ACER Opinion 12-2026 on All-Island Resource Adequacy Assessment: Ireland and Northern Ireland

No 12/2026

OPINION

on All-Island Resource Adequacy Assessment: Ireland and Northern Ireland

7 May 2026

A C E R O P I N I O N N O 1 2 / 2 0 2 6

Executive summary

Under the Electricity Regulation (EU) 2019/943, the European Network of Transmission System Operators for Electricity (ENTSO-E) conducts annual European resource adequacy assessments (ERAAs). Member States may perform national resource adequacy assessments (NRAAs) to capture local developments. If an NRAA identifies risks not reflected in the ERAA, it must be submitted for ACER’s opinion, along with an explanation of the differences between the national and European assessments. EirGrid and SONI – the Transmission System Operators (TSOs) for Ireland and Northern Ireland, respectively – conducted the All-Island Resource Adequacy Assessment 2026-2035 (AIRAA 2026- 2035), published on TSO’s websites (EirGrid, SONI). The AIRAA 2026-2035 complements ERAA 2025 by reflecting recent developments specific to the Single Electricity Market (SEM), which comprises Ireland and Northern Ireland. AIRAA 2026-2035 delivers three distinct assessments – covering Ireland, Northern Ireland, and the SEM as a whole – all of which are reviewed by ACER.

Key Findings of AIRAA 2026-2035 base (central reference) scenario

• Ireland: Adequacy risk in 2028, with LOLE of 7.95 hours – higher than in ERAA 2025 (1.62 hours) and above the 3-hour reliability standard; primarily driven by delays to new capacity entering the market and the delayed Celtic Interconnector. • Northern Ireland: Risks in 2033 and 2035, with LOLE of 7.30 and 11.26 hours, respectively – above ERAA 2025 estimates (1.15 and 2.95 hours) and exceeding the 4.9-hour standard; mainly driven by the absence of the planned second North-South Interconnector (to be commissioned in 2031) and lower projected wind generation.

ACER found the following differences justified

• Simplifications in AIRAA 2026-2035: regional scope, exclusion of flow-based market coupling, and omission of curtailment sharing - have minimal impact due to the SEM’s limited interconnection with its neighbours, while reducing modelling complexity. • Methodological improvements: more outage sample runs, detailed battery modelling, and grid feed-in constraints in Dublin – improve accuracy. • Input data update: demand, resources, delayed Celtic Interconnector - reflect recent trends.

ACER found the following differences not justified

• Assumptions restricting gas-fired peaking units to a fixed evening operating window ignore seasonal scarcity, and including the planned second North-South Interconnector in the SEM assessment only. • The Economic Viability Assessment (EVA), although developed, was not included in the base (central reference) scenario. It should be, as it is a legally required element of the analysis and is necessary to assess whether capacity is financially viable in the market and therefore likely to be built or maintained.

Conclusion and Recommendations

ACER finds most differences justified, but highlights gaps: (i) missing EVA modelling phase in the base scenario, (ii) including the second North-South Interconnector in the SEM assessment only, and (iii) restrictive operation assumptions on peaking units. To address these gaps, ACER recommends: • Including EVA in the base (central reference) scenario, ensuring it incorporates revenue stacking, and demand-side unit expansion, and maintaining revenue consistency between EVA and economic dispatch models. • Modelling increased operational flexibility for the constrained peaking units during scarcity. • Reflecting planned key infrastructure developments in all assessments.

A C E R O P I N I O N N O 1 2 / 2 0 2 6

1. Background

1 Article 20 of Regulation (EU) 2019/943 (Electricity Regulation) requires Member States to monitor resource adequacy within their territory based on the European Resource Adequacy Assessment (ERAA). It also allows them to complement ERAA by conducting a National Resource Adequacy Assessment (NRAA) according to Article 24 of the Electricity Regulation. Either of the two assessments can identify resource adequacy concerns, but both assessments must be based on the ERAA methodology. The adequacy concerns identified through these assessments should be in the first place addressed by eliminating regulatory distortions or market failures, and, where it does not suffice, by introducing capacity mechanisms.

