1. 1st Ten Year Network Development Plan: ERGEG opinion
Dr. Stefanie NEVELING - Bundesnetzagentur
Dr. Benoît ESNAULT – Commission de Régulation de l’Energie
GIF TF co-chairs
18th Madrid Forum
Madrid, 27/28 September 2010

2. Introduction ENTSOG published its first TYNDP in December 2009
ERGEG « pilot » evaluation of the 1st European TYNDP:
Aims at preparing the future task of the Agency
Methodology assessed in the light of ERGEG recommendations on the TYNDP (July 2010)
Based on the findings of the EWI Study commissioned by ERGEG : “Model-based Analysis of Infrastructure Projects and Market Integration in Europe with Special Focus on Security of Supply Scenarios” (May 2010)
Applied “TIGER”-model:  minimises gas dispatch costs subject to infrastructure, supply & demand assumptions;  represents an economic flow model assuming efficient organisation of transport & storage market (i.e. realisation of all efficient swaps & efficient capacity management)

3. Is the EU gas network sufficiently integrated? EWI’s results
Detected bottlenecks by EWI-study:
Physical need for network expansion until 2019:
“Potential” bottlenecks:
decreasing domestic production (e.g. in DK / S)  strong need for new cross- border capacity DE  DK
missing links in SE-EU for sufficient supplies during winter months (mainly HU & Balkans, somewhat eased with Nabucco or South Stream online)
resultant investments induced by new major infrastructures (e.g. Nord Stream/NEL)
preventive measure against crises  reverse flow projects mainly for Eastern EU countries in case of Russian supply disruption
For Western-EU, a potential need for capacity increases to improve market integration has been identified at several borders: DENL, DEBE, DECH, DECZ, UKBE, SKAT, ATSI, SIHR, DEFR on peak days
Such congestions are to be analysed on a case-by-case basis and might even be healed without physical capacity increases
There are some general West-to-East bottlenecks in the LNG “glut” scenario.
Consultancy study:
Short Introduction / background
One of the main results of ERGEG‘s/EWI‘s recent EU wide network modelling task – in conjunction with ETNSOG‘s establishment of a first EU wide 10 Year Network Development Plan – is, that indeed, in many regions, for Security of Supply, capacity is sufficient. But there are exemptions – for instance the connection of Germany to Denmark, but even more severely in South Eastern Europe.
I will not go into more details on the further results concerning detected bottlenecks - as they are stated in these boxes -, because they are easier to grasp when looking at a map, which I will provide on the next but one slide!
But let me focus on South Eastern Europe for a second or two:
Here (e.g. in Hungary & Balkans), integration is still lacking, especially during winter months or in crisis scenarios, as you can see on the following map:
ENTSOG: European Network Transmission System Operators for Gas
10YNDP: 10-Year Network Development Plan
FID: Final Investment Decision Taken
CAM: Capacity Allocation Mechanisms
CMP: Congestion Management Procedures
DK: Denmark
S: Sweden
SE-EU: South-Eastern Europe
HU: Hungary
NEL: Norddeutsche Anbindungsleitung

