.

1.1.2 Business model (Tool filter)

Technology and service providers generally offer three main business models:
• Instruments are purchased and used by the operator’s staff.
• Technologies are offered as a data product, whereby the technology is deployed or
installed by the technology provider, who subsequently provides data analysis/reports.
• The data product is publicly available, for example, in the case of TROPOMI satellites.

Some technologies can be deployed using either the instrument or data product business
model, while others are only available under one. A hybrid model may be possible,
including as a bespoke product. Operators can choose the “both instrument and data
product” option to filter providers who offer both options. Turnaround times and services
offered can vary and have been documented in the technology data sheets when known.

1.1.3 Sampling frequency during operation

During measurements, technologies may take samples at different time frequencies, for
example, more than every second, every minute, every 10 minutes, etc. This section will
provide further information regarding the sampling frequency of a technology while it is
deployed.

1.1.4 Deployment method (tool filter)

Deployment methods include handheld units, truck-based solutions, equipment mounted
on drones, planes or helicopters, fixed sensors on tripods, elevated mounting systems or
permanently installed on equipment, and satellite-based technology. This can be important
if certain deployment methods are challenging for a given facility, for example, plane-
mounted solutions will not be possible for a no-fly zone.

The technology filtering tool asks about the different deployment methods. The operator
should tick all the deployment methods that they wish to consider.

11
Recommended practices for methane emissions detection and quantification technologies – upstream

1.1.5 Visual/non-visual product (tool filter)

Technologies are classified as visual or non-visual products based on the output of detection
or quantification activities. A visual product may, for example, provide plume imagery
overlaid on a photo. A non-visual product would not offer imagery to identify methane
plumes. The type of product could affect the ability to follow-up on a specific source.

1.1.6 Sensor classification and types

Though not presented as a filter, the tool classifies sensors by type.

Technologies to sense methane range from metal oxide semiconductors to laser-based

methods, such as tuneable diode laser spectroscopy or laser dispersion spectroscopy
(which measures methane along a laser beam), to handheld or fixed optical gas imaging
(OGI) cameras that allow natural gas (and consequently, methane) visualization.

Sensors may be classified as using in-situ or remote sensing techniques. Sensors requiring
direct contact with the plume are classified as in-situ. Deployment methods could include
fixed sensors for stationary continuous monitoring, or mobile sensors that use ground-based
equipment, handheld monitors, or aerial solutions such as drones, planes, or helicopters.
Remote sensors could employ, for example, infrared, laser-based, or spectroscopy
technology. This does not mean all laser-based methods work remotely. Some laser-based
techniques require direct contact with the plume (such as tuneable diode laser spectroscopy
or cavity ringdown spectroscopy), so are classified as in-situ sensors in this Report.

Methane quantification approaches will vary depending on the sensor type, ranging from
dispersion-modelling to image-processing. For example, a visual product with plume
imagery overlaid on a photo or a non-visual product may be provided.

Figure 2 below presents a summary of the technologies assessed in this Report according
to the classifications described in Sections 1.1.4 to 1.1.6.

Figure 2 - Distribution of the CH4 technologies assessed by deployment method (left), sensor type
(centre) and product type (right)

12
Recommended practices for methane emissions detection and quantification technologies – upstream

1.1.7 Operating Regions

Some technology providers may not be available in all regions due to international
restrictions, lack of demand, or limited personnel availability. This section of the data sheet
covers specific areas where the technology is currently deployed or is available.

1.1.8 Operational Since

This section of the data sheet presents the age of the technology to provide an indication of
the technology provider’s experience.

1.2 Area characteristics

These criteria allow evaluation based on conditions at the site.

The first criterion (offshore applicability) enables filtering based on suitability for offshore
locations. Other criteria relate to environmental conditions. For each criterion related to an
environmental condition, the technology filtering tool and technology data sheets classify
according to the following options:
• Applicable: Performance is slightly affected or not affected by the environmental
condition.
• Not Applicable: Performance is affected, or use is impossible in those environmental
conditions.

In the data sheets, an additional criterion for “Applicable but higher detection threshold
and/or uncertainty” is included and where possible, detailed, to indicate that the technology
can be used in an area where the particular environmental condition applies; however, it is
possible that the detection threshold is higher (it may not be able to detect values as low
as its usual detection limit), its probability of detection is lower, and/or its quantification
uncertainty is higher under such circumstances.

13
Recommended practices for methane emissions detection and quantification technologies – upstream

Figure 3: Area characteristics

1.2.1 Offshore applicability (tool filter)

This criterion reflects the overall applicability to offshore conditions, which is a combination
of two different factors: technical applicability and certification.
• Technical applicability to offshore conditions: For some technologies, the capability
to monitor offshore facilities depends on sensor type. Some perform worse over
water than on land7. When technically ready and certified for offshore deployment, the
technology filtering tool categorizes the product as “Applicable”. The tool will classify
the product as “technically applicable” if the product is in the prototype phase or not
yet certified, or the provider is exploring technical and computational improvements
to take offshore conditions into consideration. Further details are presented in the
technology data sheets.
• Certification (such as, explosive atmosphere (ATEX) rating, class 1, division 1) may be
required for deployment at offshore facilities. Some technologies may be technically
suitable but waiting for certification to ensure safe use. The technology data sheets, and
technology filtering tool present the status of certification at the time of the publication of
this Report. This is likely to evolve, so an update on certification status may be obtained
from the provider. This filter should not be used if certification status is not important.

1.2.2 Access to offshore installation required

Platform access may be necessary for the deployment, installation, operation, and
maintenance of technologies at offshore facilities. This might involve obtaining hot work
permits and addressing logistical or safety measures, such as platform access and space
constraints. However, some deployments may not require platform access. This section will
provide detailed information, as specified by the technology provider or other third parties,
on a case-by-case basis.
7
Jacob D, et al, 2022.

14
Recommended practices for methane emissions detection and quantification technologies – upstream

1.2.3 Daylight (tool filter)

Some technologies, such as shortwave infrared sensors, measure spectrally resolved
back-scattered solar radiation to detect methane emissions. These cannot be used at night
because they require ample sunlight.

To consider technologies that can also operate at night, non-relevant technologies can be
filtered out using this criterion in the technology filtering tool.

1.2.4 Readings near bodies of water (tool filter)

As noted, light may be required to reach the sensor to perform measurements. Bodies
of water, such as around offshore facilities, are a dark surface and often do not provide
enough reflected radiance to allow detection of methane emissions. This is typically
more challenging for remote sensing technologies that require light reflection than for
in-situ sensors, which are not affected. New techniques are being developed that use
sun glint8 reflected off a water surface to detect and quantify emissions. Currently early
in development, this technique could improve the ability to detect and quantify methane
emissions around bodies of water.

The effect of water in the technology filtering tool is considered generally in the offshore
applicability filter (see above), while the technology data sheets provide additional
information on the specific challenges of reflected light near bodies of water.

1.2.5 Cloud cover (tool filter)

Cloud cover reduces observational ability, for example, by reducing the reflected sunlight
that passive sensors use to detect methane, while also increasing uncertainty. This
issue specifically applies to aerial technologies. Cloud cover could also affect continuous
monitoring that requires solar power. This must be anticipated to have enough power
backup (such as batteries) to operate when the meteorological conditions are not ideal.

1.2.6 Snow cover (tool filter)

Snow will impact reflectivity, affecting some laser-based technologies, for example by
increasing detection thresholds and/or the uncertainty levels for quantification. This can
affect both aerial and fence-line monitoring.

Snow can also affect continuous monitoring systems that use solar panels as a power
source, as the snow can cover the panel and prevent the charging of the battery.

Operators can filter out technologies affected by snow coverage. Details on how the
technology is affected by snow coverage is provided in the technology data sheets.

8
Glint is the specular reflection from the surface of water and occurs when the sun angle and view angle are equal and in the same
principal plane.

15
Recommended practices for methane emissions detection and quantification technologies – upstream

1.2.7 Precipitation (tool filter)

Water droplets and fog will scatter light and reduce instrument sensitivity, potentially
reducing the ability to detect or quantify emissions. Precipitation may also increase the
level of uncertainty in quantification, particularly for laser-based solutions.

Rain or snow at the time of detection can also affect the methane plume itself, including its
direction and concentration. Quantification could then result in a higher level of uncertainty.

1.2.8 Wind
Wind speed is one of the dominant factors causing uncertainty in detection and
quantification of methane emissions. While many of the technologies reviewed as part
of this project require the presence of at least some wind to transport methane from
the source to the sensor, they usually will not perform equally well at all wind speeds.
Wind speed and direction are important for use around the site. Wind can be impacted by
obstacles, such as equipment or buildings, which can affect uncertainty.

Wind speeds affect quantification of methane emissions, depending on sensor type

and deployment. Wind speed and/or direction will also impact the uncertainty of
measurements. Some recent tests evaluate the Probability of Detection (PoD) at a given
emission threshold depending on the wind conditions (see Section 4.5.2). When available,
these results are presented in the technology filtering tool. Wind direction and speed need
to be carefully considered when interpreting results.

Wind condition is not a direct filter in the technology filtering tool. However, recommended
minimum and maximum wind speeds and details about the effects of wind are provided in
the technology data sheets to detail the operating envelope in which the technology will be
able to perform reliably.

1.3 Aim of deployment

Criteria in the technology filtering tool allow the identification of deployment objective(s).
IOGP Report 661 assessed two main deployment purposes: 1) detection of methane
emissions and 2) quantification of methane emissions.

Methane emissions are detected and attributed at the site, equipment, or component level,
depending on the technology.

Quantification technologies estimate the rate of emissions, for example as a volume rate
(such as m3/h) or as a mass flow rate (such as kg/h). For some types of events, total
emissions can then be calculated by multiplying the emission rate by the duration of the
event (measured or estimated). Uncertainty can increase with duration. Some quantification
technologies (continuous monitoring) provide an estimated value for the total emissions,
subject to the uncertainty in the system design.

Some technologies can measure the methane concentration in a plume. In this case,
the data must be processed with other factors, such as wind speed and duration of the
emissions, to obtain the emission rate and total emissions. In the technology filtering tool
and technology data sheets, a technology with the capability to provide emission rates is
tagged as a quantification technology.

16
Recommended practices for methane emissions detection and quantification technologies – upstream

1.3.1 Capacity to monitor multiple sites per deployment (tool filter)

Some technologies (notably planes, helicopters, and satellites) can monitor and provide site
level estimates for multiple sites per deployment. This could be advantageous when, for
example, performing reconciliation of emissions for multiple sites (refer to Section 2.6).

Choices in the technology filtering tool for this filter are:

• Yes: The technology would be able to monitor multiple sites per deployment.
• No: The technology would not be able to monitor multiple sites per deployment.
• Maybe: No preference (the technology filtering tool will not use this criterion).

1.3.2 Detection at site level (tool filter)

This criterion captures site level emissions detection suitability. The detection threshold
affects the selection of site level technology (see Section 1.4.1 regarding thresholds).

In addition to site attribution, emissions may be attributed to specific equipment or

components. However, certain technologies that are used to detect emissions at the
equipment or component level may not be suitable for site level detection. This may be
the case, for example, if the technology is not able to visualize methane plumes. Such
technologies would be more appropriate to the identification of equipment or components
as a follow-up to the use of site level technology.

Due to the variability of cases and on-site experience reported by operators, the following
classification has been used for technology that can detect and quantify site level methane
emissions:
• Yes: Emissions can be accurately and reliably detected at site level.
• Maybe: Emissions may be detected at the overall site level, but it may be challenging
to assess the entire site if very large, or if sites are closely spaced. It may be difficult
to identify the source of the plume from one site to another.
• No: The technology is not appropriate for detecting emissions at site level.

1.3.3 Detection at equipment level (tool filter)

This criterion captures whether a technology can detect an emission source and attribute
it to a piece of equipment. The technology may be able to attribute emissions to a specific
piece of equipment at a site, for example a tank, a flare, or a compressor, but might not be
able to attribute the emissions to the emitting component.

The following classification has been used:

• Yes: Emissions can be detected at equipment level, and accurately attributed.
• Maybe: Spatially isolated equipment sources may be detected, but it may be
challenging to attribute emissions to all equipment sources in all scenarios, for
example when many sources are located close together.
• No: The technology does not have the right level of resolution to detect emissions at
the equipment level.

17
Recommended practices for methane emissions detection and quantification technologies – upstream

1.3.4 Detection at component level (tool filter)

This criterion captures whether a technology can detect an emission source and attribute it
to a specific component of a piece of equipment, for example a flange on a separator.

One of the purposes of detecting methane emissions at component level is to identify

leaking or malfunctioning components, typically during Leak Detection and Repair
(LDAR) campaigns, where the goal would be to identify emitting components and ensure
mitigation. Detection at component level can also be used for inventory: some inventory
methodologies require the operator to determine the number of leaking and non-leaking
components to estimate fugitive methane emissions.

A particular technology may detect emissions from large, isolated components (such
as a thief hatch on a storage tank) but prove less reliable in the case of equipment with
many closely placed components. In these circumstances, the technology would not be
considered a component-level detection technology.

The following classification has been used:

• Yes: Emissions can be detected at component level, and accurately attributed.
• Maybe: Components emitting may be detectable, but it may be challenging to
attribute emissions to all component levels in all scenarios, for example when many
sources are located close together.
• No: The technology can, for example, detect emissions at equipment level, but does
not have the right resolution to detect emissions from specific components.

1.3.5 Quantification at basin level (tool filter)

This criterion assesses the ability to quantify (as opposed to detect) total emissions at the
basin level.

The following classifications have been used and are available in the filtering tool:
• Yes: The technology can quantify emissions at the basin level.
• No: The technology cannot quantify emissions at the basin level.

1.3.6 Quantification at site level (tool filter)

This criterion captures the ability to provide the total emission rate for a specific site or
facility. The technology may quantify several large sources within a site but may not be able
to attribute emissions on a more granular scale, such as to specific pieces of equipment.

Quantification of methane emissions at site level is an essential input for a reconciliation

exercise (the other is source level quantification). Site level quantification ensures that
a major emission source is not missing from its source level inventory or has been
improperly quantified.

18
Recommended practices for methane emissions detection and quantification technologies – upstream

The following classification has been used:

• Yes: Emissions can be quantified at the site level. An important caveat is that,
depending on sensor placement, detection thresholds, and monitoring and emission
frequency, some technologies will not necessarily be able to confirm that total
emissions are quantified at a site level, that is, some areas of the site might not be
considered in the quantification.
• Maybe: Emissions may be quantified at the overall site level, but it may be challenging
to assess the entire site if very large, or if multiple sites are closely spaced, and
difficult to identify the source of the plume from one site to another.
• No: The technology cannot quantify emissions at site level.

1.3.7 Quantification at equipment level (tool filter)

This criterion captures whether a technology can provide the total emission rate for a piece
of equipment, such as an individual tank, flare, or compressor.

The following classification has been used:

• Yes: Emissions can be quantified at the equipment level.
• Maybe: The technology may be able to quantify spatially isolated equipment sources
but may not be able to attribute emissions to specific equipment sources in all
scenarios, such as when many sources are located close together or there are
multiple plumes present.
• No: The technology may be able to quantify emissions at a more granular scale
than site level but is not able to quantify emissions from a single piece of equipment
or equipment group. It would not be considered an equipment-level quantification
technology.

1.3.8 Quantification at component level (tool filter)

This criterion captures whether a technology can provide the total emission rate at the
component level, for example, from an individual flange on a separator.

In some cases, it will be necessary to detect the emitting components before quantifying
the volume of their emissions. For other components, prior detection may not be required
since emissions are already known to be present, such as equipment that vents methane by
design.

Component or equipment-level quantification technologies allow the operator to determine

the volume of vented emissions more accurately, as such emissions can vary widely, over
time or between similar equipment types.

19
Recommended practices for methane emissions detection and quantification technologies – upstream

The following classification has been used:

• Yes: Emissions can be quantified at the component level.
• Maybe: The technology may be able to quantify spatially isolated components but may
not be able to attribute emissions to specific sources in all scenarios, such as when
many component sources are located close together or there are multiple plumes
present.
• No: The technology may be able to quantify emissions at a more granular scale than
equipment level but is not able to quantify emissions from a single component. It
would not be considered a component-level quantification technology.

1.4 Technology characteristics

This section presents performance criteria. It should be read in conjunction with Section
1.5 (technology validation).

Figure 4 - Technology characteristics

1.4.1 Detection threshold (tool filter)

Detection threshold is the minimum amount of methane that is reliably detectable9. While
the detection threshold can be presented in several forms (for example, concentration,
concentration vs distance, volume emission rate, mass emission rate), detection thresholds
in this Report are stated in kg/h, where possible.

Detection threshold depends on the type of emissions to detect. For instance, given the
skewed distribution of emission rates10,11 a higher detection threshold will encourage focus
on higher-emitting components. Some jurisdictions set minimum detection thresholds
which could influence the selected technology.

9
IOGP-Ipieca-GIE-Marcogaz - Methane Emissions Glossary
10
Omara M et al.,2022
11
Zavala-Araiza D, et al., 2017

20
Recommended practices for methane emissions detection and quantification technologies – upstream

It should be noted that the detection threshold is a function of the distance between the
emission source and the detection technology, as well as the environmental conditions at
the time, notably wind. Some technologies have begun producing PoD curves to document
these relationships. Please see Section 4.5.2 for more detail.

In most cases, detection thresholds mentioned in the technology data sheets are supplied
by the technology provider and may not have been validated by a third party. Validation
status and the source of this information is presented in the technology data sheets. Where
available, the appropriate environmental conditions for the detection threshold are noted.

The technology filtering tool allows selection from five different detection threshold
categories, ranging from less than 1 kg/hour (most sensitive) to over 1,000 kg/hour (least
sensitive).

Since the performance of some technologies can vary due to many different parameters,
review of this criterion should be done with careful review of validation status and other
criteria in the technology data sheets.

1.4.2 Quantification at detection threshold level (tool filter)

The operator might want to quantify detected emissions for reporting purposes or to
prioritize abatement measures. Quantification provides the emission rate (for example, as
a mass rate such as kg/h, or a volumetric rate such as m3/h). Multiplying the rate by the
duration allows the estimation of total emissions.

Quantification methods often involve measuring methane concentrations in flows of gases

or ambient air but could also include a variety of other measurements, calculations, and
modelling. For most technologies, the quantification threshold will be the same as the
detection threshold. In some cases, however, quantification can only be done at a threshold
higher than the detection threshold.

The technology filtering tool allows the user to specify whether quantification is required at
the same threshold as detection.