2 According to Article 24(3) of the Electricity Regulation, where the NRAA identifies an adequacy concern that was not identified in ERAA, the NRAA must include reasons for the divergence between the two assessments, including details of the sensitivities used and the underlying assumptions. The Member State must publish the NRAA and submit it to ACER for opinion. ACER’s opinion assesses, on a case-by-case basis, whether the differences between the two assessments are justified. ACER primarily considers the differences identified and reasoned by the Member State in the submission, but may also identify further differences, if they also have a material impact on the results.

3 The national body that is responsible for the NRAA, which can be a transmission system operator (TSO) or a different body designated by the Member State, must take due account ACER's opinion, and, where necessary, amend its assessment. Where it decides not to take ACER's opinion fully into account, it must publish a report with detailed reasons.

4 The latest ERAA (ERAA 2025) was published in December 2025. To complement it, EirGrid and SONI, the TSOs for Ireland and Northern Ireland respectively, jointly carried out the All-Island Resource Adequacy Assessment 2026-2035 (AIRAA 2026-2035), published on TSO’s websites (EirGrid, SONI). The assessment constitutes an NRAA within the meaning of Article 24 of the Electricity Regulation. AIRAA 2026-2035 was published on 26 February 2026. On 10 March 2026, Ireland’s Department of Climate, Energy and the Environment submitted the AIRAA to ACER, together with the reasons for its divergence from ERAA 2025, as well as details of the sensitivities used and the underlying assumptions. On 30 March, ACER received the corresponding submission from the Department for the Economy of Northern Ireland.

5 This Opinion evaluates whether the differences between AIRAA 2026-2035 and ERAA 2025 are justified. It is addressed to Ireland’s Department of Climate, Energy and the Environment, and to the Department for the Economy of Northern Ireland, and is issued in accordance with Article 9(2) of Regulation (EU) 2019/942 and Article 24(3) of the Electricity Regulation.

2. Summary of AIRAA 2026-2035

6 The aim of AIRAA 2026-2035 is to complement the ERAA 2025 by capturing recent market trends in resource deployment and demand within the Single Electricity Market (SEM), comprising Ireland and Northern Ireland. It also applies modelling approaches that more accurately reflect the technical performance of the SEM electricity system, based on local specificities and historical observations.

7 AIRAA 2026-2035 provides adequacy results for three regions: (1) Ireland, (2) Northern Ireland, and (3) the SEM – All-Island. The modelling adequacy results of the base (central reference) A C E R O P I N I O N N O 1 2 / 2 0 2 6 scenario are shown in Table 1. AIRAA 2026-2035 identifies new adequacy concerns that were not detected in ERAA in both bidding zones – specifically, in Ireland in 2028 and in Northern Ireland in 2033 and 2035 – as the modelled Loss of Load Expectation (LOLE) exceeds the respective reliability standards, which are 3 hours for Ireland and 4.9 hours for Northern Ireland.

8 This divergence in the LOLE results is due to differences in the modelling assumptions in the base scenario between AIRAA 2026-2035 and ERAA 2025. These differences are discussed and assessed by ACER in the following section.

3. ACER’s assessment

9 ACER welcomes AIRAA 2026-2035 and considers it a well-developed assessment of the SEM electricity system, enriched with local details and capturing recent system developments. The assessment report provides a clear description of the methodological modelling assumptions.

10 ACER also appreciates EirGrid and SONI for providing a comprehensive comparison between AIRAA 2026-2035 and ERAA 2025, including a clear description and explanation of the differences and their potential impact on adequacy results.