4. EWI‘s simulation results compared to ENTSOG‘s TYNDP findingsSE DK
demand-supply gaps found by EWI & ENTSOG GB LT NL BY BE PL DE
demand-supply gaps found by ENTSOG CZ UA
Sufficient Capacity to cover demand (incl. peak-day) for all Euro-pean countries, except for DK, SE, SI, HU, BA, MK, RS
results found both by EWI & ENTSOG SK AT FR CH HU Sl PT RO BA
On the other hand, ENTSOG‘s TYNDP reports sufficient capacities to cover demand (incl. a peak demand day) for the next decade in Austria, Belgium, the Czech Republic, France, Germany, Ireland, Italy, Luxembourg,
the Netherlands, Switzerland and the United Kingdom. These results are congruent with EWI‘s simulation. (Black dotted box)
According to EWI-simulations, severe supply-demand gaps are expected to occur during the next decade from GermanyDenmark, HungarySerbia, BulgariaMacedonia (full red arrows). Further supply-demand gaps are expected in some (demand & infrastructure) scenarios in the South-Eastern region (red circles, both found by EWI & ENTSOG)
For background info only:
Not replicated in EWI study are demand-supply gaps is in Slovenia (green circle) which may be due to differing assumptions on LNG supplies to neighbouring Croatia. EWI study includes the Krk LNG terminal which then allows supplies from Croatia to Slovenia. Croatia is not explicitly considered in ENTSOGs TYNDP.
Further (economic) congestion was found by EWI in Western Europe that might lead to price differentials on some days (orange arrows) and in an hypthetical LNG „glut“ scenario. (confer previous slide)
Modelling and Analysis of the functioning of the EU system Comparison of main results: ENTSOGs TYNDP vs. EWI - study[1]
With regard to the modelling of the EU network, the aim is to develop a European perspective (top-down approach) on infrastructure needs in the coming decade taking the need for integration of national markets and the European strategy into account (i.e. Priority Interconnection Plan/TEN-E). Furthermore, the scenario-based modelling task should help to identify possible bottlenecks within the EU-27 and at its borders and to help analysing potential effects of possible supply disruptions of major sources (various security of supply “crisis” scenarios).
ERGEG commissioned a study to the consultant EWI[2] on a “Model-based Analysis of Infrastructure Projects and Market Integration in Europe with Special Focus on Security of Supply”[3]. As the TYNDP shall include the modelling of the integrated network in different scenarios (in addition to a European supply adequacy outlook and an assessment of the resilience of the system), ERGEG initiated this modelling analysis in order to gain an understanding of the European infrastructure and to provide a basis for the discussion of the impact of selected major infrastructure projects on (cross-border) gas flows, physical market integration (i.e. bottlenecks) and the potential security of supply stress scenarios. This study has to be seen as a complementary approach to GTE+/ENTSOG’s early work on modelling that should help ERGEG / ACER in evaluating ENTSOG’s TYNDP.
The EWI-study broadly confirms ENTSOG’s findings of its 1st TYNDP, supporting the interpretation, that the EU gas grid (in terms of technical security of supply) is and will be sufficiently well developed assuming that all new included projects (“final investment decision already taken”) will indeed go online and that there is an optimal/efficient functioning of the market and use of existing network (i.e. all efficient swaps are realised, efficient CAM & CMP is implemented).
In terms of concrete results, both EWI as well as ENTSOG found sufficient capacities to cover demand (incl. peak day demand) in Austria, Belgium, the Czech Republic, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Switzerland and the United Kingdom. In addition, the model-based scenario analysed by EWI also allow for the conclusion that the necessary gas volumes in all considered scenarios (with different demand projections and on the peak demand day) are there to fill the capacities with natural gas.[4] As illustrated in the figure below (red circles vs. red arrows), five of six demand-capacity gaps identified by ENTSOG are also replicated by the EWI study (Denmark and Sweden, Hungary, Bosnia and Herzegovina, Macedonia, Serbia).
Not detected by the EWI study are demand-supply gap in Slovenia found by ENTSOG. This is probably due to differing assumptions on LNG supplies to neighbouring Croatia.[5]
When applying the TIGER model and following the volume based approach (in addition to the capacity analysis), the study allows for identification (and differentiation from supply-demand gaps leading to severe security of supply issues) of congestion on pipeline routes (congestion which is not so severe as to cause demand disruption, but limit market integration - “economic bottlenecks”), as depicted in the following figure[6].
Existent or expected physical bottlenecks“weak” market integration links, “economic bottlenecks”- decreasing domestic production (e.g. in DK / S)  strong need for new cross-border capacity DE  DK - missing links in SE-EU for sufficient supplies during winter months (mainly HU & Balkans, somewhat eased with Nabucco or South Stream online) - resultant investments induced by new major infrastructures (NordStream/NEL) - preventive measure against crises reverse flow projects mainly for Eastern EU countries in case of Russian supply disruption For Western-EU, a potential need for capacity increases to improve market integration has been identified at several borders: - UKBE, DEFR, DENL, DEBE on peak demand days only - ATSI, SIHR depending on LNG prices and time of the year - DECH, DECZ, SKAT Such congestions are to be analysed on a case-by-case basis and might be solved w/o physical capacity increases. There are some general West-to-East bottlenecks in the LNG “glut” scenario.
Apart from two additional stress scenarios (4 weeks of transit disruptions via Ukraine and from Algeria), the EWI study also exceeds ENTSOG’s work in another respect: The variation of infrastructure assumptions between scenarios leads to different potential demand-supply gaps, since they depend on which (major) infrastructure projects are realised. It has been found by EWI that in south-eastern Europe, demand-supply gaps are either reduced or eliminated if one of the major new import pipelines in the region is being built. This is valid for Greece, Hungary, Bosnia and Herzegovina and Serbia.
Further value is added to the EWI analyses by actually presenting the results (e.g. flows) on maps, compared to (national balanced) capacity considerations and (highly aggregated) diagrams by ENTSOG’s 1st TYNDP.
The TIGER model’s economic approach, its resolution and satisfactory resemblance of real flows (2008 validation) turned out to be applicable and sufficient for a first-time analysis of infrastructure and gas flow development in the EU and of ENTSOG’s 1st TYNDP, although some contract-induced gas flows (or even further pipeline-operational issues necessary for a technical simulation) cannot always be replicated.
[1] Model-based Analysis of Infrastructure Projects and Market Integration in Europe with Special Focus on Security of supply Scenarios. EWI, Final Report CEER_ERGEG_PAPERS/Gas/2010/EWI_Study_ pdf
[2] Energiewirtschaftliches Institut an der Universität zu Köln
[3]
[4] One exception was found for Greece, when taking into account availability of gas volumes next to capacities: “Even though sufficient import capacity exists, high demand in Turkey might in some scenarios lead to a reduction in Turkey-to-Greece gas flows causing a supply-demand gap in Greece when demand in Greece is also very high (peak demand day). While this may be only relevant in the extreme case of very high demand in both countries, it illustrates the importance of considering both capacities and volumes.”6
[5] The EWI study includes the Krk LNG terminal which allows supplies from Croatia to Slovenia, wheras Croatia is not explicitly considered in ENTSOGs 10YNDP.
[6] This figure represents a simplification and aggregation of the bottlenecks for illustrative purposes. The scenario specific results are to be found in Table 4 (page 66) in Section 8.2. of the EWI report. ES HR RS IT BG MK GR AL

