UK ETS Guidance Note:

Uncertainty Assessments for Installations

June 2022
Note

This document is intended to provide guidance for operators of installations. If there is

any inconsistency between the guidance and legislation, the legislation prevails.

© Crown copyright 20[XX]

This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated.
To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the
Information Policy Team, The National Archives, Kew, London TW9 4DU, or email:
[email protected].

Where we have identified any third-party copyright information you will need to obtain permission from the
copyright holders concerned.

Any enquiries regarding this publication should be sent to us at:

[email protected]
Contents
Overview _________________________________________________________________ 5
1 What is uncertainty? _____________________________________________________ 6
1.1 Definitions __________________________________________________________ 6
1.2 Uncertainty in the Monitoring and Reporting Regulation _______________________ 8
1.2.1 Simplifications for low emitting installations, including hospital or small emitters _ 9
2 Uncertainty for calculation-based approaches ________________________________ 11
2.1 Activity data ________________________________________________________ 12
2.2 Measurement systems under the operator’s own control _____________________ 13
2.2.1 General aspects _________________________________________________ 13
2.2.2 Simplification ‘Route CO-1’ _________________________________________ 14
2.2.3 Simplification ‘Route CO-2a’ ________________________________________ 16
2.2.4 Simplification ‘Route CO-2b’ ________________________________________ 19
2.2.5 Full uncertainty assessment ‘Route CO-3’ _____________________________ 22
2.2.6 Specific requirements for natural gas meters that are under the operator’s control
23
2.3 Measurement systems not under the operator’s own control __________________ 24
2.3.1 General aspects _________________________________________________ 24
2.3.2 Simplification ‘Route CT-1’ _________________________________________ 26
2.3.3 Simplification ‘Route CT-2’ _________________________________________ 27
2.3.4 Simplification ‘Route CT-3’ _________________________________________ 27
2.3.5 Specific requirements for natural gas meters that are not under the control of the
operator______________________________________________________________ 27
2.4 National Legal Metrological Control for liquid fuels __________________________ 31
2.5 National Legal Metrological Control for weighing devices _____________________ 32
2.6 Uncertainty for calculation factors _______________________________________ 32
2.6.1 Frequency of analysis_____________________________________________ 32
2.6.2 Using supplier information _________________________________________ 35
3 Uncertainty for measurement-based approaches ______________________________ 37
3.1 How to demonstrate compliance with tier requirements ______________________ 38
4 Uncertainty for fall-back approaches ________________________________________ 40
5 Uncertainty in the Free Allocation Regulation _________________________________ 41

3
6 Uncertainty and quality assurance _________________________________________ 43
7 ANNEX I: Conservative measurement uncertainties for the most common measuring
instruments_______________________________________________________________ 45
8 ANNEX II: Full uncertainty assessment for source streams ______________________ 53
8.1 Introduction ________________________________________________________ 53
8.2 Error propagation laws _______________________________________________ 56
8.2.1 Uncorrelated input quantities: _______________________________________ 57
8.2.2 Correlated input quantities: _________________________________________ 60
8.3 More examples _____________________________________________________ 63
8.4 Uncertainty over the whole installation and where Article 22 of the MRR (fall-back
approach) is used ________________________________________________________ 66
9 Exemplar: installation with low emissions ____________________________________ 68
9.1 Information about the installation _______________________________________ 68
9.2 Exemplar uncertainty assessment ______________________________________ 71
9.3 Quality management for the example installation ___________________________ 74
10 Emissions from catalytic cracking using continuous emission monitoring systems
(CEMS) _________________________________________________________________ 77
10.1 Background information _____________________________________________ 77
10.2 Determination of the uncertainty ______________________________________ 79

4
Overview
Interests in emissions data vary from the public to policy makers to traders buying/selling allowances.
Whatever their interest everyone needs confidence in the data. Therefore, an assessment of
uncertainty is important.

This guidance will help operators of installations understand their obligations relating to uncertainty in
the UK Emissions Trading Scheme (UK ETS), as set out in:

• The Greenhouse Gas Emissions Trading Scheme Order 2020 as amended from time to
time (the order)] [https://www.legislation.gov.uk/uksi/2020/1265/contents/made]
• The Monitoring and Reporting Regulation (MRR) (Commission Implementing Regulation
(EU) 2018/2066 of 19 December 2018 [https://eur-lex.europa.eu/legal-
content/en/ALL/?uri=CELEX%3A32018R2066] on the monitoring and reporting of greenhouse
gas emissions pursuant to Directive 2003/87/EC of the European Parliament and of the
Council (disregarding any amendments adopted after 11 November 2020) as given effect for
the purpose of the UK ETS by article 24 of the Order subject to the modifications made for that
purpose from time to time (the Monitoring and Reporting Regulation)
• The Verification Regulation (Commission Implementing Regulation (EU) 2018/2067 of 19
December 2018 [https://eur-lex.europa.eu/legal-
content/EN/TXT/?uri=uriserv:OJ.L_.2018.334.01.0094.01.ENG] on the verification of data and
on the accreditation of verifiers pursuant to Directive 2003/87/EC of the European Parliament
and of the Council (disregarding any amendments adopted after 11 November 2020), as given
effect for the purpose of the UK ETS by article 25 of the Order subject to the modifications
made for that purpose from time to time)
• The Free Allocation Regulation (FAR) (Commission Delegated Regulation (EU) 2019/331 of
19 December 2018) (https://www.legislation.gov.uk/eur/2019/331/contents) as it has effect in
domestic law and as amended by the order
1 What is uncertainty?
Uncertainty combines the concepts of accuracy and precision, thus quantifying how close emissions
data are to the true value. It is important to understand that these terms have different meanings and
should not be interchanged. They are defined below.

1.1 Definitions
Accuracy: This means closeness of agreement between a measured value and the true value of a
quantity. If a measurement is accurate, the average of the measurement results is close to the “true”
value. For example, when measuring instruments are calibrated, a series of measurements are
compared to the measurement from a certified reference material. Inaccurate measurements can
often be overcome by calibrating and adjusting measuring instruments.

Precision: This describes the closeness of results of measurements of the same measured quantity
under the same condition. That is, how repeatable is the result if the same measurand is measured
several times. It is often quantified as the standard deviation of the values around the average. It
reflects the fact that all measurements include a random error, which can be reduced, but not
eliminated.

Uncertainty: This term characterizes the range within which the true value is expected to lie with a
specified level of confidence.

As shown in Figure 1, measurements can be accurate, but imprecise, or vice versa. The ideal
situation is precise and accurate.

Figure 1 Illustration of the concept accuracy, precision and uncertainty. The bull’s eye represents
the assumed true value, the “shots” represent measurement results

6
Source: Guidance Document 4 The Monitoring and Reporting Regulation – Guidance on Uncertainty Assessment
(https://ec.europa.eu/clima/system/files/2021-10/policy_ets_monitoring_gd4_guidance_uncertainty_en.pdf)

Article 3(6) of the Monitoring and Reporting Regulation (MRR) specifies that an uncertainty must
have a 95% confidence level (meaning that there is 95% chance the correct value lies within the
interval stated) which, assuming that the dispersion of the uncertainty is following a normal
distribution, the standard uncertainty would equal one standard deviation and correspond to a
probability of only 68% that the correct value is covered within that range. To increase this probability
to 95% the standard uncertainty is multiplied by 1.96 (the coverage factor) to calculate the
expanded uncertainty. In practice this value is often rounded to 2 (2σ).

This is illustrated in the example below.

An example: A category C installation consumes 280,000 tonnes of coal annually. For this type
of installation, tier 4 is required for the determination of the fuel quantity (uncertainty: ±1.5%).
This means that the measurement system needs to provide results that allow the ‘true value’ to
be within 280 ± 4.2 kt (±1.5%) at the 95% (2σ) confidence level, as shown in the graph below.

Source: Guidance Document 4 The Monitoring and Reporting Regulation – Guidance on Uncertainty Assessment
(https://ec.europa.eu/clima/system/files/2021-10/policy_ets_monitoring_gd4_guidance_uncertainty_en.pdf)

7
1.2 Uncertainty in the Monitoring and Reporting Regulation
Bands of uncertainty threshold, or tiers, are used to describe uncertainty levels that must be met for
specific parameters. The general principle is that operators should monitor to the highest achievable
accuracy and, if using a lower tier, strive to reach a higher tier.

The magnitude of emissions involved, and scale of operation means that installations with the largest
emissions must monitor their data with the lowest uncertainty (highest tier). Conversely, installations
with the lowest emissions can apply some simpler approaches (lower tiers). All operators can use
lower tiers if they demonstrate to their regulator that meeting a higher tier is too expensive or is
technically not possible.

Article 12(1) of the MRR requires operators to submit an uncertainty assessment with their monitoring
plan to their regulator that demonstrates that they comply with the tiers defined in Annexes II, IV and
Annex VIII of the MRR, where applicable.

The uncertainty assessment provides:

• Evidence for compliance with uncertainty thresholds for activity data for major and minor
source streams. Guidance is provided in Chapter 2.
• Evidence for compliance with uncertainty required for calculation factors for major and minor
source streams, where the sampling frequency for analyses is determined based on the rule of
1/3 of the activity data uncertainty (Article 35(2)). Guidance is provided in Chapter 2, section
2.6.
• Evidence for compliance with uncertainty requirements for measurement-based
methodologies, if applicable. Guidance is provided in Chapter 3.
• Evidence for the total emissions of the installation where a fall-back methodology is applied for
at least part of the installation, confirming that the uncertainty threshold according to Article
22(c) is met. Guidance is provided in Chapter 4.
Operators must submit an uncertainty assessment when they apply for a new monitoring plan or
when they propose changes to their approved monitoring plan if the proposed changes affect the
applied monitoring tiers. The monitoring plan must reflect the tier that is applied and not just the
minimum one that is required.

During verification, operators must retain evidence for their verifier that the information used to
calculate the uncertainty levels is valid (Article 19(1) of the Verification Regulation).

The MRR allows for several simplification options, providing operators with options to demonstrate
that the uncertainty levels correspond to certain tiers, as shown in Figure 2. Those options (or routes)
are assigned codes throughout this document. For example, if a calculation based methodology is
applied and the activity data of a source stream are monitored by a measurement system outside the
operator’s own control, chapter 2 and sections 2.3.2, 2.3.3 and 2.3.4 (routes CT-1, CT-2 or CT-3) will
provide guidance for assessing uncertainty related to that activity data.

8
 Choose one or
more approaches

Calculation Measurement- Fall-back (chapter

(chapter 2) based (chapter 3) 4)

Calculation factors EN 14181, EN uncertainty over

Activity data 15259 or other the whole
(2.6) standards installation

1/3 rule
operator's control
(2.2) Sampling and
analysis

not operator's
control (2.3)

Figure 2 Overview of chapters in this document regarding determination of

uncertainty

1.2.1 Simplifications for low emitting installations, including hospital or small emitters

Operators of low emitting installations (as set out in Article 47 of the MRR) and operators of
installations that are hospital or small emitters are exempt from:

• submitting an uncertainty assessment to their regulator

• including uncertainty relating to stock changes in their uncertainty assessment
Operators of low emitting installations may use purchasing records, such as gas bills, and estimated
stock changes to determine the amount of fuel used. They may also apply, as a minimum, tier 1 for
determining activity data and calculation factors unless they can achieve a higher accuracy without
additional effort and without providing evidence to their regulator that it is technically not feasible or
would incur unreasonable costs.

Due to the legal metrological controls upon fuel billing, low emitting installations may propose an
uncertainty value of 6% for individual fiscal gas meters and may assume that the metering of gas
complies with tier 1 1 No further evidence or assessment of uncertainty is required. The application of
tier 1 may not reflect the true tier achievable by the operator but it serves to minimise the
administrative burden.

1 See Table 3 in section 2.3.5.2, right hand column for how this is derived for natural gas meters.
9
Operators of low emitting installations may choose to read their own meters as part of their control
measures to ensure accurate records of natural gas consumption. However, this does not impact on
the validity of a low emitting installation from applying tier 1.

For major or minor liquid fuel source streams operators of low emitting installations can assume that
the source stream meets tier 1 and state that the overall uncertainty meets <7.5%. No further
evidence or assessment of uncertainty is required.

If an operator of a low emitting installation or a hospital or small emitter chooses to apply a tier that is
higher than the minimum requirement, they must retain evidence for their verifier showing how they
have derived this (unless the operator is a hospital or small emitter that chooses to self-verify their
emissions).

The operator must still be able to demonstrate to their verifier, if applicable, how they comply with the
required tiers so that their verifier can confirm the validity of the information used to calculate the
uncertainty (Article 19(1) of the Verification Regulation).

An exemplar uncertainty assessment for an installation with low emissions can be found in Chapter 9.

10
2 Uncertainty for calculation-based
approaches
The equation in the example below shows the standard calculation method for the combustion of
fuels. The parameters within the formulae are either activity data or calculation factors:

• Activity data (AD): tiers here relate to the required minimum uncertainty over the reporting
period of the amount of fuel burned (see section 2.1)
• Calculation factors (net calorific value, emission factor, carbon content, oxidation
factor, biomass fraction, preliminary emission factor, conversion factor). Tiers here
relate to specific methodologies that are set out in the MRR for the determination of each
factor, for example using default values or carrying out analyses (see section 2.6)

Example: Standard calculation method in accordance with MRR Article 24(1) for
combustion of fuels
𝐸𝐸𝐸𝐸 = 𝐴𝐴𝐴𝐴. 𝑁𝑁𝑁𝑁𝑁𝑁. 𝐸𝐸𝐸𝐸 . 𝑂𝑂𝑂𝑂. (1 − 𝐵𝐵𝐵𝐵)

Where,

𝐸𝐸𝐸𝐸 …. Emissions (t CO2)

𝐴𝐴𝐴𝐴…...Activity data (fuel quantity) (t or 𝑁𝑁𝑚𝑚3)

𝑁𝑁𝑁𝑁𝑁𝑁 …Net calorific value (TJ/t or t J/𝑁𝑁𝑚𝑚3)

𝐸𝐸𝐸𝐸 …. Emission factor (t CO2/TJ, t CO2/t or t CO2/ 𝑁𝑁𝑚𝑚3)

𝑂𝑂𝑂𝑂 …. Oxidation factor (dimensionless)

Activity data and/or calculation factors can be determined using measuring instruments that are either
under the operator’s control (preferred approach) or under the control of another party, such as the
supplier of the fuel or material. If a supplier’s measuring instruments are governed by legal
metrological control, operators may assume that the uncertainty associated with the measurement is
reasonably low. If the supplier’s measuring instruments are not governed by legal metrological
control, the operators could include quality assurance for the measuring instruments, such as
maintenance and calibration in the purchase contracts. However, the operator must seek a
confirmation of the uncertainty applicable for such meters to assess if the required tier can be met.

If the operator decides to use a supplier’s measuring instruments, even if they could use their own
instruments, they must provide evidence to their regulator that the supplier’s measuring instruments
allow compliance with at least the same tier, give more reliable results and are less prone to control
risks than applying the methodology based on their own measuring instruments. This evidence must
be accompanied by a simplified uncertainty assessment.
11
2.1 Activity data
This section applies to both the input or output material of a source stream monitored by a mass
balance approach, as well as source stream monitored by a calculation approach.

Operators may determine their activity data:

• Based on continual metering of the process which causes the emissions, or

• Based on aggregation of metered amounts separately delivered taking into account relevant
stock changes.
The tiers for activity data of source streams are defined using thresholds for a maximum uncertainty
allowed for the determination of the quantity of fuel or material over a scheme year (which is a
calendar year).

The uncertainty assessment must account for all sources of uncertainty, including uncertainty of
measuring instruments, of calibration, any additional uncertainty connected to how the measuring
instrument is used in practice, and of environmental impacts, unless some simplifications are
applicable. The impact of the determination of stock changes at the beginning and end of the period
must be included, where applicable (see example 7 in section 8.3).

Table 1 illustrates the tier definitions for the combustion of fuels. A full list of the tier thresholds can be
found in section 1, Table 1 in Annex II of the MRR.

Table 1 Typical definitions of tiers for activity data based on uncertainty for the combustion of
fuels.

Tier No. Definition

1 Amount of fuel [t] or [𝑁𝑁𝑚𝑚3] over the scheme year is determined

with a maximum uncertainty of less than ± 7.5 %.

2 Amount of fuel [t] or [𝑁𝑁𝑚𝑚3] over the scheme year is determined

with a maximum uncertainty of less than ± 5.0 %.

3 Amount of fuel [t] or [𝑁𝑁𝑚𝑚3] over the scheme year is determined

with a maximum uncertainty of less than ± 2.5 %.