1.4.3 Frequency of technology deployment (tool filter)

The recommended frequency of deployment may be specified, though only from a technical
perspective. Section 2.7 provides information on other elements that can influence choices
regarding frequency of deployment.

Technologies have been classified as follows:

• Continuous monitoring: This could be at site level, equipment level or component
level. Continuous monitoring can be affected by gaps in network connectivity or
environmental conditions, leading to downtime of the system.
• Periodic monitoring: This concerns technologies such as handheld devices and aerial
monitoring, which may require assistance in deployment. The actual frequency is then
selected by the operator (refer to Section 2.7).

21
Recommended practices for methane emissions detection and quantification technologies – upstream

1.4.4 Quantification uncertainty

While a sensor may be highly precise, the quantification method using that sensor may be
more uncertain. Technologies with stated uncertainties consider quantification algorithms,
environmental conditions, and emission rates. Quantification uncertainty may be reported
in terms of a 1σ or 2σ uncertainty (68% and 95% confidence intervals, respectively), in
relative or absolute values. Care should be taken when evaluating uncertainties. Please
refer to Section 4.1 for details.

1.5 Technology validation

There is no international standard to measure and compare the performance of detection
and quantification technologies (see Section 4.4). To improve transparency regarding third-
party validation, technologies have been assessed against several types of validation that
have been presented in the datasheets. The database helps operators select technologies
based on the validation performed. This criterion may be useful for operators who are not
planning to perform internal technology validation.

Figure 5 - Technology validation

For the purposes of this project, “validation” means that test results are publicly available.
It does not necessarily mean that the technology will perform “as advertised” under all site
conditions. The validation criteria are independent from the performance criteria.

Four technology validation options are available in the technology filtering tool:
• Not applicable for this technology: Filter for technologies that can perform either
detection or quantification. For example, some are able to detect methane but not
quantify it, in which case verification of quantification performance is not relevant.
• Not Validated: Tests may have been performed by the technology provider, either in
the lab or field, with the presence and size of the emission source either known or
unknown to the technology operator. Care should be taken when considering the
conditions under which in-house testing took place, since these may not reflect field
conditions. Technologies are considered “not validated” if they have only undergone
in-house testing or results are not publicly available.

22
Recommended practices for methane emissions detection and quantification technologies – upstream

• Validated: Validation has been done by peer-reviewed papers prepared by independent

academic researchers, or validation has been done using partially or fully blinded
tests performed with a third party such as academics, independent researchers or by
oil and gas operators.

In the data sheets, information has been provided about the type of validation. The following
categorizations have been presented in the data sheets, apart from the ones specified above:
• Validated: academia: The information comes from a peer-reviewed paper prepared
by independent academic researchers and may include results from fully or partially
blind testing (see below).
• Validated: partial/fully blind tests: Validation can be done using partially or fully
blinded tests performed with a third party such as academics, independent
researchers or by oil and gas operators. For fully blind tests, the presence, location,
and size (if any) of the controlled test release(s) were unknown to the technology
provider at the time of the test. This is the closest approximation of field conditions,
with the least amount of inherent bias. For partially blind tests, the technology
provider was aware that controlled release testing was taking place but was unaware
of the size or location of the release. Partially blind tests offer improved validation of
technology performance over scenarios where the emission source size was known
but may still introduce bias. For instance, the operator performing the test may have
taken more proactive steps than normally to detect or quantify emissions.

Some validation work is ongoing. The technology filtering tool and technology data sheets
should be regularly updated to account for results of new tests and research. The following
cases are highlighted:
• Testing may have already been performed, but the results not yet made public.
Information about such cases, where known, are indicated in the technology data
sheets. The technology will still be considered “not validated,” since the results
were not publicly available at the time of publication. This does not imply anything
regarding performance, but only the availability of the information.
• Some validation may have been performed, but there are no plans to make the results
public. In such cases, the technology has been classified as “not validated”, even if
the results of such validation were communicated orally. This does not imply anything
regarding performance, but only the availability of the information.

Where relevant, information in the technology data sheets is provided regarding the layout
of the testing site, environmental conditions, and limitations of the validation. The user
should consider the test conditions and setup relative to those in which the technology is
likely to be used (see Section 4.5). For example, a partially blind test performed in a desert
with a single point emission source may not be relevant if the operator intends to use the
technology for multiple, small sources in dense foliage.

23
Recommended practices for methane emissions detection and quantification technologies – upstream

1.5.1 Validation of detection threshold/quantification threshold (tool filter)

Validation of detection and quantification thresholds refers to the ability to correctly detect/
quantify the smallest amount of methane that is claimed by the provider.

Probabilities of detection and quantification are ideally based on fully blind test results
and consider sensor performance as well as environmental variables that can affect
measurements, offering the closest conditions to the field.

The options for validation of detection and quantification thresholds are those indicated
above in Section 1.5.

1.5.2 Quantification performance and uncertainty (tool filter)

Quantification performance refers to the ability to give measurement values for the
emission rate that match the actual emissions. Quantification performance may be
described by comparing measurements to true emission rates. Ideally, the linear
regression between measurements and actual emissions is a unit-slope line.

Quantification performance may be based on emission rates, wind speeds, and/or distances
of measurement technology from the source, all of which can impact quantification
performance. Robust, defined, and publicly available analyses increase transparency
regarding the abilities. Technologies that have published results for these parameters offer
a more reliable indication of performance than those for which results are not publicly
available.

Providers that have published results of quantification performance typically provide a

range of emission rates for which the technology is able to perform quantification and
a quantification uncertainty at a specified emission rate either under typical operational
conditions or, for example, in terms of wind speed. This type of information helps users
understand the performance envelope of the technology.

The technology filtering tool allows selection where the presented quantification
uncertainty is validated. Where available, more details on the technology’s quantification
performance are presented in the technology data sheets.

The options for quantification performance and uncertainty are those indicated above in
Section 1.5.

1.5.3 Validation of false positives (tool filter)

False positives are reports of methane emission detection where no methane emissions
occurred. False positives may lead to unnecessary follow-up and alarm fatigue. Tests for
false positives are reported in the technology filtering tool and technology data sheets,
where available.

The options for validation of false positives are those indicated above in Section 1.5.

24
Recommended practices for methane emissions detection and quantification technologies – upstream

1.6 Deployment aspects

Qualitative aspects to consider when selecting suitable technology include ease of
deployment, time required to deploy, and training required. While these criteria are not
used in the technology filtering tool, qualitative information on these and other deployment
aspects is presented in individual technology data sheets.

1.6.1 Time considerations for technology deployment

This section provides detailed information about the amount of time required for initial
setup, installation lead times, and other temporal aspects of the technology. This may
include battery lifetime, charging time (if applicable), maintenance duration, and other
relevant time-related factors.

1.6.2 Ease of deployment

Some technologies, such as handheld analysers, require the site to be manually assessed
for emissions. Depending on safety certifications, this could require obtaining hot work
permits.

In the case of aerial monitoring, such as with drones or airplanes, the safety of the
pilot and the site operators must be considered. This could require permits, as well as
significant coordination on the part of the operator. These requirements differ by country.
For satellites, deployment depends on the orbital path of the satellite and on environmental
conditions, such as cloud cover.

1.6.3 Training
Training required for deployment is likely to be closely associated with the business model
of the technology provider. Some providers handle everything from installation to post-
processing of data. In such cases, the operator would receive the estimated emissions
data from the provider, so little training would be required for the staff of the oil and gas
operator. However, some providers train the operator to use their handheld devices, drones,
or other equipment. Training time required will vary, depending not only on the equipment
but, for example, on staff experience and field/site characteristics.

25
Recommended practices for methane emissions detection and quantification technologies – upstream

1.7 Other factors when selecting technology

The following sections present additional factors that may need to be considered when
selecting a methane emissions detection or quantification technology. These are not
covered by the technology filtering tool, or the technology data sheets, as they do not
apply globally, are not relevant for most technologies, or are unique to a particular region,
operator, or site.

1.7.1 Presentation of output and results

Data generated and reported by detection and quantification technologies and services are
not standardized. Output can vary significantly, including in terms of format, scale, unit, and
scope.

Some providers offer online platforms or other tools to help assess and use the output. The
operator will need to consider how actionable these deliverables are, measured against
its needs. For example, an operator that is trying to identify components that need to be
mitigated may require an output that includes clear and precise localization of the methane
plume, whereas figures for methane concentration downwind could be sufficient for an
operator that is trying to prioritize efforts across several sites.

1.7.2 Safety, regulation, and social responsibility

Some regulations can directly impact the technology itself, either by requiring a particular
type of technology, carrier, or sensor, or by requiring technology performance standards,
such as a detection threshold or PoDs (see Section 4.5.2).

Other regulations may not directly target methane emissions detection and quantification
technologies but could impact deployment. This is typically the case with airborne detection
and quantification. For example, flight restrictions for aircrafts, weight or altitude limits for
drones, or a ban on drone flights all together, can limit deployment options.

Beyond regulation, questions of sustainability and social responsibility can come into
play. The overall impact of deployment on the local population and the environment
can influence the decision. This aspect can be very site-specific and cover many topics.
Examples include limiting the number of aircraft flying over inhabited areas (for reasons
of safety and noise pollution), limiting disruption to local flora and fauna, and avoiding
increased road traffic.

1.7.3 Scalability of technologies

Operators will typically need to deploy technologies across many assets, which may be
widely dispersed. In such instances, scalability is likely to be important.

Potential constraints are twofold. The first is availability, including whether the technology
is available in the operator’s geographic area, and the provider’s manufacturing and
deployment capacity.

26
Recommended practices for methane emissions detection and quantification technologies – upstream

The scale of deployment will also depend on budgetary and labour resources. For example,
if the deployment of a particular technology requires months of work from a full team for a
single site to obtain conclusive results, challenges will likely arise when looking to deploy
this technology across all company assets, including operational and logistical constraints,
such as ongoing maintenance.

1.7.4 Third-party deployment/service providers

In some cases, equipment can be deployed by either the operator or third parties,
depending on aim and scale.

Hiring a third-party service provider usually means that the operator does not need to
acquire the technology directly or train personnel to deploy it, and fewer employees need
to be redirected (only for managing or supervising the deployment). In addition, for leak
detection and repair, personnel experience plays a significant role in leak identification.12

On the other hand, relying on a third party for methane emissions detection and
quantification implies that the operator either has a long-term contract with a monitoring
service provider or hires each time the technology is deployed. This could impact the
scalability of the deployment. It may also require site access for external personnel, which
can increase administrative burden.

The operator may need to consider whether a service provider is able to support the
deployment of the specific technology within the operator’s required region(s) and
timeframe(s).

12
Zimmerle, D, et al., 2020

27
Recommended practices for methane emissions detection and quantification technologies – upstream

2. Deployment – decision trees

While the technology filtering tool uses criteria to help select appropriate technologies
for use at a particular type of site for a given purpose, additional factors should be
considered when it comes to deployment, such as part of a methane emissions detection,
quantification, and reporting programme. This section addresses some deployment
considerations using “decision trees”.

The following deployment decision trees have been developed:

• Tree 0: A general decision tree that organizes the different processes into a coherent
framework
• Tree 1: Screening of components and sites
• Tree 2: Quantification of emissions at source level
• Tree 3: Quantification of emissions at site level
• Tree 4: Reconciliation for a single site
• Tree 5: Reconciliation for a group of sites and/or a single site with multiple
measurements over time
• Tree 6: Reconciliation to produce a single Measurement Informed Inventory (MII) as
per the GTI Veritas protocols

The final section presents some elements of deployment frequency.

2.1 Tree 0 - General Tree

The general tree is used to identify which other tree(s) should be used, depending on the
main objective of the deployment. The two main deployment objectives covered by the
general tree are:
• reducing methane emissions
• reporting methane emissions, following an international standard or otherwise

The “reporting” side of the decision tree combines both objectives since most operators are
also aiming to reduce the emissions they report.

In some instances, deployment will respond to a regulatory requirement. The operator

is invited to use this decision tree to evaluate whether complementary technologies
or processes that are not already included in the regulatory requirements should be
considered, taking into account objectives and site characteristics.

2.1.1 If the objective is to reduce emissions

Where the focus of deployment is reducing emissions, the first step is to perform a
screening exercise at either the site, equipment-group, or component level (refer to Section
2.2). This allows the operator to identify priority areas for resources.

28
Recommended practices for methane emissions detection and quantification technologies – upstream

0 General tree

What is the main objective?

The main objective is to report emissions on

The main objective is
a voluntary basis (following an international
to reduce emissions
standard, or otherwise) and reduce emission.

What is the targeted level of detail for the emissions inventory?

Source level inventory Source level inventory and

Simple, source level
based on engineering calc reconciliation with site
inventory based on
and measurement on a level, measurement-based
generic EF (e.g., in line
representative sample quantification (e.g., in line
with OGMP level 3)
(e.g., in line with OGMP level 4) with OGMP level 5)

1 Screening of components and sites to list emission sources

Source-level Source-level
2 quantification 2 quantification 2 Source-level quantification
(optional) (simplified)

Measurement based emissions

Assess, prioritize and implement emission reduction 5 quantification – site/group of
equipment level

Group
Single site
3 4 of sites
reconciliation
reconciliation

Assess, prioritize and implement

emission reduction
Report emissions

Figure 6: General tree

29
Recommended practices for methane emissions detection and quantification technologies – upstream

An optional follow-up is the quantification of emissions from the identified sources. There are
several ways to quantify methane emissions at source level, as presented in the source level
quantification tree (refer to Section 2.3). This step is relevant if the operator wishes to use results
as part of an inventory based on detected emissions, or report emission reductions achieved
through mitigation. Mitigation prioritization and implementation are not covered in this Report.

2.1.2 If the objective is to report and reduce emissions

The creation of a robust inventory of methane emissions will always require an
understanding of potential emission sources and a screening of those sources (refer to
Section 2.2). Inventories can have various levels of accuracy. Deployment to aid in this
process depends on the targeted level of detail.

The simplest form of inventory at source level relies on generic emission factors (EF),
such as in line with OGMP13 Level 3 or the baseline inventory in the GTI Veritas14 Source-
Level and Measurement Reconciliation Protocol. Source level inventories based on generic
emission factors can be a first step in assessing methane emissions (see Section 2.3).

Operators can develop a more specific source level inventory by using engineering
calculations or measurements performed on a sample in place of generic emission
factors, such as in line with OGMP Level 4. For such an inventory, the detailed process for
implementation may be found in Section 2.3.

One option where operators can take the development of their inventory a step further
is by comparing a source level inventory (see Section 2.3) with site level measurements
(see Section 2.4). One purpose of site level measurement is to help ensure that source
level quantification has considered all large emission sources, and that source level
quantification of major sources is accurate. This process is an example of reconciliation
between source and site measurements and can be done for either a single site or group
of sites with similar characteristics. It can also be used for measurements of the same
site over time. Decision trees are available describing the reconciliation process for each
of these cases (see Section 2.5 for a single site and Section 2.6 for a group of sites). This
option is the one considered for this Report and detailed in the decision trees.

Operators may also combine measurements, EFs, and engineering calculations to

produce a Measurement-Informed Inventory (MII), in which both source- and site level
measurements, and engineering calculations may be combined to produce a single
emission inventory estimate for a group of equipment, site, or group of sites (see Section
2.7).Reconciliation and measurement-informed inventories are emerging areas, where
methodological refinement is on-going around how to integrate different types of
measurements into inventories over time.

2.2 Tree 1 - Screening of components and sites

Screening of methane emissions, even without quantification, can already be an important
source of information. The aim of this decision tree is to guide operators through source
level and site level screening.
13
https://www.ogmpartnership.com/
14
https://veritas.gti.energy/protocols

30
Recommended practices for methane emissions detection and quantification technologies – upstream

1 Screening of components and sites to list emission sources – not to quantify

1 Do we need site screening?

Is it logistically challenging or very expensive Yes Perform a screening of the many sites – to prioritize efforts
to perform source screening at many sites? Please refer to the technology database for the selection of a technology
Are there many similar sites? for site-level screening based on site characteristics

Is there a large share of sites where no emissions were detected?

Yes No
No
Are there reasonably available
Revisit technology selection for Prioritize sites for
technologies with a lower
site screening source screening
Yes detection threshold? No

2 Source screening

For one site: Is an exhaustive list of emissions sources available?

No

Develop an exhaustive list of all the emission sources

Create list of ALL potential emission sources. Emissions can be considered and divided into three broad categories:

Leaks – always a potential source Routine and process emission sources Non-routine emissions and incidents
Leaks are the unintentional releases Equipment or processes that emit methane as part Incidents and emergency stops are unintended
of natural gas from equipment of regular operation (e.g. process and design vents and unplanned events/venting which are
emitting at expected levels) not part of routine operations.

Perform a component level field Desktop screening Prepare a list of all potential non-routine
screening of the sources Prepare a list of design and process emission sources emissions and incidents. In particular:
(i.e. Detection step in LDAR) from equipment/events, including but not limited to: • Unlit flare?
Yes

Please refer to the technology database • Flaring and incomplete combustion • Operational issues on the storage tank?
for the selection of the technology • Compressors (e.g., open thief hatch)
• Tanks • Equipment maintenance, or equipment being
• Well activities stopped/started/purged?
• Pneumatics • Equipment upsets/malfunctions?
• Gas treatment (e.g. glycol dehydrators, AGR)
• Other venting and purging

And optionally

Perform a component level field

screening of the sources
Please refer to the technology database for the
selection of the technology

Is it also required to quantify emissions (e.g. reporting)?

Qualitative or semi qualitative assessment of No Yes

emissions only to prioritize mitigation 2 Source level quantification

Prioritize and implement mitigation

3 Continuous improvements: update or improve existing list of emission sources on a continuous basis

What may trigger updates of source screening?

Further emissions reduction Issues with reconciliation (ref trees 3 and 4) An additional potential source
of emission is identified
(e.g., unexpected upset)
Known

Update only when there are changes in

vents

operations or equipment on the site.

Regular screening of all components or

Unintended

Continuous monitoring Perform a new source screening,

Leaks

Prioritization, frequency and approach selected will if possible, at the time of the
depend on the site characteristics and on the ambition. site level measurement

Depending on the ambition:

Unexpected

Permanent tracking of parameters Stop the emission, record it and quantify

events

or it based on best available information and

Continuous measurement to identify
and address emissions
then add it to the list of emissions sources.

Figure 7: Screening of components and sites

31
Recommended practices for methane emissions detection and quantification technologies – upstream

2.2.1 Do we need site screening?

Source level screening should be performed at all sites. In some cases, site level screening
may be used to identify the sites at which to prioritize source level screening. Site level
screening is recommended where there are many similar sites and it is logistically challenging
or expensive to perform source level screening, or to increase the frequency of screening.