11 The differences between the AIRAA 2026-2035 base scenario and the ERAA 2025 central reference scenario fall into two categories: (i) methodological and (ii) input data. Each difference is listed in Table 2, along with its rationale, potential impact on LOLE and the assessment in which it is applied. A detailed description of these differences is provided later in this section. A C E R O P I N I O N N O 1 2 / 2 0 2 6

12 ACER notes that the 2 North-South interconnector is considered only in the SEM assessment, distinguishing it from the two other assessments. The project is expected to become operational by the end of 2031. The SEM adequacy results presented in Table 1 demonstrate its importance: inclusion of the interconnector ensures that LOLE in the SEM remains below the reliability standard in the modelled target years 2033 and 2035, thereby strengthening the security of electricity supply across the island. This contrast also highlights that increased cross-zonal capacity supports resource adequacy.

3.1. Method

Difference #1 Regional scope

13 Description: AIRAA 2026-2035, compared to the pan-European ERAA, has a smaller geographic scope, focusing on the Single Electricity Market (SEM), comprising Ireland and Northern Ireland, and its neighbouring bidding zones of Great Britain and France. The neighbouring electricity systems are calibrated to their respective reliability standards, ensuring sufficient standardcompliant resource capacity. This calibration is performed through the iterative addition of perfect generation (or demand) until each region meets the target LOLE (3 hours) defined by its reliability standard.

14 Power imports from non-modelled countries are represented as fixed exchanges derived from the results of ERAA 2024. These exchanges are specific to both the target year and the weather scenario.

15 Relevance: Reducing the geographical scope in electricity system modelling reduces complexity and modelling time. This enables faster and more resource-efficient analyses while allowing for a deeper focus on a specific region.

16 Assessment: The reduced spatial coverage in AIRAA 2026-2035 includes the key relevant regions: the SEM and its direct neighbours, Great Britain and France. As the neighbouring countries are significantly larger than the SEM, their inclusion provides an adequate representation of the external systems most relevant for cross-border exchanges. By calibrating neighbouring regions to their reliability standard, the model ensures the SEM neither relies excessively on imports nor is constrained by energy-dependent neighbouring regions. When combined with fixed, target-year- and weather-scenario-specific exchanges with non-modelled bidding zones, this approach ensures consistency with the ERAA framework and is therefore considered justified. Difference #2 No flow-based market coupling

17 Description: Flow-based market coupling (FBMC), while present in ERAA, is absent in AIRAA 2026-2035. Instead, net transfer capacity (NTC) limits are applied among the modelled bidding zones.

18 Relevance: FBMC models cross-border electricity trade by reflecting the physical constraints of the interconnected transmission networks within bidding zones rather than treating interconnectors as independent links. In European markets these constraints are incorporated in the EUPHEMIA, which determines day-ahead prices and exchanges. In ERAA, modelling FBMC helps capture how neighbouring systems can support each other during scarcity events, influencing estimates of LOLE and the contribution of interconnections to system adequacy.

19 Assessment: AIRAA 2026-2035 focuses on the SEM, rather than the meshed AC network of continental Europe where FBMC provides the greatest benefits. The SEM is not part of the Core A C E R O P I N I O N N O 1 2 / 2 0 2 6 Capacity Calculation Region in which FBMAC is applied. In addition, the geographic scope of AIRAA 2026-2035 does not extend to the full Core region, and its modelling approach already captures support from Great Britain and France through calibration to their respective reliability standards. Thus, excluding FBMC in AIRAA 2026-2035 is considered justified. Difference #3 No curtailment sharing

20 Description: The optional curtailment sharing feature, while present in ERAA, is not included in AIRAA 2026-2035. The curtailment sharing ensures that energy shortages are fairly distributed across bidding zones whose domestic capacity is not sufficient to meet local demand. This reflects the actual functioning of the spot markets.

21 Relevance: Curtailment sharing is embedded in the EUPHEMIA algorithm for day-ahead electricity trade among EU bidding zones. It spreads more evenly the impact of network constraints on scarcity experienced in the interconnected bidding zones. When applied to future electricity system modelling, this feature is relevant in multi-bidding-zone simulations to ensure accurate LOLE results in markets facing resource scarcity.