4 Amount of fuel [t] or [𝑁𝑁𝑚𝑚3] over the scheme year is determined

with a maximum uncertainty of less than ± 1.5 %.

12
2.2 Measurement systems under the operator’s own control
2.2.1 General aspects

Operators must ensure that the applicable uncertainty threshold of the tiers set out in Articles 26 and
41 of the MRR are met. Uncertainty assessments are a means to provide robust evidence that the
uncertainty of the measurement system meets the required tier.

According to Article 3(29) of the MRR, a measurement system is ‘a complete set of measuring
instruments and other equipment, such as sampling and data-processing equipment, used to
determine variables such as the activity data, carbon content, the calorific value or the emission
factor of the greenhouse gas emissions.’ The measurement system includes instruments such as
temperature and pressure probes, flow meters and telemetry systems.

There can be many sources of uncertainty in a measurement system, such as errors that are caused
by a lack of precision (for example, the meter’s uncertainty as specified by the manufacturer for use
in an appropriate environment, and certain conditions related to the installation of measuring
equipment (such as the length of straight piping before and after a flow meter) and a lack of accuracy
(for example, caused by ageing or corrosion of the instrument, which may result in drift). The MRR
requires that the uncertainty assessment takes account of the measuring instrument’s uncertainty,
the influence from calibration and all other possible influencing parameters. This can mean that, in
practice, uncertainty assessments can be demanding and so operators should focus on the most
relevant parameters that contribute to the uncertainty.

The MRR allows several pragmatic simplifications to assessing uncertainty. These are illustrated in
Figure 3.

The operator can choose to simplify the uncertainty assessment, if

• The measuring instrument(s) is subject to legal metrological control (Route CO-1). This is the
simplest approach. In this case the maximum permissible error in service laid down in the
relevant national legal metrological text can be used as the overall expanded uncertainty (read
section 1.1 for an explanation of expanded uncertainty).
• The measuring instrument is not subject to national legal metrological control (NLMC) but is
installed in an environment appropriate for its use specifications. In this case the operator may
assume that the expanded uncertainty over the whole reporting period, as required by the tier
definitions for activity data in MRR Annex II, equals:
o the maximum permissible error specified for that instrument in service (Route CO-2a),
or
o where available and lower, the expanded uncertainty obtained by calibration, multiplied
by a conservative adjustment factor (this is explained in more detail in section 2.2.4.4)
for taking into account the effect of uncertainty in service (Route CO-2b).

13
Where the simplifications described above are not applicable, or do not show that the required tier is
met, a full uncertainty assessment in accordance with Route CO-3 and Annex II of this document
must be carried out.

If the operator cannot demonstrate that the tier required by the MRR has been met, they must take
the necessary steps to comply with the MRR by:

• Carrying out corrective action, such as installing a measurement system that meets the tier
requirements, or
• Provide evidence to their regulator that meeting the required tier is not technically feasible or
would incur unreasonable costs and use the next lower tier in accordance with the result of the
uncertainty assessment.

Measuring instrument is not

subject to national legal
metrological control

Measuring instrument is
installed in an environment Full uncertainty assessment
Measuring instrument is appropriate for is use and (Route CO-3)
specification
subject to national legal
metrological control
(Route CO-1)
Uncertainty = maximum
permissible error specificed for
that measuring instrument in
service
Uncertainty = maximum (Route CO-2a)
permissible error in service
allowed by national legal
metrological control
Uncertainty = Uncertainty obtaind
by calibration multipled by a
conservative adjustment factor
(Route CO2b)

Figure 3 Activity data for calculation-based approaches: approaches for determination of

the uncertainty achieved (‘C’ means calculation based, ‘O’ means instrument is under
operator’s own control)

2.2.2 Simplification ‘Route CO-1’

Measuring instrument is subject to national legal metrological control (NLMC)

Overall expanded uncertainty = maximum permissible error in service (MPES)

14
Article 3(24) of the MRR defines ‘legal metrological control’ to mean the control of the measurement
tasks intended for the field of application of a measuring instrument, for reasons of public interest,
public health, public safety, public order, protection of the environment, levying of taxes and duties,
protection of the consumers and fair trading.

Measuring instruments subject to national legal metrological control (NLMC) are regulated for
consumer protection purposes. NLMC instruments must meet minimum ‘essential’ requirements for
their manufacture and use before they can be sold. The stringent requirements around measuring
instruments subject to NLMC mean that they are considered more reliable than measuring
instruments not subject to NLMC. For further information see ‘background information on maximum
permissible errors under NLMC’.

The most appropriate evidence for being under NLMC is a certificate of official verification of the
instrument.

Read section 2.3.5 on the specific requirements for natural gas meters for guidance on how to
demonstrate that natural gas meters comply with NLMC. Sections 2.4 and 2.5 provide guidance on
NLMC for liquid fuels and weighing devices.

Background information on maximum permissible errors under NLMC

Under legal metrological control calibration is considered valid where the expanded uncertainty
resulting from the calibration procedure is lower than the maximum permissible error (MPE) in
verification. The term ‘in verification’ is a metrological term here and must not be confused with
verification under the UK ETS.

Furthermore, it is considered that the equipment under regular service (that is, in use) is exposed to
measurement conditions that might have an impact on the measurement result (such as temperature,
vibration, pressure). This aspect led to the introduction of a parameter called the maximum
permissible error in service (MPE in service = MPES). This value represents a fair estimation of the
uncertainty of a device under regular operation, which undergoes regular legal metrological control
complying with the associated regulations. It sets a threshold for simplified checks which could be
applied during regular operation and has therefore to be considered as the uncertainty which needs
to be attributed to the daily operation of the measurement equipment. This means that the MPES is
more appropriate for use to ensure a fair exchange of goods, the ultimate objective of legal
metrological control.

For some measuring instruments the MPE ‘under rated operating conditions’ (values for the
measurand and influence quantities making up the normal working conditions of an instrument) are
regulated in The Measuring Instruments Regulations 2016 (SI 2016 No. 1153)
(https://www.legislation.gov.uk/uksi/2016/1153/contents/made) and The Non-automatic Weighing
Instruments Regulations 2016 (SI 2016 No. 1152)
(https://www.legislation.gov.uk/uksi/2016/1152/contents).

Metrological control systems usually apply a factor of 2 to convert the maximum permissible error
derived in verification into the maximum permissible error in service (MPES). Note that this factor is
15
not derived from statistics (unlike the difference between standard and expanded uncertainty) but
follows from general experience in legal metrology with measuring instruments which have
undergone successful type approval tests.

For further background information Annex I of the training material of the M&R Training Event on
Uncertainty Assessment 2 may be helpful.

2.2.3 Simplification ‘Route CO-2a’

Measuring instrument is not subject to national legal metrological control but is installed in an
environment appropriate for its use specifications.

Overall expanded uncertainty = maximum permissible error in service

The second simplification allowed by the MRR, applies to measuring instruments that are not subject
to national legal metrological control but are installed in an environment appropriate for their use
specifications. Please note that MPE and MPES values for instruments under NLMC are based on
experience and they are not transferable to industrial measurement. However, the terms ‘MPE’ and
‘MPES’ for instruments not subject to NLMC are used here for simplicity reasons.

Article 28(2) of the MRR allows the operator to use the ‘maximum permissible error in service’
specified for the instrument as the overall uncertainty, ‘provided that measuring instruments are
installed in an environment appropriate for their use specifications’. The MPE in service is
significantly higher than the MPE of a new instrument and is often expressed as a factor (usually 2)
times the MPE of a new instrument.

Where no information is available for the MPES, or where the operator can achieve better values
than the default values, the uncertainty obtained by calibration may be used, multiplied by a
conservative adjustment factor for taking into account the higher uncertainty when the instrument is
‘in service’. The latter approach reflects route CO-2b.

The MRR does not define the information sources for MPES or what ‘appropriate use specifications’
must be. However, operators may assume that the following are suitable sources for MPES:

• the manufacturer’s specifications,

• specifications from legal metrological control, and
• guidance documents such as the UK ETS authority guidance (see Chapter 8 (Annex II) for
illustrative conservative values for uncertainty ranges of common measuring instruments and
additional conditions).

2 https://ec.europa.eu/clima/system/files/2020-02/uncertainty_assessment_en.pdf
16
The uncertainties given in these sources may only be taken as the overall uncertainty if the
measuring instruments are installed in an environment appropriate for their use specifications and
steps 1 to 4 below are met. If this is the case values taken from these sources can be considered as
representing the MPES and no further corrections to that uncertainty value are necessary.

The operator can assume they meet the requirements of Article 28(2) of the MRR, if they provide
evidence that all the requirements of the following 4 steps are met:

Step 1: Operating conditions regarding relevant influencing parameters are available

The manufacturer’s specification for that measuring instrument contains operating conditions that
specify the environment appropriate for its use regarding relevant influencing parameters (for
example, flow, temperature, pressure, medium etc.) and maximum permissible deviations for these
influencing parameters. Alternatively, the manufacturer may have declared that the measuring
instrument complies with an international standard (CEN or ISO standard) or other normative
documents (such as recommendations by Organisation Internationale de Métrologie Légale (OIML)
http://www.oiml.org/), which lay down acceptable operating conditions for the relevant influencing
parameters.

If the manufacturer’s specifications do not contain requirements for operating conditions regarding
relevant influencing parameters, the operator must carry out an individual uncertainty assessment
(Route CO-3). However, in simple cases, expert judgement might be sufficient, especially for minor,
de-minimis and marginal source streams and for installations with low emissions.

Step 2: Operating conditions regarding relevant influencing parameters are met

The operator provides evidence that the operating conditions regarding relevant influencing
parameters are met. Evidence could take the form of a checklist of the relevant influencing
parameters for different measuring instruments and compare for each parameter the specified range
with the used range (see tables 15, 16 and 17 in section 8.1 for examples of influencing parameters).
This list should be provided to the regulator as part of the uncertainty assessment when submitting a
new or updated monitoring plan.

The result for this step should be an assessment that

• the measuring instrument is installed appropriately

• the measuring instrument is appropriate to measure the medium of interest
• there are no other factors that could have adverse consequences on the uncertainty of the
measurement instrument.
Only if all of this is the case, can it be assumed that the MPES described in the ‘suitable source’ (see
above) is appropriate for use without further correction.

Step 3: Performing quality assured calibration procedures

17
Operators must compare the uncertainty of their instrument against the requirements of the MRR at
least once per year and after each calibration. The uncertainty may be taken from the manufacturer’s
specifications, or initial calibration certificate if measurement instruments have been installed and
used as the manufacturer intended.

Regular calibration (see route CO-2b for more information on ‘calibration’,) is performed by an
institute accredited in accordance with EN ISO/IEC 17025, employing CEN, ISO or national
standards where appropriate. Alternatively, if calibration is performed by a non-accredited institute or
by a manufacturer's calibration, the operator must provide evidence (for example with a calibration
certificate) of the suitability of the calibration method, that the calibration is performed using the
instrument manufacturer’s recommended procedure, by competent personnel and that the results
comply with the manufacturer's specifications.

One way of checking that the measurement device can still meet the required tier level on an annual
basis (assuming that it was installed and used correctly) is to demonstrate that the manufacturer’s
recommended maintenance procedures have been followed.

Complying with this requirement in Article 28 does not necessarily mean the device, for example a
gas meter, needs to be removed for calibration every year 3. However, for other measuring
instruments such as weighbridges, annual calibration may be appropriate. Operators can check the
manufacturer’s recommendations/ technical specification or technical performance standards to help
identify appropriate calibration intervals (see Annex I).

Step 4: Further quality assurance procedures for measuring activity data

Article 59(3) of the MRR requires operators to establish, document, implement and maintain various
written procedures to ensure an effective control system, including quality assurance of relevant
measurement equipment, and handling of resulting data. Where certified quality or environmental
management systems are in place, such as EN ISO 9001, EN ISO 14001, EN ISO 15001 to ensure
that control activities (calibration, maintenance, surveillance and loss/failure management etc.) are
carried out, it is recommended that these systems also include the quality assurance for measuring
activity data under the UK ETS. If, through monitoring the effectiveness of control systems, the
control system is found to be ineffective, the operator must improve the control system, updating the
monitoring plan or underlying written procedures, as appropriate.

If the requirements of all 4 steps cannot be fulfilled, overall uncertainties may be calculated by
combining the uncertainties provided in the ‘suitable sources’ and a conservative estimate of the
uncertainty for the parameter(s) preventing compliance with one or more steps (for example, the flow
rate is partially outside the normal operating range) by applying error propagation (see route CO-3
and section 8.2 of Annex II).

3 Gas meters are not routinely removed for calibration but are often chosen on the basis of statistical sampling.
(17/04/2012 note of ETG WG 3/7 following presentation to UK Emissions Trading Group by Mark Way, National Grid)
18
2.2.4 Simplification ‘Route CO-2b’

Measuring instrument is not subject to national legal metrological control but is installed
in an environment appropriate for its use specifications

Overall expanded uncertainty = expanded uncertainty from calibration multiplied

by a conservative adjustment factor

Operators may choose this simplification if the results provide a lower uncertainty than if they follow
route CO-2a.

2.2.4.1 Calibration 4
Article 28(1)(b) of the MRR requires that ‘the operator must ensure at least once per year, and after
each calibration of measuring instruments, that the calibration results multiplied by a conservative
adjustment factor based on an appropriate time series of previous calibrations of that or similar
measuring instruments for taking into account the effect of uncertainty in service, are compared with
the relevant uncertainty thresholds.’

The performance of regular calibration is the process where metrology (the infrastructure that covers
the accuracy, precision and repeatability of a measurement) is applied to measurement equipment
and processes to ensure conformity of measuring instruments in use with a known international
measurement standard. It involves traceability or comparison with a ‘standard’ or between different
measuring systems. This is achieved by using calibration materials or methods that ensure a closed
chain of traceability to the ‘true value’ performed as a measurement standard.

Calibration should, if possible, be carried out by an EN ISO/IEC17025 accredited calibration

laboratory. Appropriate calibration procedures and intervals may be found in the manufacturer’s
specification, standards provided by accredited laboratories, etc. 5
Note 1: A calibration may be expressed by a statement, calibration function, calibration diagram, calibration curve, or
calibration table. In some cases, it may consist of an additive or multiplicative correction of the indication with associated
measurement uncertainty.

4see ‘EA 4/02 - Guidance to Expression of Uncertainty of Measurement in Calibration https://european-

accreditation.org/wp-content/uploads/2018/10/ea-4-02-m-rev01-september-2013.pdf

5see ‘International vocabulary of metrology’

http://www.bipm.org/utils/common/documents/jcgm/JCGM_200_2008.pdf

19
Note 2: Calibration should not be confused with adjustment of a measuring system, often mistakenly called “self-
calibration”, nor with ‘metrological’ verification of calibration.

Example: Requirements for calibration of a flow meter for nonaqueous liquids with static
start/stop measurement

For calibration the following aspects need to be considered:

• The flow meter is installed in accordance with the manufacturer’s specifications

• The flow meter as well as the rest of the whole calibration system are filled completely
and are free from gases
• The flow meter is at operating temperature
• All parameter settings, to the extent available, are documented
• During zero flow rate before and after the measurement no signal indicating a flow is
detected
• The calibration conditions (for example, flow rate, temperature, pressure, liquid type) are
within the operating conditions
• The flow rate is stable
• The pressure is high enough to avoid gasification or cavitation (the formation and then
immediate implosion of cavities in a liquid, which may occur when a liquid is subjected to
rapid changes of pressure, for example in turbines). Density and viscosity have an
influence on the calibration curve as well. Therefore, it is optimal to calibrate under the
same conditions as during (intended) normal operation and to use the same, if available,
or similar liquids
• Adjusting to zero (“zeroing”) is to be done before and not during a measurement series.
Conditions of the liquid (temperature, pressure) are to be documented at the time of
zeroing. Zeroing is not required if the output signal for zero flow rate is lower than the
range for the zero-value provided by the manufacturer

The core element of each calibration procedure is the comparison of measurement results with a
reference standard by applying a procedure which enables the determination of a calibration function
and of measurement uncertainties. The result of calibration will be a reliable assessment of the
calibration function, its linearity (where this is a requirement) and the measurement uncertainty. The
uncertainty obtained by calibration should, to the extent possible, relate to the operating range of the
measuring instrument in actual use. Thus, the calibration procedure should reflect to the extent
possible the operating conditions where the instrument is installed.