If no emissions are detected at a large share of sites, it may be worthwhile to revisit the
technology selected for site screening and determine whether one with a lower detection
threshold can be used instead (this may not be necessary if the technology selected is in
line with local minimum detection threshold regulations, as discussed in Section 1.4.1).
If no technology with a lower detection threshold is available, sites can be prioritized for
source level screening based on the small share of sites where emissions were detected or
on other relevant parameters, such as number of components and age of installations.

Some site level screening technologies provide source level detection (see the technology
filtering tool and technology data sheets).

2.2.2 Source screening

The first step in source level screening is to develop an exhaustive list of emission sources
at each site, regardless of whether methane emissions have been confirmed.

Emission sources can be divided into three broad categories:

• Leaks (always a potential source): unintentional releases of natural gas from
equipment
• Routine and process emission sources: equipment or processes that emit methane
regularly
• Non-routine emissions and incidents: unintended events/venting

Since leaks can arise anywhere and at any time, a complete component-level screening for
such emissions15 is useful.

Routine and process emissions are more predictable. Sources can often be determined
based on facility design and operational practices. It is recommended to start the analysis
with a desktop study to prepare a complete list of all design and process emission sources
from equipment and events, including but not limited to:
• Flaring and incomplete combustion from power and heat generation, such as
engines, turbines, and boilers
• Compressor seals, for example, rod packing of reciprocating compressors, and wet/
dry seals for centrifugal compressors
• Hydrocarbon storage tanks
• Well activities, such as liquids unloading, casinghead gas venting, well completion
and workover, and well drilling and testing
• Pneumatic controllers and pumps
• Gas treatment, namely glycol dehydrators, acid gas removal (AGR)

15
https://www.iogp.org/workstreams/environment/environment/methane-emissions-detection-and-quantification/methane-detection-
and-quantification-technology-filtering-tool/tool/

32
Recommended practices for methane emissions detection and quantification technologies – upstream

• Blowdown and pressure-control releases of vessels and pipes

• Other venting and purging

Optionally, screening in the field can be performed in addition to a desktop study to ensure
that all potential emission sources have been considered. A field screening may not be
sufficient for some of these sources since some may be intermittent.

Finally, all potential non-routine emissions and incidents should be listed, based on
equipment and operational practices on site. This could include:
• Unlit, malfunctioning, or inefficient flares
• Operational issues on the storage tank, such as an open thief hatch, typically in the
case of onshore operations
• Maintenance or equipment stopped/started/purged
• Upsets/malfunctions

Since screening is also an essential first step to quantification, the operator should
consider whether the objective includes quantification. If so, the exhaustive list of all
emission sources can be used for source level quantification (refer to Section 2.3). If
emission quantification is not required, then a qualitative or semi-qualitative assessment
can be used to prioritize mitigation. This process can support mitigation action.

2.2.3 Continuous improvement: update or improve existing source level

inventory continuously
Source level screening of methane emissions should not be a one-time event, but rather
viewed as a snapshot of a situation that can change. Different things can trigger an update
of source level screening, for example:
• Further emissions reduction, such as, the operator aiming to reduce or eliminate
emission sources (in particular, leaks) following a schedule
• Inconsistencies noted from reconciliation, such as variability in emissions not
captured properly at the source level inventory (refer to Sections 2.5 and 2.6)
• Identification of an additional potential source of emissions, such as an unexpected upset

When the aim of the update is to mitigate emissions, how screening should be considered
depends on the category of the emission source. For known vents, for example, source
screening only needs to be updated when there are changes in operations or equipment on
the site, as their status tends to remain relatively consistent. This is not the case for leaks,
which can arise at any moment. For this type of emission source, it is recommended, for
example, to perform regular component screening (that is, the detection component of
LDAR) or continuous monitoring such as for larger, more frequent events. The frequency
and approach will depend on the site(s), previous screening, and the level of operator
ambition to reduce methane emissions.

The objective also plays a role in the screening of unexpected events. At a first level,
operators are encouraged to institute permanent tracking of relevant parameters (such as,
SCADA, online meters, and so on) that could indicate when unexpected events occur. To
go further, operators are encouraged to implement continuous monitoring to identify and
address the source of unexpected emissions.

33
Recommended practices for methane emissions detection and quantification technologies – upstream

When source level screening is performed to resolve issues with reconciliation, all source
categories may be treated in a similar way, i.e., by performing a new source level screening.
This should take place at the same time as the site level measurement to provide potential
explanations for emissions found at that time.

Finally, the aim of the screening may be to add a potential source to the list of emission
sources at a facility. This typically only applies to unexpected events. Once the emission source
has been addressed, the operator is encouraged to record it and, if relevant, quantify it.

2.3 Tree 2 - Source level quantification

Source level quantification can be done by quantifying emissions from individual sources
and summing them. A list of all potential emission sources is required to perform accurate
source level quantification. This should be informed by the steps presented in Section 2.2
(Tree 1) to identify all potential sources at a facility. Without this, emissions may be under-
or over-estimated.

For source level, there are four methods to quantify emissions, each of which is covered in
this section:
• Generic emission factors
• Measurement-based emission factors
• Engineering calculations
• Measurements

This decision tree helps operators identify the appropriate quantification method for each
source at a facility.

If the goal of the inventory is a simple, source level inventory, generic emission factors
(such as in OGMP Level 3) may be used. It is important to note that using generic emission
factors may result in higher uncertainty or errors and may not provide accurate results.
However, this approach can be used in a first, high-level assessment to develop a baseline,
which can be improved by adding measurements or engineering calculations. In addition,
or alternatively, this approach may be used to prioritize mitigation or to pinpoint emission
sources that could represent interesting mitigation projects.

If the goal is to develop a source level inventory based on measurements or engineering

calculations, or to perform reconciliation with site level, measurement-based
quantification, a more detailed approach should be taken. For any emission source, an
operator must first determine if the source is material, that is, if it is likely to contribute a
non-negligible share of site level emissions. This can be informed by operator knowledge
of site processes and previous quantification methods, such as generic emission factors.
If the emission source is non-material, a generic emission factor can be used to quantify,
as any error is unlikely to significantly contribute to the uncertainty of quantification for
the facility. If the emission source is material, or may be material, a more conservative,
source-specific quantification approach is recommended. This can involve measurements,
engineering calculations, or other methods equivalent to OGMP Level 4 quantification, as
presented in the OGMP technical guidance documents.16

16
https://ogmpartnership.com/guidance-documents-and-templates/

34
Recommended practices for methane emissions detection and quantification technologies – upstream

2 Emission inventory source-level quantification

Follow the tree to identify appropriate quantification methods for each emission source identified

Information required

• A list of all potential emission sources to perform a conclusive source-level quantification (See tree 1 for process to create this list)
• While no recommendations on the percentage of components to sample, it is recommended to use Measurements, Engineering Calculations or
measurement-based EF where possible.

Is the goal to develop a simplified

source-level emission inventory?1

No Yes Simplified

Is the emission source material?2

(Optional) Yes, or uncertain No

Generic EF
Is a statistically representative It is part of a simplified source-level inventory, or a
measurement dataset on like Is it possible to measure? non-material source, difficult to quantify otherwise,
No satisfactory generic EF available, equivalent to
system available?
OGMP level 3 (Refer to OGMP TGDs)

Yes Yes No

Use measurement-based EF4 Are engineering calculations

(equivalent to OGMP level 4) possible for the emission source?

Yes

Do engineering calculations provide a reasonable level of uncertainty?

Ex. lower uncertainty than measurements

No Yes
Can contribute to
No Is it unsafe, prohibitively
expensive or logistically
difficult to measure?
No Yes

Perform Measurement
Perform Engineering
(equivalent to OGMP level 4) – refer to the technology database
calculations
(equivalent to OGMP level 4)
What is the level of variability of the emission source? (Informs measurement timing)

Highly intermittent (e.g. unlit

flare, pressure relief valves)

Is it possible to know where and when it is occurring,

its frequency, and duration?

Continuous or near Continuous or near Yes, it can be observed by No, it has a random occurrence
continuous, constant emissions continuous/cyclical, variable monitoring some specific or/and it cannot be monitored
(e.g. baseload turbine) emissions (e.g. routine flaring) parameters with specific parameters

Measure emissions at the

time they are occurring,
Measure emissions at
Measure anytime and frequency, duration and Measure emissions with
different times of the
extrapolate over full volume (Including flare continuous monitoring
cycle, measure frequency
operating time ignition monitoring, solution3
and duration of the cycle
flow meters, presence
sensors, etc.)

1 If the source level inventory is a simplified, high level assessment, a user can choose the simplified source-level quantification method using generic emission factors with the
knowledge that the estimates may be associated with high uncertainty or errors and may not provide accurate results, which can be improved over time with the supplementation of
measurements or engineering calculations.
2 Material emissions are estimated to contribute non-negligible emissions with respect to facility level emissions
3 May be associated with larger emission uncertainties, which can be a function of ex. wind conditions, background methane emission sources, or emission source attribution.
However, implementing continuous monitoring is better than having no measurements.
4 Measurement-based emission factors can be developed as part of level 4 quantification for like systems. Generally, events or equipment with similar operational, environmental or design
characteristics can be considered as like systems. Variations around some characteristics are acceptable, if it can be demonstrated that these do not significantly affect methane emissions.

Figure 8: Source-level quantification

35
Recommended practices for methane emissions detection and quantification technologies – upstream

If the source is material, the first question is whether it is possible to reliably measure
emissions. If the methane emissions cannot be measured for technical reasons, it
is recommended to rely on engineering calculations for quantification. Engineering
calculations can be preferred if taking measurements could be unsafe, expensive, difficult,
or results in greater uncertainty. However, this requires that engineering calculations exist,
are possible, and provide a reasonable level of uncertainty for quantification at the source
level.

When it is not prohibitively challenging to do so, it is recommended to measure. This Report

provides general principles for measuring several different types of emission sources.
Additional factors should also be considered, such as the complexity of the emissions and
operations, topography, or meteorological conditions.

When measuring, variability should be considered to inform the measurement timing

and to assess total emissions over the relevant timeframe. If the emission source is
continuously or near-continuously emitting at the same rate, measurements can be
performed anytime and extrapolated over the full period of operation.

If the source is cyclical with variable emissions, measurements should be taken at different
times in the cycle and attributed to the different operating modes of the source that would
reflect overall emissions.

If the source is highly intermittent or event-based, such as in the case of an unlit flare, and
it is possible to know the frequency, duration and timing of such emissions, measurements
should be performed to capture volume, frequency, and duration.

If the event occurs at random or is not monitorable, it is recommended to measure

emissions using continuous monitoring.

An alternative, optional quantification route is to use emission factors derived from previous
measurements performed on site or on other sites with similar operating conditions,
though only if a representative dataset, based on similar sources, is available. In this case,
measurement-based emission factors can be used, such as in line with OGMP Level 4, in
place of measurements, engineering calculations or generic emission factors. Generally,
events or equipment with similar characteristics – referred to as ‘like systems’ – can be
considered representative. Variations around some characteristics are acceptable if they do
not significantly affect the volume of methane emissions.

Since emissions are rarely consistent over time, source level quantification should be
updated regularly, and particularly in the case of newly identified sources, modifications
of site design and operations, or changes in materiality, operating conditions, or
characteristics of existing sources.

36
Recommended practices for methane emissions detection and quantification technologies – upstream

2.4 Tree 3 – Measurement based emissions quantification – site/

group of equipment
This section guides the selection and deployment of technologies for site level
quantification. For the purpose of this report, site level quantification is defined as
emissions measurement at the scale of the site and is independent of site measurements
at source level. This definition is in line with the one considered in OGMP 2.017. Other
definitions for site level quantification have also been presented as part of other guidelines
or standards, such as GTI Veritas which defines site level measurements as “Methane
measurements taken at spatial scales greater than the component or equipment scale,
capable of detecting and/or quantifying emissions without the knowledge of a source level
inventory.”18 Tree 3 is also applicable for measurements of a group of equipment (also
definable as a functional element), which may be defined as spatially separable areas
related to different identified processes.19 The decision tree below is applicable for site level
measurements following the OGMP 2.0 definition, as well as for sites as defined by GTI
Veritas which focuses on site level measurements for a spatially distinct set of equipment.
The tree does not follow the exact structure of a decision tree since what it presents should
be considered simultaneously rather than sequentially. The tree is complementary to the
technology filtering tool, highlighting constraints to consider when selecting appropriate
site level quantification technologies in the technology filtering tool (see Section 1).

Collecting information on site characteristics, such as location and environmental

conditions, helps optimize selection and is necessary for avoiding non-applicable
technologies, such as those intended for use onshore when the site is offshore.

Prior to site level quantification, it is recommended to have estimates of total site level
emission rates based on source level quantification (as described in Section 2.3), including
from routine and non-routine sources and ideally obtained under different operational
modes. It is important to define the goal of the site level quantification, which is the entry
point for the processes described in this section. The question of frequency of site level
quantification is addressed in Section 2.7.6.

2.4.1 What is the aim of the site level quantification?

There are two essential parameters of the site level measurement technology to consider:
• The detection and quantification threshold, above which emissions should be
measured.
• The uncertainty of the quantification.

The extent to which these criteria should be fulfilled depends on the goal of the site level
quantification.

17
OGMP 2.02022, https://ogmpartnership.com/wp-content/uploads/2023/02/OGMP-2.0-UR-Guidance-document-SG-approved.pdf
18
https://veritas.gti.energy/protocols
19
Innocenti et al.,2023

37
Recommended practices for methane emissions detection and quantification technologies – upstream

3 Site-level quantification measurement

The main tool for selecting site-level quantification technology is the technology database. The different aspects present in this
document are to be considered simultaneously (as filters) rather then sequentially.

Information required

• Information on site characteristics (location, environmental conditions, other co-located industrial activity …)
• Objective of site level quantification (reconciliation with source-level inventory, screening assessment for anomalous
emissions…)
• Source-level assessment of total emission rates (in different operational mode, if possible) – is recommended to be done
prior to site-level measurements, including knowledge of both routine and non routine emission sources – ref Tree 2

What is the objective of the site-level quantification?

Inform inventory / validation /

Monitor and address potential super Build understanding of temporal
reconciliation
emitters / unexpected sources variability – continuous
(equivalent to OGMP level 5)

Select a technology with a Select a technology with a Select technology with

threshold well below expected threshold higher than the total of quantification threshold (or
emission rate determined by continuous source, and in line with alarm threshold) that does not
source-level inventory – within either super emitter definition for generate alarm fatigue (i.e quality
Threshold reasonable costs, logistical and your site or proportionately large degradation due to repetition).
labor efforts with regards to the emission sources. Detection threshold can be slightly
absolute level of emissions.
Very high probability of detection higher than the total of
Very high probability of detection
required for the threshold target. the continuous sources.
required for the threshold target.

Requirement on the uncertainty

Technologies with documented of the quantification depends
uncertainties that consider on whether the quantification NA - Currently high to very
Uncertainty uncertainty of the sensor and of will directly be used for inventory high uncertainty for all
the method depending on or whether the measurement technologies assessed.
environment conditions. will be combined with other
estimation methods.

Technology constraints to consider when selecting site-level quantification technology:

Validation Source localization

Safety Selection of technologies that can
Documented, transparent validation
Technologies that respect company and attribute emissions to desired level
of emissions (third party testing,
local safety requirements, e.g. ATEX (e.g. site or equipment level) and that are
public availability of information,
certification, civil aviation requirements, appropriate with respect to the facility
controlled release testing in
IOGP/company aviation requirements. characteristics (e.g. small/large, congested/
representative conditions).
geographically dispersed assets).

Environment
Operational data Technologies may be impacted by
Availability Ensure field data collection at the time of environmental conditions (e.g. cloud cover,
Import/export, commercial availability monitoring (operational mode, events, …) snow, precipitation) that undermine
in-country and other restrictions and to improve the understanding of their ability to monitor emissions at
logistical constraints for technologies. operational factors and correlate them desired frequency.
to measured levels of emissions. Location offshore may also make some
technologies not applicable.

Figure 9: Site-level quantification

38
Recommended practices for methane emissions detection and quantification technologies – upstream

2.4.1.1 Inform inventory, validation, or reconciliation

When the aim is to develop an inventory or to validate or reconcile emissions (for example,
equivalent to OGMP Level 5 or measurement-only reconciliation pathway in the GTI Veritas
Protocols20), the operator is encouraged to choose technologies with a minimum detection
threshold that will capture the majority, such as 90% of expected emissions by the source
level inventory. This increases the likelihood of obtaining accurate site level measurements.

Measurement technologies with low detection thresholds should be selected, bearing in

mind costs, logistical and labour efforts relative to the level of emissions at the site. At a
minimum, they must have a very high probability of detecting known emissions, based on
the source level emission rates determined by the inventory, at an aggregate level.

The operator is encouraged to choose technologies for which uncertainties are well
documented, including both the uncertainty of the sensor and the uncertainty of the
method, which may be impacted by environmental conditions, because uncertainty analysis
is at the core of the validation and reconciliation process. Where important temporal
variability is proven or expected, it can be interesting to select a technology that can easily
be deployed multiple times over the observation period to reduce temporal uncertainty.

2.4.1.2 Monitor and address high emitters and/or unexpected sources

Another objective could be to monitor and address super emitters and/or unexpected
sources, which would correspond to a High Minimum Detection Limit (MDL) measurement
technology according to GTI Veritas Protocols. Threshold requirements in this case depend
on the operator’s definition of a “super emitter,” as well as what is considered a “large”
emission source for the site. Selected technologies need an adequate detection threshold
and a very high probability of detection above this threshold. If the goal is only to detect
abnormally large emissions, it is recommended to choose a threshold higher than the total
of known continuous emissions sources (as assessed by source level quantification).

The requirements regarding uncertainty depend on the goal of the estimates. Uncertainty
requirements will differ, for example, depending on whether the quantification will also be
used to develop an inventory or whether the measurements will be combined with other
quantification methods, such as engineering calculations or process simulation.

2.4.1.3 Build an understanding of temporal variability – continuous quantification

Finally, operators could choose to deploy site level quantification technologies to
understand temporal variability of site emissions. This is typically done through continuous
site level quantification.

It is recommended to select a quantification threshold (or alarm threshold) that is low

enough to measure emissions from the targeted emission sources a sufficient share of
the time, but high enough to avoid alarm fatigue, that is, deterioration in the quality of the
alarm follow-up due to too many alarms. Usually, the threshold can be slightly higher than
the total of the continuous emissions sources.

20
https://veritas.gti.energy/protocols

39
Recommended practices for methane emissions detection and quantification technologies – upstream

In this case, no consideration of uncertainty is required because all assessed technologies

that allow for continuous quantification have a high quantification uncertainty21. Time series
analysis of site emissions should consider this.