22 Assessment: Curtailment sharing can have a significant impact on LOLE results. In ERAA, its application roughly doubled the average LOLE per bidding zone, as explained in ACER’s ERAA 2025 Decision. However, the impact of curtailment sharing in the SEM is limited due to the island’s geographic position, being connected only to the bidding zones of Great Britain and France. According to ERAA 2025 , the SEM experiences a minor change in adequacy outcomes when curtailment sharing is applied: the feature increases LOLE by 1 to 2 hours in Ireland and up to 1 hour in Northern Ireland depending on target year. This confirms the low impact of the curtailment sharing in the SEM. Thus, its omission in AIRAA 2026-2035 is considered justified. Difference #4 No economic viability assessment

23 Description: The Economic Viability Assessment (EVA), while included in AIRAA 2026-2035, is absent from the base (central reference) scenario. As the result, the Economic Dispatch (ED) adequacy model operates using the initial resource capacity assumptions without evaluating the economic feasibility of either existing assets or new resource candidates.

24 Relevance: EVA assesses whether existing and prospective power resources are financially sustainable within an energy-only market. It serves as a proxy for investor behaviour, to identify which assets are likely to remain operational, be decommissioned, or be newly developed. In doing so, it provides a more realistic basis for evaluating future supply adequacy.

25 Assessment: EVA is a useful component of ERAA, as it goes beyond simply checking whether enough electricity capacity exists – it evaluates whether that capacity is financially sustainable and likely to be built, maintained, or retired. Without EVA, ERAA risks misestimating security of supply and potentially leading to suboptimal policy decisions. Therefore, its omission from AIRAA 2026-2035 is not justified.

26 Although EVA is not included in the base scenario of AIRAA 2026-2035, it is incorporated in several sensitivity scenarios. These scenarios assess resource entry and exit based on profitability, indicating the retirement of gas turbines in the SEM across all target years and in all sensitivity scenarios. In the low battery cost sensitivity, battery storage is deployed.

27 ACER recommends including the EVA modelling phase in the base scenario of the assessment.

28 ACER recommends that the assessment ensures strong consistency between EVA and ED models, so that investment decisions (i.e. capacity adjustments) derived from EVA align with the security of supply outcomes assessed in ED. Particular attention should be paid to the selection of weather scenarios in EVA. These scenarios should capture the overall climatic variability in A C E R O P I N I O N N O 1 2 / 2 0 2 6 the SEM present across the complete ED scenario set. This will help ensure that the EVA model identifies the consistent level of investment. In addition, revenues observed under the selected EVA scenarios should be representative of revenues across the entire ED scenario set, ensuring alignment between the two models. Also, technical constraints should well align in two models.

29 ACER recommends that the assessment considers incorporating modelling revenue stacking within the EVA, with particular relevance for battery storage while remaining applicable to all generation and demand-side resources. Revenue streams may include, for example, energy market trade (arbitrage) – including intraday and balancing markets – as well as provision of ancillary/system services under the Delivering a Secure, Sustainable Electricity System (DS3) programme, as highlighted in the Electricity Storage Policy Framework for Ireland.

30 ACER recommends that the assessment considers demand side units (DSU) as potential expansion candidates within EVA. Relevant economic parameters and expansion potentials can be sourced from publicly available data, including datasets consulted on in March 2026 and to be published by ACER. EirGrid and SONI White Paper highlights demand-side flexibility as a key resource for achieving renewable energy and emissions targets while enhancing security of supply, signalling a clear policy direction to expand and incentivise DSU deployment across both Ireland and Northern Ireland. Difference #5 More forced outage samples

31 Description: In ERAA 2025, each weather scenario within each target year is simulated with 15 different forced outage runs. In AIRAA 2026-2035 this is increased to 30 forced outage runs per scenario, which is considered sufficient under the ERAA methodology, as it provides a formal convergence criterion to verify that additional samples no longer yield a meaningful improvement in modelling accuracy.

32 Relevance: Multiple forced outage runs provide robust LOLE results accounting for uncertainty, capturing a wide range of potential outcomes and providing a more realistic analysis. This approach is especially important for smaller power systems, where a higher ratio of single‑plant capacity to peak demand makes the system far more sensitive to random unit outages.