In many cases the measurand of interest is not measured directly but rather calculated from other
input quantities with a functional relationship, for example, a volumetric flow (fV) is calculated by
measuring inputs like density (ρ) and pressure difference (∆p) through the relationship fV=fV(ρ, ∆p).
20
The uncertainty related to the measurand of interest will then be determined as the combined
standard uncertainty via error propagation 6 (see Annex II). For the combined standard uncertainty
associated with the measurement result, uncertainty contributions of long-term drift and operational
conditions are also important influences which must be considered (along with the uncertainty
associated with calibration itself).

Calibration certificates typically express uncertainty as the expanded uncertainty, meaning that there
is no need to multiply the uncertainty from the calibration certificate by a further coverage factor (see
section 1.1 for an explanation of coverage factors). However, if the calibration certificate states a
standard uncertainty (i.e. a 68% confidence interval) then operators must multiply that value by a
coverage factor, which is usually 2. This coverage factor is not the same as the conservative
adjustment factor (see section 2.2.4.4).

2.2.4.2 Frequencies of calibration

Depending on the type of measuring instrument and the environmental conditions the uncertainty of a
measurement might increase over time (drift). To quantify and to mitigate the increase of uncertainty
resulting from drift an appropriate time interval for recalibration is necessary.

In the case of a measuring instrument subject to NLMC (Route CO-1) the frequency of calibration (re-
calibration) is regulated by the relevant legal text.

For other measuring instruments that are not subject to NLMC (for example flare gas meters and fuel
gas meters) re-calibration intervals should be based on information provided by manufacturer’s
specifications or other suitable sources. As the result of every calibration allowing quantification of the
drift that has occurred, a time series analysis of previous calibrations may also be helpful to
determine the relevant calibration interval. Based on this information the operator should use
appropriate calibration intervals subject to the regulator’s approval.

In any case the operator must check annually if the measuring instruments used still comply with the
tier required (as set out in Article 28(1)(b) of the MRR).

2.2.4.3 Industry practice

Various situations need to be guarded against when it comes to calibration in industrial
circumstances, including

• simplifications for applications that do not then meet requirements for calibration according to
legal standards
• single-point-tests or short checks that may be designed, for example, for checking the zero
value and for providing day to day quality assurance, but which do not constitute full calibration

6 It is more appropriate to call it “propagation of uncertainty” although “error propagation” is more frequently used.
21
 • postponement of calibrations due to favourable ad-hoc checks (suggesting proper operation of
monitoring equipment) and due to the costs involved
• failure to follow-up the results of the calibration by making adequate corrections.
Problems may occur when a device is not easily accessible for calibration. For example, it can’t be
de-installed for checks or calibration during operation of the installation and the process cannot be
shut down without major disruption to the installation or to the security of supply associated with the
product. There may be long periods between shutdowns of the production process and in such cases
a periodic calibration using shorter intervals may not be feasible.

Where only limited possibilities for calibration exist, the operator must seek approval from their
regulator for an alternative approach (as set out in Article 60(1) of the MRR), enclosing alongside the
submission of the monitoring plan any relevant evidence with regards to technical feasibility or
unreasonable costs. The application of alternative standards must be considered within the hierarchy
as set out in Article 32(1) of the MRR.

2.2.4.4 Conservative adjustment factor

The expanded uncertainty obtained from calibration is multiplied by a conservative adjustment factor
to consider random as well as systematic errors in service. The operator may determine the
conservative adjustment factor based on experience, other sources of information, or use a typical
value of 2 as a pragmatic yet appropriate approach. The result obtained may be used as the overall
uncertainty without further corrections.

However, a conservative adjustment factor is only applicable if the measuring instrument is used
within the use specifications (as set out in the last sub-paragraph of Article 28(2)) of the MRR
Consequently, the requirements described for route CO-2a (step 1 to step 4) must be met. If those
requirements are not met neither route CO-2a or route CO-2b are applicable and a specific
uncertainty assessment described under route CO-3 and Annex II of this document is required.

Implementing effective measuring instrument control and management procedures can minimise the
risk of the instrument performance varying over time while in service. In these circumstances
operators could propose a conservative adjustment factor value of 1 if they can demonstrate to their
verifier and their regulator that they have good control over their measuring system; otherwise, a
typical value of 2 is appropriate.

2.2.5 Full uncertainty assessment ‘Route CO-3’

Sections 2.2.2 (‘Route CO-1’ measurement systems subject to NLMC), 2.2.3 (‘Route CO-2a’
measurement systems not subject to NLMC but is installed in an environment appropriate for its use
specifications) and 2.2.4 (‘Route CO-2b’ measurement systems not subject to NLMC but is installed
in an environment appropriate for its use specifications and subject to calibration) describe routes
that help simplify how operators comply with their uncertainty obligations under the MRR. Operators
22
don’t have to apply a simplified approach and can choose to carry out a full uncertainty assessment,
especially if it provides better results. However, the operator must carry out a full uncertainty
assessment if none of the simplification routes are possible. Even then, the operator can use the
outputs of the simplification routes as starting points for further calculations, for example if using error
propagation (see Annex II, section 8.2). This approach not only presents a more pragmatic and less
burdensome way for operators to assess uncertainty, but it may also in most cases provide more
reliable results.

Example: An operator uses a turbine meter subject to national legal metrological control for the
consumption of a liquid source stream. The operator must determine the density of the liquid to
convert the volumetric flow into mass flows. Simplification routes CO-1 or CO-2a/2b can’t be
applied for the source stream because density is regularly determined by an aerometer if
expressed in tonnes. However, the operator could use the uncertainty laid down in the relevant
national legal metrological text related to the determination of the volume in the overall
uncertainty calculation by error propagation (see example 7 in section 8.3).

2.2.6 Specific requirements for natural gas meters that are under the operator’s control

Operators that own the gas meter(s) used to determine their activity data may, if the gas meter is
used for billing purposes, use the maximum permissible error in service allowed by UK legislation or
UK rules (such as industry codes of practice produced by the Office for Gas and Electricity Markets,
Ofgem) as their uncertainty value.

For natural gas metering, to make use of this simplification the operator must identify the type of
meter to identify the maximum permissible error in service (MPES). See sections 2.3.5.1 and 2.3.5.2
for how to do this.

2.2.6.1 Annual checks to comply with Article 28

Operators must compare the uncertainty of their measurement device(s) against the requirements of
the MRR at least once per year and after each calibration. Operators that have installed and used the
measurement device as the manufacturer intended may, for example, take the uncertainty from the
manufacturer’s specifications, or initial calibration certificate.

Complying with this requirement that is set out in Article 28(1) of MRR does not necessarily mean
that gas meters need to be removed for calibration every year. Operators should check the
manufacturer’s recommendations, technical specification or technical performance standards to help
identify appropriate calibration intervals. See Annex I of this document for examples of calibration
intervals for typical gas meters.

If the operator calibrates their measurement instruments, they must apply a conservative adjustment
factor to the uncertainty determined from the results of the calibration (as described in section
2.2.4.4).

23
2.3 Measurement systems not under the operator’s own control
2.3.1 General aspects

Operators may use a measurement system outside their own control to determine activity data, if that
system complies with at least as high a tier, gives more reliable results and is less prone to control
risks 7 than using their own instruments, if available. For these cases, activity data may be determined
either by

• amounts taken from invoices issued by the trading partner, or

• direct readings from the measurement system.
Whichever approach is used, the same tiers for activity data are required as for systems under the
operator’s own control. The only difference is how the operator demonstrates that they comply and
what simplifications may be applied.

If operators use invoices as the primary data source for determining the quantity of material or fuel,
Article 29 of the MRR requires the operator to demonstrate that the trade partners are independent.
In principle, this should be considered a safeguard for ensuring that meaningful invoices exist. In
many cases it will also be an indicator of whether national legal metrological control (see route CO-1
in section 2.2.2) is applicable.

Note that there is a ‘hybrid’ possibility allowed by the MRR: The instrument is outside the control of
the operator, but the operator reads the instrument for their monitoring purposes. The owner of the
instrument is responsible for maintenance, calibration and adjustment of the instrument, and
ultimately for the applicable uncertainty value, but the quantity of fuel or material used can be directly
checked by the operator. This is a situation frequently found for natural gas meters.

7For guidance on risk assessment see Guidance document No. 6 (Data flow and control activities).
https://ec.europa.eu/clima/system/files/2021-10/policy_ets_monitoring_gd6_dataflow_en.pdf
24
Figure 4 shows the way provided by the MRR to comply with the tier requirements in case of
measurement systems not under the operator’s control.

Use amounts from invoices, provided

that a commercial transaction
between two independent trade
partners takes place
Use of direct readings from the
measurement sytem

Measuring instrument is not subjec to

Measuring insturment is subject to national legal metrological control
national legal metrological control
(Route CT-3)

Requirements under national legal Requirements under national legal

metrological control are at least as metoroligcal control are less stringent
stringent as the required tier than the required tier
(Route CT-1) (Route CT-2)

Uncertainty = maximum permissible

Obtain evidence on the applicable
error in service allowed by national
legal metrological control uncertinaty from the trade partner

Figure 4 Activity data for calculation-based approaches: Approaches for determination of the
uncertainty achieved (‘C’ means calculation-based, ‘T’ means instrument is under trading
partner control)

The operator can simplify the uncertainty assessment:

• If the measuring instrument is subject to legal metrological control, the maximum permissible
error laid down in the relevant national legal metrological text can be used as the overall
expanded uncertainty for assessing whether the tier requirements in accordance with Article
26 of MRR are met (route CT-1).
• If the applicable requirements under national legal metrological control are less stringent than
the uncertainty threshold of the tier required in accordance with Article 26 of the MRR, the

25
 operator may obtain evidence from the trade partner concerning the expanded uncertainty that
is actually applicable (route CT-2).
• If the measuring instrument is not subject to national legal metrological control, the operator
may obtain evidence from the trade partner relating to the uncertainty concerned (route CT-3).
If sufficient evidence cannot be obtained from the trading partner (for example, the fuel supplier) the
operator may take the following steps:

Step 1. Can the operator provide evidence that the uncertainty in a reasonable worst-case scenario
is still better than using the operator's own meters and is at least meeting tier 1? Such evidence may
be obtained by demonstrating that, for example, this measuring instrument is subject to national legal
metrological control and even the least stringent requirements will meet a defined tier (the
background information in the exemplar in section 9.1 explains this further).

Step 2. If yes, then the tier met in the worst case should be used for further assessments.

Step 2a. If the tier met in the worst case meets at least the required tier then no further evidence is
needed.

Step 2b. If the tier met in the worst case is lower than the required tier, the operator must
demonstrate that using their own meters to meet the required tiers would incur unreasonable costs or
is not technically feasible.

Step 3. If no, then the operator is not meeting at least tier 1 and is applying a fall-back approach (as
referred to in Article 22 of the MRR). The operator must demonstrate that using their own meters to
meet the required tiers would incur unreasonable costs or is not technically feasible.

2.3.2 Simplification ‘Route CT-1’

Measuring instrument of the trade partner is subject to national legal metrological control
(NLMC).

Overall expanded uncertainty = maximum permissible error in service (MPES)

This simplification is applicable for the same reasons and under the same conditions as described in
section 2.2.2 (route CO-1). The operator must demonstrate that the trade partner's measuring
instrument complies with at least as high a tier as an instrument available under the operator's own
control, gives more reliable results, and is less prone to control risks.

26
2.3.3 Simplification ‘Route CT-2’

Measuring instrument of the trade partner is subject to national legal metrological control but the
requirements under national legal metrological control are less stringent than the required tier

‘The operator shall obtain evidence of the applicable uncertainty from the trade partner
responsible for the measurement system.’

If the applicable requirements under national legal metrological control are less stringent than the tier
requirements set out in Article 26 of the MRR, the operator must obtain evidence from the trading
partner that the required tiers are met. The operator must be able to demonstrate that the trade
partner's measuring instrument complies with at least as high a tier as an instrument available under
the operator's own control, gives more reliable results, and is less prone to control risks.

This may also be based on an uncertainty assessment as explained in Annex II of this guidance,
using information on the measuring instruments obtained from the trade partner. Please also see the
information given in section 2.2.5 (route CO-3).

2.3.4 Simplification ‘Route CT-3’

Measuring instrument is not subject to national legal metrological control

‘The operator shall obtain evidence of the applicable uncertainty from the trade partner
responsible for the measurement system.’

This route is similar to route CT-2 above. Where the transaction is not subject to NLMC, the operator
must obtain evidence from the trading partner that the tiers set out in Article 26 of the MRR are met.
The operator must be able to demonstrate that the trade partner's measuring instrument complies
with at least as high a tier as an instrument available under the operator's own control and gives more
reliable results and is less prone to control risks.

This may also be based on an uncertainty assessment as explained in Annex II of this document,
using information on the measuring instruments obtained from the trade partner. Please also see the
information given in section 2.2.5 (route CO-3).

2.3.5 Specific requirements for natural gas meters that are not under the control of the
operator

Operators that do not own the gas meter used to determine their activity data, may, if that meter is
used for billing purposes, use the maximum permissible error in service allowed by UK legislation or

27
UK rules as their uncertainty value. To make use of this simplification operators must identify the type
of meter (see section 2.3.5.1) in order to identify the MPES (see section 2.3.5.2).

2.3.5.1 Identify the meter type

Operators can find out what type of gas meter they have by asking their gas supplier (the business
that they pay for gas they consume) for written confirmation of their gas meter's specifications or use
the specifications given in a supplier calibration certificate, or supplier contract documentation. The
specifications must correspond with the gas bill, the gas meter and the volume converter used (if they
are separate instruments). Corresponding information may include a reference number (that matches
the reference on their gas bill), the meter serial number and meter type, the date of manufacture, the
maximum capacity, as well as details of any secondary instruments such as volume converters.

2.3.5.2 Identify the maximum permissible error in service

Gas meters used for billing purposes must be accurate and will have been approved as suitable for
use under one of two UK regulations: the Measuring Instruments (Gas Meters) Regulations 2006 (SI
2006 No 2647) (https://www.legislation.gov.uk/uksi/2006/2647/contents/made) or the Gas (Meters)
Regulations 1983 (SI 1983 No 684) (https://www.legislation.gov.uk/uksi/1983/684/made).

The Measuring Instruments (Gas Meters) Regulations 2006 apply to newer gas meters but also
principally to small low pressure/low volume gas meters8, 9, including domestic meters and light
industrial/commercial uses. Gas meters that were type approved under the Gas (Meters) Regulations
1983 can continue to be used so long as they meet the legal requirements. Gas meters that have
been ‘stamped’ under the Measuring Instruments (Gas Meters) Regulations 2006 stipulate the
maximum permissible error (MPE) and the maximum permissible error in service (MPES), depending
on the accuracy class10 and flow rate range. Table 2 shows the relationship between the MPE and
the MPES.

Table 2 The relationship between the MPES and MPE for different meter types

Measurement device Relationship between MPES and MPE

class 1 meter MPES = MPE

class 1.5 meter MPES = 2*MPE

8 Section 17 of the Gas Act 1986 (https://www.legislation.gov.uk/ukpga/1986/44/section/17) and the ‘stamping of meters’
that displays evidence of conformity only applies to gas meters that supply a quantity of gas at a rate of flow which, if
measured at a temperature of 15 °C and a pressure of 1013.25 millibars, does not exceed 1600 m3/h or the equivalent
quantity in kilograms.
9 This is equivalent to a net thermal input of about 16.6 MW (net) or 18.5 MW (gross) which is equivalent to a gas turbine

output of about 5 MWe (JEP, 2014).

10 Regulation 28 of the Measuring Instruments (Gas Meters) Regulations 2006.

https://www.legislation.gov.uk/uksi/2006/2647/regulation/28/made
28
 Volume converter 11 MPES = MPE

Diaphragm meter/other 12 meter MPES = MPE

(1983 Regulations)

Standard letters, such as those provided by National Grid Metering may not use the terminology
'MPES' but may state wording such as ‘Rotary and Turbine Meters have an accuracy of ± 1% from
20% to 100% of the flow range, and ± 2% below 20%.’ From this information we can deduce that this
is an accuracy class 1 meter or is regulated under the Gas (Meters) Regulations 1983 (see Table 2).
Tables 3 and 4 show the MPES of various meter classes and compares them to the tiers that they
would achieve (operators should refer to Article 26 of the MRR to identify what tier is needed to apply
to major, minor, de-minimis and marginal source streams). It is inevitable that the flow rate will be
less than 20% of the maximum flow rate at some point. However, if the meter operates above 20% of
the maximum flow rate during normal plant operation, the overall uncertainty is assumed to be within
the higher flow.