2.4.2 Other constraints when selecting site level quantification technology

In addition to adequate threshold and uncertainty, some constraints can impact selection.
When evaluating technologies for continuous monitoring of site level emissions, operators
may consider constraints regarding:
• validation
• safety
• source localization
• availability
• operational data collection
• environmental conditions

These constraints are included in the technology filters that are described in Section 1.

2.5 Tree 4 - Reconciliation for a single site

This section covers reconciliation between a source level inventory and site level
measurement, from either a single site or group of equipment (reconciliation for a group
of sites or for a single site with multiple measurements over time is covered in Section
2.6). Many operators have reported that it can be helpful to initially perform a reconciliation
exercise on a group of equipment or a small site, as opposed to a large site, since starting
small can improve understanding of the connection between source level and site level
emissions.

2.5.1 Information required

Before a reconciliation exercise, it is necessary to collect methane emissions data for the
target site, including:
• The conclusive results of a site level measurement, namely:
– The detection threshold of the technology, considering geographical conditions
(such as high latitudes) and applicable environmental conditions (for example,
high windspeeds measured at the time of the site level quantification) which
may increase the detection threshold.
– The rate of the site level quantification (if emissions are detected), e.g., 23 kg/h.
– Uncertainty (e.g., +/- 30%) covering not only the sensors but also the method
at the time of the measurement, including consideration of environmental
conditions that could have an impact (refer to Section 4.1).
• A source level inventory (refer to Section 2.3 to create one) with uncertainty
assessment, if relevant.

21
See the list of technologies assessed in Appendix B

40
Recommended practices for methane emissions detection and quantification technologies – upstream

Reconciliation between source level inventory and site level measurement

4
– single site or group of equipment1 for a single point in time
1 Determine a source level inventory at the time of the site level measurement

• For each emission source present in the inventory:

Uncertainty
– To the best ability, determine if the source was present at the time of the measurement. Information required

range
– Determine the emission rate at the time of the measurement (note that the approach
is different between continuous sources and intermittent/event-based sources) • Source level inventory (refer to Tree 2)
– A detection device (e.g., OGI) present on site at the time of the site level • Conclusive result of a site level

Uncertainty
quantification may inform if an emission source was emitting when the measurement including:

range
measurement was performed. Source – Detection threshold of the
• Determine the expected total emission rate at the time of the site level measurement quantification technology deployed Detection
considering all continuous emissions and intermittent/event-based sources occurring at the time. – If detected, emission rate Threshold

Notes: of the site level quantification

• If step 1 is performed at an early stage, the estimate can inform the technology selection for site – Uncertainty of site level Site
level measurement quantification measurement
• If it is not possible to determine a source level inventory at the time of the site level measurement (e.g. if – Consideration of weather
inventory is limited to annual reporting), it is possible to skip and go directly to step 2. However, caution should conditions and geographical site setup
be taken as this may result in larger uncertainties on the reconciliation performed .

Comparison between source level inventory at the time of the measurement and site level
2
measurement-based quantification
Were emissions detected during the site-level measurement?
No (i.e. emissions above the detection threshold of the measurement technology) Yes

Is the source quantification expected Is there an overlap between the site-level measurement and the
to be above the detection threshold? source quantification when considering the uncertainty ranges?

No Yes No Yes

Is the source quantification at the time of site measurement

greater than the measured site level emissions?

Yes No

Overlap
Site Source Site Source
measurement quantification measurement quantification

Reconciliation completed – reconsider

site-level measurement technology
selection (e.g. lower detection
threshold)
Repeat exercise over time (refer to
Site Source Site Source Site Source
Frequency section in report)
measurement quantification measurement quantification measurement quantification

Is there a risk that the site level measurement did not capture all emission Reconciliation successful2 – repeat exercise over
sources? Is there a potential issue with the site level quantification? time (refer to Frequency section in report)

Yes No
Are there emission sources that may be overlooked during source quantification? In particular:
• Unlit flare?
• Operational issues on the storage tank? (e.g., open thief hatch)
• Equipment maintenance, or equipment being stopped/started/purged?
• Equipment upsets/malfunctions?

No Yes

Identify the emission sources which may be a source of discrepancy in the source quantification. In particular:
• Emission sources where the quantification is based on generic EF which may not be representative of the site
(level 3 in OGMP 2.0)
• Emission sources with high variability over time
• Emission sources which represent an important source of emission and have large uncertainty.

3 Recommended additional measurements/quantification

Reconsider site level measurement

Improve source level quantification with additional measurement Perform source level quantification for
technology selection
or engineering calculations for emission sources relevant additional sources
AND/OR
AND/OR OR
Ensure source level quantification and
Review the quality of the site measurement/uncertainty range Review source quantification to ensure all
site level measurement cover same
and reconsider site level measurement technology emission sources are accounted for
emission sources
Repeat from step 1 Repeat from step 1
Repeat from step 1

1 Depending on the site level measurement technique. It is recommended to use group of equipment if possible.
2 Consideration should be taken if the site-level measurement technology results in a large uncertainty range. In that situation, it is recommended to consider an alternate site-level
measurement technology.

Figure 10 - Reconciliation for a single site

41
Recommended practices for methane emissions detection and quantification technologies – upstream

2.5.2 Source level quantification

Once the data has been gathered, the next step is determining a source level inventory at
the time of the site level measurement. This will be used to calculate an expected value for
the site level emission rate. This should be done in a unit that allows comparison with the
actual site level measurement, for instance, both measurements might be expressed in
mass of methane per hour.

For each of the emission sources in the inventory, the operator needs to:
• Determine whether the source was present at the time of the site level measurement.
For example, liquids unloading may be a large source of emissions for a site over
the year but may not have taken place at the time of the site level measurement. As
another example, a particular compressor may not have been running at the time of
the site level measurement, so should be discarded for the reconciliation.
• Determine the emission rate at the time of the site level measurement:
– For continuous sources with limited variability, rates could be considered as
constant (total of yearly emissions divided by the operational time). Or, the rate
could be used directly, for example, when the flow of a specific vent is measured
continuously with a flowmeter.
– For variable, intermittent or event-based sources, such as liquids unloading,
storage tank loading, equipment blowdown, gas driven pneumatic controllers
and pumps, maintenance activities and well casinghead gas venting, it is
important to understand if these were occurring at the time of the site level
measurement. Other relevant parameters monitored using SCADA, online
meters, etc., may also be leverageable and that could indicate how these
sources may be emitting. The emission rate may be determined based on
either the duration of the event and total emissions, or use of the emission rate
directly.

With this data, the expected total emissions rate at the time of the site level measurement
can be determined, considering all continuous emissions and intermittent/event-based
emissions at the time. A detection device, such as an OGI camera, deployed during the
site level quantification can be a helpful tool to ensure that all potential sources are in the
inventory.

If the source level inventory is available from an earlier period, this can inform the
technology selection for the site level measurement (refer to Section 2.4).

If it is not possible to determine a source level inventory at the time of the site level
measurement, some operators have used the yearly average for the reconciliation. This
may cause larger uncertainties for reconciliation if a large share of source level emissions
at the site is variable, which in turn reduces the relevance of performing reconciliation.

42
Recommended practices for methane emissions detection and quantification technologies – upstream

2.5.3 Comparison between source level inventory at the time of the

measurement and site level, measurement-based quantification
Once the site level quantification has been successfully performed, the next step is to
determine whether any emissions were detected or whether they were below the detection
threshold.

If none were detected, it is still possible to draw meaningful conclusions in some cases. For
example, if site level emissions were expected to be below the detection threshold based on
source level quantification, the fact that no emissions were detected at the site level would
suggest that the source level inventory does not exclude a major source of emissions.
In that case, reconciliation could be considered completed. It is still recommended to
regularly review the appropriateness of the site level measurement technology (see Section
2.4), and repeat the exercise over time (refer to Section 2.7), since reconciliation only
represents a particular moment and its validity is therefore time-limited.

If no emissions were detected at the site level, but source level quantification leads the
operator to expect emissions above the detection threshold during site level quantification,
reconciliation may be indicating issues with either source level or site level quantification,
which would need to be assessed.

The first element to consider would be the risk that the site level measurement did not
capture all emission sources, or that there is some other problem with the site level
quantification. If the risk of this is high, or if another issue with the site level quantification
is found, the operator may need to reconsider use of the particular site level measurement
technology and/or further ensure that source- and site level measurements cover the same
sources. This should be followed up by a new reconciliation exercise.

Alternatively, the operator may ensure source level quantification and site level
measurement cover the same emission sources by excluding those from the source level
inventory which were not covered by the site level measurement (see Section 2.5.2 above
for intermittent events).

If, however, the risk that the site level measurement failed to capture all emission sources
is low, and if no other issues with the site level quantification have been identified, a more
detailed analysis of the source level inventory would be required. This involves identifying
the emission sources which could cause a discrepancy. Priority should be given to
reviewing the following types of sources, which have been shown to be more likely than
others to cause discrepancies between source- and site level quantification22,23
• Sources for which quantification is based on generic emission factors, which may not
be representative of actual emissions (ref. Level 3 in OGMP 2.0).
• Sources that are highly variable over time.
• Sources which are expected to represent an important share of site level emissions,
but which have a high level of uncertainty associated with their quantification.

22
Vaughn T, et al.,20187
23
Zavala-Araiza D, et al.,, 2015

43
Recommended practices for methane emissions detection and quantification technologies – upstream

It is recommended to review the source level quantification, supplementing this with

measurements or engineering calculations, and to review the reconciliation process from
the beginning.

In addition, or as an alternative, before starting the reconciliation process from the

beginning, the quality of the site level measurement or its uncertainty range could be
reviewed, focusing on the measurement threshold and uncertainty range.

On the other hand, when emissions are detected and successfully quantified by site level
measurement, there are three possible scenarios:
• Measured site level quantification is lower than what source level quantification would
suggest.
• Measured site level quantification is higher than what source level quantification
would suggest.
• Site level and source level quantification are aligned.

For the third scenario, in which there is an overlap between the uncertainty ranges of
site level and source level quantification, the reconciliation exercise is considered to be
successful. A one-off successful reconciliation is an indication of a satisfactory inventory at
a given point in time. Nevertheless, the exercise should be repeated periodically to confirm
the validity of the inventory over time and across different conditions. Elements to consider
when assessing the frequency of site level measurements can be found in Section 2.7.

If the site level measurement is below the lower uncertainty range for the emissions that
could be expected based on the source level inventory, reconciliation has indicated potential
issues with source level or site level quantification which would need to be assessed.
Like the case in which no site level emissions are detected, the first possibility is that the
measurement may not have captured all sources, or that there could be some other issue
with the site level quantification. If the risk of either of these is high, or if another issue is
detected, an operator should review the selection of the site level measurement technology
and/or ensure that source- and site level measurements cover the same emission sources.
This step should be followed by a new reconciliation exercise.

However, if the risk of missing sources in the source level inventory is low and no issues
with site level quantification have been identified, a detailed analysis of the source
level inventory should be conducted to identify the sources which may be causing the
discrepancy. As described previously, it is important to review emission sources for which
the quantification is based on generic emission factors, that are highly variable over time,
and which have a high level of uncertainty associated with their quantification to identify the
source of the discrepancy.

It is recommended to review the source level quantification with additional measurements

or engineering calculations for relevant sources, and to review the reconciliation
process from the beginning. In addition, or as an alternative, the quality of the site level
measurement or its uncertainty range could be reviewed by reconsidering the choice of site
level measurement technology before redoing the reconciliation process.

If the site level measurement is above the upper-bound of the uncertainty range of the
emissions expected from the source level inventory, the reconciliation exercise has
indicated potential issues with the source level quantification. Identifying the source(s) of

44
Recommended practices for methane emissions detection and quantification technologies – upstream

the discrepancy will allow the operator to improve the source level inventory and potentially
reduce these emissions.

The first thing to consider when looking at potential reasons that the site level
measurement is higher than expected is whether there are any emission sources which
could have been overlooked during source level quantification. As mentioned in Section
2.2.2, the following sources tend to lead to large emissions but are not always captured by
source level inventories:
• Unlit or malfunctioning flare, or other issues with flare ignition.
• Operational issues with storage tanks, such as an open thief hatch (typically, in the
case of onshore operations).
• Maintenance, or equipment being stopped/started/purged during the site level
measurement.
• Equipment upsets/malfunctions.

If large sources that could explain the discrepancies are identified, a source level
quantification can be performed for the additional sources. The site level quantification
should then be reviewed against the revised source level inventory, which should now
include the additional sources. The alternative is to review the source level quantification
to ensure that all sources are accounted for before comparing it with the site level
measurement.

If no large sources that could explain the discrepancy are identified, other, smaller, sources
of discrepancies should be reviewed, including the following:

• Sources for which the quantification is based on generic emission factors that may
not be representative of actual emissions for the site, such as Level 3 in OGMP 2.0 or GTI
Veritas bottom-up inventory methods.

• Sources that are highly variable over time.

• Sources which would be expected to represent an important share of site level

emissions, and which have high quantification uncertainty.

It is recommended to review the source level quantification by supplementing with

additional measurements or engineering calculations, and to review the reconciliation
process. In addition, or alternatively, the quality of the site level measurement or its
uncertainty range could be reviewed by reconsidering the site level measurement
technology before redoing the reconciliation process.

2.5.4 Additional considerations

The relative size of the uncertainty range affects the outcome of the reconciliation
exercise. When looking only at a reconciliation outcome, there is an incentive to favouring
technologies with larger uncertainty ranges that may increase the chances of an overlap
between the source- and site level quantification uncertainty ranges. Therefore, whether
the site level measurement technology results in a large uncertainty range should
be established. If so, it is recommended to select alternative site level measurement
technology. In some cases, the selected technology may be the most suitable technology
and is expected to be able to capture the majority of emissions, or if it is in line with local

45
Recommended practices for methane emissions detection and quantification technologies – upstream

regulation minimum detection threshold requirements. It may also be recommended to

perform additional measurements with the same technology within similar conditions when
emissions are expected to be similar. Either of these could reduce the uncertainty of the
site level measurement and increase confidence in reconciliation.

In general, there may be limitations to performing these snapshot reconciliation exercises.

For example, this may be due to the challenges with deriving source level uncertainty
estimates, which can be technically challenging and time consuming. Methane emissions
can be highly variable over time (see Section 2.7.3), such that conducting site/facility-
level measurements may result in high levels of uncertainty. The extrapolation of these
measurement beyond the measured time frames will also introduce additional uncertainty.
Therefore, this typically does not allow an annual inventory or reporting with a satisfactory
level of certainty. Reconciliation is intended to be repeated. The main objectives are to
learn from the reconciliation to improve the quality of the quantification and reporting
and to achieve emission reductions. Operators can conduct site level measurements and
reconciliation that cover various conditions and patterns. Reconciliations do not need to
be performed using the same technologies. The process described above can be adjusted
to reflect different detection thresholds and uncertainties of the site level measurement
technologies used as part of the time-series analysis. Combinations of technologies can
be used in measurement and reconciliation over time. For example, multiple site level
measurement technologies may be combined to increase measurement coverage of site
level emissions. Examples can be found in Section 3.

2.6 Tree 5 – Reconciliation for a group of sites

This section covers the decision tree for reconciliation between a source level inventory and
a site level measurement from a group of sites for which there are measurements from a
point in time, or from a single site with measurements over time. The aim is to understand
reconciliation exercises involving many site level measurement data points (Reconciliation
of a single site with site level measurement performed at a single point in time is covered
by Section 2.5.). This is intended as a simplified approach, however, other approaches for
reconciliation exist, such as the one proposed by GTI Veritas.

If sites are not similar, the grouping of sites should define subgroups of ‘like’ systems.
In addition, given the loss of precision compared to single-site reconciliation, it is
recommended to follow this methodology only in cases where it is challenging or expensive
to perform many single-site reconciliations (refer to Section 2.5).

These multi-site measurements can provide a wealth of additional or more granular

observations and interpretations. Some groups, organizations, and researchers have
developed more advanced methodologies to reconcile emissions across large samples.

46
Recommended practices for methane emissions detection and quantification technologies – upstream

Reconciliation between source level inventory and site level measurement –

5
multiple sites/multiple measurements1
1 Use this tree?
Information required
• Conclusive result of a sample of site level measurement including:
Perform multiple, single site – Detection threshold of the technology deployed
Are there many similar sites or many site reconciliations (refer to tree 3) – If detected, emission rate of the site level quantification
level measurements of the same site? Is it – Uncertainty of site level quantification
logistically challenging or very expensive to OR – Consideration of weather conditions and geographical site setup
perform many, single site reconciliations? No • It is assumed that the operator has a source level inventory available for all sites
Review site grouping in the sample

Yes

2 Compare site level and source level estimates

Source quantification for the sites = Average of source-level quantifications of the sites or of the site (refer to tree 2)
Overlap between site- and
source-level quantification? Uncertainty of source quantification Total uncertainty or statistical analysis of the source-level quantification of the sites
=
(optional) – Propagation of uncertainty
Site-level measurements of the sites = Average of site level measurements
Uncertainty

Uncertainty
range

Total uncertainty or statistical analysis2 of the site level measurements –

range

Uncertainty of site-level measurements =

Propagation of uncertainty

Detection Large uncertainty?

Threshold Yes The uncertainty range is very wide, the Yes Consider an alternate site level
reconciliation may be successful in theory quantification technology
Total site Total source but without robust insight
measurement quantification Total site Total source
measurement quantification

No
No Reconciliation successful
Consider performing site level (tree 3) reconciliation on a few sites, which will provide additional insight

3 Root cause analysis of discrepancies in reconciliation – source level quantification

Not capturing all emissions?

