Consultation outcome

Proposed best available techniques (BAT) implementation guidance for mid-merit gas generators: summary of consultation responses

Updated 16 April 2026

1. Introduction

The purpose of the consultation was to get feedback on proposals for amendments for flexible gas generation in England using Article 14(6) of the Industrial Emissions Directive (IED) as the framework where no best available techniques (BAT) conclusions are published.

This document summarises the responses we received and explains how we have considered and addressed the feedback and what we will do next.

2. How we ran the consultation

We ran the consultation for 12 weeks from 3 November 2025 to 26 January 2026. It was hosted on Citizen Space, our consultation website. It was open to all stakeholders with an interest in mid merit gas generators and their emissions to air.

The consultation was run in line with our legal requirements to consult following the Cabinet Office’s consultation principles guidelines.

We promoted the consultation through direct engagement with industry stakeholders, including operators, trade associations, and public sector bodies. We also shared information through our existing networks and published details on GOV.UK to encourage broad participation.

Respondents were invited to provide feedback on the proposed amendments to our guidance on techniques for flexible gas generation in England. Responses could be made by using the online consultation tool or by submitting a completed response form by email.

All responses received within the consultation period were reviewed and analysed. This document summarises the feedback and explains what we will do next.

We received 4 responses in total from:

• 2 operators
• 1 trade body
• 1 member of the public

3. Summary of key findings and actions we will take

We consulted on a proposal to allow open cycle gas reciprocating engines and open cycle gas turbines (OCGTs) to operate up to 4,000 hours per year as a new ‘mid-merit’ category, where operators meet a set of criteria on efficiency, emissions and fast start capability. The intent was to incentivise lower oxides of nitrogen (NOx) and higher efficiency by linking environmental performance to additional operating hours under Article 14(6) of the IED. We sought views via 11 questions covering each criterion, application of the 4,000 hours per year cap within installations, unintended consequences, cost impacts and any further comments. We have summarised the responses to these questions and given our response.

3.1 Proposed criteria listed in table 1 for the energy efficiency parameter

Industry respondents did not agree, and 1 individual did agree with the criteria for the energy efficiency parameter we proposed.

The industry respondents fed back that a single efficiency threshold was not aligned with BAT which is explicitly technology specific under the IED, Large Combustion Plant Best Available Techniques Reference document (LCP BRef) and Medium Combustion Plant Directive (MCPD).

They noted that the 40% energy efficiency requirement would be at the lower end of BAT for gas engines (39.5 to 44% for new plant) and at the upper end of BAT for OCGTs (36 to 41.5% for new plant), making compliance significantly harder for OCGTs. This difference was seen as creating an uneven playing field and potentially excluding OCGTs from mid-merit eligibility despite their established role in flexible generation.

Several respondents highlighted that OCGTs have much lower methane slip than engines, meaning their overall greenhouse gas (GHG) intensity can be comparable even if their thermal efficiency is lower.

Some suggested that the criteria should instead align directly with LCP BRef Associate Energy Efficiency Levels (AEELs) for each technology.

One respondent supported the criterion, viewing it as a reasonable way to incentivise efficiency gains, provided proper monitoring and periodic review are maintained.

Our response

We acknowledge that BAT is technology specific and that applying a single efficiency threshold across all combustion technologies could create uneven compliance pressures, particularly for OCGTs. We also recognise the view that, when methane slip is considered, the overall GHG intensity of engines and turbines can in some cases be comparable.

If we were to take forward amendments, we would explore the option of revising the criteria for existing OCGTs to potentially allow access to mid-merit operation where a plant is able to demonstrate an energy efficiency of equal to or greater than 35.3%, the BAT range being 33 to 41.5% for existing plant, and NOx emissions of equal to or less than 35 milligrams per normalised cubic metre (mg/Nm³) annual average. The 35.3% OCGTs efficiency delivers a carbon dioxide intensity equivalent to that of gas engines at 40% efficiency, taking into account the methane (CH4) slip.

For new OCGTs, we would also consider whether an efficiency threshold of equal to or greater than 36%, aligned with the lower end of the BAT AEEL range, together with an annual average NOx limit of equal to or less than 35 mg/Nm³, would be an appropriate basis for mid-merit access.

For gas engines, we would look at retaining the equal to or greater than 40% energy efficiency requirement.