For large meters, that are out of scope of section 17 of the Gas Act 1986
(https://www.legislation.gov.uk/ukpga/1986/44/section/17), operators can use the uncertainty values
(MPES) quoted by the gas supplier (for example, in contract documentation or standard National Grid
Metering letter).

All instruments that are applicable to the measurement must be considered. So, if there is a separate
instrument for converting the volume, for example, then operators must also take the MPE for this
into account also. The MPE for temperature correction devices is ±0.7% and ±1% for other
conversion devices. Operators may have more accurate data than this; if operators use their own
data they must be able to demonstrate it to their verifier. Tables 3 and 4 show the calculations with
temperature and pressure correction devices.
Note: operators taking gas consumption from invoices must check their invoice to see if their supplier has applied a
correction factor. If their suppler has applied a standard factor then exclude the volume converter accuracy from their
calculation; there is no impact on the tier achieved.

11 Paragraph 4 of Schedule 1 of the Measuring Instruments Regulations 2016 (SI 2016 No.1153)

https://www.legislation.gov.uk/uksi/2016/1153/schedule/1/paragraph/4
There is no information in this regulation on the requirements for maximum permissible error in service. Operators may
assume that the maximum permissible error in service is the same as the maximum permissible error.
12 There is no information in this regulation on the requirements for maximum permissible error in service. Operators may

assume that the maximum permissible error in service is the same as the maximum permissible error.
29
Table 3 Comparison of tier thresholds with various natural gas meter classes (Measuring
Instruments (Gas Meters) Regulations 2006).

Accuracy class Accuracy class Accuracy class Accuracy class

1.0 1.0 1.5 1.5

(high flow rate, (low flow rate, 0% (high flow rate, (low flow rate, 0%
20%-100%) to 20%) 20%-100%) to 20%)

Meter ±1% ± 2% ±1.5% ±3%

accuracy/MPE

MPES ±1% ± 2% ±3% ±6%

Volume ±1% ±1% ± 1% ± 1%

converter
accuracy

Uncertainty √ (12 + 12 ) = √ (22 + 12 ) = √ (32 + 12 ) = √ (6 2 + 12 ) =

calculation 1.41 2.24 3.16 6.08

Tier threshold 4 3 2 1

Table 4 Summary of MPE and MPES required by the Gas (Meters) Regulations 1983 for various
natural gas meter types.

Other meter type Other meter type

Diaphragm Meter (high flow rate, 20%- (low flow rate, 0% to
100%) 20%)

Meter accuracy/MPE ± 2% ±1% ± 2%

MPES ± 2% ±1% ± 2%

Volume converter ±1% ±1% ±1%

accuracy

Uncertainty calculation √ (22 + 12 ) = 2.24 √ (12 + 12 ) = 1.41 √ (22 + 12 ) = 2.24

Tier threshold 3 4 3

30
Without any evidence from their gas supplier of the gas meter accuracy class, the actual meter type
and the flow rate range, regulators must take a conservative approach and assume that the gas
supplier has installed an accuracy class of 1.5 and therefore will apply a MPES of 6%. This means
that the tier compliance is tier 1, which is acceptable for a low emitter and hospital or small emitter.

The operator’s verifier will note a non-conformity with the MRR if the operator is required to apply a
higher tier than tier 1 but cannot demonstrate it before the end of the verification process. The
operator will have to submit an improvement report to their regulator explaining the steps that they
are taking to obtain the necessary information.

2.4 National Legal Metrological Control for liquid fuels

Example 7 in section 8.3 describes the uncertainty assessment for stored gas oil. The legal
metrological controls 13 upon liquid fuel supply for most liquid fuels used in the UK ETS (regardless of
whether the fuel is delivered by vessels or road tanker) means that the MPE and the MPES of the
meter measuring systems (defined as the meter and all devices required to ensure correct
measurement or intended to facilitate the measurement operations) used by independent trade
partners can be assumed to be at least 0.5%. No further evidence of uncertainty is required.

However, this is only one part of the uncertainty assessment, and the uncertainty of the liquid fuel
must be determined by assessing the whole metering system, as described in Example 7. Stock
leaving the storage tanks may be measured in a variety of ways, such as by using volumetric meters,
automatic tank level gauges or manual dips. If these devices are maintained in accordance with the
manufacturer’s recommendations, technical specifications or codes of practice 14 operators can use
the uncertainty quoted in the manufacturer’s specifications and calibration certificates as part of their
uncertainty assessment.

If the operator’s liquid fuel is a de-minimis or marginal source stream and if an uncertainty
assessment has not been undertaken which involves both the uncertainty of the tanker meter AND
the bulk tank meter/dip their regulator will accept an overall uncertainty as ‘not applicable’ (N/A) and
list the source stream category as no tier.

13 Schedule 6 to the Measuring Instruments Regulations 2016 (SI 2016 No 1153)

https://www.legislation.gov.uk/uksi/2016/1153/contents and section 4.10.2 HMRC Reference: Notice 179 (February

2014). https://www.gov.uk/guidance/motor-and-heating-fuels-general-information-and-accounting-for-excise-duty-and-
vatexcise-notice-179#accuracy-requirements

14Such as those mentioned in section 4.7 and 4.10 of the HMRC Reference: Notice 179 and produced by the Energy
Institute.
31
2.5 National Legal Metrological Control for weighing devices
The maximum permissible error for weighing devices is dependent on many factors, including the
type of weighing device, the accuracy class and the measurement intervals and range 15. This is too
complex to include in this document and operators should refer to the legislation.

A simpler option for weighing devices is to use the calibration route that is described in section
2.2.4.1. The value of the conservative adjustment factor, as described in section 2.2.4.4, is
dependent on the risk that the performance of the weighing device deteriorates while in use. This
risk can be reduced through maintenance and control procedures.

2.6 Uncertainty for calculation factors

2.6.1 Frequency of analysis

The tiers for calculation factors (as defined in Article 3(7) of the MRR) are not based on uncertainty
thresholds being met but are defined on the basis of using default values (tier 1 or 2) or values
derived from laboratory analyses (usually tier 3). However, determinations involving laboratory
analyses are linked to required frequency for analyses, as set out in Article 35 of the MRR and Annex
VII. One option allowed for determining the required frequency is expressed in terms of the
‘uncertainty’ related to the frequency of analyses. Article 35(2) of the MRR states:

“The regulator may allow the operator to use a frequency that differs from those referred to in
paragraph 1, where minimum frequencies are not available or where the operator demonstrates one
of the following:

a) based on historical data, including analytical values for the respective fuels or materials in the
reporting period immediately preceding the current reporting period, any variation in the analytical
values for the respective fuel or material does not exceed 1/3 of the uncertainty value to which the
operator has to adhere with regard to the activity data determination of the relevant fuel or material…“

This means that the operator may apply a different frequency to that listed in Annex VII of the MRR if
any variation in the analytical values for the respective fuel or material does not exceed one third
(1/3) of the uncertainty value. The determination of this variation must be based on historical data,
including analytical values for the respective fuel or material in the reporting period immediately
preceding the current reporting period.

The changes to the MRR from 1 January 2021 put the indirect analysis of the emission factor and
carbon content on equal footing with direct analysis if the operator can justify it with an uncertainty

15 For details see Schedule 6 of the measuring Instruments Regulations 2016 (SI 2016 No 1153)

https://www.legislation.gov.uk/uksi/2016/1153/contents for automatic weighing devices and the Non-automatic Weighing

Instruments Regulations 2000 (SI 2000 No. 3236) for non-automatic weighing devices
https://www.legislation.gov.uk/uksi/2000/3236/contents/made. It is important to note that Regulation 4.2 of the Non-
automatic Weighing Instruments Regulations 2000 specifies that the MPES is twice the MPE. The MPES for automatic
weighing devices is the same as the MPE.
32
assessment. Sections 2.1 and 3.1 of Annex II of the MRR both state for tier 3, point (b): [The operator
may use] ‘…the empirical correlation as specified for tier 2b, where the operator demonstrates to the
satisfaction of the regulator that the uncertainty of the empirical correlation does not exceed 1/3 of the
uncertainty value to which the operator has to adhere with regard to the activity data determination of
the relevant fuel or material.’

Any variation in the analytical value may be determined as the overall uncertainty of uncorrelated
input quantities (see section 8.2.1 of this document).

�(𝑢𝑢1 ⋅ 𝑥𝑥1 )2 + (𝑢𝑢2 ⋅ 𝑥𝑥2 )2 +. . . +(𝑢𝑢𝑛𝑛 ⋅ 𝑥𝑥𝑛𝑛 )2

𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 =
|𝑥𝑥1 + 𝑥𝑥2 +. . . +𝑥𝑥𝑛𝑛 |

Where:

𝑢𝑢𝑖𝑖 ….. relative uncertainty of the analytical value of sample 𝑖𝑖

𝑥𝑥𝑖𝑖 ….. samples size of sample 𝑖𝑖

Assuming that the uncertainty of the analytic value of each sample is the same and all sample sizes
are similar, the formula simplifies to:

√𝑛𝑛 𝑢𝑢𝑖𝑖
𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑢𝑢𝑖𝑖 ⋅ =
𝑛𝑛 √𝑛𝑛

Where:

𝑛𝑛 …. Number of samples

If the total uncertainty related to the analytical values is known (in most cases it is a direct result of
the standard deviation of the analytical values) the requirement minimum number of samples can be
determined as:

𝑢𝑢𝑖𝑖2
𝑛𝑛 = 2
𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡

Alternatively, operators can use the Excel frequency of analysis tool16.

16 https://ec.europa.eu/clima/document/download/0d1499ab-1808-413d-92c3-af85bfbee9b2_en
33
Example

A category B installation is burning heavy fuel oil. In the monitoring plan heavy fuel oil is listed
as a major source stream to be monitored by a calculation-based approach. The MRR (and
approved monitoring plan) requires it to meet tier 4 (±1.5%) for activity data and to determine
the calculation factors (emission factor (EF) and net calorific value (NCV)) by laboratory
analyses in accordance with Articles 32 to 35 of the MRR. The ‘1/3’ rule requires that the
uncertainty related to the determination of the calculation factors does not exceed 0.5% (𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡
– this is the input parameter for determining the number of samples).

Annex VII of the MRR requires a minimum frequency of analysis for heavy fuel oil of at least
six times a year. The operator inputs details of historic analyses into the Excel tool to
demonstrate that the uncertainty related to the determination of the NCV is 1.00%. The
following table displays the results from historic samples.

Number of samples NCV [GJ/t]

1 42.28

2 42.41

3 42.35

4 42.68

5 42.44

6 42.4

7 42.68

8 42.6

9 42.02

10 42.33

11 42.41

12 42.2

average 42.4

Uncertainty 𝑈𝑈𝑖𝑖 1.00%

34
 The uncertainty is determined as the standard deviation of the data series (0.45%)
multiplied by the Student t-factor for 12 values and a 95% confidence interval
(coverage factor = 2.201). The minimum frequency of analysis to meet the
requirements of the 1/3 rule is then calculated by:

1.0%2
𝑛𝑛 = =4
0.5%2

In this case, for NCV determination, the operator may be allowed to apply a lower
frequency of analysis of 4 times per year instead of 6 times. For the emission
factor a similar analysis can be carried out to show whether these requirements
are also fulfilled with 4 samples analysed per year.

2.6.2 Using supplier information

Operators choosing to outsource the determination of calculation factors, by using information

provided by the trade partner supplying the fuel or material, are still responsible for to demonstrating
compliance with the required tiers (including the requirements of Articles 32 to 35 of the MRR). The
example below shows some steps that an operator could follow to use their supplier’s data for
calculation factors to comply with tier 3.

Step 1. Can evidence be provided that an appropriate sampling plan is in place and that
analyses are carried out by a suitably accredited laboratory (EN ISO/IEC 17025) for that
determinant or by a laboratory meeting the equivalent requirements? If the answer is ‘yes’ go to
‘step 2’, otherwise go to ‘step 3’.

Step 2. the operator is deemed to meet tier 3 for all relevant calculation factors for which this
evidence has been provided and there is no need to apply step 3.

Step 3. If the answer to step 1 is ‘no’, then the analytical values obtained from the supplier
cannot be considered to meet tier 3. The operator can choose:

a. To take their own samples and analyse in accordance with Articles 32 to 35, or

b. To use available default values and justify to the regulator’s satisfaction that carrying out
sampling and analysis is technical not feasible or would incur unreasonable costs. As part of an
unreasonable cost justification, the operator should consider whether it is possible to apply tier 3
but with a lower frequency of analysis (see section 2.6.1)

Step 4. If the operator can’t meet step 3 and meet at least tier 1 (step 3b), they must justify to
their regulator’s satisfaction, based on unreasonable costs and/or technical feasibility, that
applying no tier is the only remaining option, as set out in MRR Article 22.
35
Operators must manage their use of supplier(s) data according to their written procedure for control
of out-sourced processes, as set out in Articles 59(3)(f)) and 65 of the MRR. See Chapter 6 of this
document for more guidance on uncertainty and quality assurance.

36
3 Uncertainty for measurement-based
approaches
Emissions of nitrous oxide (N2O) and transfers of carbon dioxide (CO2) and N2O must be
determined using measurement-based methodologies (CEMS). Emissions of CO2 may also be
monitored using measurement-based methodologies if the operator can demonstrate that the tiers
set out in Article 41 and Annex VIII of the MRR are complied with.

Operators applying measurement-based approaches must list of all relevant equipment in their
monitoring plan, indicating its measurement frequency, operating range and uncertainty. Unlike
calculation-based approaches, there are no simplifications for measurement-based approaches to
determine the uncertainty.

Article 42 of the MRR requires all measurements to be carried out applying methods based on the
following standards:

• EN 14181 Stationary source emissions – Quality assurance of automated measuring systems

• EN 15259 Air quality – Measurement of stationary source emissions – Requirements for
measurement sections and sites and for the measurement objective, plan and report
• EN ISO 16911-2 – Manual and automatic determination of velocity and volume flow rate in
ducts.
And other corresponding EN standards referred to in these standards, such as
• EN ISO 14956 Air quality – evaluation of the suitability of a measurement procedure by
comparison with a required measurement uncertainty. This is required by EN 14181 and
describes the QAL 1 procedure.
• EN 15267-3 Air quality – certification of automated measuring systems for monitoring
emissions from stationary sources. This standard is required to carry out the quality assurance
level 1 (QAL 1) procedure. It is an application of EN ISO 14956 and is used to define testing
procedures for CEMS and the determination of uncertainties in the measurement.
Article 42 further states: ‘Where such standards are not available, the methods shall be based on
suitable ISO standards or national standards. Where no applicable published standards exist,
suitable draft standards, industry best practice guidelines or other scientifically proven methodologies
shall be used, limiting sampling and measurement bias.

The operator shall consider all relevant aspects of the continuous measurement system, including the
location of the equipment, calibration, measurement, quality assurance and quality control.’

EN 14181 contains information about quality assurance procedures (QAL 2 and 3) to minimise the
uncertainty as well as guidelines on how to determine the uncertainty itself. Guidance for QAL 1 can
be found in EN ISO 14956.

37
As a minimum, quality assurance checks, including parallel measurements with standard reference
methods, must be performed once a year by competent staff. In England and Wales, for example,
competence can be demonstrated by certification of the individual to the Monitoring Certification
Scheme (MCERTS) but this is not mandatory throughout the UK. See Monitoring emissions to air
land and water https://www.gov.uk/government/publications/mcerts-personnel-competency-standard-
manual-stack-emissions-monitoring

Operators must notify their regulator if quality assurance requirements are not met and take
appropriate corrective action as soon as is reasonably practicable.

3.1 How to demonstrate compliance with tier requirements

The tiers for CEMS relate to the maximum permissible uncertainties for the annual average hourly
emissions. They are calculated in accordance with the equation below, which is derived from
equations 2 (a, b and c) in Annex VIII, section 3 of the MRR:

∑𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑜𝑜𝑜𝑜𝑜𝑜 𝑖𝑖 [𝑔𝑔⁄𝑁𝑁𝑚𝑚3 ] . 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑔𝑔𝑔𝑔𝑔𝑔 𝑓𝑓𝑓𝑓𝑓𝑓𝑤𝑤𝑖𝑖 [𝑁𝑁𝑚𝑚3⁄ℎ]

𝐺𝐺𝐺𝐺𝐺𝐺 𝐸𝐸𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑎𝑎𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 ℎ𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 [𝑘𝑘𝑘𝑘⁄ℎ] =
ℎ𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑜𝑜𝑜𝑜 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 .1000

Values for the greenhouse gas (GHG) concentration and the flue gas flow must be consistent and
relate to the same conditions, for example to dry flue gas at standard conditions.