5. Scenario: „Reference“ (only Nord Stream I)
Is the EU gas network sufficiently integrated in case of a crisis?
ERGEG/EWI study: Detected bottlenecks in “Ukraine crisis” scenario
Scenario „South Stream“ (Reference + South Stream)
Scenario: „Reference“ (only Nord Stream I)
The precondition for living solidarity is sufficient network integration and interconnection:
Regarding EU gas network integration and development, two important recent projects are worth to mention:
1.) ENTSOG‘s first 10 Year Network Development Plan
2.) An ERGEG/EWI study on EU gas flow modelling with varying infrastructure assumptions and different demand/supply scenarios, which is meant to als evaluate the TYNDP (as it is a future ACER task).
One of the main results of the EWI-study was, that indeed, in many regions, in terms of technical Security of Supply, capacity is sufficiently developed. But there are exemptions, where bottlenecks are existent and capacity increases are essential – for instance the connection of Germany to Denmark, but even more severely in South Eastern Europe:
Here (e.g. in Hungary & Balkans), integration is still lacking, especially during winter months or in a crisis scenario, as you can see on this map:
The results of an exemplary crisis simulation of a 4-week Russian Gas supply disruption via Ukraine in a reference scenario (only Nord Stream 1 in place) are illustrated, suggesting disruptions of customers (violet color) and significant supply cost increases in the South Eastern region (especially Romania).
In case South Stream will go online, the situation is somewhat eased, as visible (less violet) in the map on the left side. But still, bottlenecks remain, as indicated by the (black and dotted) arrows, and marginal supply costs are significantely increased (red colored area) in such a (admittedly extreme) crisis scenario.