33 Assessment: Unlike Europe-wide system in ERAA, where 15 forced outage runs are sufficient, the smaller SEM is more sensitive to conventional generator outages, requiring additional modelling runs to ensure convergence and robust adequacy assessment, providing a more accurate representation of system reliability under various outage scenarios. Therefore, the 30 forced outage runs in AIRAA 2026-2035 are both appropriate and justified. Difference #6 More planned maintenance samples

34 Description: ERAA 2025 applies a single maintenance profile per target year reflecting the average level of planned outages. In AIRAA 2026-2035, each simulation run – defined by the combination of target year, weather scenario, and forced outage sample – uses a slightly different reoptimized planned maintenance profile. Consequently, a new optimised maintenance schedule is generated for every run. Maintenance events are predominantly scheduled during low-demand seasons, but they may also occur during scarcity periods at times of peak demand.

35 Relevance: Capturing the historical seasonal pattern of planned maintenance – especially the unavoidable maintenance near scarcity periods – is crucial for accurate future electricity system modelling. This feature ensures realistic reliability assessment, as such maintenance can directly limit resource availability during scarcity.

36 Assessment: Historical data from the SEM indicates that planned maintenance does not always occur strictly within the optimal window. Modelling the historical frequency and randomness of A C E R O P I N I O N N O 1 2 / 2 0 2 6 these events during scarcity – using 30 samples per target year and per weather scenario – therefore brings the simulation results closer to real system behaviour. This approach is justified and aligns with the ERAA methodology, which requires planned outage schedules to be optimised to avoid maintenance during likely ENS periods, while still respecting the share of capacity that must undergo maintenance in winter. Difference #7 Restricted operating hours

37 Description: Operating hours for peaking gas-fired generation units in Northern Ireland are capped under the Pollution Prevention and Control regulatory framework: units summing to 270 MW are limited to 500 hours per year, while units summing to 510 MW may operate up to 1500 hours annually.

38 In ERAA, dispatch of these constrained units is optimized assuming perfect foresight across the full modelling horizon. In contrast, AIRAA 2026-2035 imposes operational restrictions: units subject to the 1500-hour limit may run for up to six hours per day within a fixed evening peak window (16:00–22:00), while those under the 500-hour limit are restricted to a maximum of four hours per day within a narrower window (17:00–21:00).

39 Relevance: Restrictions on the operating hours of power plants can affect the availability of resource capacity during periods of system stress. Units subject to limited operating hours provide less reliability value than fully available plants. Incorporating these constraints properly ensures that reliability assessments capture the operational challenges associated with emissions-driven restrictions.

40 Assessment: The concern is not the existence of an annual emissions-related running-hour limit as such, but the rigid daily allocation of those hours in the adequacy modelling. The fixed daily operating window for the constrained units may restrict their contribution during times of system scarcity. Scarcity events are often seasonal and can extend outside the restricted operational window. Applying a uniform daily constraint throughout the year fails to account for this seasonality, while a rigid schedule limits the units’ ability to respond to longer-duration or off-peak scarcity events . For these reasons, the constraint is not justified.

41 ACER recommends that the assessment provides greater flexibility for thermal units with limited annual operating hours. This could include allowing operators to allocate a larger share of their operating hours to periods when resource scarcity is expected, thereby aligning the use of constrained units more closely with the seasonal pattern of scarcity events. In addition, operational frameworks should permit these units to run for longer continuous durations – exceeding daily hour limit where necessary – since scarcity events may occur outside the predefined window and last longer , while thermal units are capable of sustained operation provided that fuel supply is maintained. Difference #8 More constrained battery model

42 Description: Compared with ERAA 2025, which models a single aggregated battery per bidding zone, AIRAA 2026-2035 models individual battery units with unit-specific capacity and duration. It further categorises batteries into three sub-groups based on duration: less than one hour, 1 to 4 hours, and more than 4 hours. Battery duration indicates how long a fully charged battery can operate at full power.