Are there emission sources that may be overlooked during source quantifications? In particular: Add sources to source
• Unlit flare? Yes quantification
• Operational issues on the storage tank? (e.g., open thief hatch) (refer to tree 2)
• Equipment maintenance, or equipment being stopped/started/purged? and start step 2
Total site Total source
• Equipment upsets/malfunctions? measurement quantification

No
Miscalculated emission source?
Review existing source
Identify the emission sources which may be a source of discrepancy in the source quantification. In particular: Yes quantification
• Emission sources where the quantification is based on generic EF which may not be representative of the sites (e.g. level 3 in OGMP 2.0) (refer to tree 2)
• Emission sources with high variability over time (liquids unloading, equipment blowdown, starts and stops, casinghead gas venting, …) and start at step 2
• Emission sources which represent an important source of emission and have large uncertainty

No

Root cause analysis of discrepancies in reconciliation – site level quantification

Detection threshold too high?
Occurrence

Missing an important share of the total distribution of emissions, including smaller emission sources. This can result Yes Consider an alternate
in both an over- or an under-estimation of site-level measurements. If technologies with different detection thresholds site level quantification
are used, this can be built into the statistical analysis of the emissions curve and identified super emitters can be used technology
to inform and improve source-level inventory. Emission Rate

No
Either perform more site
Not capturing all emissions? Yes level measurements or
Site level measurements may not capture all the emission sources (for example an equipment a bit on the side) or the exclude the emission sources
distribution of different operational modes or upsets, neighboring emission sources, intermittency etc. missed from the source level
Total site Total source
measurement quantification estimate and start from 2

No

Root cause analysis of discrepancies in reconciliation – sampling strategy

Occurrence

Sampling? Yes Do more site level

There is not enough points to fully understand the distribution. This can result in an unsatisfactory estimation of the measurements
average and uncertainty of the site-level measurements.
Emission Rate

No

Grouping of sites considered as similar? Review site grouping,

Occurrence

Yes separate the original

After statistical analysis, the distribution indicates that the population of initially considered similar sites may
population into two or more
not have similar enough emission characteristics. groups and restart from
Note: not relevant for multiple site quantification for a single site step 2 for each group
Emission Rate

1 The tree is presented for multiple sites – the approach is similar if the operator has performed many site level measurements for one site
2 Where site quantification technologies with different detection thresholds are used, this should be reflected in the total uncertainty and/or statistical analysis

Figure 11 - Reconciliation between source level inventory and site level measurement

47
Recommended practices for methane emissions detection and quantification technologies – upstream

2.6.1 Required information

Before reconciling, collect methane emissions data for the target sites in the group (or for
the single target site over time), including:
• The conclusive results of a site level quantification, including:
– The detection threshold of the technology, considering geographical conditions
(such as high latitudes) and environmental conditions at deployment (for
instance, high windspeeds measured at the time of the site level quantification)
which may increase the detection threshold.
– The rate of the site level quantification (if emissions are detected), e.g., 23 kg/h.
– The uncertainty of the site level quantification, covering the uncertainty of the
sensors and of the method, including consideration of environmental conditions
which can have an impact (refer to Section 4.1).
• A source level inventory (refer to Section 2.3) with uncertainty assessment if relevant.

2.6.2 Compare site level and source level estimates

Once measurements have been taken and results are available, the site level and source
level quantifications can be compared to identify if the results are consistent.

Four different elements are required for the analysis:

• Source quantification for the sites is the average of the source level quantification
exercises of the sites (or site) included in the analysis (refer to Section 2.4).
• Uncertainty of source level quantification is the total uncertainty of the source level
quantification exercises of the sites, taking into consideration spread of uncertainty
throughout the measurements. This parameter is optional. The analysis can be
conducted without this information, which may be difficult to obtain depending on the
approach. However, use of uncertainty levels increases the rigour of the analysis and
the reliability of the results.
• Site level measurement of the sites is the average of all the site level measurements
considered in this study, relying on statistical analysis where relevant.
• Uncertainty of site level measurements is the total uncertainty of the site level
measurements, taking into consideration the spread of uncertainty throughout the
measurements. Uncertainty typically decreases when the number of measurements
increases. When the result of a site measurement has been inconclusive or below the
detection threshold, the operator should consider this uncertainty as well. If site level
measurement technologies with different detection thresholds are used, this can be
reflected in the total uncertainty using statistical analysis, by considering the share of
emissions potentially not captured by the site level measurements. Alternatively, two
analyses can be conducted, segregating the data from the different technologies.

Once these data have been obtained, one should look for an overlap between:
• The total of site level measurements and its uncertainty range (from now on, referred
to as “total site measurements”).
• The source level quantification and its uncertainty range, if available (from now on,
referred to as “total source quantification”).

48
Recommended practices for methane emissions detection and quantification technologies – upstream

Where the total source quantification falls within the range of the total site measurements,
the reconciliation exercise can be considered successful for the group of sites. For more
information, one could perform single-site level reconciliation (refer to Section 2.5) for a
sample of sites included in the analysis.

When total source quantification falls within the uncertainty range of total site
measurements due to a very wide uncertainty range of the total site measurements, the
reconciliation can, in theory, be considered successful. However, the value it provides might
be limited. When this happens, it is recommended to re-evaluate the selection of the site
level quantification technology, or to perform additional measurements with the same
technology.

2.6.3 Root causes of discrepancies in reconciliation

This section presents possible explanations for unsuccessful reconciliation and how these
can be addressed.

2.6.3.1 Root causes associated with source level quantification

Where the total source quantification is less than the total site measurements, the first
potential root cause to explore is whether any emission sources were not properly captured
in the total source quantification, in particular:
• Unlit, malfunctioning, or inefficient flares.
• Operational issues on the storage tank, such as an open thief hatch (in the case of
onshore operations).
• Maintenance or equipment being stopped/started/purged.
• Equipment upsets/malfunctions.

If certain sources have been overlooked in the total source quantification, these should be
added to source quantification (refer to Section 2.3). From there, the comparison between
total site measurement and total source quantification can be re-evaluated (refer to Section
2.6.2).

If no sources have been overlooked for source level quantification, another possible cause
for discrepancy between total site measurement and total source quantification is an error
in the quantification of an emission source. To remedy this, identify the emission sources
which may be a cause of discrepancy in the source level quantification. As noted in previous
sections, sources that could cause such discrepancies include:
• Sources for which the quantification is based on generic emission factors that may
not be representative of the sites (e.g., Level 3 in OGMP 2.0).
• Sources with high variability over time (such as liquids unloading, equipment
blowdown, starts and stops, and casinghead gas venting).
• Sources which are expected to represent an important share of site level emissions
but have a high level of uncertainty associated with their quantification.

If such sources are identified, their quantification should be reviewed (refer to Section 2.3).
After this, the comparison between total site measurement and total source quantification
should be re-evaluated (refer to Section 2.6.2).

49
Recommended practices for methane emissions detection and quantification technologies – upstream

2.6.3.2 Root causes associated with site level quantification

Another path to explore when looking for root causes of discrepancies between total site
measurements and total source quantification is problems with the site measurement
itself.

First, the operator should assess if the detection threshold for the site level measurement
technology is well adapted to the sites included in the analysis. An unsuited threshold can
lead to an under-estimation of site level emissions. If the detection level is high compared
to the distribution of emissions across the group of sites, an important share of total
emissions could be missing, including smaller sources. If statistical analysis establishes
that the detection level of the selected site level technology is too high, an alternate site
level measurement technology should be selected, or additional measurements can be
carried out to reduce uncertainty. If several site level measurement technologies with
varying detection thresholds are used, this can be included in the analysis to evaluate if a
sufficient share of emissions are captured by the combination of technologies. Site level
technologies can be deployed for other purposes, such as super-emitter monitoring. Where
super-emitters are identified, the data can be used to improve source level inventory.

Other possible explanations for discrepancies linked to site level quantification are that the
site level measurement may not have captured all emission sources, such as equipment
located far from the main facility. Or, it may not have properly accounted for the distribution
of the different operational modes or upsets, neighbouring emission sources, intermittency,
and so on. In such situations, it is recommended to perform more site level measurements
or exclude the missed emission sources from the source level estimate and review the
comparison between site level and source level estimates.

2.6.3.3 Root causes associated with sampling strategy

Another root cause of discrepancies in reconciliation may be found in the sampling
strategy. For example, the site level measurement sample may not have enough points to
portray the distribution of site level emissions, resulting in unsatisfactory estimates for the
average and uncertainty of the site level measurements. To correct this situation, additional
site level measurements would be required.

The way the sites have been grouped for the analysis could provide an explanation.
Statistical analysis would be required to identify whether a group of sites initially considered
“similar” might not be similar enough in terms of their emissions characteristics. In such
a case, the site grouping should be reviewed to separate the original single group into two
or more groups, followed by a review of the site level and source level comparison for each
group.

The size of the sample is dependent on several factors, including, but not limited to, the
population size, the shape of the distribution, the complexity of the site, the variability of
emissions and operations. Statistical analysis is required to ensure the full distribution of
site level emissions is captured in the sample.

This exercise can be repeated to provide a better understanding of emissions over time
(refer to Section 2.7 for frequency of reconciliation).

50
Recommended practices for methane emissions detection and quantification technologies – upstream

2.7 Tree 6 – Reconciliation to produce a single Measurement

Informed Inventory (MII)
This section covers the decision tree for reconciliation to produce a single, Measurement
Informed Inventory (MII) that relies on both measurements and calculations. The decision
tree follows the GTI Veritas Measurement and Reconciliation Protocol. An MII is defined in
the February 2024 Source-Level Measurement and Reconciliation Protocol for the upstream
segment as follows: “An inventory that is predominantly informed by data from methane
measurements of the assets and sources in the inventory, where predominantly means
methane emissions quantification informed by measurement can be based on 100%
sample size or based on a statistically representative subset of samples. This definition is
different from the Veritas Protocol.24 The aim is to develop an annual inventory estimate of
total methane emissions for a group of equipment, site, or group of sites.

2.7.1 Step 1: Define Scope and Identify Emission Sources

The first step to develop an MII is to define the scope of sources to be included. The scope
may be established at different levels, including:
• A single site.
• All sites within a certain production region.
• All operated sites.

Once the scope is defined, the operator should identify all emission sources within the
scope boundary. Operators can refer to Tree 1 in Section 2.2 to assist in the development
of a list of emission sources. Otherwise, the operator can continue to Tree 2 or another
source-specific inventory protocol.

2.7.2 Step 2: Categorize and Stratify Emission Sources

All emission sources identified in Step 1 may now be categorized as one of the following:
• Best Calculated: sources whose emissions are not well characterized by snapshot
measurements and/or are more accurately estimated by engineering approaches or
EFs, for example:
1) Sources whose activity is tracked, or emission times are bounded by
independent means like SCADA systems or other activity records.
2) Sources that are expected to be below detection limits of deployed
measurement technologies, intermittent, and/or short in duration.
• Best Measured: sources whose annual emissions are accurately estimated using
measurements. Technologies or methods for direct measurements of sources in
the ‘Best Measured’ category can be selected by the user. Sources that combine
measurements with engineering calculations can be categorized as ‘Best Measured’.

The operator may also refer to Tree 2 in Section 2.3 for further guidance on determining the
best method for quantifying each emission source.25

24
OGMP 2.0, 2022, https://ogmpartnership.com/wp-content/uploads/2023/02/OGMP-2.0-UR-Guidance-document-SG-approved.pdf
25
Higgins S, et al., 2024doi: https://doi.org/10.2118/219445-PA

51
Recommended practices for methane emissions detection and quantification technologies – upstream

After categorization, the operator will stratify emission sources. Stratification may assist
in the development of emission distributions (Section 2.7.3) and developing sampling and
measurement strategies (Section 2.7.4). Stratification can be performed using various
approaches, including:
• Per emission sources that are best measured.
• Using natural groupings, such as business units or sites within discrete regions.
• By facilities or emission sources that have similar characteristics, such as the age of
equipment, expected emissions, site complexity, temporal variability, or intermittency
of emissions.

Stratification can be performed at different levels of granularity, such as at site, group, or

equipment-level, and it is up to the operator to choose an approach based on their own
operations. Since stratification is meant to influence the sampling and measurement
strategy, stratification should be influenced by the emission characteristics of the Best
Measured sources.

2.7.3 Step 3: Establish Initial Inventory and Expected Emissions Distribution

(EED)
An initial inventory should be developed to provide an estimate of annual emissions of all
emission sources included in the analysis and can be performed for each stratum created
in Step 2. If this is the first year of developing an MII, an EED can be developed based on
several options:
1) Publicly available datasets, such as the US EPA Greenhouse Gas Reporting Program
(GHGRP), the Greenhouse Gas Index (GHGI) or the Natural Gas Sustainability Initiative
(NGSI).
2) A source level inventory developed using Tree 2 or a similar program.
3) An important note is that the GTI Veritas Protocols allow an exception to constructing
an EED in the first year of reporting. As there are many ways to develop an EED, it
can be both time and resource-consuming for some operators, and this step may
therefore be skipped in the first year.

In subsequent years of developing an MII, an EED may be developed using the previous
year’s MII to help inform sampling and measurement strategies. This allows an operator to
have a better understanding of anticipated frequency or intermittency of emission rates for
their own assets based on information specific to their sites and emission sources.

2.7.4 Step 4: Develop Sampling and Measurement Strategies

The GTI Veritas Source-Level Measurement and Reconciliation Protocol requires an MII
to be based on either a 100% sample size (i.e., all of the sources are measured) or a
“representative unbiased sample of all sources within an asset”.

52
Recommended practices for methane emissions detection and quantification technologies – upstream

Other objectives may also be set by the operator, such as quantifying the MII with a given
uncertainty, quantifying a certain percentage of emissions from high-emitting sources,
improving understanding of certain emission sources, developing an MII with lower total
emissions or uncertainty each year, or to screen with a sufficiently low detection limit
technology such that inventories are accurate.

Based on the objectives, the operator can identify methane detection and quantification
technologies that may be relevant using the technology filtering tool26 and accompanying
guidance in Section 1. Consideration should be taken regarding the ability of technologies
to detect, quantify, and localize emissions. Operators may use a combination of
technologies, including technologies that can detect and quantify at different levels (site,
equipment, or component).

The technologies should be subsequently selected for deployment, based on the objectives
defined in Step 4a, the initial inventory developed in Step 3, and Tree 2 (source level
quantification).

Next, a survey plan should be established. All sites included within the scope should be
surveyed at least once. Operators can also refer to Section 2.8 for elements of frequency
for site level measurement. The survey plan should be documented, including a description
and rationale for selection of the measurement campaign objectives, the deployment
technologies, and the frequency of deployments.

2.7.5 Step 5: Deploy Technologies and Collect Data

Next, the measurement technologies should be deployed following the survey plan. During
measurements, additional non-measurement data such as operational data, occurrence
of emissions that are best calculated, results of follow-up investigations of leak indications
or alerts, activities to support the estimation of event durations, and environmental data
should be collected to analyse the data and evaluate data quality.

2.7.6 Step 6: Analyse Data and Evaluate Quality

The next step in developing an MII is to start data analysis to ensure quality measurements
are being used. Reliable emissions inventories require accurate data, which can be
compromised by sensor failures, transcription errors, omissions, or other issues. Before
using the data, it must be checked for quality, completeness, and accuracy. As part of
this process, operators should begin analysing key data properties like the number of
detections, average emission rates, standard deviation, event duration, and emissions
distribution.

Operators should perform data quality checks to look for unusual or untrustworthy
measurements, including anomalies or outliers, or measurements of emissions that are
not attributable to the operator’s own sites or activities.

26
https://www.iogp.org/workstreams/environment/environment/methane-emissions-detection-and-quantification/methane-detection-
and-quantification-technology-filtering-tool/tool/

53
Recommended practices for methane emissions detection and quantification technologies – upstream

Specific environmental conditions that may influence the measurements should also
be considered, including high or variable wind speeds; presence of precipitation, cloud
cover, snow cover, or humidity; very high or low temperatures; complex topography; or
obstructions of measurements.

The evaluation can be supported by the non-measurement data collected at the same
time as measurements were performed, such as data gathered using SCADA or other
continuous monitoring systems, periodic LDAR, or an OGI camera deployed at the time of
measurements.

2.7.7 Step 7: Choose Reconciliation Pathway

Operators have two options to select from when performing reconciliation. The first option
is the measurement-only pathway, which is based entirely on measurements. To select this
pathway, operators must use a technology that is sensitive enough to detect and quantify
over 90% of emissions with measurements and also provides full spatial coverage of the
inventory. In the measurement-only pathway, the MII is the sum of all measurements, ERm.
This option can be selected without completing cause analysis or collecting operational
data. However, this option is not currently widely implemented and is therefore more
intended for the future when technology capabilities improve. It is not expected that
operators will select this option, and will instead use the second option, the hybrid pathway.

The hybrid pathway includes emission estimates using both measurements and
calculations. Measurements can be performed with a measurement technology that
supports the attribution of emissions to sources.

2.7.8 Step 8: Reconcile Inventories and Estimate Measurement Informed

Inventory (MII)
For the measurement-only pathway, the MII is calculated as the sum of ERm. No additional
EFs of engineering calculations are required for this pathway, and the user can proceed to
Step 9.

For the hybrid pathway, emission sources are determined as one of the following:
1) If an emission source is Best Calculated, the emission is estimated using engineering
calculations or engineering factors (as detailed in Tree 2 or another source-specific
inventory protocol) and the emission source is defined as ERc.
2) If the emission source is not Best Calculated (i.e., it is Best Measured), the next step
would be to determine if a quantification technology is able to quantify the emission rate.
a) If so, the emission source is defined as ERm, and the measured emission rate
is used.
b) If the emission rate was not measured by the quantification technology (for
example, if the emission rate was below the technology’s detection threshold
or not measured), the emission source is defined as ERb, and the emission rate
will be determined using relevant emission factors or engineering calculations.

The MII is then calculated as the sum of all ERm, ERc, and ERb.

54
Recommended practices for methane emissions detection and quantification technologies – upstream

6 Reconciliation to produce a Measurement Informed Inventory (MII)

Follow the decision tree to identify appropriate quantification methods for each emission source identified. This decision tree
follows the 10-step approach outlined by the GTI Veritas Protocols. The decision tree summarizes the approach in the Upstream
Protocols for measurement and reconciliation. When necessary, please refer to the main report guidance to use the tree, and the
GTI Veritas document for comprehensive information.

1 Define Scope and Identify Emissions Sources

No
Has a list of emission sources been developed? 1 Develop an emissions inventory

Yes

2 Categorize and Stratify Emission Sources

For each source: determine if source is best calculated or best measured. Operators can Source level quantification
determine which quantification method is the most suitable for each source 2 OR Veritas Guidance Documents

Best Calculated:
Best Measured:
Sources whose emissions are not well characterized by snapshot
• Sources whose annual emissions are most accurately estimated by
measurements and/or are more accurately estimated by engineering
measurements
approaches:
• Technologies or methods for direct measurements of sources in the Best
1. Sources whose activity is tracked or bounded by independent information
Measured category can be selected by the user, provided it is consistent
(SCADA systems or other activity records)
with the guidance presented in the protocol
2. Sources that are expected to be below the detection limits of the deployed
• Sources that combine measurements with engineering calculations can be
measurement technologies, as well as those considered intermittent or
categorized as Best Measured
short in duration

Stratify:
Stratification may inform sampling and measurement strategy. Can be performed based on:
• Per source types that are Best Measured
• Natural groupings (e.g., business units)
• Facilities with similar characteristics (e.g., age, types of equipment, expected emissions, site complexity, temporal variability or intermittency of emissions)
Can be performed at different granularity (e.g., site-level, group of equipment, or equipment level)
Stratification of sources is performed to develop emissions distributions (see Step 3) and for planning the sampling strategy (see Step 4)

3 Establish Initial Inventory and Expected Emissions Distribution (EED)

For each stratum determined in Step 2: No

Is this the first inventory/reconciliation activity?