This potential adjustment could provide a more proportionate route to mid-merit access across the main technology types, while continuing to prioritise environmental protection.

3.2 Proposed criteria listed in table 1 for the oxides of nitrogen (NOx) emissions (concentration) parameter

Most respondents did not support the proposed NOx limit.

The proposed less than 30 mg/Nm³ was viewed as more stringent than OCGTs BAT and inconsistent with LCP BRef Associated Emission Levels (AELs) (15 to 35 mg/Nm³ annual, 25 to 50 mg/Nm³ daily for new OCGTs).  

As OCGTs can typically achieve approximately 35 mg/Nm³ without Selective Catalytic Reduction (SCR), the proposed limit was seen as effectively mandating SCR retrofits.

Respondents said that SCR is:

• technically challenging for OCGTs due to high exhaust temperatures
• costly, with significant capital expenditure
• a source of ammonia slip, increasing nitrogen deposition

Several respondents highlighted that ammonia slip from engines fitted with SCR could consume available critical load headroom, potentially constraining future projects.

Respondents also noted that for large OCGTs, the binding regulatory limit is the annual average, because they are required to fit continuous emissions monitoring system (CEMS).

Views highlighted a need for clarity on whether an air quality impact assessment would be required when applying for mid-merit operation. One respondent supported the inclusion of a NOx limit to protect air quality.

Our response

Any permit application to operate greater than 1,500 hours per year would require an Air Quality Impact Assessment (AQIA), including assessment of critical loads and ammonia slip (where SCR is used).

Open cycle gas turbines

We recognise that BRef AELs for new OCGTs are equal to or less than 35 mg/Nm³ (annual average) and equal to or less than 50 mg/Nm³ (daily average).

We also acknowledge the nitrogen deposition considerations associated with ammonia slip where SCR is applied.

For new OCGTs, we will consider aligning the NOx limits with BRef AELs equal to or less than 35 mg/Nm³ annual average and equal to or less than 50 mg/Nm³ daily average.

For existing plant, we will consider NOx emissions of equal to or less than 35 mg/Nm³ (annual average) and 55 mg/Nm³ daily average. We consider the tighter annual average reflects the incentivisation of cleaner combustion plant.

In both cases annual averages should continue to be verified through CEMS.

Spark ignition engines

For engines, we will look at retaining a more stringent NOx limit such as equal to or less than 30 mg/Nm³ while reconsidering whether daily averaging, or the average over the sampling period, is appropriate where compliance is demonstrated without CEMS.

This more stringent approach for engines takes into account the ammonia slip associated with SCR and its implications for nitrogen deposition.

3.3 Proposed criteria listed in table 1 for the methane emissions parameter

Most respondents did not support the methane criterion as written in our consultation proposal.

Respondents stated that methane slip is negligible for OCGTs and therefore a methane emission limit value (ELV) is not appropriate for turbines.

Several respondents noted that the proposed limit reflects the lowest end of the LCP BRef BAT range for engines (215 to 500 mg/Nm³ as carbon) and that using a single strict value does not reflect normal BAT practice.

Stakeholders recommended aligning methane limits for engines with BRef defined BAT ranges rather than selecting the most stringent value.

One respondent supported having a methane limit as a measure to reduce climate impacts more broadly.

Our response

We recognise that methane slip is primarily an issue associated with gas engine technologies rather than OCGTs.

If we proceed with further development of the proposals, we may consider not applying a methane limit to OCGTs under normal operating conditions, reflecting their very low contribution to methane emissions.

For engines, we may look at retaining a methane related control but aligning it with the BAT associated ranges set out in the BRef, to ensure any requirement remains proportionate and technically consistent. We are minded to retain the lower level of the BRef range as this aligns the carbon intensity with that from OCGTs.

3.4 Proposed criteria listed in table 1 for the ammonia emissions parameter

Most respondents did not support ammonia limits being applied to all technologies.

Multiple respondents stated that ammonia emissions are only relevant where SCR is used.

If OCGTs are not expected to install SCR, then an ammonia ELV would be unnecessary for those technologies.

Some respondents noted that ammonia slip is already managed through existing regulatory processes and risk assessments requiring consideration of sensitive habitats.

One respondent supported ammonia limits due to risks to air quality and ecosystems.

Our response

We recognise that ammonia slip is inherently linked to the use of SCR and that its relevance varies between technologies.