The uncertainty associated with the determination of the concentration is combined with the
uncertainty associated with the determination of the flue gas flow:

𝑢𝑢𝑎𝑎𝑎𝑎 ℎ𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = �𝑢𝑢𝐺𝐺𝐺𝐺𝐺𝐺 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑛𝑛2 + 𝑢𝑢𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑔𝑔𝑔𝑔𝑔𝑔 𝑓𝑓𝑓𝑓𝑓𝑓𝑤𝑤2

The combined standard uncertainty is multiplied by a coverage factor to obtain the expanded
uncertainty. The resulting expanded uncertainty associated with the average hourly emissions is
compared to the uncertainty associated with the tier required by the MRR for the relevant emissions
source (see Annex VIII section 1 of the MRR).

This calculation can be performed using the uncertainty associated with the determination of the
concentration obtained by the QAL1 procedure. For some CEMS this uncertainty is readily available
where a QAL1 calculation is attached to an EN 15267-3 certification.

If the CEMS fails to meet the uncertainty threshold of the tier required by the MRR using the
uncertainty obtained by QAL1, the operator should either

• use another CEMS

• demonstrate that it is technically not feasible or they would incur unreasonable costs
However, the uncertainty associated with the determination of the concentration obtained by the
QAL2 procedure is the relevant input parameter for demonstrating compliance with the MRR. Only if
the CEMS also fails to meet the uncertainty threshold of the tier required by the MRR, obtained by

38
QAL2, do the bullet points above then become mandatory. Note that QAL2 does not take into
consideration uncertainty resulting from drift since this is addressed by QAL1 and QAL3.

Further guidance on CEMS and the application of EN 14181 can be found in specific guidance on
Continuous Emissions Monitoring Systems17.

17 https://ec.europa.eu/clima/system/files/2021-10/policy_ets_monitoring_gd7_cems_en.pdf
39
4 Uncertainty for fall-back approaches
Operators may apply a monitoring methodology not based on tiers for selected source streams or
emissions sources (also called a fall-back methodology) if all the conditions set out in Article 22 of the
MRR are met:

• to apply at least tier 1 under the calculation-based methodology for one or more major source
streams or minor source streams and a measurement-based methodology for at least one
emissions source related to the same source streams is not technically feasible or would incur
unreasonable costs
• the operator assesses and quantifies each year the uncertainties of all parameters used for
the determination of the annual emissions in accordance with the ISO Guide to the Expression
of Uncertainty in Measurement (GUM) (JCGM 100:2008) 18 or another equivalent
internationally accepted standard, and includes the results of the uncertainty assessment in
the annual emissions report
• the operator demonstrates to the satisfaction of the regulator that by applying such a fall-back
monitoring methodology, the overall uncertainty threshold for the annual level of greenhouse
gas emissions for the whole installation does not exceed:
o 7.5% for category A installations
o 5.0% for category B installations
o 2.5% for category C installations
Further guidance for assessing the uncertainty of a fall-back approach can be found in section 8.4 of
Annex II of this document.

18https://isotc.iso.org/livelink/livelink/fetch/2000/2122/4230450/8389141/ISO_IEC_Guide_98-3_2008%28E%29_-
_Uncertainty_of_measurement_--
_Part_3%2C_Guide_to_the_expression_of_uncertainty_in_measurement_%28GUM%2C1995%29.pdf?nodeid=8389142
&vernum=-2
40
5 Uncertainty in the Free Allocation
Regulation
The requirements for taking uncertainty into account are different for complying with the Free
Allocation Regulation (FAR) compared to the Monitoring and Reporting Regulation (MRR). Annex VII
section 4 of the FAR lists a hierarchy of preferred data sources ranked according to accuracy instead
of defining tiers.

Article 7 of the FAR requires the operator to use ‘data sources representing highest achievable
accuracy pursuant to section 4 of Annex VII’. Use of other data sources is allowed in cases where the
use of most accurate data sources is technically not feasible, would incur unreasonable costs, or
where the operator can provide evidence that another chosen method exhibits equivalent to or lower
uncertainty.

Unlike in the MRR where uncertainty assessments are mandatory, for the FAR a (simplified)
uncertainty assessment is required only to provide a reason to deviate from the main hierarchy of
data sources. However, like the MRR, the uncertainty also uses a 95% confidence interval (Article
2(16) of the FAR).

A full uncertainty assessment, as discussed in Chapter 2, must consider:

1. How the instrument’s readings are used for calculating the parameter under consideration
(for example, how individual measurements contribute to the uncertainty over the whole
reporting year). In the case of indirect determinations, the error propagation law must be
applied accordingly for individual measurements.

2. The instrument’s specified uncertainty (based on maximum permissible error given in

legislation, or the producer’s specifications, or taken from a calibration certificate, etc.)

3. Factors that influence the uncertainty in use (for example, whether the use environment is in
accordance with the specifications, whether ageing, corrosion, or other systematic sources of
error play a role, etc.)

4. Further factors, such as conservative adjustment factors for unknown sources of error.

When carrying out a simplified uncertainty assessment, operators should use expert judgement (such
as their experience gained from assessing uncertainty for the purpose of annual emissions
monitoring and reporting) to decide which of the factors mentioned in points (3) and (4) above can be
disregarded, if not easily accessible. For example, if there is information available about the
‘maximum permissible error in service’, there is no need to include a safety margin as environmental
and use factors affecting the instrument have already been accounted for. If the instrument is not
installed in an environment appropriate for its use specifications, the operator should apply
reasonable efforts to assess at least some more important influencing factors.

41
For more information, see Guidance on Monitoring and Reporting in Relation to the Free Allocation
Rules 19.

19 https://ec.europa.eu/clima/system/files/2019-02/p4_gd5_mr_guidance_en.pdf
42
6 Uncertainty and quality assurance
All operators must establish, document, implement and maintain effective control systems to ensure
that their data conforms to the MRR and their approved monitoring plan, including the quality
assurance of measurement equipment (Articles 59(3)(a) and 60 of the MRR) to ensure that their
reported data are accurate and reliable.

Information on quality assurance of measurement-based instruments can be found in Chapter 3.

Operators can demonstrate that measuring instruments subject to national legal metrological meet
the requirement set out in Article 60 of the MRR to check against traceable international standards by
referring to the official calibration certificate.

If components of the measurement systems can’t be calibrated, the operator must propose
alternative control activities in their monitoring plan and carry out a full uncertainty assessment (route
CO-3/CT-3) (see sections 2.2.5 and 2.3.4).

Section 9.3 gives an example of a quality assurance procedure for measurement equipment.

Procedures, such as those for managing outsourced processes (Article 65 of the MRR) are a
practical and flexible way of managing quality assurance if data is obtained from multiple suppliers,
without having to modify the monitoring plan each time a supplier change. The only constraint is that
the overall procedure must stay within the description of the procedure set out in the approved
monitoring plan, as shown in the example below.

Example:

Heavy fuel oil is delivered on trucks owned by different suppliers. The volume flow meters used
for determining the purchased amounts are all installed on the trucks, hence outside the
operator’s own control.

A procedure will be established for keeping track of all measuring instruments involved for
determining the activity data of this source stream. A summary of this procedure may contain
the following elements:

a. Responsible post or department: for example, the shift manager in charge accepting the fuel
delivery.

b. For each delivery at least the following will be documented:

c. Truck number plate

d. Name of the truck’s company

e. ID of the volume flow meter

43
 f. Delivered amount

g. Responsible person checks if this truck and/or volume flow meter already has an account in
the internal database

h. Responsible person checks once a month whether evidence for flow meter uncertainties has
been provided by all suppliers, for example, the latest (official) calibration certificate. If not,
responsible person will request such evidence from those suppliers where evidence is
missing.

i. Where relevant information is stored.

Please note that this procedure must allow tracking of all measuring instruments involved to an extent
allowing calculation of the uncertainty over the whole period and to demonstrate compliance with the
required tier. If this is not achieved, the operator must propose alternative monitoring methods (see
section 2.6.2) or provide justifications, for example, unreasonable costs.

Notwithstanding the continued need for suitable procedures, an alternative option for demonstrating
compliance with the tier requirements can be achieved by providing documents clearly demonstrating
which accuracy classes can be used, for example, a contractual arrangement with the supplier
demonstrating that only measurement instruments with certain accuracy classes are to be used.

44
7 ANNEX I: Conservative measurement
uncertainties for the most common
measuring instruments
The following tables provide an overview of conservative measurement uncertainties (expressed as
expanded uncertainties) for certain categories of common measuring instruments as far as putting
the instrument on the market and into use is concerned.

Due to the hostile nature of the environment, the recalibration intervals listed in the tables are not
suitable for instruments on offshore platforms.

The uncertainty values and additional conditions presented in the tables below should be considered
only if more specific information is not available from the manufacturer of the measuring instrument,
or from normative documents such as those published by OIML 20. Where standards are dated,
operators must check the latest version of the standards for up-to-date information. Also, these
uncertainty values should be considered only if steps 1 to 4 in section 2.2.3 are met. If this is not the
case route CO-2a (where the maximum permissible error in service can be used as the overall
expanded uncertainty) can’t be applied. For measuring instruments suitable for gases and liquids
relevant OIML documents are R137 and R117. For measuring instruments for solids R76 is a suitable
source.

Please also note that an interval for recalibration is advised for each instrument and is subject to the
regulator’s approval. This implies that after each calibration the requirements to apply simplification
route CO-2b (where the expanded uncertainty from calibration multiplied by a conservative
adjustment factor can be used as the overall expanded uncertainty) in section 2.2.4 might be
applicable and provide more reliable results. This option should always be considered before
applying standard values listed below.

The uncertainty values provided in this Annex covers most but not all the factors that could influence
uncertainty as they do not take account the effects of the measuring instrument being in service. For
example, the values do not take drift into account that may arise from being used. For example, the
uncertainty could increase due to aging of the instrument or being used in a hostile environment that
leads to corrosion between calibration/maintenance intervals

The overall uncertainty ( 𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 (𝑖𝑖𝑖𝑖 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) ) must consider the drift by applying the following formula:

𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 (𝑖𝑖𝑖𝑖 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) = �𝑈𝑈𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑖𝑖𝑖𝑖 𝑡𝑡ℎ𝑖𝑖𝑖𝑖 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 + 𝑈𝑈𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑

where 𝑈𝑈𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑖𝑖𝑖𝑖 𝑡𝑡ℎ𝑖𝑖𝑖𝑖 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 refers to the uncertainty figures provided in this annex and 𝑈𝑈𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 refers to
the additional uncertainty caused by the drift. The 𝑈𝑈𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 should be determined based on robust data

20Documents containing technical specifications adopted by the Organisation Internationale de Métrologie Légale
(OIML). http://www.oiml.org/
45
on common drift observed for similar instruments between calibration/maintenance intervals. If such
values cannot be obtained, the operator should assume conservative values for 𝑈𝑈𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑡𝑡 between 5%
(for example in a non-corrosive, low-dust environment) and 7.5% (for example in a corrosive, high-
dust environment).

Table 5

Rotor meter

Medium: gas

Relevant standards: EN 12480:2015+A1:2006

Uncertainty for 0-20% of the measurement range: 3%

Uncertainty for 20-100% of the measurement range: 1.5%

Conditions:

• Once per 10-year cleaning, recalibration and if necessary adjusting

• Annual inspection of the oil level of the carter
• Application filter for polluted gas
• Life span 25 years

Medium: liquid

Uncertainty for 0-10% of the measurement range: 1%

Uncertainty for 10-100% of the measurement range: 0.5%

Conditions:

• Once per 5-year cleaning, recalibration and if necessary adjusting (or at an earlier time when
flow liquid of 3500 hours × maximum range of the meter has run through the meter
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 25 years

46
Table 6

Turbine meter

Medium: gas

Relevant standards: EN 12261:2002 + A1:2006

Uncertainty for 0-20% of the measurement range: 3 %

Uncertainty for 20-100% of the measurement range: 1.5%

Conditions:

• Once per 5-year cleaning, recalibration and if necessary adjusting - Annual visual inspection
• Once per three months lubrication of bearings (not for permanent lubricated bearings)
• Application filter for polluted gas
• No pulsating gas stream
• Life span 25 years
• No overload of longer than 30 minutes › 120% of maximum measurement range

Medium: liquid

Uncertainty for 10-100% of the measurement range: 0.5%

Conditions:

• Once per 5-year cleaning, recalibration and if necessary adjusting

• Once per three months lubrication of bearings (not for permanent lubricated bearings)
• Application filter for polluted liquid
• Life span 25 years
• No overload of longer than 30 minutes › 120% of maximum measurement range

Table 7

Bellows meter / diaphragm meter

Medium: gas

Relevant standards: EN 1359:1998 + A1:2006

47
Uncertainty for 0-20% of the measurement range: 7.5%

Uncertainty for 20-100% of the measurement range: 4.5%

Conditions:

• Once per 10-year cleaning, recalibration and if necessary adjusting

• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 25 years

Table 8

Orifice meter

Medium: gas and liquid

Relevant standards: EN ISO 5167

Uncertainty for 20-100% of the measurement range: 3%

Conditions:

• Annual calibration of the differential pressure transmitter

• Once per 5 years calibration of the orifice meter
• Annual inspection for abrasion orifice and fouling
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 30 years
• No corrosive gases and liquids
• Guidelines for building in orifices, if not stated otherwise by the manufacturer: minimum of
50D free input flow length before the orifice and 25D after the orifice: smooth surface of
inner wall.

48
Table 9

Venturi meter

Medium: gas and liquid

Relevant standards: EN ISO 5167

Gas: Uncertainty for 20-100% of the measurement range: 2%

Liquid: Uncertainty for 20-100% of the measurement range: 1.5%

Conditions:

Annual calibration of the pressure transmitter

• Once per 5 years calibration of entire measuring instrument

• Annual visual inspection
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 30 years
• No corrosive gases and liquids

Table 10

Ultrasonic meter

Medium: gas and liquid

Relevant standards: ISO 17089-1:2010

Medium: gas

Gas: Uncertainty for 1-100% of the measurement range: 2%

Gas (clamp on): Uncertainty for 1-100% of the measurement range: 4%

Medium: liquid

Uncertainty for 1-100% of the maximum measurement range: 3%

Conditions:

49
 • Once per 5 years cleaning, recalibration and, if necessary, adjusting
• Annual inspection of contact between transducer and tube wall. When there is not sufficient
contact, the transducer assembly must be replaced according to the specifications of the
manufacturer.
• Annual inspection on corrosion of wall
• Annual inspection of transducers
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 15 years
• No disturbances in frequencies
• Composition of medium is known
Guidelines for building in ultrasonic meters, if not stated otherwise by the manufacturer: minimum
of 10D free input flow length before the meter and 5D after the meter.

Table 11

Vortex meter

Medium: gas and liquid

Gas: Uncertainty for 10-100% of the measurement range: 2.5%

Liquid: Uncertainty for 10-100% of the measurement range: 2%

Conditions:

• Once per 5 years cleaning, recalibration , recalibration and, if necessary, adjusting

• Annual inspection of sensors
• Annual inspection of bluff body
• Annual inspection on corrosion of wall
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 10 years
• Set-up is free of vibration
• Avoid compressive shocks

50
 Guidelines for building in vortex meters, if not stated otherwise by the manufacturer: minimum of
15D free input flow length before the meter and 5D after the meter

Table 12

Coriolis meter

Medium: gas and liquid

Gas: Uncertainty for 10-100% of the measurement range: 1.5%

Liquid: Uncertainty for 10-100% of the measurement range: 1%

Conditions:

• Once per 3 years cleaning, recalibration and, if necessary, adjusting

• Stress-free installation
• Monthly control of adjusting zero point
• Annual inspection of corrosion and abrasion
• Annual check on sensors and transmitters
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 10 years

Table 13

Oval gear meter

Medium: liquid

Uncertainty for 5-100% of the measurement range: 1%

Conditions:

• Viscid liquids (oil): Once per 5 years cleaning, recalibration and, if necessary, adjusting
• Thin liquids: Once per 2 years cleaning, recalibration and, if necessary, adjusting
• Annual inspection of abrasion

51
 • Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 30 years

Table 14

Electronic Volume Conversion Instrument (EVCI)

Medium: gas

Relevant standards: EN 12405-1:2005+A1:2006

Uncertainty for 0.95 -11 bar and -10 – 40°C: 1%

Conditions:

• Once per 4 years recalibration and, if necessary, adjusting

• Replace batteries (frequency is dependent on instructions manufacturer)
• Annual maintenance according to instructions of manufacturer / general instructions
measurement principle
• Life span 10 years

52
8 ANNEX II: Full uncertainty assessment for
source streams
8.1 Introduction
This annex provides an overview of the general approach to assess uncertainties if no simplifications
are applicable. For further details operators may consult the ISO Guide to the Expression of
Uncertainty in Measurement (GUM) (JCGM 100:2008) (see footnote 18).