43 In addition, AIRAA 2026-2035 introduces two refinements, not present in ERAA 2025. First, it incorporates forced outage rates for batteries. Second, based on an analysis of connection A C E R O P I N I O N N O 1 2 / 2 0 2 6 contract data, it limits battery charging power to 50% of the installed capacity. This limitation applies only to batteries already in the system and not to new batteries in the generation forecast.

44 Relevance: Accurately representing battery size, availability and operational constraints is crucial for modelling power scarcity events. These constraints are key parameters in adequacy assessments, as they directly determine resource ability to cover scarcity periods and, consequently, affect the reliability of the results.

45 Assessment: Incorporating detailed characteristics of both existing and planned batteries in the SEM – reflecting both actual duration and availability – enhances modelling accuracy and is therefore justified. Difference #9 Limited grid feed-in in Dublin

46 Description: Maximum electricity generation from conventional power plants in the Dublin area is constrained, as not all available generation can be dispatched simultaneously without overloading power grid.

47 Relevance: These generation limits restrict the effective utilisation of available resource capacity, particularly during periods of system stress. As a result, power plants subject to such constraints contribute less to overall system reliability compared to an unconstrained generation fleet, since their dispatch may be capped below their technical availability.

48 Assessment: High short-circuit levels – often exceeding equipment design limits – can arise during peak winter evening demand or when renewable output is low and more conventional generation is online. This issue is particularly pronounced in the Dublin region due to rapid demand growth. AIRAA 2026-2035 accounts for these operational constraints, limiting the simultaneous operation of conventional generators in Dublin. In doing so, it captures the specific characteristics of the local transmission system, thereby justifying the adopted approach. Difference #10 No 2 North-South Interconnector

49 Description: The second North–South Interconnector (linking Northern Ireland and Ireland) is assumed to be operational from 2031 in ERAA 2025. However, it is not included in the national adequacy assessments for either Ireland or Northern Ireland. The project has been under development for over 15 years and is now fully consented and entering a critical stage of delivery. Excluding the second North-South interconnector at this juncture from the jurisdictional assessments illustrates potential adequacy concerns should the two jurisdictions continue to share constrained interconnection while the project is under delivery. Notably, this differs from the overall SEM (All-Island) assessment, which does consider the interconnector from 2031.

50 Relevance: Accurately reflecting the deployment of planned transmission infrastructure is essential for producing credible resource adequacy projections, as it represents the most likely evolution of the power system.

51 Assessment: Acknowledging the interconnector is included in the All-Island assessment, the exclusion of the interconnector from jurisdictional assessments is not justified. As its purpose is to relieve a bottleneck on the grid between the Irish and Northern Ireland jurisdictions , omitting this planned development prevents the model from realistically representing future system conditions and is likely to result in an overestimation of adequacy risks.

52 ACER recommends that the assessment includes the second North-South Interconnector in the jurisdictional assessments. A C E R O P I N I O N N O 1 2 / 2 0 2 6

3.2. Input data

Difference #11 Reduced electricity demand

53 Description: Compared to ERAA 2025, in AIRAA 2026-2035 both annual demand volumes and peak demand in the two bidding zones – Ireland and Northern Ireland – are slightly lower reflecting recent demand developments and national targets. Depending on the target year, annual demand decreases by between nearly 0% and 3% in Ireland, and between 4% and 7% in Northern Ireland. Peak demand, meanwhile, declines by 1% to 2% in Ireland and by 2% to 6% in Northern Ireland.

54 Relevance: Accurately capturing the most recent historical developments in electricity demand is essential for producing credible resource adequacy projections. It ensures that the initial years of the forecast reflect actual system behaviour.

55 Assessment: AIRAA 2026-2035 incorporates updated demand profiles that reflect the latest observed trends in the SEM, including economic growth, the deployment of heat pumps and electric vehicles, overall load growth (including the expansion of data centres), and changes in demand flexibility affecting both baseload and peak demand. Incorporating these developments is well justified, as it enables the model to better capture both the magnitude and temporal profile of future demand, leading to more robust adequacy conclusions.