6. Comments on TYNDP: scenario & modeling requirements
Results of ENTSOG’s supply, demand & capacity analysis:
Demand assumptions higher than EU COM’s Primes Baseline Scenario
Supply assumptions lower than forecasts by IEA
Aggregated capacity will be sufficient to satisfy peak-day & annual demand except for 3 regions w/ demand potentially > transport capacity
ERGEG comments:
Profound & detailed database on existing and FID infrastructure projects
But: lacking homogeneity in terms of details per country, identification of bottlenecks by regions is too broad (cannot be pinpointed)
Security of supply dimension not sufficiently studied yet (“crisis sc.”)
no enhanced EU infrastructure modeling or simulations (“Top-down perspective”) in varying infrastructure / supply / demand scenarios yet
Next TYNDP shall include maps presenting congestion in EU
Demand assumptions are higher than EU COM’s Primes Baseline Scenario
Supply assumptions are lower than forecasts by IEA
( from the infrastructure perspective on the „safe side“)
Aggregated Capacity will be sufficient to satisfy peak-day & annual demand ( assuming equivalent gas volume being available and FID projects are realised) except for 3 regions with demand potentially exceeding transport capacity
Supply, demand and capacity analysis
ENTSOG analysis of supply, demand and capacity developments from 2010 to 2019 includes peak day and annual scenarios; it also includes a comparison with alternative scenarios published by other organisations (IEA, European Commission, etc.).
According to ENTSOG’s peak-day demand analysis in terms of capacity, the report concludes, that during the next decade, there will always be sufficient (supply) capacity to satisfy peak-day demand. This is true under the assumption that, in total, not only the sum of all indigenous production, LNG, and storage capacity is sufficient to satisfy the aggregated peak-day demand, but that there are also equivalent gas volumes readily available.
Looking at annual demand scenarios (ENTSOG demand), the aggregated analysis concludes, that with respect to infrastructure, capacity will be sufficient until 2019, if all FID projects are realised. The annual potential supply scenario was split by potential supplies from existing and FID infrastructure and potential supplies from mature projects. This was compared with the annual demand scenario with the addition of 5 pipeline imports projects (Galsi, ITGI, Nabucco, South Stream and White Stream). Further “mature projects” or prominent ones such as “Nabucco” would increase the “safety margin” in terms of capacity.
ENTSOG compared also its forecasts on EU Indigenous production, the Russian pipeline import capacity, the Norwegian production and the North African production with the forecasts of the International Energy Agency (IEA), the Russian Energy Ministry and the Norwegian Petroleum Directorate. Concerning demand, the yearly aggregated demand scenarios for the EU27 was compared to EU demand scenarios of the EU Commission (Primes), Cedigaz, Eurogas and the IEA. The ENTSOG annual demand scenario is amongst the higher scenarios, with Cedigaz and Eurogas.
Regarding congestion in the network, the report identifies several regions (aggregated capacities) of different sizes, where demand potentially exceeds transport capacity:
- region Denmark/Sweden (from 2014 onwards)
- region AUS, GER, B, CZ, F, IRL, IT, LUX; NL; SUI, UK ( )
- region HU, MAC, SERB, SLO.
At the same time, there are some excess capacities in other countries that could increase security of supply in the above mentioned countries as soon as investments in interconnection capacities (e.g. “reverse flows”) are realised.
ERGEG comments and proposals
ERGEG welcomes ENTSOG’s previous and ongoing efforts towards an EU-wide TYNDP during the interim period, before the 3rd Package becomes applicable. The first TYNDP, which focuses on the existing and expected infrastructure enhancements from , creates a profound database for all interested market parties on existing infrastructure and projects, where the Final Investment Decision (FID) was already taken”.
The first TYNDP represents a detailed database on the European gas infrastructure, even if it still lacks of homogeneity in terms of details per country. It combines peak day and annual analyses, which enables to estimate two important dimensions of the European gas system in a reference situation elaborated by ENTSOG.
As first TYNDP, the result must be welcome. However, ERGEG expects some improvements about the content and analysis provided. First, the TYNDP shall investigate market integration, namely identify where the European system lacks of capacity in order to achieve the single gas market. Second, security of supply dimension should be studied further by testing the reaction of the system over longer periods of exceptionally high demand, as stated in the security of supply regulation, for instance. Third, scenario comparisons should include simulations of the behaviour of the EU gas infrastructure.
Concerning the presentation of the results, next versions shall include clear maps presenting congestions in Europe and consistent national descriptions. In particular, potential capacity gaps shall be easy to identify in the report. In addition, an analysis including the assumptions used should be included in order to provide a clear diagnosis on the future gas balances in Europe.
Scenario development and network simulation
ENTSOG’s supply projections are comparably lower than the ones from IEA or EU Commission. On the demand side, ENTSOG’s demand projections are rather higher than PRIMES baseline (ENTSOG). These relatively low supplies combined with high demand can be considered as a careful analysis with respect to infrastructure sufficiency. Nevertheless, this approach does not allow for a more sensitive scenario analysis with variable demand/supply assumptions and the analysis of the impact on the European gas infrastructure and security of supply in general.
In addition, the comparison of ENTSOG scenarios with scenarios of other international organisations should be explored and analysed more deeply, notably by testing their different stakes in terms of congestions and security of supply. Analysing different infrastructure scenarios would indeed be worthwhile to compare and evaluate different - potentially competing - (major) infrastructure projects (such as Nabucco, South Stream, Nord Stream etc.) and their impact on the existing system. ENTSOG did not elaborate such analysis.
A more sophisticated EU infrastructure modelling would therefore be welcome to enhance the top-down perspective on necessary European infrastructure development. It would also enable to analyse the interdependencies of gas flows within Europe and facilitate the identification of potential bottlenecks. Obviously, the applied ENTSOG model does not sufficiently address such requirements and does not deliver concrete elements on structural bottlenecks that can be geographically located or isolated. Furthermore, the current model does not include security of supply scenarios and does not evaluate the potential impact in terms of customer cut offs.
Since EU-wide technical flow models are not readily developed yet, an economic based network flow model, such as the TIGER model used in the EWI/ERGEG study, can serve as a first starting point in the discussions on modelling of European infrastructure development.