If we take these proposals further, we will consider applying ammonia limits only in situations where SCR is installed.

For OCGTs, which typically achieve BAT aligned NOx levels without the need for SCR, we will explore an approach where an ammonia ELV is not applied.

3.5 Proposed criteria listed in table 1 for the number of start-ups and shutdowns parameter

Most respondents did not support the inclusion of this parameter.

The parameter was widely seen as not environmentally relevant, since start‑up capability does not drive significant emissions changes.

Respondents noted that all modern technologies capable of flexible operation already meet or exceed this capability.

One respondent supported the parameter to ensure flexibility.

Our response

We agree the criterion provides little environmental benefit.

We propose removing the startup and shutdown criterion from eligibility requirements if we take them forward. This would also remove any requirement to keep combustion plant hot between generating runs in order to meet this criterion.

3.6 Criteria listed in table 1 and the application across all open cycle combustion technologies for example reciprocating engines and open cycle gas turbines

There was a strong consensus that BAT is inherently technology specific and must remain aligned with IED, MCPD and LCP BRef structures.

Applying the same criteria to engines and OCGTs was seen as creating technology bias:

• gas engines with SCR would qualify
• OCGTs which are still essential for grid stability would often be excluded

Respondents stressed that engines less than 50 megawatt thermal (MWth) have not undergone a full BAT derivation, whereas OCGTs equal to or greater than 50 MWth have and this must be reflected in regulatory design.

One respondent supported common criteria across all technologies for consistency.

Our response

We recognise that BAT is technology specific and that applying uniform criteria across engines and OCGTs could create unintended or disproportionate outcomes.

If we decide to take the proposals forward, we will explore separating the criteria into technology specific schedules to more accurately reflect the different BAT associated performance levels and operational characteristics of each technology type.

This could include developing distinct schedules for gas engines and OCGTs, potentially drawing on the LCP BRef ranges as a reference point, particularly for OCGTs where these values are well established.

Taking this approach would help ensure greater proportionality and alignment with the structure of the BRef, while maintaining fairness across technologies, should we choose to progress the proposals.

3.7 Open cycle combustion plants permitted under Chapter II and Chapter III of the IED should be able to operate up to limit of 4,000 hours in a calendar year subject to meeting the criteria in table 1

Responses were mixed, but most agreed with the concept in principle.

Several respondents supported the principle of mid-merit operation (up to 4,000 hours per year) but not with the criteria as proposed, particularly in relation to OCGTs.

Concerns centred on the risk that only SCR equipped engines would qualify, excluding other flexible generation technologies.

Others supported the 4,000 hours per year limit as long as environmental performance requirements are achieved.

Our response

We recognise the value of retaining a mid-merit concept, but also that any eligibility conditions would need to reflect technology specific BAT and performance characteristics.

If we decide to develop the proposals further, we will consider whether separate mid-merit eligibility conditions for OCGTs and engines would provide a more proportionate and technically aligned approach.

We recognise that the 4,000 hours per year operational mode has an important role in providing operational flexibility, but that eligibility requirements may need to reflect technology specific BAT and ensure environmental robustness.

If we choose to pursue this further, we may consider retaining the 4,000 hours per year mode while exploring revised, technology specific eligibility criteria, supported by AQIA.

3.8 Where a Chapter II or Chapter III installation is comprised of more than one combustion plant, when should the hourly limit up to 4,000 hours in a calendar year apply?

Views were divided, reflecting different interpretations of operational practice.

Several respondents supported applying the limit at plant level, defined by a common stack. This reflects long established regulatory practice and avoids ambiguity.

One respondent supported applying the limit to the installation.

One respondent preferred applying the limit to each individual combustion plant, ensuring each meets the criteria independently.

There is strong operator support for continuing plant level accounting aligned to current practice.

Our response

We recognise that a plant level (common stack) approach to hours accounting is already well established, widely understood by operators and consistent with existing regulatory practice. It also provides a clear and enforceable basis for determining operational hours on installations with multiple combustion plant.

If we take the proposals forward, we will consider retaining this plant level approach while also exploring the option of enhancing guidance with worked examples to support operators in applying the methodology consistently.