Yes

Develop EED based on:

Option 1
Develop EED based on:
• Publicly available datasets Option 2
Option 3 • Previous years Measurement Informed Inventory (MII)
• Use a known probability Source level inventory
Exception to constructing an (Step 8)
distribution developed using Tree 2 or
EED in first year • The EED describe the anticipated frequency of
• Previous measurement Veritas Protocols
emission rates and will influence actions in Step 4
campaigns

4 Develop Sampling and Measurement Strategies

Complete steps from a-e:

a) Review and establish b) Identify available c) Select measurement d) Establish Survey Plan e) Document sampling and
objectives measurement technologies methods for deployment Considerations: frequency, measurement strategy
e.g., survey 100% of Use Technology Filtering Informed by the objectives, duration, expected Describes the objectives
facilities within the Tool EEDs, and selected emissions distribution of the measurement
implementation scope, Important to consider the measurement technology See Section 2.7 of Report campaign and the deployed
and collect enough detection, localization, capabilities technologies. It should
measurements that at least quantification, and spatial/ also provide the rationale
50% of total emissions temporal coverage of used in setting objectives,
are determined by suitable technologies selecting technologies,
measurements and determining the
See Section 2.7 of Report scale and frequency of
measurements

Figure 12 - Reconciliation to create a single Measurement Informed Inventory (MII)

55
Recommended practices for methane emissions detection and quantification technologies – upstream

5 Deploy Technologies and Collect Data

• Deploy measurement technologies and implement sampling plan. This can be a single technology, or combination of detection devices, source-level
quantification, or site-level quantification technologies
• In addition, collect non-measurement data (operational data, occurrence of emissions that are best calculated, results of follow-up investigations of leak
indications or alerts, mitigation action to support estimation of duration of events, environmental data)

6 Analyze Data and Evaluate Quality

Data quality checks, cleaning, and analysis Specific environmental considerations: Use of operational data to determine causes and
Consider looking for: • Variable wind speed/direction durations:
• Unusual observations • Precipitation, snow cover, and humidity • SCADA or continuous monitoring system data
• Untrustworthy measurements/outliers • Proximity to a body of water • Periodic LDAR
• Measurements that are not attributable to the • Temperature
operator’s own operations and/or facilities • Cloud cover
• Complex topography, obstructions

7 Choose Reconciliation Pathway

Measurement-Only Pathway1 OR Hybrid Pathway

• Inventory based on measurements and optionally supplemented for • User will use emission estimates from both measurement and
sources below a technology detection limit calculations
• Requires sufficiently sensitive technology to detect >90% of total • Measurements can be performed using source level or site level
emissions, provides full spatial coverage, and produces estimated measurements. If from site level measurements, cause analysis
emission rates for all detection events should be performed to attribute emissions to emission sources (as
Can be implemented without cause analysis for each source, and can be determined in Step 6) which requires measurement approaches with
implemented without operational data sufficient spatial resolution for source-level attribution

8 Reconcile Inventories and Estimate Measurement Informed Inventory (MII)

Yes
For Measurement-Only pathway Is the emission source best calculated?

𝐌𝐌𝐌𝐌𝐈𝐈 = ∑𝑬𝑬𝑬𝑬
𝒎𝒎
No

Was emission rate quantified using

𝑬𝑬𝑬𝑬𝒄𝒄 quantification technology?
Determined using engineering calculations/emission factors
(refer to Tree 2) Yes No

𝑬𝑬𝑬𝑬𝒃𝒃
Note on uncertainty quantification: 𝑬𝑬𝑬𝑬𝒎𝒎
Determine emission rate
Determine emission rate
ERb: based on uncertainty associated with the frequency and/or duration of events, depending using emission factors or
using Measurements as
on calculation method engineering calculations
determined by Tree 2 or
ERc: based on the uncertainties associated with the underlying emissions modeling and as determined by Tree 2 or
Veritas Protocols
calculations used to develop the estimate Veritas Protocols
ERm: should be based on the uncertainty associated with emission source frequency, duration,
and emission rate
As per the GTI Veritas Protocols, uncertainty estimates are optional. There is no one-case- For hybrid pathway
fits-all method to determine uncertainty. Please refer to GTI Veritas Protocols for an in-depth
analysis of potential methods to use to determine uncertainty 𝐌𝐌𝐌𝐌𝐌𝐌= ∑𝑬𝑬𝑬𝑬 + ∑𝑬𝑬𝑬𝑬 + ∑𝑬𝑬𝑬𝑬
𝒎𝒎 𝒄𝒄 𝒃𝒃

9 Evaluate Objectives

Were objectives set in Step 4a met?

Yes No

Objectives met. Repeat in next reporting period Update objectives for next reporting period

10 Develop Report: Refer to GTI Veritas Protocols

1 Note the measurement-only pathway is not currently realistic or widely implemented. It is expected that most operators follow the hybrid pathway.

Figure 12 (continued) - Reconciliation to create a single Measurement Informed Inventory (MII)

56
Recommended practices for methane emissions detection and quantification technologies – upstream

2.7.9 Step 9: Evaluate Objectives

The results of the MII are evaluated in comparison to the objectives set in Step 4. If
objectives were met, the operator is encouraged to repeat the activities in the following
reporting period. If the objectives were not met, the objectives should be updated, and the
same reconciliation exercise should be repeated in the following period. Reconciliation is an
iterative process, and inability to meet objectives is not seen as a failure: rather, it should
be used to better understand and continuously improve the emissions inventory.

2.7.10 Step 10: Develop Report

While not covered in this Report, Step 10 is to develop a report in line with the GTI Veritas
Protocols. Please refer to the protocols for full report guidance.

2.8 Elements of frequency – site level measurement-based

quantification
An essential question of methane emissions detection and quantification is the frequency of
deployment, or how often quantification connected with reconciliation should be conducted.
There is no single answer to this question, which depends on many factors including
site configuration, local legislation, operational characteristics, expected emissions
patterns or persistency, the type of technology used (for example, facility level vs. source
level), monitoring of other parameters already in place, historical record of successful
reconciliation, and the costs and benefits of technology deployment.

This section covers how to determine the frequency of reconciliation between source level
and site level emissions quantification. This is different from the frequency of LDAR, which
may depend on other factors such as local regulation and the emissions reduction targets
and guidelines of the company. Furthermore, LDAR is a mitigation tool independent of site
level measurement and reconciliation.

Frequency is linked to a cost-benefit assessment which identifies the frequency that allows
the operator to maximize knowledge within acceptable costs. Interviews and literature
review have highlighted several drivers that could impact a decision to change the
frequency of reconciliation, notably:
• History of successful/unsuccessful reconciliation.
• The presence of potential super emitters.
• Degree of understanding of emissions variability during different operational modes
and of factors that could increase variability.
• Advantages from combining detection/quantification technologies.

57
Recommended practices for methane emissions detection and quantification technologies – upstream

2.8.1 History of successful reconciliation

A history of successful reconciliation at a site or group of sites demonstrates an
understanding of the variability of the methane emissions over time. It also indicates that
the quantification methods have a good track record of capturing all emissions at those
sites and that there are likely no continuous, unexpected emissions. In such cases, the
operator may consider reducing the frequency of the reconciliation exercise.

Sites where the operator has only recently begun to investigate methane emissions
reconciliation, or where they have conducted reconciliation exercises with unsatisfactory
results, may require more frequent reconciliation for the operator to better understand
those sites’ emission sources and their variability.

2.8.2 Potential super emitters

One of the main challenges of any methane inventory is to understand the “fat-tail”
distribution, i.e., the presence, likelihood, and time variability of “super emitters” or
super-emitting events. Some equipment, processes, and operational practices have been
documented in peer-reviewed literature to be important potential sources of super emitters
or generators of super-emitting events.27,28, 29, 30 These notably include, but are not limited to:
• Gas flares.
• Un-stabilized condensate or crude oil storage tanks.
• Upset/malfunctioning process conditions.

If equipment, processes, or operational practices that are likely to lead to super-emitting

events, or to become super emitters, are present, more frequent reconciliation may be
needed. Operators may also consider direct or permanent monitoring to identify the
potential cause of a super emitter. For example, the operator could continuously monitor
the flare ignition to ensure that unlit-flare events are tracked and understood, reducing the
need for frequent site level measurements.

2.8.3 Understanding emissions variability

The time variability of emissions can be impacted by variations in operating modes,
including but not limited to:
• Full operation.
• Partial operation.
• Starts and stops.

Clear understanding of how emissions vary over the different operating modes of a site
or equipment reduces uncertainty related to the reconciliation exercise. The link between
emissions and operating mode can provide useful input on when to perform reconciliation
that most effectively covers the different operating modes. It can also help the operator
demonstrate a good understanding of the time variability of emissions at the site in

27
Zavala-Araiza D, et al., 2017
28
Brandt A, Heath, G. A., Cooley, D., 2016
29
Tyner D, and Johnson M., 2021
30
Cusworth D, et al., 2021

58
Recommended practices for methane emissions detection and quantification technologies – upstream

question. This in turn helps justify performing a reconciliation exercise less often at that
site than at a facility where such an understanding is more limited.

Some factors can increase the variability of emissions within an operating mode, increasing
the range of conditions that reconciliation should cover. This could lead to more frequent
reconciliation to ensure that such factors are properly captured. Such factors include, but
are not limited to:
• Seasonal/climatic variations impacting processes.
• Variability of key processes, such as, variability of the load of the turbines, number of
compressors in operation, volume of flared gas.
• Non-continuous processes, such as loading and unloading.
• Processes with operating pressure close to or above design pressure.
• History of incidents, malfunctions, or other super-emitting events.

2.8.4 Combining different technologies31

Factors linked to detection and quantification technologies can impact the frequency
of reconciliation. For example, some technologies are better suited to detecting and
quantifying smaller emission sources, while others are better at picking up larger ones.
Combining different source level detection and quantification technologies can help gain
a fuller picture of methane emissions throughout the site. This is important, as site level
quantification is intended to capture methane emissions from all sources. If a section of the
emissions distribution curve is not captured, it could lead to unsuccessful reconciliation.

A history of successful reconciliation is an important criterion to consider when

determining the frequency of reconciliation. Combining technologies can increase the
chance of successful reconciliation and therefore reduce the frequency of this exercise.

When source level quantification technologies are combined with continuous measurement
systems, the continuous measurement systems help identify intermittent emissions events
while source level quantification helps identify the expected sources of such events. This
can be done alongside operational data collection, which can indicate the source of the
high-emitting events. If no such combination is in place, it may be required to perform the
reconciliation exercise more often to successfully capture those events in the analysis.

One source of added complexity is combining measurements from different technologies,

or repeating measurements consecutively. Recent peer-reviewed research concluded that
the variability of emissions from multiple site level measurements do not always agree with
one another32. Therefore, aerodynamic effects, site layout, and unconsidered algorithm or
model uncertainties may contribute to additional challenges in assessing the results of the
site level measurements.

31
Reference Section 3 below for examples of technologies combinations
32
Brown et al., 2023

59
Recommended practices for methane emissions detection and quantification technologies – upstream

2.8.5 Large-scale reconciliation

Finally, reconciliation can be performed for a single site or a group of sites. Conducting the
analysis for a group of sites can reduce the influence of operating modes and of variability
within operating modes. This is because taking site level measurements of a large number
of sites is likely to capture a broad range of operating modes, as well as a broad range of
factors impacting emissions within different modes, such as variability of key processes
and non-continuous processes. However, capturing the variety of operating modes this way
will lead to limited insights at the site level, making identification of mitigation options for
processes or pieces of equipment at specific sites more difficult.

Not all problems related to time variability can be reduced through this process. Factors
such as climatic variations that impact processes, as well as processes operating at
pressures close to or above design levels, will typically not be reflected using this approach.

2.8.6 Overview of factors influencing frequency of reconciliation

The following table summarizes qualitative factors influencing the frequency of
reconciliation.

Table 1 - Qualitative factors influencing site level quantification and reconciliation.

Factors that could justify a lower frequency: Factors that could justify a higher frequency:

There is a long history of successful reconciliation. There is not a long history of successful reconciliation.

No equipment, processes, or operational practices are There are equipment, processes, or operational practices
likely to become super emitters or to generate super- likely to become super emitters or to generate super-
emitting events. emitting events.

The operator has a good understanding of how emissions The operator has limited understanding of how emissions
vary across the different operating modes. vary across the different operating modes.

The operator has a good understanding of factors The operator has limited understanding of factors
impacting the variability of emissions within operating impacting the variability of emissions within operating
modes. modes.

There is a continuous monitoring system of emissions There is no continuous monitoring system of emissions
and/or of key parameters influencing the variability of and/or of key parameters influencing the variability of
emissions. emissions.

Reconciliation is performed for a group of sites. Reconciliation is performed for a single site.

60
Recommended practices for methane emissions detection and quantification technologies – upstream

3. Technology combinations

Most methane emissions detection and quantification technologies are not well suited for
every type of emission source, size, or deployment purpose. Combinations are often used to
cover some shortcomings. No “one-size-fits-all” combination has been identified since the
needs of operators and the conditions at facilities vary.

For example, a source with variable emissions may require more frequent measurements
to properly characterize its emissions. It may require different characteristics from those
required for monitoring sources with continuous emissions. Other factors impacting
selection include expected emission patterns, which can inform the proper capture of the
fat-tail emissions distribution.33,34,35 Finally, the expected quantity of methane emissions
from a source can affect the selection of an appropriate detection threshold.

This section presents several examples of combining technologies for methane emissions
detection and quantification, showing the selection process, criteria, and challenges
experienced by operators. These examples are not intended to serve as guidelines or
recommendations, but to share experiences between operators regarding specific situations.

3.1 Example 1 – Framework for combination of numerous

technologies
A peer-reviewed article36 published in 2022 presents a quantification, monitoring, reporting,
and verification framework that uses periodic monitoring (such as satellites, aircraft-based
measurements, or drones) along with continuous monitoring of emissions to reconcile
measurements with inventory estimates. The framework also considers intermittent
emissions, which may have high intra-day variability.37 A quantification, monitoring,
reporting, and verification program with up to two phases is outlined:
• The first phase uses monthly, systematic detection surveys (which consider all
emission sources, not just leaks) or periodic, aerial- or drone-based measurements.
The addition of audio, visual, and olfactometry surveys, United States Environmental
Protection Agency (US EPA) Method 21 or other techniques are used to capture
intermittent emissions.
• Over time, a second phase is launched using continuous monitoring solutions to
capture intermittent or short-duration events, providing real-time verification of large
events that may be missed by periodic monitoring.

The article demonstrates how continuous monitoring, paired with an increased

understanding of site level events, are key to an accurate accounting of short-duration,
intermittent, and high-volume events that are often missed in periodic surveys and to
annualize these measurements.
33
Frankenberg, C., et al., 2016
34
Irakulis-Loitxate I et al., 2021
35
Zavala-Araiza D, et al., 2017
36
Jiayang Lyra Wang, et al., 2022
37
Stokes S, et al., 2022

61
Recommended practices for methane emissions detection and quantification technologies – upstream

3.2 Example 2 – Framework for combination of numerous

technologies
A recent peer-reviewed article38 presents an 11-month methane measurement campaign
at oil and gas production sites to improve conventional, bottom-up inventories by
incorporating aerial-based site level measurements and continuous monitoring. Basin and
operator-level, aerial-based, top-down measurements show lower methane emissions
at end-of-project than during the baseline nine months earlier. This is potentially due to
temporal variability, or emission reduction activities and monthly LDAR. The paper presents
a case study investigating a 94% difference in bottom-up vs. top-down measurements at
an unspecified production site. Top-down estimates were 1.8 times higher than an average
emission rate estimate using continuous monitoring, suggesting temporal variability of
measurements contributes to the estimate discrepancy. Further analysis showed the
bottom-up inventory overestimated emissions for five months and underestimated them
in the final two months. This was attributed to a gas processing unit swap, which matched
with observed emission rates from continuous monitoring systems. The study also
highlights the importance of record keeping of one-time events, maintenance, or upsets to
help interpret continuous monitoring data when performing reconciliation with site level
measurements.

The case serves as proof of concept to use continuous monitoring solutions to assess
validity of periodic, top-down measurements and determine their relation to the temporal
emission profile of a given site.

3.3 Example 3 – Simulations of technology combinations

A recent peer-reviewed article39 shows the benefits of combining satellite, aerial, and
continuous monitoring with OGI in a tiered approach, compared to OGI inspections alone.

The paper simulated combinations of methane detection technologies for facilities

representative of the Permian Basin, where extensive datasets are available. Emission
distributions in this region follow highly skewed emission rates with many high emitters
and emissions spanning six to eight orders of magnitude. These datasets may not be
representative of distributions and measurement capabilities in all regions.

Results found that combinations of technologies achieve larger reductions than

single technologies. For example, a combination of satellites with daily surveys, aerial
technologies with surveys at intervals of months, and OGI done once a year reached higher
reductions than quarterly and monthly OGI inspections, and more than only aerial surveys
at intervals of months plus OGI once a year. The application of continuous monitoring at
priority sites with high potential to emit (sites with tanks and flares) reduced the time large
leaks were emitting and achieved higher reductions than monthly OGI inspections alone.

Quick identification and repair of high emitters while maintaining periodic inspections of
smaller leaks by combining OGI, continuous monitoring, aerial, and satellite inspections
can achieve much higher reductions than quarterly or monthly OGI inspections alone.

38
Daniels W, et al., 2023
39
Cardoso-Saldaña, F. J, 2023

62
Recommended practices for methane emissions detection and quantification technologies – upstream

The paper focused on detection (LDAR), but these combinations could also be applied for
quantification. Frequent surveying of super-emitters can help reduce the contribution of
emissions from super-emitters to annual emissions, by either detecting and constraining
event durations, or by monitoring and confirming the lack of emissions.

3.4 Example 4 - Aerial measurement combined with OGI and

permanent sensors
One operator reported that it previously used only OGI for methane emissions detection. It
then added aerial detection methods to identify larger leaks and prioritize follow-up. While
the aerial method was useful for site/facility-level monitoring, the operator followed up
almost all aerial detections using OGI to attribute emission sources to specific equipment
or components. The operator noted that incorporating both aerial measurement and OGI
saved time while also improving safety. Based on the prioritization established by the
aerial surveys, the operator was able to consider the best place for continuous monitoring
solutions. Although not able to detect every methane release, the addition of low-cost
sensors proved useful for finding large leaks quickly.