56 This approach is also in line with the trends and projections scenario defined in the amended ERAA methodology. This scenario aims to capture the actual pace of the energy transition, which develops further from current levels but could remain below the trajectories envisioned in the NECPs .

57 In light of the rapid deployment of data centres in Ireland the Irish energy regulator has introduced a grid connection policy for data centres. The primary aim of this policy is to ensure security of supply while promoting renewable energy targets. Under this policy, data centres must provide onsite (or proximate) generation capacity equivalent to their peak grid demand.

58 ACER recommends that the assessment – including inputs to forthcoming ERAAs – continue to account for demand flexibility in the context of rapid demand growth. This should include modelling the expansion of explicit demand-side resources across the entire time horizon. Such approach should capture emerging sources of flexible demand, such as data centres with demand response capabilities (e.g. onsite dispatchable generation that can also participate in electricity markets), as well as the effects of national policies and incentives aimed at activating additional flexibility .

59 In AIRAA 2026-2035, however, the expansion of explicit demand-side resources is only considered up to 2029, with capacity assumed to remain constant through 2035. Although demand forecast does include flexibility from electric vehicles and smart meters, contributing to peak demand reduction, the lack of continued growth in explicitly modelled demand-side resources may underrepresent full future flexibility potential and could lead to an overestimation of system risks. A C E R O P I N I O N N O 1 2 / 2 0 2 6 Difference #12 Higher non-renewable generation capacity

60 Description: Compared to ERAA 2025, AIRAA 2026-2035 incorporates recent resource capacity updates. Higher conventional generation capacity in Ireland – by 61 MW in 2028 and by 461 MW over the period 2030-2035. Battery storage capacity is also higher in AIRAA 2026-2035 – by 2 MW in 2028 and by 432 MW between 2030 and 2035 in Ireland. In Northern Ireland, battery storage capacity is higher by 20 MW in 2028 and by 40 MW over the period 2030-2035.

61 Relevance: Accurately capturing the most recent historical developments in resource capacity is essential for producing credible resource adequacy projections. It ensures that the initial years of the forecast reflect actual system behaviour.

62 Assessment: AIRAA 2026-2035 incorporates updated resource capacities that reflect outcomes from the most recent capacity auctions, as well as revised expectations for the commissioning of new resources within the SEM. Incorporating these developments is well justified, as it enables the model to better capture capacity of future resources, leading to more robust adequacy conclusions.

63 Note, resource viability assessment in AIRAA 2026-2035 (applied only in sensitivity scenarios) indicates potential decommissioning of gas-fired power plants in the SEM from 1000 to 2500 MW, depending on the target year and sensitivity assumptions.

64 ACER recommends that the assessment – including inputs to forthcoming ERAAs – ensure that initial assumptions on battery storage (prior to EVA investment modelling) account for expected developments in the electricity system that may enable greater deployment of storage outside of capacity mechanisms in the coming years. In AIRAA 2026-2035, battery deployment in the SEM does not extend beyond 2030; installed capacity in 2030 is assumed to remain unchanged through to 2035. Difference #13 Reduced renewable generation

65 Description: Compared to ERAA 2025, AIRAA 2026-2035 projects lower renewable capacity in Northern Ireland, with reduction of 90 MW in 2028, 566 MW in 2030, 765 MW in 2033 and 380 MW in 2035. These adjustments are based on updated information regarding the pipeline of planned renewable energy projects.

66 In addition, AIRAA 2026-2035 applies downward scaling to onshore and offshore wind generation profiles used in ERAA 2025, based on historically observed availability in the last 12 years. In Ireland, onshore wind capacity factors are reduced from an average of 45% to 30% across all weather scenarios, while offshore wind capacity factors are reduced from 54% to 46%. In Northern Ireland, onshore wind capacity factors are adjusted from 42% to 29%, and offshore wind capacity factors from 47% to 45%.