In principle the uncertainty assessment must include:

• the specified uncertainty of the applied measuring instrument

• the uncertainty associated with the instrument’s calibration
• any additional uncertainties connected to how the measuring instrument is used in practice
• the application of the appropriate coverage factor (for example, a factor of 2 to obtain the
expanded uncertainty)
If additional measurements such as pressure and temperature are required, the uncertainty of these
measurements must also be considered. If the uncertainty information from the manufacturer cannot
be applied, the operator must substantiate and justify that any deviations from the specification do not
influence the uncertainty. If this is not possible, they must make conservative and substantiated
estimations of the uncertainty. Possible influences on the uncertainty include:

• Deviation from specified working range

• Different uncertainties subject to load or flow rate
• Atmospheric conditions (wind, temperature variation, humidity, corroding substances)
• Operational conditions (adhesion, density and viscosity variation, irregular flow rate, in-
homogeneity)
• Installation conditions (raising, bending, vibration, wave)
• Using the instrument for a medium other than the one it is designed for
• Calibration intervals
• Long-term stability
The general focus should be on the most significant parameters such as temperature, pressure
(difference), flow rate, viscosity, etc., whichever applicable. Significant influences on the uncertainty
must be considered and evaluated. The uncertainty can be calculated with the appropriate error
propagation formula. Examples for the calculation of a specific uncertainty are given in this annex.

Tables 15 and 16 list influencing parameters that might be relevant for uncertainty assessment.
There may be other parameters that are not listed here and there may be some parameters that can
53
be dismissed from consideration as they are likely to have minimal impact upon the results. However,
it is a useful first starting point when running a risk assessment about the uncertainty of activity data
and may help operators focus on the most relevant influencing parameters. Tables 16 and 17 provide
some measuring instrument specific influencing parameters, depending on whether the substance
being measured is a gas, liquid or solid.

Table 15 Influencing parameters on the determination of activity data

Gaseous source
Fluid source stream Solid source stream
stream

Influencing parameter turbulences in gas turbulences in fluid exposure to wind and

related to the stream impacts of stream, bubbling of radiation
instrument and its cladding temperature dissolved gases
installation of environment temperature of
temperature of environment
long-run behaviour environment
(calibration and long-run behaviour
maintenance long-run behaviour (calibration and
frequency) (calibration and maintenance
maintenance frequency)
acceptable frequency)
measurement range position on scale
acceptable
electromagnetic fields measurement range electromagnetic fields

electromagnetic fields storage capacities /

volumes
storage capacity and
monitoring slope of conveying
belts
phase changes
start and stop
behaviour

acceptable
measurement range

storage capacity and

monitoring

vibration

54
 Influencing parameter temperature temperature purity / humidity
related to the medium
being measured pressure density accessibility as net
weight (e.g.
compressibility factor viscosity packaging)

dewpoint (for some boiling or melting point handling of medium

gases only) (for some rare
circumstances only) impacts by drying
corrosiveness density
corrosiveness
flow characteristics
(e.g. related to grain
size)

adhesiveness

melting point (for some

rare constellations
only)

Table 16 Measuring instrument specific influencing parameters and ways to validate/mitigate

them: metering of gases and liquids

Measuring instrument Influencing parameter Validation/mitigation option

Turbine meter Intermittent flow, pulsation Appropriate operating

parameters, avoid pulsation,
e.g. by using controlling
instruments

Bellows meter Correct detection of Use Electronic Volume

temperature and pressure Conversion instrument (EVCI)

Orifice meter, Venturi meter Damages, roughness of the Satisfy EN ISO 5167
pipe, stability of pressure requirements
difference detectors

Ultrasonic meter Strong noise signals Reduce noise

55
 Vortex meter Pulsation Avoid pulsation

Coriolis meter Stress, vibration Build in compensators

Oval gear meter Resonances, pollution Dampers, filters

Table 17 Measuring instrument specific influencing parameters and way to validate/mitigate them:
metering of solids

Measuring instrument Influencing parameter Validation/mitigation option

Conveyor belt weighing Adhesion, sliding if belt is Use horizontal belt

slanted

Wheel loader scale Adhesion Zeroing after each

measurement

Wagon weigh bridge Weighed object not fully on Use big enough scales
scale

Hopper weigher, truck weigher, Wind Use wind protection sites

crane weigher

8.2 Error propagation laws

In many cases the measurand of interest is not measured directly but calculated through a functional
relationship from other input quantities being measured. For example, a volumetric flow (f𝑉𝑉) is
calculated by measuring inputs like density (ρ) and pressure difference (∆p) through the relationship
fV=fV(ρ, ∆p). The uncertainty related to the measurand of interest will then be determined as the
combined standard uncertainty via error propagation.

For input quantities it is necessary to distinguish between:

• Uncorrelated (independent) input quantities

• Correlated (interdependent) input quantities

56
If the operator uses different measuring instruments to determine the activity data of parts of the
source stream, the associated uncertainties can be assumed to be uncorrelated 21.

However, this assumption must be carefully assessed for each case as there may be significant
correlation between two input quantities if the same measuring instrument, physical measurement
standard, or reference datum having a significant standard uncertainty is used.

Example: A gas flow measurement is converted from m³ to Nm³ by taking into account
temperature and pressure which are measured by separate measuring instruments. These
parameters can generally be considered as uncorrelated (see section 8.2.1).

Example: The annual consumption of coal of a coal-fired power plant is determined by

weighing the batches delivered during the year with the same belt weigher. Due to drift-
effects during operation in practice and due to uncertainties associated to the calibration of
the belt weigher, the uncertainties associated with the results of weighing are correlated
(see section 8.2.2).

8.2.1 Uncorrelated input quantities:

If uncorrelated input quantities 𝑋𝑋1 ,..,𝑋𝑋𝑛𝑛 are being used to calculate the measurand 𝑌𝑌 = 𝑌𝑌(𝑋𝑋1,..,𝑋𝑋𝑛𝑛 ) the
uncertainty of 𝑌𝑌 can be determined by:

2 2 2
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
�
𝑈𝑈𝑌𝑌 = � 𝑈𝑈 � + � ⋅ 𝑈𝑈 � + ⋯ + � ⋅ 𝑈𝑈 � 𝑈𝑈𝑌𝑌
𝜕𝜕𝑋𝑋1 𝑋𝑋1 𝜕𝜕𝑋𝑋2 𝑋𝑋2 𝜕𝜕𝑋𝑋𝑛𝑛 𝑋𝑋𝑛𝑛
2 2 2
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
= �� 𝑈𝑈𝑋𝑋1 � + � ⋅ 𝑈𝑈𝑋𝑋2 � + ⋯ + � ⋅ 𝑈𝑈𝑋𝑋𝑛𝑛 �
𝜕𝜕𝑋𝑋𝑖𝑖 𝜕𝜕𝑋𝑋2 𝜕𝜕𝑋𝑋𝑛𝑛

Where

𝑈𝑈𝑌𝑌 …..uncertainty (absolute value) of the measurand Y

𝑈𝑈𝑋𝑋𝑖𝑖 …..uncertainty (absolute value) of the input quantity 𝑋𝑋𝑖𝑖

21Whether or not input quantities are correlated, and if yes to what extent, is not always straightforward to tell. One
statistical approach to identify correlation is to calculate covariances. Further guidance can be found for example in
sections 5.1, 5.2 and F.1.2 of the GUM
57
Example 1 Uncorrelated input quantities
𝑌𝑌 = 𝑌𝑌(𝑋𝑋1 , 𝑋𝑋2 ) is defined by the relationship 𝑌𝑌 = 𝑋𝑋1 . 𝑋𝑋2

The partial derivatives are:

𝜕𝜕𝑌𝑌 𝜕𝜕𝜕𝜕
𝜕𝜕𝑋𝑋1
= 𝑋𝑋2 𝜕𝜕𝑋𝑋2
= 𝑋𝑋1

The absolute uncertainty is then given by:

2 2
𝑈𝑈𝑌𝑌1 = ��𝑋𝑋2 ⋅ 𝑈𝑈𝑋𝑋1 � + �𝑋𝑋1 ⋅ 𝑈𝑈𝑋𝑋2 �

Where:

𝑈𝑈𝑌𝑌 …….absolute uncertainty of measurand Y

𝑈𝑈𝑋𝑋𝑖𝑖 …….absolute uncertainty of input quantity 𝑋𝑋𝑖𝑖

The relative uncertainty is given by:

2 2 2 2
𝑈𝑈𝑌𝑌 �𝑋𝑋2 ⋅𝑈𝑈𝑋𝑋1 � + �𝑋𝑋1 ⋅𝑈𝑈𝑋𝑋2 � 𝑈𝑈𝑋𝑋 𝑈𝑈𝑋𝑋
= 𝑢𝑢𝑌𝑌 = � = �� 𝑋𝑋 1 � + � 𝑋𝑋 2 � = �𝑢𝑢𝑋𝑋21 + 𝑢𝑢𝑋𝑋22
𝑌𝑌 𝑋𝑋12 ⋅ 𝑋𝑋22 1 2

Where:

𝑈𝑈𝑌𝑌 …….relative uncertainty of measurand Y

𝑈𝑈𝑋𝑋𝑖𝑖 …….relative uncertainty of input quantity 𝑋𝑋𝑖𝑖

The square sum of the relative uncertainty of the measurand is therefore simply
determined as the sum of the squares of the relative uncertainties of the input quantities.

58
Example 2: Independent uncertainties of a sum

A steam boiler that produces process steam is operated by burning gas as fuel. The gas
used is supplied to the boiler by ten different pipes. The amount of gas is determined by
ten different standard orifice plates according to EN ISO 5167. The uncertainty associated
with the determination of the annual consumption of gas (uncertainty of a sum) for the
steam boiler is calculated by following formula:

�(𝑈𝑈1 )2 + (𝑈𝑈2 )2 + ⋯ (𝑈𝑈10 )2

𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 =
|𝑥𝑥1 + 𝑥𝑥2 + ⋯ 𝑥𝑥10 |

Where

𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 …total (relative) uncertainty associated with the determination of the gas

𝑈𝑈𝑖𝑖 …….uncertainty (absolute) value of the individual standard orifice plates

𝑋𝑋𝑖𝑖 …….quantities of gas that are measured annually by the different orifice plates

Example 3: Independent uncertainties of a product

A combined heat and power plant has several boilers, all fired by natural gas and no other
fuels. The annual quantity consumed is determined by a measurement system at the
central transfer station (before distribution to the individual boilers) which consists of a
turbine meter, a separate pressure measurement and a separate temperature
measurement. The turbine meter determines the flow rate at operating conditions.

For emissions reporting the standard volume of natural gas is relevant. For the conversion
of operating m³ into standard m³, measurements of pressure and temperature must be
considered. The uncertainty associated with the determination of the natural gas in
standard m³ (uncertainty of a product) is calculated by following formula:

𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = �𝑢𝑢𝑣𝑣2 + 𝑢𝑢𝑡𝑡2 + 𝑢𝑢𝑝𝑝2

Where

𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 ….total (relative) uncertainty associated with the determination of natural gas

𝑈𝑈𝑣𝑣 ……..(relative) uncertainty of the volume measurement

𝑈𝑈𝑡𝑡 ………(relative) uncertainty of the temperature measurement

𝑈𝑈𝑝𝑝 ……..(relative) uncertainty of the pressure measurement

8.2.2 Correlated input quantities:

If correlated input quantities 𝑋𝑋1,..,𝑋𝑋𝑛𝑛 are being used to calculate the measurand Y=Y(𝑋𝑋1,..,𝑋𝑋𝑛𝑛 )
the uncertainty of Y can be determined by:

𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

𝑈𝑈𝑦𝑦 = �� � ⋅ 𝑈𝑈𝑥𝑥1 � + �� � ⋅ 𝑈𝑈𝑋𝑋2 � + ⋯ �� � . 𝑈𝑈𝑋𝑋𝑛𝑛 �
𝜕𝜕𝑋𝑋1 𝜕𝜕𝑋𝑋2 𝜕𝜕𝑋𝑋𝑛𝑛

Where

𝑈𝑈𝑌𝑌 …..uncertainty (absolute value) of the measurand Y

𝑈𝑈𝑋𝑋𝑖𝑖 …..uncertainty (absolute value) of the input quantity 𝑋𝑋𝑖𝑖

Example 4: Correlated input quantities

𝑌𝑌 = 𝑌𝑌(𝑋𝑋1 , 𝑋𝑋2 )is defined by the relationship 𝑌𝑌 = 𝑋𝑋1 . 𝑋𝑋2

If the example above was calculated for correlated input quantities, the relative
uncertainty would be obtained as:

𝑈𝑈𝑦𝑦 = 𝑈𝑈𝑋𝑋 1 + 𝑈𝑈𝑋𝑋2

The relative uncertainty of the measurand is therefore simply determined as the sum of
the relative uncertainties of the input quantities.

Please note that this is only applicable for the very special case where all the input
estimates are correlated with correlation coefficients of 1. If the coefficient is different from
1, more complex functions for covariances are to be considered which are not within the
scope of this document.

For further reading please consult the GUM.

Example 5: Correlated uncertainties of a sum

A power plant is coal-fired. The annual consumption of coal is determined by weighing the
batches delivered during the year with the same belt weigher. Due to drift-effects during
operation in practice and due to uncertainties associated to the calibration of the belt
weigher, the uncertainties associated with the results of weighing are correlated.

Therefore, the uncertainty associated with the determination of the coal (uncertainty of a
sum) is calculated by following formula:

𝑈𝑈1 + 𝑈𝑈2 + ⋯ + 𝑈𝑈𝑛𝑛

𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 =
|𝑥𝑥1 + 𝑥𝑥2 + ⋯ + 𝑥𝑥𝑛𝑛 |

Where:

𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 …..total (relative) uncertainty associated with the determination of coal

𝑈𝑈𝑖𝑖 ……….uncertainty (absolute value) of the belt weigher (𝑈𝑈1 = 𝑈𝑈2 = 𝑈𝑈𝑛𝑛 )

𝑥𝑥𝑖𝑖 ……….quantities of coal of the relative batches

In this case the (relative) uncertainty associated with the determination of coal is equal to
the (relative) uncertainty of the belt weigher.
Example 6: Correlated uncertainties of a product

A mineral industry determines the loss on ignition by weighing the product on a table
scale before and after the burning process. The loss on ignition is the mass difference
after the burning process compared to the initial weight. The uncertainties associated
with the results of weighing are correlated, because the same table scale is used.

Therefore, the uncertainty associated with the determination of the loss on ignition
(uncertainty of a product) is calculated by the following formula:

𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑢𝑢1 + 𝑢𝑢2

Where:

𝑢𝑢𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 ..... is the total (relative) uncertainty associated with the determination of the loss on
ignition

𝑢𝑢1,2 ...... (relative) uncertainty of the mass measurement before and after heating
8.3 More examples

Example 7: Uncertainty of the amount of stored fuel

The overall annual consumption of gasoil is calculated from the aggregated deliveries by
tank trucks. The trucks are equipped with a flow meter on the truck subject to national
legal metrological control with a maximum permissible error of 0.5%. One truck can
deliver 25,000 litres of gasoil. After the annual forecast the operator expects to require
an average of 750,000 litres annually over the next year. Therefore, 30 tank truck
deliveries per year are expected.

The storage tank for gasoil at the installation has a capacity of 40,000 litres. With a
cross section of 8m² the expanded uncertainty of level reading is 2.5% of the total
capacity.

If the storage facilities were incapable of containing more than 5% of the annual quantity
used of the fuel or material being considered, or this was a low emitting installation, the
uncertainty of stock changes can be omitted from the uncertainty assessment (MRR
Article 28(2) and Article 47(5)). However, the storage tank can contain 40,000/750,000 =
5.3% of the annually used quantity and therefore must be considered for the uncertainty
assessment.