3.5 Example 5 – Aerial measurements combined with permanent

sensors
One operator considered the use of several methane emissions detection and quantification
technologies, not only for compliance, but also to reduce emissions and costs. The
operator hired an aircraft-based technology to perform measurements that would confirm
compliance. Unfortunately, they found that measured emissions were many times greater
than expected. OGI was used to follow up identified sources, though in many cases, the
aircraft-based technology was able to sufficiently identify the source so that OGI follow-up
was unnecessary. The addition of aircraft-based measurements helped reduce methane
intensity by 75%.

The company also deployed several continuous monitoring technologies to cover sites
or pieces of equipment with great potential for emissions, in particular, tank batteries
and wellheads. Due to the increased frequency of monitoring and measurements,
the operator began relying increasingly on measurement-based quantification using
continuous monitoring technologies rather than periodic monitoring with aircraft-based
measurements.

When performing reconciliation between the two measurement methods, successful

reconciliation was found to be sensitive to wind conditions. The operator reported false
negatives if continuous monitoring solutions were set up too close to emission sources,
which could also affect successful reconciliation.

63
Recommended practices for methane emissions detection and quantification technologies – upstream

3.6 Example 6 – Site level quantification combined with OGI

One operator reported using a “layered” approach to technology deployment. It first
deployed site level measurement technology, such as continuous monitors, drones, aircraft,
or satellites. The operator selected the site level quantification approach based on the
level of emissions expected and geographical considerations, including availability of the
technology. However, the operator reported that none of the technologies it tested were
found to be suitable for quantification. Insights from site level measurements were used to
perform follow-ups with OGI cameras to identify the emission sources.

Depending on the quality of the site level measurements, the operator reported difficulties
reconciling emissions between bottom-up inventories and site level quantification.
Different results in repeated measurements made reconciliation challenging, particularly
when technology performance did not match the providers’ specifications. Site level
measurements were useful, particularly when including imagery to assist with source
attribution. However, they could also be misleading when the performance characteristics
of the site level measurement were not properly documented and communicated, or if
source attribution was inaccurate, as it would affect follow-up and prioritization of leak
repair or mitigation. It seems that proper selection of site level measurement technology
and consideration of all associated parameters is critical to successfully combining
technologies.

64
Recommended practices for methane emissions detection and quantification technologies – upstream

4. Other recommendations and

overarching elements
The following cross-cutting elements are covered in this section:
• Uncertainty.
• Data management and security.
• Internal practices and processes that are independent of the technology provider.
• Lack of independent standards for technologies.
• Interpretation of test results.

4.1 Understanding uncertainty

Uncertainty of methane measurements is a complex topic, and methods to calculate these
uncertainties are not fully resolved and implementable with a widely agreed upon method.
Therefore, many open research questions remain in this area. This section attempts
to identify common sources of uncertainty, and documents one potential pathway for
navigating uncertainty as part of this framework. However, other methods exist and will
continue to be developed.

It is important to distinguish different types of uncertainties related to methane emissions

detection and quantification, such as:
• Sensor uncertainty.
• Methodology uncertainty.
• Methodology uncertainty for a given measurement at a given time.
• Uncertainty related to aggregated emissions.

Sensor uncertainty refers to the accuracy of the measurement compared to the true
concentration of methane in the air. Sensor uncertainty is often called precision error.
Uncertainty related to the sensor can be much smaller than uncertainty related to the
method for quantifying the methane emission rate.

Total uncertainty related to the methodology consists of uncertainties in the measurement

by the sensor (sensor uncertainty) and how the results are used to quantify the emission
rate. Methodologies may use assumptions to quantify the emission rate, which can also
introduce uncertainty into the measurement. For example, using wind data from regional
meteorological stations instead of from the site itself increases uncertainty,40 since such
readings may not reflect true wind conditions at the site.

Uncertainty of a measurement at a given time may be influenced by many factors. These

can include the position of both the technology and the emission source, as well as the
distance between them (greater distances increase uncertainty). Other factors that could
increase this kind of uncertainty include methane emissions from background sources, as
well as environmental factors such as wind conditions, precipitation, and cloudiness.

40
Sherwin E, et al., 2021

65
Recommended practices for methane emissions detection and quantification technologies – upstream

Developing quantitative uncertainty estimates can be challenging for both source and site
level measurements. One may get statistical variance measurements that could be used as
a proxy for uncertainty. However, this depends on knowing all activities/sources of the group
at the time of site-level measurement.

Some technology providers may document uncertainties or errors in quantified methane

emissions on an aggregated basis rather than on a point-by-point measurement basis.
In other cases, technology providers may provide uncertainty estimates from idealized
tests (e.g., from controlled releases) rather than uncertainties that are applicable to the
actual measurements at the facility of interest. Depending on the desired outcome of
quantification and uncertainty, this may be misleading. If actual measurements have a
larger uncertainty or error range than what is quoted by technology providers or from
controlled release tests, this can make it difficult to reconcile the results from source level
and site level measurements. Moreover, higher uncertainties on individual sources can
impact the prioritization of emissions mitigation.

Care should be taken when evaluating technology uncertainties, including what the
uncertainties represent. They may refer to individual or aggregated emissions, provide
different confidence intervals (e.g., 1σ or 2σ), and may refer to relative or absolute
uncertainty. Since different technologies may quantify emission rates using different
methods, documented potential factors that introduce uncertainty should be considered.

The recognition of random variation in intra-estimate variability highlights the importance

of incorporating robust uncertainty models into emissions quantification methodologies.
Controlled release experiments provide a useful baseline for understanding these
uncertainties, but care must be taken to acknowledge their limitations when applied to field
conditions.

4.2 Data management and security

A challenge that may emerge as an increasing number of measurement and detection
campaigns are performed is the management of the data, particularly with an increase in
continuous monitoring.

The collection of data does not, by itself, lead to effective methane management. Operators
need to ensure that the data collected are actionable and can inform the mitigation
strategy or other objective. Some technology providers have started to offer data-analysis
software that translates the data into relevant information. This can help operators better
understand the methane landscape of their facilities and identify where action is required.
It would otherwise be necessary for the operator to deploy internal procedures and systems
to address the volume of data generated by the sensors, especially those involved in
continuous monitoring.

Some data might be considered sensitive for oil and gas operators, whether directly related
to methane emissions or other parameters in the context of methane management. Many
interviewees highlighted that it is important to ensure data security and confidentiality,
particularly if data is to be stored by the technology provider or on their cloud service.

66
Recommended practices for methane emissions detection and quantification technologies – upstream

4.3 Internal practices and processes independent of the provider

Internal practices and processes are needed for safe deployment, ensuring employees
with the correct qualifications are available and trained. This is even more important for
technologies that require access to the site.

Since technologies and detection/quantification needs are evolving, interviewees

have recommended that operators develop internal practices and processes that are
independent of a specific provider, as this allows for a smoother transition between brands
or versions.

4.4 Lack of independent standards for comparing technologies

There are no enforced protocols or standards to test and report performance in a consistent
and comparable format. This is not surprising, since the industry is still relatively new.
Different technologies sense methane differently, quantify methane in different ways,
attribute emissions to specific sources using different formats, and report methane
emissions detection and quantification differently. Facilities like the Methane Emissions
Technology Evaluation Centre (METEC) and the Total Anomaly Detection Initiative (TADI)
testing complex have developed test protocols for all deployment methods excluding
satellites, which are currently beyond testing capabilities. However, there are no
requirements or standards for reporting. Much of the data is anonymized, presented in the
best possible light, or only made public at the discretion of the participating technology
provider. Technology testing may also be performed in a variety of environments (see Section
4.5). The lack of standards makes comparison challenging for companies, both in terms
of selecting technologies and reporting emissions. It is recommended that the industry
develop consistent practices that allow robust and comparable testing of different methane
emissions detection and quantification technologies. This could include, for example, a
unified definition of detection threshold and probability of detection using comparable
metrics, such as probability of detecting emissions of 10 kg/h from 20 m at 3 m/s wind.

4.5 Interpretation of test results

4.5.1 Site layout

Technologies may perform well at a testing site; this does not guarantee similar
performance in all locations and conditions. For example, in a realistic field scenario,
with potentially multiple sources and plumes, background methane emissions can impair
source attribution and increase uncertainty in quantified emission rates. Results may
differ, for example, because the number of sensors used in semi-controlled environments
exceeds the number that will be deployed in the field.

The variation of site layouts means that testing facilities will not be representative of all
field conditions. Results from a test run on a well pad with spread out, discrete emission
sources are likely to differ from an offshore platform with densely packed equipment, for
example.

67
Recommended practices for methane emissions detection and quantification technologies – upstream

Placement of the sensor is important. If measurements are taken too close to the emission
source, the plume may not be properly formed and may not be adequately detected,
negatively impacting quantification algorithms. Deployment far from the source could also
reduce the probability of detection.

While third-party testing sites may not match an operator’s conditions, they do offer a more
rigorous testing and validation process. Conducting controlled-release testing at third-
party sites can demonstrate dedication to improving abilities and transparency. Results
may still vary when deploying the same technology at a site with different characteristics
from the testing site.

4.5.2 Probability of Detection (PoD)

In addition to minimum detection threshold, methane emissions detection and
quantification technologies may include a PoD. This refers to detection sensitivity and is
used to help understand the chances of detecting an emission, considering a number of
factors. For example, a technology may have a stated PoD of 90% for sources emitting 10
kg/h, at a wind speed of 2 m/s, at 50 m from the source. This means that the technology
is expected to detect 9 out of 10 sources in those specific conditions. A PoD provides more
confidence that the technology will be able to detect emissions than the minimum detection
threshold alone. A higher PoD may also be associated with a higher false-positive rate.41
PoD can be determined through partially42 or fully blinded43 testing.

It is important to consider the extent to which conditions and other factors during testing
represent the conditions at the operator’s site. For example, testing may have taken place
in open fields, large, simplified, or sparse sites, or using a large quantity of sensors,
compared to the characteristics of the sites where the operator intends to deploy the
technology. Large discrepancies can lead to significant differences in the probability of
detection.

Future efforts should focus on refining uncertainty and probability of detection models to
better capture the effects of aerodynamic influences, instrument-specific variability, and
algorithmic processing, particularly for complex emission environments and infrastructure.

41
Bell C, et al.,. 2023
42
Qube Technologies,2022 https://highwoodemissions.com/wp-content/uploads/2022/09/2022-08-25_Qube-Probability-of-Detection-
White-Paper.pdf.
43
Johnson M, et al.,2021

68
Recommended practices for methane emissions detection and quantification technologies – upstream

5. Conclusion

Methane emissions detection and quantification is a well-known challenge for oil and gas
operators. An increasing number of technologies are available to tackle this essential
aspect of greenhouse gas emissions inventory and mitigation. The aim of this Report is to
help operators turn the current knowledge into actions at their facilities. The technology
filtering tool and the technology data sheets provide a centralized and standardized
database to help operators select and compare technologies. The decision trees offer
guidance on deployment and data interpretation, depending on objective. There is currently
no “one-size-fits-all” technology available, and a combination of solutions is required for
methane emissions management. Some examples are presented in the Report. Selection
and deployment cannot be fully summarized in a technology filtering tool or in decision
trees. Other overarching elements should be considered by operators, some of which are
detailed in the last sections of this Report.

This Report and its accompanying technology filtering tool, technology data sheets, and
decision trees do not recommend or impose one technology or approach over another. They
have been developed to provide a framework of detailed technology characteristics so that
operators can make informed decisions when selecting and deploying the technology (or
technology combinations) best suited to their circumstances, considering their objectives
and their operating environment. It is hoped that this framework will help operators achieve
their goals relating to upstream methane emissions management and reporting.

69
Recommended practices for methane emissions detection and quantification technologies – upstream

Glossary

Term Definition

Basin Wide geographical area with a collection of sites.

Bottom-up Bottom-up estimates sum up individual emission sources within a facility

to produce a single value. Bottom-up estimates can be synonymous with
source level inventories.

Detection of Process of identification of methane emissions from potential sources,

methane emissions without the measurement of the mass quantity (flow rate, e.g., kg/h). The
detection is typically performed above a threshold, and above ambient
levels.

Detection threshold The minimum [flow rate] of a gas, e.g., methane, which is reliably
detectable by detection equipment. This is sometimes called a Minimum
Detection Limit (MDL).

Equipment A mechanical system where a single process or action takes place.

Examples of equipment: compressor, tank, controller, pump, dehydrator,
separator. A piece of equipment may include different components.

Equipment A part or element of a larger whole. In the context of equipment

Component [emissions], components are individual sealed surfaces on pressurized
equipment such as flanges, valves, connections, pressure relief valves,
open ended lines, etc. This is typically the most granular level of fugitive
emissions reporting.

Equipment Group A collection of equipment located in proximity, often within a delimited

area. Examples of an equipment group: tank battery, group of
compressors, dehydration units.

Measurement The process of taking a reading of a methane emission. Measurement

can be of any variable (volume, concentration, mass, frequency, and so
on) that allows for detection or for an estimate of emission rate.

Measurement An inventory that is predominantly informed by data from methane

Informed Inventory measurements of the assets and sources in the inventory, where
predominantly means methane emissions quantification informed
by measurement can be based on 100% sample size or based on a
statistically representative subset of samples. Note: This definition is
different from the Veritas Protocol.

Quantification Determining an emission rate, such as mass per time or volume per
time. This can be done directly through measurement of the emissions,
or indirectly through estimations, calculations, and modelling.

70
Recommended practices for methane emissions detection and quantification technologies – upstream

Term Definition

Reconciliation Reconciliation is the process of comparing source level inventories with

independent site level measurements to produce emissions estimates
(Oil and Gas Methane Partnership (OGMP) Uncertainty and Reconciliation
guidance document). Other definitions of reconciliation exist, however,
this is the one that is referred to throughout the document.

Screening Evaluations with the main purpose of identifying sources of emissions.

However, in some regulatory contexts, screening applies only to less
rigorous or less sensitive detection approaches, such as AVO (Audio,
Visual, and Olfactory).

Screening typically refers to identifying emissions within a wide area,

while detection typically refers to identifying emissions from specific
sources.

Site Collection of emission sources with some relation to one another

within a delimited geographical area. Emissions from a site combine
emissions from different equipment and components. Examples of sites:
compressor station, offshore production platform. In the body of the
report, site/facility will be referred to as “site” but can be interpreted as
synonymous with “facility”.

Site level Methane measurement applied to a site, without identifying specific

measurement sources at the equipment or component level. A site level measurement
can be synonymous with a top-down estimate.

Source A component within a process or equipment that releases methane to

the atmosphere either intentionally or unintentionally, intermittently, or
persistently.

Source level A record of all known sources of emissions and emission rates. An
inventory inventory provides a summary of emissions over a given period of time. A
source level inventory can consist of measurement-based quantification,
engineering calculations, or emission factors. Total emissions are
calculated by summing data from each emission source. Source level
inventory can be synonymous with bottom-up estimate.

Super emitter Methane emission source that represents a disproportionate amount of

the total methane emissions released from all sources44.

Top-down Top-down estimates measure methane at a facility level that may

combine multiple emission sources, without being able to resolve them to
specific sources. Top-down estimates can be synonymous with site level
measurements.

44
Ipieca-IOGP-GIE-Marcogaz Methane Emissions Glossary Methane Emissions Glossary

71
Recommended practices for methane emissions detection and quantification technologies – upstream

List of Acronyms

Acronym Meaning

ATEX Explosive Atmosphere

EED Expected Emissions Distribution

EF Emission Factors

EPA Environmental Protection Agency

GTI Gas Technology Institute

LDAR Leak Detection and Repair

MDL Minimum Detection Limit

MII Measurement Informed Inventory

OGI Optical Gas Imaging

OGMP Oil and Gas Methane Partnership

PoD Probability of Detection

SCADA Supervisory Control and Data Acquisition

72
Recommended practices for methane emissions detection and quantification technologies – upstream

Appendix A:
Methodology and data sources
The review of technologies performed as part of the recommended practices for methane
emissions detection and quantification technologies relied on data sources with varying
levels of independent validation. Data sources include (from most to least independent):
• peer-reviewed academic literature
• public datasets
• interviews with operators and service providers
• interviews with technology providers

Results were included in the technology data sheets and technology filtering tool.45 In each
case, the type of source is clearly identified.

For the development of the decision trees, data sources such as methodologies presented
in international framework or protocols (such as OGMP 2.0 and GTI Veritas Protocols) were
considered, together with the project team’s extensive experience in methane emissions
from the oil and gas sector, supplemented by input from operators and academic
researchers. All decision trees were critically reviewed by the IOGP working group, whose
comments were incorporated.

A.1 Literature review

A review of the literature was performed to collect information on the performance of
detection and measurement technologies. Technology providers were asked to share case
studies, company reports, and academic studies featuring their technology.

Academic papers covered independent comparisons of the performance of different

technologies, often through semi or fully blind testing and controlled releases, in varied
geographical locations. Technologies in these studies were typically segregated by type,
such as handheld, drone, and aerial since the size and level of emissions identified by each
type tend to differ. While results from studies often use different types of indicators, making
direct comparisons difficult, the literature review informed an understanding of methane
emissions detection and quantification practices, best-available knowledge, successful
implementation, and related challenges, at the time of the Report.

Over 60 independent peer reviewed academic papers were reviewed. The complete list is
available in Appendix C.

45
https://www.iogp.org/workstreams/environment/environment/methane-emissions-detection-and-quantification/methane-detection-
and-quantification-technology-filtering-tool/tool/

73
Recommended practices for methane emissions detection and quantification technologies – upstream

A.2 Interviews with technology providers

For all relevant technologies, providers were contacted for an interview to review publicly
available data collected from their websites, industry testing, reports, and peer-reviewed
literature, to confirm the technologies’ capabilities. The interviews were semi-structured
discussions following a template to ensure consistent collection of data from each provider.
Interviews were conducted with over 30 technology providers, and multiple interviews for a
single provider were carried out where required.

Some providers did not reply to requests for an interview, despite multiple attempts.
Technologies whose providers were unable to be interviewed were not included in the
analysis.

A.3 Interviews with service providers and operators

Interviews were conducted with service providers and operators, that is, technology
users. These interviews were used to supplement information from technology providers
to present a holistic picture of deployment in practice, and to identify advantages and
limitations in diverse operating conditions. Interviews were conducted as open discussions,
and the results incorporated into the technology data sheets and the main body of this
Report.