7. Methodology analysis: need for a top-down supervision
Combining top-down and bottom-up approaches is a key ERGEG recommendation
Data collection and assumptions
The first TYNDP: mainly built on data collection from TSOs and ministries (bottom-up approach)
Lack of data homogeneity and transparency of assumptions
Need for a transparent top-down process which would facilitate the assessment of the TYNDP
For the future TYNDP: ENTSOG should ensure that
TSOs elaborate coherent and consistent national capacity reports
Adjacent TSOs consistently address common cross-border infrastructure projects

8. Role of stakeholders and market consultation
Extensive consultation process conducted by ENTSOG: bilateral meetings and workshops with relevant market participants
Stakeholders should contribute to data collection
Collection of data for the 1st TYNDP focused on TSOs and ministries
Lack of involvement of project sponsors
Data on domestic European production was not provided by producers
Lack of data on non-EU gas production
ERGEG recommendation on the role of stakeholders:
Ensure a regular dialogue
Data communication to ENTSOG
Consultation on the Community-wide TYNDP

9. Addressing market integration and security of supply
Should be formally included in the analysis
Identify where the European system lacks of capacity
The TYNDP should contribute “to non-discrimination, effective competition, the efficient functioning of the market or a sufficient level of cross-border interconnection open to third-party access”.
Security of supply
TEN-E projects, the EEPR and the reverse flow study: no details on how these projects were taken into account in the TYNDP
Scenario comparisons should include simulations of the behaviour of the EU infrastructure – test different scenarios of infrastructure
Need to simulate system’s reaction over longer periods of exceptionally high demand