The annual quantity Q of gasoil is determined by the following formula

𝑄𝑄 = 𝑃𝑃 − 𝐸𝐸 + (𝑆𝑆𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 − 𝑆𝑆𝑒𝑒𝑒𝑒𝑒𝑒 )

Where:

P ......... Purchased quantity over the whole year

E ......... Exported quantity (e.g. fuel delivered to parts of the installation or other
installations which are not included in the UK ETS)

𝑆𝑆𝑏𝑏𝑏𝑏𝑏𝑏𝑖𝑖𝑖𝑖 ... Stock of the gasoil tank at the beginning of the year

𝑆𝑆𝑒𝑒𝑒𝑒𝑒𝑒 ..... Stock of the gasoil tank at the end of the year

As the quantity of purchased gasoil over the whole year (P) is not determined by a
single measurement but as the sum of many measurements, i.e. 30 truck deliveries, P
can be written as:

𝑃𝑃 = 𝑃𝑃1 + 𝑃𝑃2 + ⋯ 𝑃𝑃30

Where:

𝑃𝑃𝑖𝑖 …..Purchased quantity from one truck

All input quantities for the determination of Q can be considered as uncorrelated.

The level reading on the storage tank cannot be considered as being within one
measurement series because of the long time between the measurements (beginning and
end of the year). However, as it is still the same measuring instrument that is being used, it
is possible for there to be correlation. For the purposes of this example, the assumption is
that it is uncorrelated. To know for sure, an assessment should be carried out, for example
by determining correlation coefficients, in accordance with the GUM.

Assuming that no gasoil is being exported (E=0) the uncertainty can be determined in
accordance with section 8.2.1 as an uncorrelated uncertainty of a sum:

2 2
��𝑈𝑈𝑆𝑆,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 � + �𝑈𝑈𝑆𝑆,𝑒𝑒𝑒𝑒𝑒𝑒 � + (𝑈𝑈𝑃𝑃1 )+ . . + (𝑈𝑈𝑃𝑃 30 )2
𝑈𝑈𝑄𝑄 =
�𝑠𝑠𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 − 𝑠𝑠𝑒𝑒𝑒𝑒𝑒𝑒 + 𝑃𝑃1 + ⋯ + 𝑃𝑃30 �

𝑈𝑈𝑄𝑄 ........ total (relative) expanded uncertainty associated of Q

𝑈𝑈𝑆𝑆,𝑃𝑃 ....... (absolute) expanded uncertainty of the stock level reading or quantity provided by
one tank

The uncertainty related to the stock level reading is the same for both readings. As the
difference between 𝑆𝑆𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 and 𝑆𝑆𝑒𝑒𝑒𝑒𝑒𝑒 cannot be predicted 𝑆𝑆𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 − 𝑆𝑆𝑒𝑒𝑒𝑒𝑒𝑒 can be assumed as
zero. Furthermore, if all 𝑃𝑃𝑖𝑖 are considered as equal quantities having equal absolute
uncertainties the equation simplifies to:

�2 ⋅ (𝑈𝑈𝑠𝑠 )2 + 𝑛𝑛 ⋅ (𝑈𝑈𝑃𝑃𝑃𝑃 )2
𝑈𝑈𝑄𝑄 =
𝑃𝑃

�2 ⋅ (40,000 ⋅ 2.5%)2 + 30 ⋅ (25,000 ⋅ 0.5%)2

𝑈𝑈𝑄𝑄 = = 0.21%
750,000

As mentioned above, this example assumes that input quantities, stock level readings and
meters on all trucks, are not correlated. If the uncertainty of a ‘worst-case’ scenario is to be
calculated (that is, the measurements are correlated) the overall expanded uncertainty
would be 0.57%:

�2 ⋅ (40,000 ⋅ 2.5%)2 + (30 ⋅ 25,000 ⋅ 0.5%)2

𝑈𝑈𝑄𝑄 = = 0.57%
750,000

As the activity data related to gasoil consumption must be expressed in tonnes the density
of the fuel must be taken into account. The expanded uncertainty for determining the bulk
density using representative samples is around 3%. Using the formula from section 8.2.1
for uncorrelated uncertainties of a product leads to:

2 2
𝑈𝑈𝑄𝑄(𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡) = �𝑈𝑈𝑄𝑄(𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣) + 𝑈𝑈𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 = �0.21% + 0. 3%2 = 3.007%
Although the flow metering in this example has a rather low uncertainty, the conversion into
tonnes shows that the influence of the uncertainty of the density determination is the most
significant contribution to the overall uncertainty. Future improvements should therefore
focus on determination of the density with lower uncertainty.

Example 8: Uncertainty for source streams partly transferred to connected

installations not falling under the UK ETS

When the installation is partly covered by UK ETS (for example, because not all parts of
that installation fall within the scope of the scheme), the quantity measurement determined
by an internal sub-meter (for this example the expanded uncertainty is assumed to be 5%)
for the non-UK ETS part may have to be subtracted from the quantity of the source stream
that is measured by the main meter which falls under national metrological control
(maximum permissible error is 2%).

Assuming the installation uses 500,000 Nm³ natural gas per year. Out of that amount of
natural gas 100,000 Nm³ will be transferred and sold to an installation not falling under UK
ETS. To determine the consumption of natural gas of the UK ETS installation, the
consumption of natural gas by that connected installation must be subtracted from the total
natural gas consumption of the installation. To assess the expanded uncertainty for the
natural gas consumption of the UK ETS installation, following calculation is performed:

�(2% ⋅ 500,000)2 + (5% ⋅ 100,000)2

𝑢𝑢𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = = 2.8%
|500,000 + (−100,000)|

Please note, that the uncertainty of the main gas meter under national metrological control
does not have to be assessed. The uncertainty of the internal submeter that is not
guaranteed by national metrological control must be assessed and confirmed before being
able to determine the expanded uncertainty associated with the source stream.
8.4 Uncertainty over the whole installation and where Article
22 of the MRR (fall-back approach) is used
Example 9: Overall uncertainty with a fall-back approach

A category A installation with annual emissions of 35,000 t CO2 exclusively burns natural
gas. The fuel is obtained by a commercial transaction subject to national legal
metrological control so the expanded uncertainty related to the activity data is 2.0%,
using the maximum permissible error allowed by the relevant national legislation. The
2.0% is also the uncertainty related to the total emissions as all calculation factors
applied are default values and, for reasons of simplicity, not influencing uncertainty (see
note below).

Due to an extension of on-site activities, an additional source stream must be added to

the monitoring plan and monitored. The operator provides evidence to the satisfaction of
the regulator that applying at least tier 1, for example installing a measurement system,
is technically not feasible and proposes to use the fall-back approach. The operator
provides evidence in accordance with the GUM that an uncertainty assessment for that
source stream gives an expanded uncertainty of 18%. The expected emissions from that
source stream are 12,000 t CO2 annually.

When applying the fall-back approach to a category A installation the operator must
demonstrate that the expanded uncertainty of the emissions for the whole installation
does not exceed 7.5%. In this example the operator calculates the uncertainty using the
equation

𝐸𝐸𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝐸𝐸𝐸𝐸𝑁𝑁𝑁𝑁 + 𝐸𝐸𝐸𝐸𝐹𝐹𝐹𝐹

where:

𝐸𝐸𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 .. total emissions of the installation

𝐸𝐸𝐸𝐸𝑁𝑁𝑁𝑁 .. emissions resulting from natural gas burning (35,000 t CO2)

𝐸𝐸𝐸𝐸𝐹𝐹𝐹𝐹 ... emissions resulting from the source stream monitored by the fall-back approach
(12,000 t CO2)

As the (relative) uncertainty of the overall emissions can be interpreted as the

uncertainties of a sum, the overall uncertainty is calculated by:

�(2.0% ⋅ 35,000)2 + (18% ⋅ 12,000)2

𝑈𝑈𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = = 4.8%
|35,000 + 12,000|

The expanded uncertainty related to the emissions over the whole installation doesn’t
exceed 7.5%. Therefore, the proposed fall-back approach is appropriate.
Note that default values have uncertainty values attached to them which must be taken into
account by calculating the uncertainty of the source stream from the independent uncertainties
of the product (see example 3) using error propagation.
9 Exemplar: installation with low
emissions
This section describes an uncertainty assessment that is commensurate for a low emitting
installation (as set out in Article 47 of the MRR) and for hospital or small emitters.

9.1 Information about the installation

The example installation produces bricks and pavers and emits on average 15,000 t CO2 per
year. The table below contains details of the source streams that need monitoring, the
estimated emissions, and the minimum tier requirements for monitoring activity data.

Estimated Minimum
Fuel/Material Category emissions (t CO2 / monitoring
year) requirements for
activity data

Light fuel oil Commercial standard fuel 6,500 Tier 1 (± 7.5%)

Clay Ceramics: Method A 8,000 Tier 1 (± 7.5%)

Lignite Other solid fuels (pore- 498 De-minimis

forming agent)

Diesel Other gaseous and liquid 2 De-minimis

fuels (auxiliary power unit)

Light fuel oil:

Fuel is delivered by trucks and stored in tanks (the storage capacity of the tanks is less than
5% of the annual total consumed). In this example the operator can show, by using invoices,
that there are clearly commercial transactions between independent parties and the
measurements used for trading are subject to national legal metrological control (see route
CO-1 or CT-1). Therefore, the maximum permissible error in service, as allowed by relevant
national legislation, can be used.
 Note:

To show compliance the operator must demonstrate evidence that the uncertainty
threshold of the required tier is not exceeded, for example by requesting the trade partner
to provide the official calibration certificate/protocol for the volume flow measurement
instruments installed on the trucks. This evidence will allow verifiers to confirm the validity
of data used to determine that the tier is met.

Although it is highly likely that national legislation will require an uncertainty of at least
7.5% or better, a document confirming that this instrument is subject to national legal
metrological control is needed.

If the national legal metrological legislation also allows measurement instruments with a
higher uncertainty for that purpose, further evidence would be needed. Such evidence
may be documents clearly demonstrating which accuracy classes can be used, for
example, contractual arrangement with the supplier demonstrating that only
measurement instruments with certain accuracy classes are to be used.

Clay

The operator takes the clay directly from the clay pit. No commercial transaction takes place
and any available measurement instrument used is not subject to national legal metrological
control. The operator transports the clay from the pit to the installation by truck and it is
possible that those trucks could be weighed on a weigh bridge owned by the operator.

The operator can simplify the uncertainty assessment here if the measurement instrument is
used in an environment appropriate for its use specifications (see steps 1 to 4, route CO-
2a/2b).

Note

To apply the proposed route CO-2a/2b, the operator must demonstrate that:

1. Information on operating conditions regarding relevant influencing parameters are

available

2. Operating conditions regarding relevant influencing parameters are met

3. It performs quality assured calibration procedures

4. It has further, appropriate, quality assurance procedures for measuring activity data

Compliance with these 4 steps is also relevant for light fuel oil (see above). However, the
obligations for compliance with national legal metrological control will assure that those 4 steps
are met.

The application of these steps is outlined in section 9.2.

In this example, step 1 is met because the manufacturer’s specifications for this weigh bridge
contains information about the appropriate operating conditions.

To demonstrate that the requirements of step 2 are satisfied, the operator could prepare a
simple checklist like the table displayed in section 9.2.

To demonstrate to their verifier compliance with steps 3 and 4, the operator must have an
appropriate procedure for quality assurance of the measurement equipment and ensure that all
relevant measuring equipment is calibrated, adjusted and checked at regular intervals
including prior to use, and checked against measurement standards traceable to international
measurement standards (see the requirements set out in Articles 59(3) and 60(1) of the MRR
in Chapter 6).

According to Article 47(5) of the MRR, the operator of an installation with low emissions is
exempt from taking stock changes into account in the uncertainty assessment. However, this
example includes stock changes in the uncertainty assessment to demonstrate how simple the
calculation is and how marginal the impact of the associated uncertainty is on the overall
uncertainty.

The consumed quantity of clay is calculated as:

𝑄𝑄 = 𝑃𝑃 − 𝐸𝐸 + �𝑆𝑆𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 − 𝑆𝑆𝑒𝑒𝑒𝑒𝑒𝑒 �

Example 7 shows how the uncertainty related to stock changes can be calculated. Section 9.2
demonstrates how the operator of the example installation uses this approach.

When determining CO2 emissions, activity data and all calculation factors must relate to the
same state of the material stream. Annex IV Section 12 of the MRR refers to ‘dry’ clay but
because ‘moisture content’ is not a ‘calculation factor’ as defined in the MRR the moisture
content must be considered as part of the determination of the uncertainty of the activity data
(see calculation in section 9.2). In this example, the clay has the same moisture level (it is all
‘dry’) see example 3 in section 8.2.1 for uncorrelated uncertainties of a product).

Because the determination of moisture content and the emission factor is carried out by
laboratory analyses, a sampling plan covering these parameters must be written, submitted to
and approved by the regulator.

Lignite:

The operator can apply an estimation method to determine the annual emissions from this de-
minimis source stream. They may use purchase invoices to determine the annual activity level.
The UK ETS authority has not published default values for lignite which would allow the use of
tier 2 without any additional effort. Therefore, emissions are obtained by multiplying the amount
lignite used by the net calorific value and emission factor provided in Annex VI of the MRR
(Tier 1).

Diesel:
Diesel is also a de-minimis source stream. Precise measurement would be demanding but fuel
invoices can’t be used because diesel is also used for mobile machinery such as truck loaders,
forklifts etc. The operator can, however, apply an estimation method to determine the diesel
used in the auxiliary power unit. In this example a common formula is proposed:

𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 = 𝐴𝐴𝐴𝐴𝐴𝐴 𝑥𝑥 𝐶𝐶𝐶𝐶𝐶𝐶 𝑥𝑥 (3600 / 109 ) 𝑥𝑥 (1 / 𝑁𝑁𝑁𝑁𝑁𝑁)

𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = 𝐴𝐴𝐴𝐴 𝑥𝑥 𝑁𝑁𝑁𝑁𝑁𝑁 𝑥𝑥 𝐸𝐸𝐸𝐸

Where

AOH .... Annual operating hours

CAP .... Installed capacity of the auxiliary power unit (kW)

AD ....... Activity data (t)

NCV .... Net calorific value (TJ/t, taken from, for example, MRR Annex VI or the UK national
greenhouse gas inventory, if available)

EF ....... Emission factor (t CO2/TJ, taken from, for example, MRR Annex VI or the UK national
greenhouse gas inventory, if available)

9.2 Exemplar uncertainty assessment

Light fuel oil

• Tier applied for activity data: Tier 2 (± 5.0%), based on invoices

• Evidence for complying with the tier requirements: see attached the latest official
calibration certificates for the rotary flow meters on the trucks from three suppliers.
Clay:

• Tier applied for activity data: Tier 2 (± 5.0%), uncertainty achieved = 4.5% (see
calculation below)
• Evidence for complying with the requirements of the tier: route CO-2a/2b is used.
Evidence for complying with the requirements of ‘Step 1’:

See manufacturer’s specification (‘MPES ± 4.0%’) in the weigh bridge’s operating manual; see
sampling plan for determination of the moisture content of the (raw) clay.

Error propagation considering stock changes:

• storage capacity: 7,000 t

• uncertainty related to stock estimation at end of year (conservative estimate): 10%
• average annual amount of clay consumed: 125,000 t
 • maximum permissible error in service laid down in manufacturer’s specifications: 4%
• uncertainty related to determine the moisture content: 2%
Calculation:

2
�2⋅(𝑈𝑈𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 )2 +�𝑈𝑈𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 � �2⋅(7,000⋅10%)2 +(125,000⋅4%)2
𝑢𝑢𝑤𝑤𝑤𝑤𝑤𝑤 = 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 = = 4.08%
125,000

2 2
𝑢𝑢𝑑𝑑𝑑𝑑𝑑𝑑 = �𝑢𝑢𝑤𝑤𝑤𝑤𝑤𝑤 + 𝑢𝑢𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = √4.08%2 + 2%2 = 4.5%

Evidence for complying with the requirements in ‘Step 2’:

Checklist for relevant parameters of the weigh bridge:

Parameter listed in
Value specified by Actual applied
manufacturer’s Compliant?
manufacturer ranges/conditions
specifications

Temperature -15 – 50 °C -15 – 40 °C Yes

Measurement range 2 - 50 tonnes 10 - 35 tonnes Yes

Wind speed < 20 m/s < 15 m/s Yes

Calibration interval Every two years Every two years Yes

Evidence for complying with the requirements in ‘Steps 3 and 4’:

See attached the latest calibration certificates for the truck weigh bridge WB-XYZ123 and the
quality management procedures in section 9.3.

Lignite:

• Tier applied for activity data: Tier 3 (± 2.5%), based on invoices

Evidence 22: see attached the latest official calibration certificates requested from the trading
partners delivering lignite.

Diesel:

• Tier applied for activity data: de-minimis

Approach: Emissions are calculated based on the annual operating hours, the auxiliary power
unit’s installed rated thermal input and the national inventory emission factor for diesel.
Conservative estimates of emissions are typically found to be in the range of 1 to 5 t CO2 per
year.