Operators were also contacted for interviews and to share case studies and results from
independent or internal benchmark testing. In total, 13 interviews with operators and
service providers were conducted.

74
Recommended practices for methane emissions detection and quantification technologies – upstream

Appendix B:
List of technologies assessed
Table B1 - List of CH4 technologies assessed

Technology name Technology provider Datasheet up to date as of:

GFM 2.0 AddGlobe January 2023
Charm Adlares January 2023
PRISMA ASI January 2023
D-fenceline Atmosfir January 2023
Gas Mapping Lidar (GML) Bridger Photonics January 2023
Carbon Mapper Carbon Mapper - Planet January 2023
MetCam CI Systems December 2024
Autonomous 365 Clean Connect January 2023
Worldview3 DigitalGlobe January 2023
Sentinel-2 ESA January 2023
TROPOMI ESA January 2023
XPLOROBOT Laser OGI Exploration Robotics December 2024
Technologies
Fixed Wing Drone Flylogix December 2024
GHGSat Constellation GHGSat December 2024
Remote Methane Leak Detector Heath Consultants December 2024
(RMLD-CS)
Detecto-Pak Infrared+
(DP-IR+) Heath Consultants December 2024
DISCOVER Advanced Mobile Leak Heath Consultants December 2024
Detection (AMLD)
HETEK Flow Sampler HETEK Solutions Inc. December 2024
Leaks Surveyor Insight M December 2024
Kuva Daylight Kuva January 2023
Longpath Laser System Longpath Technologies December 2024
SPOT Robotic Dog MFE Instruments January 2023
ORION Mirico January 2023
Landsat-8 NASA/USGS January 2023
UAS Drone Net Zero Aerial December 2024

75
Recommended practices for methane emissions detection and quantification technologies – upstream

Technology name Technology provider Datasheet up to date as of:

MPS Methane Gas Sensor NevadaNano December 2024
EyeCgas 24/7 Opgal January 2023
Hyperspectral monitoring solutions Orbital Sidekick January 2023
ALMA Pergam-Suisse December 2024
LMS (Laser Methane Scanner) Pergam-Suisse December 2024
SELMA Duo Pergam-Suisse December 2024
SELMA Roof-Dome Pergam-Suisse December 2024
Laser Falcon Pergam-Suisse December 2024
G4301 Gas Concentration Analyser Picarro December 2024
Canary X Project Canary December 2024
Mantis Flare Monitor Providence Photonics January 2023
QL320 Providence Photonics January 2023
Axon Qube Technologies December 2024
SOOFIE Scientific Aviation January 2023
Scientific Aviation Manned Aircraft Scientific Aviation January 2023
DJI Matrice Scientific Aviation January 2023
Multi rotor drone SeekOps January 2023
Fixed wing drone SeekOps January 2023
Agni Sensia January 2023
Mileva 33 Sensia January 2023
Mileva 33F Sensia December 2024
NuboSphere Sensirion December 2024
Hi-Flow 2 Sensors Inc December 2024
Ventus OGI Sierra Olympia December 2024
LWIR OGI Camera Sierra Olympia December 2024
G300a Teledyne FLIR January 2023
GF77 Teledyne FLIR January 2023
GF77a Teledyne FLIR January 2023
GFX 320 + QL 320 Teledyne FLIR January 2023
PoMELO UCalgary December 2024

76
Recommended practices for methane emissions detection and quantification technologies – upstream

Appendix C:
Selected peer-reviewed articles
Alden C, et al. “Bootstrap inversion technique for atmospheric trace gas source detection and
quantification using long open-path laser measurements”. Atmospheric Measurement Techniques
11:3. 2018. p. 1565–1582.

Alden C, et al. “Temporal Variability of Emissions Revealed by Continuous, Long-Term Monitoring of

an Underground Natural Gas Storage Facility”. Environmental Science & Technology 54:22. 2020. p.
14589-14597.

Alden C, et al. “Single-Blind Quantification of Natural Gas Leaks from 1 km Distance Using
Frequency Combs”. Environmental Science & Technology 53:5. 2019. p.2908-2917.

Allen D, et al. “Measurements of methane emissions at natural gas production sites in the United
States”. Proceedings of the National Academy of Sciences of the United States of America 110:44. 2013.
p.17768–17773.

Ayasse A, et al. “Methane remote sensing and emission quantification of offshore shallow water oil
and gas platforms in the Gulf of Mexico”. Environmental Research Letters 17:8. 2022.

Bell C, et al. “Single-blind determination of methane detection limits and quantification accuracy
using aircraft-based LiDAR”. Elementa 10:1. 2022.

Bell C, et al. “Performance of continuous emission monitoring solutions under single-blind

controlled testing protocol.” Environmental Science & Technology 57:14. 2023. p. 5794-5805.

Bell C, Vaughn T. L., & Zimmerle D. J. “Evaluation of next generation emission measurement
technologies under repeatable test protocols”. Elem Sci Anth. 8:32. 2020. p. 32.

Brandt A, Heath, G. A., & Cooley, D. “Methane Leaks from Natural Gas Systems Follow Extreme
Distributions”. Environmental Science and Technology 50:22. 2016. p. 12512–12520.

Brantley H, et al. “Assessment of Methane Emissions from Oil and Gas Production Pads using
Mobile Measurements”. Environmental Science & Technology 48:24. 2014. p. 14508–14515.

Brown et al. “Informing Methane Emissions Inventories Using Facility Aerial Measurements at
Midstream Natural Gas Facilities”. Environmental Science & Technology 57:39. 2023. p 14493-14786.

Cardoso-Saldaña F. J. “Tiered Leak Detection and Repair Programs at Simulated Oil and Gas
Production Facilities: Increasing Emission Reduction by Targeting High-Emitting Sources”.
Environmental Science & Technology 57:19. 2023. p. 7382-7390.

Chen Q, et al. “Assessing detection efficiencies for continuous methane emission monitoring
systems at oil and gas production sites”. Environmental Science & Technology 57:4. 2023. p. 1788-
1796.

Chen Y, et al. “Quantifying Regional Methane Emissions in the New Mexico Permian Basin with a
Comprehensive Aerial Survey”. Environmental Science & Technology 56:7. 2022. p. 4317–4323.

Coburn S, et al. “Regional trace-gas source attribution using a field-deployed dual frequency comb
spectrometer”. Optica 5:4. 2022. p. 320.

77
Recommended practices for methane emissions detection and quantification technologies – upstream

Cogliati S, et al. “The PRISMA imaging spectroscopy mission: overview and first performance
analysis”. Remote Sensing of Environment 262. 2021. p. 112499.

Conrad B, Tyner D, and Johnson M. “Robust Probabilities of Detection and Quantification

Uncertainty for Aerial Methane Detection: Examples for Three Airborne Technologies”. Remote
Sensing of Environment 288. 2023. p. 113499.

Cusworth D, et al. “Intermittency of Large Methane Emitters in the Permian Basin.” Environmental
Science and Technology Letters 8:7. 2021. p. 567–573.

Cusworth D, et al. “Strong methane point sources contribute a disproportionate fraction of total
emissions across multiple basins in the U.S.” Preprint submitted to EarthArXiv. 2022.

Daghestani N, Brownsword R, and Weidmann D. “Analysis and demonstration of atmospheric

methane monitoring by mid-infrared open-path chirped laser dispersion spectroscopy”. Optics
Express 22:7. 2014. p. 1731.

Daniels W, et al. “Towards multi-scale measurement-informed methane inventories: reconciling

bottom-up inventories with top-down measurements using continuous monitoring systems.”
ChemRxiv. Cambridge: Cambridge Open Engage. 2023. This content is a preprint and has not been
peer-reviewed.

Duren R, et al. “California’s methane super-emitters”. Nature 575:7781. 2019. p. 180–184.

Erland B, et al. “Comparing Airborne Algorithms for Greenhouse Gas Flux Measurements over the
Alberta Oil Sands”. Atmospheric Measurement Techniques 15:19. 2022. p. 5841-5859.

Feingersh T and Dor E. “SHALOM - A Commercial Hyperspectral Space Mission. In S.-E. Qian (Ed.),
Optical Payloads for Space Missions” in Optical Payloads for Space Missions, S. Qi (ed). John Wiley &
Sons, Ltd, 2015. p. 247-263.

Foulds A, et al. “Quantification and assessment of methane emissions from offshore oil and gas
facilities on the Norwegian continental shelf”. Atmospheric Chemistry and Physics 22:7. 2022. p
4303–4322.

France J, et al. “Facility level measurement of offshore oil and gas installations from a medium-
sized airborne platform: Method development for quantification and source identification of
methane emissions”. Atmospheric Measurement Techniques 14:1. 2021. p. 71–88.

Frankenberg C., et al. “Airborne methane remote measurements reveal heavy-tail flux distribution
in Four Corners region.” Proceedings of the National Academy of Sciences of the United States of
America 113:35. 2016. p. 9734–9739.

Guanter L, et al. “Mapping methane point emissions with the PRISMA spaceborne imaging
spectrometer”. Remote Sensing of Environment 265. 2021. p. 112671.

Guha A, et al. “Assessment of Regional Methane Emission Inventories through Airborne

Quantification in the San Francisco Bay Area”. Environmental Science & Technology 54:15. 2020. p.
9254–9264.

Higgins S et al. “A Practical Framework for Oil and Gas Operators to Estimate Methane Emission
Duration Using Operational Data”. SPE J. 29:05. 2024. p 2763–2771. https://doi.org/10.2118/219445-PA

78
Recommended practices for methane emissions detection and quantification technologies – upstream

Innocenti et al. “Comparative Assessment of Methane Emissions from Onshore LNG Facilities
Measured Using Differential Absorption Lidar”. Environmental Science & Technology 57:8. 2023. p.
3301-3310.

Irakulis-Loitxate I et al. “Satellite-based survey of extreme methane emissions in the Permian

basin”. Science Advances 7:27. 2021. https://doi.org/10.1126/sciadv.abf4507

Irakulis-Loitxate I, et al. “Satellites Detect Abatable Super-Emissions in One of the World’s Largest
Methane Hotspot Regions”. Environmental Science and Technology 56:4. 2022. p. 2143–2152. https://
doi.org/10.1021/acs.est.1c04873

Jacob D, et al. “Quantifying methane emissions from the global scale down to point sources using
satellite observations of atmospheric methane”. Atmospheric Chemistry and Physics 22:14. 2022. p.
9617–9646. https://doi.org/10.5194/acp-22-9617-2022

Jiayang Lyra Wang, et al. “Multiscale Methane Measurements at Oil and Gas Facilities Reveal
Necessary Frameworks for Improved Emissions Accounting.” Environmental Science and Technology
56:20. 2022. p. 14743-14752.

Johnson D, Covington, A, and Clark N. “Methane Emissions from Leak and Loss Audits of Natural
Gas Compressor Stations and Storage Facilities”. Environmental Science and Technology 49:13. 2015.
p. 8132–8138.

Johnson M, Tyner D, and Szekeres A. “Blinded evaluation of airborne methane source detection
using Bridger Photonics LiDAR”. Remote Sensing of Environment 259. 2021. p.112418.

Littlefield J, et al. “Synthesis of recent ground-level methane emission measurements from the U.S.
natural gas supply chain”. Journal of Cleaner Production 148. 2017. p.118–126.

Morales R, et al. “Controlled-release experiment to investigate uncertainties in UAV-based emission

quantification for methane point sources”. Atmospheric Measurement Techniques 15:7. 2022. p.
2177–2198.

Omara M et al. “Methane emissions from US low production oil and natural gas well sites”. Nature
Communications 13:1. 2022. p. 2085.

Plant G, et al. “Inefficient and unlit natural gas flares both emit large quantities of methane”.
Science 377:6614. 2022. p. 1566–1571. https://doi.org/10.1126/science.abq0385

Ravikumar A, et al. “Single-blind inter-comparison of methane detection technologies – results

from the Stanford/EDF Mobile Monitoring Challenge”. Elem Sci Anth. 7:37. 2019.

Ravikumar A, et al. “Are Optical Gas Imaging Technologies Effective for Methane Leak Detection?”
Environmental Science and Technology 51:1. 2017. p. 718–724.

Ravikumar A, et al. “’Good versus Good Enough?’ Empirical Tests of Methane Leak Detection
Sensitivity of a Commercial Infrared Camera.” Environmental Science & Technology 52:4. 2018. p.
2368–2374.

Riddick S, et al. “A cautionary report of calculating methane emissions using low-cost fence-line
sensors”. Elementa 10:1. 2022.

Rieker G, et al. “Frequency-comb-based remote sensing of greenhouse gases over kilometer air
paths”. Optica 1:5. 2014. p. 290.

79
Recommended practices for methane emissions detection and quantification technologies – upstream

Robertson A, et al. “New Mexico Permian basin measured well pad methane emissions are a factor
of 5−9 times higher than U.S. EPA estimates”. Environmental Science and Technology 54:21. 2020. p.
13926–13934.

Schwietzke S, et al. “Aerially guided leak detection and repair: A pilot field study for evaluating
the potential of methane emission detection and cost-effectiveness”. Journal of the Air and Waste
Management Association 69:1. 2019. p. 71–88.

Shen L, et al. “Satellite quantification of oil and natural gas methane emissions in the US and
Canada including contributions from individual basins”. Atmospheric Chemistry and Physics 22:17.
2022. p. 11203–11215.

Sherwin E, et al. “Single-blind test of airplane-based hyperspectral methane detection via

controlled releases”. Elementa: Science of the Anthropocene 9:1. 2021.

Sherwin E, et al. “Single-blind validation of space-based point-source methane emissions detection

and quantification”. Sci. Rep 13:3836. 2023.

Singh D, et al. “Field Performance of New Methane Detection Technologies: Results from the
Alberta Methane Field Challenge.” Non-peer reviewed pre-print submitted to EarthArXiv. 2019.

Stokes S, et al. “An aerial field trial of methane detection technologies at oil and gas production
sites”. Non-peer reviewed pre-print submitted to ChemRxiv. 2022.

Stokes S, et al. “Reconciling Multiple Methane Detection and Quantification Systems at Oil and Gas
Tank Battery Sites”. Environmental Science & Technology 56:22. 2022. p. 16055-16061.

Sun S, Ma L, and Li Z. “Methane emission estimation of oil and gas sector: A review of
measurement technologies, data analysis methods and uncertainty estimation”. Sustainability 13:24.
2021. p. 13895.

Thorpe M, et al. “Gas mapping LiDAR for large-area leak detection and emissions monitoring
applications”. 2017 Conference on Lasers and Electro-Optics. San Jose, USA. 14-19 May 2017.

Tratt D, et al. “Airborne visualization and quantification of discrete methane sources in the
environment”. Remote Sensing of Environment 154:1. 2014. p. 74–88.

Tullos E, et al. “Use of Short Duration Measurements to Estimate Methane Emissions at Oil and Gas
Production Sites”. Environmental Science & Technology Letters 8:6. 2021. p. 463-467.

Tyner D, and Johnson M. “Where the Methane Is—Insights from Novel Airborne LiDAR
Measurements Combined with Ground Survey Data”. Environmental Science & Technology 55:14.
2021. p. 9773–9783.

Varon D, et al. “Quantifying methane point sources from fine-scale satellite observations of
atmospheric methane plumes”. Atmospheric Measurement Techniques 11:10. 2018. p. 5673–5686.

Vaughn T, et al. “Temporal variability largely explains top-down/bottom-up difference in methane

emission estimates from a natural gas production region”. Proceedings of the National Academy of
Sciences 115:46. 2018. p. 11712-11717.

Wang J, et al. “Multiscale Methane Measurements at Oil and Gas Facilities Reveal Necessary
Frameworks for Improved Emissions Accounting”. Environmental Science & Technology 56:20. 2022.
p. 14743–14752.

80
Recommended practices for methane emissions detection and quantification technologies – upstream

Waxman E, et al. “Intercomparison of Open-Path Trace Gas Measurements with Two Dual
Frequency Comb Spectrometers”. Atmos Meas Tech 10:9. 2017. p. 3295-3311.

Yacovitch T, Daube C, and Herndon S. “Methane Emissions from Offshore Oil and Gas Platforms in
the Gulf of Mexico”. Environmental Science and Technology 54:6. 2020. p. 3530–3538.

Zang K, Zhang G, and Wang J. “Methane emissions from oil and gas platforms in the Bohai Sea,
China”. Environmental Pollution 263. 2020. p. 114486.

Zavala-Araiza D, et al. “Super-emitters in natural gas infrastructure are caused by abnormal

process conditions”. Nature Communications 8:14012. 2017.

Zavala-Araiza D, et al. “Reconciling divergent estimates of oil and gas methane emissions”.
Proceedings of the National Academy of Sciences 112:51. 2015. p. 15597–15602.

Zavala-Araiza D, et al. “Toward a Functional Definition of Methane Super-Emitters: Application to

Natural Gas Production Sites”. Environmental Science and Technology 49:13. 2015. p. 8167–8174.

Zavala-Araiza D, et al. “A tale of two regions: Methane emissions from oil and gas production in
offshore/onshore Mexico”. Environmental Research Letters 16:2. 2021. https://doi.org/10.1088/1748-
9326/abceeb

Zeng Y, et al. “Methods to determine response factors for infrared gas imagers used as quantitative
measurement devices”. Journal of the Air and Waste Management Association 67:11. 2017. p.1180–
1191.

Zhang Y, et al. “Quantifying methane emissions from the largest oil-producing basin in the United
States from space”. Science Advances 6:17. 2020. p. 1–10.

Zimmerle, D, et al. “Detection Limits of Optical Gas Imaging for Natural Gas Leak Detection in
Realistic Controlled Conditions”. Environmental Science & Technology 54:18. 2020. p.11506–11514.

81
 This Report provides oil and gas
operators with a framework and
guidelines to help select and deploy
methane emissions detection and
quantification technologies that are
tailored to their sites and objectives.
It is accompanied by an online
technology filtering tool, detailed
technology data sheets covering
over fifty technologies, and decision
trees to guide deployment.

IOGP Headquarters IOGP Europe www.iogp.org

City Tower, 40 Basinghall Street, London EC2V 5DE, United Kingdom T: +32 2 882 16 53
T: +44 (0)20 4570 6879 E: reception-europe@iogp.org
E: reception@iogp.org

Ipieca www.ipieca.org
City Tower, 40 Basinghall St, London EC2V 5DE, United Kingdom
T: +44 (0)20 7633 2388
E: info@ipieca.org

OGCI www.ogci.com
25 Argyll Street, London W1F 7TS, United Kingdom
T: +44 (0)20 3922 0853

Energy Institute www.energyinst.org

61 New Cavendish Street, London W1G 7AR, United Kingdom
T: +44 (0)20 7467 7100