3.9 Unintended consequences for future power generation scenarios, in particular changes to operating patterns and technology mixes

All respondents believed unintended consequences were likely. Themes given in the consultation response included:

• technology distortion: criteria would favour SCR equipped engines and disadvantage OCGTs
• environmental effects: widespread use of SCR on engines could increase ammonia slip, potentially consuming nitrogen deposition headroom and preventing other projects from being permitted
• market impacts: if OCGTs are excluded, the technology mix available for grid balancing could be narrowed, reducing operational resilience
• decarbonisation risks: one respondent suggested it could inadvertently delay uptake of low carbon technologies

Our response

We recognise the concerns raised and understand how they relate to the need for technology specific criteria and a more consistent, evidence based approach to assessing local environmental impacts.

If we choose to take the proposals forward, we will retain the requirement for AQIA, including assessment of critical loads and nitrogen deposition, particularly where SCR is used or where site specific risks justify it. This would help ensure that any technology specific criteria are applied in a way that avoids unintended distortions and maintains appropriate environmental protections.

We will also carry out a post‑implementation review after the first year to assess how the revised approach is operating in practice, identify any emerging issues and adjust if needed.

We will engage with the Department for Energy Security & Net Zero (DESNZ) and the Department for Environment, Food and Rural Affairs (Defra) with respect to any unintended consequences as outlined.

3.10 Estimating the financial costs of the proposed changes on operations

Most respondents provided no cost data. One offered detailed information, estimating that adding SCR to a typical 50 megawatt electrical (MWe) OCGTs could cost £5 to 10 million in capital expenditure with additional parasitic load and operational costs. Others acknowledged that costs were likely but did not provide figures. There was broad consensus that mandatory SCR for OCGTs would be economically prohibitive and inconsistent with BAT.

Our response

We acknowledge the cost issues raised and recognise that SCR should not ordinarily be expected on OCGTs, given BAT requirements and the need to avoid disproportionate economic impacts.

If we choose to develop the proposals further, we may consider ensuring that any revised criteria for OCGTs allow compliance without reliance on SCR, consistent with established BAT expectations and proportionate cost burdens.

3.11 Further comments

Respondents provided a range of further comments.

Several respondents asked the Environment Agency to confirm that the proposals apply only to mid-merit (1,500 to 4,000 hours per year) and that the less than 1,500 hours per year framework remains unchanged.

One respondent suggested allowing multi‑year averaging of operating hours (for example 3 to 5 years) or the ability to roll over unused hours.

One emphasised the need for clear public communication and transparent monitoring of environmental improvements.

Our response

We confirm that the proposals under discussion relate only to the mid-merit band (1,500 to 4,000 hours per year), and that the framework for plant operating below 1,500 hours per year would remain unchanged.

If we decide to progress this work, we will consider reviewing the scope of the hours accounting approach and assessing whether alternative options such as multiyear averaging could be appropriate, proportionate and enforceable. As part of this, we may also seek views on different accounting methodologies to ensure that any eventual approach is clear, workable, and aligned with regulatory expectations.

These considerations would help inform any future consultation on hours accounting options, should we choose to take the proposals forward.

4. Next steps

We will engage with DESNZ and Defra to ensure the proposals are aligned with wider government policy objectives, including:

• reducing NOx emissions
• compliance with the Energy Efficiency Directive (EED)
• supporting a secure and resilient electricity supply
• avoiding unintended barriers or distortions in future generation mixes
• ensuring that any divergence from legacy and new EU derived frameworks does not create unforeseen policy or environmental consequences, and that the overall approach remains aligned with the UK’s current regulatory and strategic direction

We will also discuss the responses and revised proposals with colleagues in the devolved administrations to support consistency of approach where appropriate and to understand any nation specific considerations.

At this stage we cannot give any certainty about whether the proposals and revisions we are now proposing will go ahead but we will continue to update our stakeholders regularly on progress including the engagement outlined.

We thank all respondents for their constructive feedback and technical evidence. The potential changes outlined aim to balance strong environmental protection (air quality and habitats), fairness across technologies, and practical enforceability, while enabling mid-merit flexibility where it can be environmentally justified and meets wider government policy objectives.

If you wish to discuss your consultation response or any of the points raised in this document, please email combustion@environment-agency.gov.uk.

5. List of respondents

• A member of the public
• Progressive Energy Ltd
• Joint Environmental Programme – co-funded programme of research into the environmental impacts of electricity generation jointly funded by Drax Power, EPUKi, RWE, SSE Thermal, Uniper, VPI and WBE
• SSE Thermal