10.  Insufficient information to the market on necessary investment
Conclusions
ERGEG welcomes ENTSOG’s challenging work
Instructive process for all actors involved
Comprehensive database on infrastructures and projects
Findings on the status of the European gas system
According to EWI-Study and ENTSOG’s TYNDP, the EU gas grid - in technical terms - is well developed (under assumptions taken)
But some (physical and potential) bottlenecks identified
Precondition to avoid inefficient network expansion:  effective capacity management
Room for improvements:
Methodology & homogeneity of assumptions
Results (identification of bottlenecks) more precise (map)
Security of supply dimension (missing “crisis scenarios”)
Gas flow modeling for varying scenarios
Room for improvements according to ERGEG recommendations & pilot evaluation of ENTSOG’s 1st TYNDP, especially concerning:
Gas flow modelling with varying scenarios concerning infrastructure / demand / supply (3 Dimensions)
ERGEG/EWI-Study generally confirms ENTSOG’s TYNDP, that the EU gas grid - in technical terms – is/will be sufficiently well developed
But: Some bottlenecks have been simulated in the EWI study suggesting necessary network expansion for improved EU network integration!
Precondition to avoid inefficient network expansion:  effective capacity allocation mechanisms and congestion management
 results are only aggregated at a very broad regional level  no identification of specific physical bottlenecks (e.g. location / certain interconnections) possible  ENTSOG rather wants to provide the 10YNDP as a “communication tool” to the market without directly interfering in investment decision making processes
 Insufficient information to the market on necessary investment
ENTSOG’s first TYNDP can be regarded as a wide-ranging documentation (“communication tool”) on infrastructure (pipelines, LNG terminals, gas storages), “firm” infrastructure projects, production deliverability and import capacities. Confronting those with an annual and a peak day demand scenario allows for broad conclusions on the level of capacities compared to demand-capacity balances for each European country.
The methodology followed by ENTSOG essentially corresponded to a bottom-up approach. ERGEG recommendation is to combine it with a “top-down” model-based scenario analysis in order to increase validity and have a final outcome fitting with the legal requirements and having the characteristics stakeholders could expect from an investment plan. Addressing European security of supply issues (disruptions) and their potential consequences on EU gas infrastructure in different scenarios would add value to the TYNDP as well.
The implementation and application of an infrastructure based model for the development of the next TYNDP seems also essential for identification or indication of concrete existing and potential infrastructure bottlenecks that can be illustrated and pinpointed on a map and translated into specific projects.
Even though both ENTSOG and EWI generally report that the EU gas grid (in terms of technical security of supply) is and will be sufficiently well developed under the assumptions taken, some capacity increases have to be realised nevertheless (e.g. resulting from decreasing domestic production, missing links or measures against crises). This is true under the assumption that gas flows are not hampered by inefficient capacity allocation / congestion management.
ENTSOG is already working on improving their European analysis model in At the past workshops, participants delivered positive remarks a well as critiques on the approach. This constructive dialogue with stakeholders shall be continued. Not only the Madrid Forum qualified as a good platform, but also the established links of working groups of European regulators and ENTSOG. ERGEG/ACER is looking forward to further fruitful cooperation with ENTSOG on that matter. Therefore, the experience with EWI’s TIGER model and their analyses will be gladly shared with all stakeholders via a public consultation and may contribute some aspects (such as methodology, scenario assumptions etc.) to ENTSOG’s next TYNDP, which will be published in In the meantime, results of the ENTSOG’s and ERGEG’s work will surely be discussed on a national level, thereby potentially influencing national network development plans.

11. Way forward
Analysis of consultation results / comments by stakeholders on EWI-study
Providing input & advice to the
EU COM - on Energy Infrastructure Package
ENTSOG - on TYNDP methodology for the 2nd TYNDP and 2010 Network modeling improvements; next milestone: ENTSOG workshop 7. October
ACER - on its future tasks related to TYNDP, refining further aspects, such as coordination & relation to national /regional network development plans

12. Thank you!