22 If the certificates are not available, activity data can still be determined using invoices, but it would be a no-tier

approach if the operator can’t demonstrate that it is complying with a tier. This is acceptable in this example as it
is a de-minimis source stream.
9.3 Quality management for the example installation
The example below shows a quality assurance procedure for the measurement equipment.

Example for a procedure

1. The installation is normally shut down between December and February.

Measurement equipment (including that for UK ETS) is usually calibrated during that
time.

2. Responsible person (deputy manager of O&M) maintains a calendar of appropriate

calibration and maintenance intervals for all ETS instruments listed in the measuring
instruments table of the monitoring plan. An alert is set for 30 November of each year.

3. Responsible person (deputy manager of O&M) checks which QM activities are

required according to the calendar within the next 4 weeks. As appropriate, they reserve
resources required for this task in meetings with the plant manager.

4. Calibration and maintenance of ETS instruments is tracked and documented in file

“Z:\ETS_MRV\QM\calibr_log.xls” electronically and hardcopy: Office HS3/27, shelf 3,
Folder identified “QM 27-ETS -nnnn”. (nnnn=year). Information documented contains: ID
of instrument, date when instrument was installed, last calibration, meter reading after
last calibration, laboratory hired for the last calibration, statement of the last calibrations,
date until next calibration is due.

5. For all measurement instruments for which calibration in that particular year is due
the responsible person follows the procedure:

a. Responsible person (deputy manager of O&M) commissions external experts

(calibration laboratory).

b. Responsible person ensures that QM tasks are carried out on the agreed dates.

c. Responsible person keeps records of the above QM activities.

d. Responsible person reports back to plant manager on corrective action required.

Corrective action is handled under procedure QM 28-ETS

<End of procedure>

The procedure is a document independent from the monitoring plan. However, a summary of
the procedure must be included in the monitoring plan in a standardised table. This could be as
follows:
Item according to Article 12(2) Possible content (examples)

Title of the procedure Quality manual (QM) for ETS instruments

Traceable and verifiable reference for QM 27 ETS

identification of the procedure

Post or department responsible for Q&M office

implementing the procedure and the post or
department responsible for the management
of the related data (if different)

Brief description of the procedure • Responsible person maintains a

calendar of appropriate calibration and
maintenance intervals for all
instruments listed in the measuring
instruments table of the monitoring
plan
• Responsible person checks which QM
activities are required. As appropriate,
they reserve resources required for
these tasks in meetings with the plant
manager.
• Responsible person orders external
experts (calibration laboratory and/or
manufacturer service technicians).
• Responsible person ensures that QM
tasks are carried out on the agreed
dates.
• Responsible person keeps records of
the above QM activities.
• Responsible person reports back to
plant manager on corrective action
required, if any.
• Corrective action is handled under
procedure QM 28-ETS, if relevant.

Location of relevant records and information Hardcopy: Office HS3/27, shelf 3, Folder
identified ‘QM 27-ETS – nnnn’. (nnnn=year)
 Electronically:
‘Z:\ETS_MRV\QM\calibr_log.pst’

Name of the computerised system used, MS Outlook calendar, also used for storing
where applicable documents as attachments chronologically

List of EN standards or other standards In the instrument list (document ETS-Instr-

applied, where relevant A1.xls) the applicable standards are listed.
This document is made available to the
verifier upon request.
10 Emissions from catalytic cracking using
continuous emission monitoring systems
(CEMS)
10.1 Background information
Fluid catalytic cracking is used in refineries to convert high-boiling hydrocarbons into lower-
molecular weight but higher value products. During this catalytic reaction part of the carbon-
containing feedstock forms carbonaceous deposits on the catalyst which causes its
inactivation. Therefore, the catalyst must be regenerated by burning off the deposited carbon in
air in a separated reactor, called the regenerator. The carbon in the flue gas formed from this
regeneration is converted into CO2 either during the regeneration or during a subsequent post-
combustion.

For the monitoring of emissions stemming from catalytic cracker regeneration Annex IV section
2 of the MRR states: ‘[..] by way of derogation from Article 24 and 25, emissions from catalytic
cracker regeneration, other catalyst regeneration and flexi-cokers shall be monitored using a
mass balance, taking into account the state of the input air and the flue gas. All carbon
monoxide (CO) in the flue gas shall be accounted for as carbon dioxide (CO2), applying the
mass relation: t CO2 = t CO * 1.571. The analysis of input air and flue gases and the choice of
tiers shall be in accordance with the provisions of Articles 32 to 35 of the MRR. The specific
calculation methodology shall be approved by the regulator.’ This provision clarifies that the
determination of emissions from catalytic cracker regeneration in general requires the use of
appropriate analytical standards and accredited laboratories following the provisions set out in
Articles 32 to 35 of the MRR.

One way to satisfy those criteria can be by application of continuous emissions monitoring
systems (CEMS) following the rules set out in Articles 40 to 46 of the MRR. It must be noted
that the mass balance mentioned in section 2, Annex IV of the MRR is not a ‘real’ mass
balance as defined in Article 25 but rather a flue gas volume balance according to Article
43(5)(a) of the MRR According to Annex IV of the MRR, section 2, the determination of the
annual emissions from the regeneration of catalytic converters from cracking and re-forming
processes must be monitored using a balance, taking into account the CO2, CO, oxides of
nitrogen (NOx) and sulphur dioxide (SO2) contents in the flue gas from the regeneration and in
the amount of air supplied in accordance with Article 43(5)(a) of the MRR.

For CEMS, the annual emissions of the emitted greenhouse gases (GHG) are calculated by
the equation provided in equation 1, Section 3, of Annex IV in the MRR ):

𝐸𝐸𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 [𝑡𝑡] = � 𝐺𝐺𝐺𝐺𝐺𝐺 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑜𝑜𝑜𝑜𝑜𝑜 𝑖𝑖 [𝑔𝑔/𝑁𝑁𝑚𝑚3 ] ⋅ 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑔𝑔𝑔𝑔𝑔𝑔 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑖𝑖 [𝑁𝑁𝑚𝑚3 /ℎ𝑟𝑟] ⋅ 10−6 [𝑡𝑡⁄𝑔𝑔]
𝑖𝑖

where:
GHG𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑜𝑜𝑜𝑜𝑜𝑜 𝑖𝑖 ....... concentrations of GHG in the flue gas flow measured during operation
hour 𝑖𝑖

𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 𝑔𝑔𝑔𝑔𝑔𝑔 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑖𝑖 ........ flue gas flow determined for each hour 𝑖𝑖

For each hour 𝑖𝑖, therefore, the emissions from coke are determined as the product of the
GHG𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑜𝑜𝑜𝑜𝑜𝑜 𝑖𝑖 and the 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑔𝑔𝑔𝑔𝑔𝑔 𝑓𝑓𝑓𝑓𝑓𝑓𝑤𝑤𝑖𝑖 . As the same measurement equipment is usually used
through-out the year for each hour 𝑖𝑖, the uncertainties associated with the emissions calculated
for each hour should be treated as correlated (see section 8.2.2). Consequently, the
uncertainty of the annual emissions is the same as for the emissions of each hour 𝑖𝑖.

In the subsequent processes, a complete conversion of CO to CO2 is assumed:

𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 . 𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑

44.01
𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = �𝑎𝑎𝐶𝐶𝑂𝑂2 + 𝑏𝑏𝐶𝐶𝐶𝐶 � ⋅
22.41 . 1000

𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 ……. overall CO2 emissions from coke burned off in t CO2

𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 …….. greenhouse gas (CO2) concentration in the dry flue gas in g/Nm³

𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 ........ calculated annual volume of the dry flue gas (see calculation below) in Nm³

aCO2 .......... measured carbon dioxide content in dry flue gas in % by volume

bCO ........... measured carbon monoxide content in dry flue gas in % by volume

The volume flow rate of the flue gas to be used in the equation above is usually not measured,
so it must be calculated by a balance. In the regeneration, the coke-loaded catalyst is
regenerated by an air supply and all combustible constituents are converted to CO2, CO, H2O,
NOx and SO2. The calculation of the amount of dry flue gas from the amount of air supplied is
done according to the following formula, assuming a constant inert gas content of 79.07% by
volume:

79.07
𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 = ⋅ 𝑉𝑉
100 − 𝑎𝑎𝐶𝐶𝐶𝐶2 − 𝑏𝑏𝐶𝐶𝐶𝐶 − 𝑐𝑐𝑂𝑂2 − 𝑑𝑑𝑁𝑁𝑁𝑁𝑁𝑁 − 𝑒𝑒𝑆𝑆𝑆𝑆2 𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑟𝑟𝑟𝑟

𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 .......... volume of dry air supplied in Nm³

𝑎𝑎𝐶𝐶𝐶𝐶2 .......... measured carbon dioxide concentration in dry flue gas in % by volume

𝑏𝑏𝐶𝐶𝐶𝐶 ........... measured carbon monoxide concentration in dry flue gas in % by volume

𝑐𝑐𝑂𝑂2 ............ measured oxygen concentration in dry flue gas in % by volume

𝑑𝑑𝑁𝑁𝑁𝑁𝑁𝑁 .......... measured NOx concentration in dry flue gas in % by volume

𝑒𝑒𝑆𝑆𝑆𝑆2 .......... measured SO2 concentration in dry flue gas in % by volume

A prerequisite for the balance shown is that coke contains hardly any nitrogen compounds, or
they are converted into NOx (which is usually the case).

10.2 Determination of the uncertainty

For source streams the uncertainty thresholds set out in the MRR commonly refer to the
determination of activity data. In contrast to that, the uncertainty threshold for emissions from
cracking activities relate to the total annual emissions. Therefore, the uncertainty of 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
must be assessed and compared against the thresholds of the required tier listed in table 1 of
Annex II of the MRR.

As 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 depends on two input quantities, 𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 and the 𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 , the uncertainty
associated with these two components must be assessed:

79.07
𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 = ⋅ 𝑉𝑉
100 − 𝑎𝑎𝐶𝐶𝐶𝐶2 − 𝑏𝑏𝐶𝐶𝐶𝐶 − 𝑐𝑐𝑂𝑂2 − 𝑑𝑑𝑁𝑁𝑁𝑁𝑁𝑁 − 𝑒𝑒𝑆𝑆𝑆𝑆2 𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑

44.01
𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = �𝑎𝑎𝐶𝐶𝑂𝑂2 + 𝑏𝑏𝐶𝐶𝐶𝐶 � .
22.41 . 1000

Step 1: determination of the uncertainty of 𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑

In order to determine the flow rate of the dry flue gas, 𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 , the volume flow of dry air at
standard conditions (𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 ) is needed as well as the composition of the components in the
flue gas, namely the concentrations of CO2, CO, O2, NOx and SO2.

Step 1.1: uncertainty of 𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑

𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 is not measured directly. What is measured is the volume flow of the air supplied at
operating conditions and in the wet state. To convert this parameter into the volumetric flow of
dry air at standard conditions the measurements must be corrected for temperature, pressure
and water vapour content. Therefore, the uncertainty associated with the parameter
𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 can be calculated as the product of uncorrelated input quantities from the measuring
uncertainties of the air flow, the temperature, the pressure and the water vapour content using
the following equation as independent uncertainties of a product (see example 3 in section
8.2.1):

𝑈𝑈𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 = �𝑢𝑢𝑉𝑉2𝑎𝑎𝑎𝑎𝑎𝑎,𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 + 𝑢𝑢2𝑇𝑇 + 𝑢𝑢𝑝𝑝2 + 𝑢𝑢𝑤𝑤

2
Each parameter would be determined using appropriate standards, respectively, which also
cover the determination of associated uncertainties. For this example, the table below
describes the relative uncertainty of each parameter:

Relative uncertainty 𝒖𝒖𝒊𝒊

Parameter (expanded at the 95% confidence
level)

𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎 measured ± 2.0%

Temperature T ± 0.5%

Pressure p ± 0.5%

Water vapour content w ± 1.5%

Using these figures in the formula above leads to an uncertainty related to 𝑉𝑉𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 of ±2.6%.

Step 1.2: uncertainty of flue gas concentration components

Out of the five parameters listed, CO2 and CO usually show the highest concentrations; the
concentration of NOx and SO2 are always very low by comparison. Therefore, related
uncertainties are negligible and can be omitted from the uncertainty assessment without any
significant impact on the result. Oxygen (O2) can be ignored if the measurement is performed
before of the CO post combustion unit but not if the measurement is performed after the CO
post combustion unit.

The concentration of CO2 and CO is determined while emitted from the stack. The applicable
standard for the continuous emissions monitoring system (CEMS) is EN ISO 14181 which also
covers the determination of the associated measurement uncertainty of the concentration.

The table below describes the resulting uncertainties of the CO2 and CO concentrations
obtained by CEMS are as follows:

Parameter Relative uncertainty 𝒖𝒖𝒊𝒊 Absolute 23 uncertainty 𝑼𝑼𝒊𝒊 =

(𝒙𝒙𝒊𝒊 *𝒖𝒖𝒊𝒊 ) (expanded at the 95%
(𝒙𝒙𝒊𝒊 upper end of the (expanded at the 95%
confidence level)
measured range) confidence level)

23 Note that despite figures given in percentages these are labelled as absolute uncertainties as they indicate

percentage points related to the parameter concentration. For instance, using the figures in the table the
concentration of CO2 would be 16% ± 0.48%, i.e. between 15.52% and 16.48% at the 95% confidence level.
 CO2: 16%vol ± 3.0% ± 0.48vol%

CO: 2%vol ± 3.0% ± 0.06vol%

The uncertainty is determined using the following equation as independent uncertainties of a

sum 24 (see example 2 in section 8.2.1):

2 2
�𝑈𝑈𝐶𝐶𝑂𝑂 + 𝑈𝑈𝐶𝐶𝐶𝐶
2 √0.48%2 + 0.06%2
𝑢𝑢𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = = = 0.6%
100 − 𝑎𝑎𝐶𝐶𝑂𝑂2 − 𝑏𝑏𝐶𝐶𝐶𝐶 82%

Using these figures leads to an uncertainty related to the measurement of the concentrations
of the flue gas components of ±0.6%.

Step 1.3: combined uncertainty of 𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑

The combined uncertainty of 𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 is 2.7% as determined by using the formula for
independent uncertainties of a product:

𝑢𝑢𝑉𝑉𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓,𝑑𝑑𝑑𝑑𝑑𝑑 = �𝑢𝑢𝑉𝑉2𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 + 𝑢𝑢𝐺𝐺𝐺𝐺𝐺𝐺

2
𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
= √2.6%2 + 0.62 = 2.7%

Step 2: uncertainty of 𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐

The combined uncertainty of 𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 can then be determined using the following formula for
independent uncertainties of a sum (see example 2 in section 8.2.1):

2 2
�𝑈𝑈𝐶𝐶𝑂𝑂 + 𝑈𝑈𝐶𝐶𝐶𝐶
2 √0.48%2 + 0.06%2
𝑢𝑢𝐺𝐺𝐺𝐺𝐺𝐺𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = = = 2.7%
𝑎𝑎𝐶𝐶𝑂𝑂2 + 𝑏𝑏𝐶𝐶𝐶𝐶 18%

24 Note that this formula is not fully correct here as parameters are in the denominator which leads to different
results when calculating partial derivatives. However, the formula used is simpler but still provides very similar
results. Furthermore, it is assumed that uncertainties associated with the concentrations of CO2 and CO are
uncorrelated. However, if for example the same equipment (analyser, sampling system, etc.) is used or
measurements are performed simultaneously, these assumptions may not be valid and uncertainties would have
to be treated as correlated, thus higher uncertainties.
Using the figures from the table in step 1.2 leads to an uncertainty related to the GHG
concentrations of ±2.7%.

Step 3: overall (combined) uncertainty of 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐

The uncertainty related to 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑘𝑘𝑘𝑘 is calculated as independent uncertainty of a product as

follows:

𝑢𝑢𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = �𝑢𝑢𝑉𝑉2𝑎𝑎𝑎𝑎𝑎𝑎,𝑑𝑑𝑑𝑑𝑑𝑑 + 𝑢𝑢𝐺𝐺𝐺𝐺𝐺𝐺

2
𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
= = √2.7%2 + 2.7%2 = 3.8%

The overall uncertainty of 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 is ±3.8%. Note that because expanded uncertainties were
used in each step, this overall uncertainty also corresponds to the expanded uncertainty, i.e.
the uncertainty at the 95% confidence level as required by the MRR. This value of 3.8% must
be compared against the tier thresholds in table 1 of Annex II of the MRR.
If you need a version of this document in a more accessible format, please email
[email protected]. Please tell us what format you need. It will help us if you say what
assistive technology you use.