67 Relevance: Accurately capturing the most recent historical developments in resource capacity deployment is essential for producing credible resource adequacy projections. It ensures that the initial years of the forecast reflect actual system behaviour.

68 Assessment: The primary recent driver of renewable capacity deployment in Northern Ireland is the development of a new government-backed support mechanism, the Renewable Electricity Price Guarantee (REPG). This scheme, together with the broader Path to Net Zero strategy and its 2030 target of achieving at least 80% renewable electricity, establishes the auction framework for future renewable capacity deployment.

69 AIRAA 2026-2035 incorporates updated renewable capacity projections for Northern Ireland, reflecting the current project pipeline. Combined with the adjustment of wind generation profiles in the SEM based on historical observations, the renewable generation assumptions in the AIRAA 2026-2035 are both appropriate and well justified. A C E R O P I N I O N N O 1 2 / 2 0 2 6 Difference #14 Delayed Celtic Interconnector

70 Description: While the Celtic Interconnector (Ireland-France) is assumed to be available from the start of 2027 in ERAA 2025, AIRAA 2026-2035 reflects a delay, with availability now expected from April 2028.

71 Relevance: Accurately capturing the most recent updates on infrastructure development is essential for producing credible resource adequacy projections. It ensures that the initial years of the forecast reflect actual system behaviour.

72 Assessment: AIRAA 2026-2035 incorporates construction delays affecting the Celtic Interconnector, shifting its availability to April 2028. Including this update is well justified, as it allows the model to more accurately represent the planned commissioning timeline of the interconnector, thereby supporting more robust adequacy conclusions.

4. Conclusions

73 Table 3 presents ACER’s assessment of whether the differences in the AIRAA 2026-2035 base scenario, compared with the ERAA 2025 central reference scenario, are justified.

74 The differences between AIRAA 2026-2035 and ERAA 2025 assessments contribute to divergent estimates of adequacy results.

75 For Ireland, the most significant driver of adequacy concern in 2028 is likely the delayed commissioning of the Celtic Interconnector, now expected in April 2028.

76 For Northern Ireland, the consistently higher LOLE across all target years is primarily driven by the restricted operating hours of the peaking gas-fired units, combined with the reduced wind generation due both lower installed capacity and a lower average capacity factor. Additionally, the exclusion of the second North-South Interconnector – expected to be commissioned in 2031 – further contributes to increased adequacy concerns in the years following its commissioning.

77 ACER considers that most of the differences in AIRAA 2026-2035, compared with ERAA 2025 edition, are justified. However, it finds that the omission of the economic viability assessment (EVA), the inclusion of the second North-South Interconnector in the SEM assessment only, and the allocation of restricted operation hours in the base (central reference) scenario are not justified. The latter two differences primarily affect the adequacy situation in Northern Ireland, A C E R O P I N I O N N O 1 2 / 2 0 2 6 whereas the absence of EVA has implications for both Ireland and Northern Ireland systems. As set out in the previous section, ACER recommends addressing these gaps to better reflect the possible evolution of the SEM system.

78 ACER recommends including a viability assessment of adequacy resources in the base scenario and ensure strong revenue consistency between the economic viability and economic dispatch modelling phases. EVA should also incorporate revenue stacking and consider demand-side units as potential expansion candidates. In addition, capacity planning – prior to the EVA stage – should account for the growth of explicit demand-side units and battery storage across all target years, reflecting rapid demand increases and renewable deployment.

79 Furthermore, the assessment should (i) include the planned second North-South Interconnector (to be commissioned in 2031) in all assessments, and (ii) allocate restricted operation hours that better reflect the seasonality and duration of potential resource scarcity events.

80 By implementing the recommendations set out above, ACER considers that AIRAA 2026-2035 would be on a clear path to aligning with ERAA. This Opinion is addressed to the Department of Climate, Energy and the Environment of Ireland, and to the Department for the Economy of Northern Ireland. Done at Ljubljana, on 7 May 2026. — SIGNED — V. ZULEGER, ACER Director ad interim