Can open models replace PRIMES?


#1

BrusselsThink Tanks want to discuss with the commission to rely their climate strategies on more open/transparent models than PRIMES and invited to an expert roundtable on “Modelling for the new Climate Strategy” mids September. To understand themselves the alternatives that exists they invited various modelling experts (not only/not a lot from open models). If you have any input for them about which open models could best be used for which purpose please let’s gather that here and take it along.
Further concrete questions they will discuss:

  • challenges and approaches in modelling net zero emissions
  • crucial aspects to be included in the new climate strategy’s modelling
  • leveraging relevant recent research for the new climate strategy’s modelling

It would be great to have insights and possibilities from various open models

If you have some “central messages” from the openmod side => I’m happy to include them.

Best
Berit


#2

Berit, thanks for this gathering of inputs!
I think with respect to the second bullet point, crucial aspects for any model / modeling suite to be considered for climate strategy modelling would be:

  • be aware of reporting requirements under the Energy Union Governance; including the level of detail and the reporting formats.

  • being able to provide integrated climate and energy modelling; i.e. also output GHG emissions, split by gas; with a high sectorial level of detail (including a split into ETS and non-ETS emissions per sector modeled)

  • being able to do so on EU-28 level and individually on Member State level, following the same formats in order to account for Member State specific policies and measures and national climate and energy targets that may be in place.

Best
Hannah


#3

Assuming that the EU-28 becomes the EU-27 on 29 March 2019, the following reports might be of use when considering both models and modellers.

References

Boffey, Daniel (30 May 2018). “Brexit: UK may get poorer access than Israel to EU science scheme”. The Guardian. London, United Kingdom. ISSN 0261-3077.

Pye, Steve, Carole Mathieu, and Paul Deane (January 2017). The energy sector implications of Brexit. Europe: INSIGHT_E.


#4

Thanks Hannah,
I will take it along!


#5

hi Robbie,
thanks
Steve will be there, too. I think he will contribute about Brexit implications


#6

That are key messages I gathered:
challenges and approaches in modelling net zero emissions
• Openness is crucial so that all stakeholders can see and understand the assumptions, and participate in modelling exercises as well as in the implementation of transformation measures
• Typical day approaches and copper-plate grid models miss to model important grid bottlenecks (that block the transport of renewable energy) as well as the full variability of renewables
• Integration of behaviour (which is basic to evaluate feasibility of the pathways) is done through to explicit assumptions but mostly miss to take into account the functional chains (Wirkungsketten). Other approaches are agent based models and adaptation of the modelling processes.
• Just to emphasise that: Integration of behaviour also means not only focusing on the supply side but also focusing more on mitigation options of the demand side
• Biggest challenges: high complexity versus manageability of the models; data availability

crucial aspects to be included in the new climate strategy’s modelling
• Bring together EU and national (as well as national and regional) climate strategy modelling
crucial aspects for any model / modelling suite to be considered for climate strategy modelling would be:
• be aware of reporting requirements under the Energy Union Governance; including the level of detail and the reporting formats.
• being able to provide integrated climate and energy modelling; i.e. also output GHG emissions, with a high sectorial level of detail (including a split into ETS and non-ETS emissions per sector modelled)
• being able to do so on EU-28 level and individually on Member State level, following the same formats in order to account for Member State specific policies and measures and national climate and energy targets that may be in place.

leveraging relevant recent research for the new climate strategy’s modelling
• MEDEAS project (open world/EU model); taking into account material flow (I’m not sure how far they realised the last point, but that was the program)
• Opening of JRC TIMES (EU) (2019)
• With PYPSA (focusing on the electrical grid) T.Brown and J. Hörsch show the Energy Transition in Europe is possible without grid reinforcement, but is 10% more expensive than with grid reinforcement
• Awareness raising and first steps to transparent data (IDEES, OEP, OPSD,…)


#7

Hi @berit.mueller It may be worth adding two more thoughts:

  • the current mismatch between the IPCC IAMs (sometimes “integrated models”) and most energy models, possibly also reflective of a lack of contact between the two communities (please correct me if I am wrong, others might wish to comment?). In any case, the IAM scenarios and the energy modeling scenarios are highly disjoint at present. The IAMs are full of CCS and BECCS with allegedly unfavorable renewables characterizations, while the reverse applies for energy modelers. Moreover CCS has serious public acceptance problems at present.

  • the next batch of models is not only a question of increasing model fidelity and scope (including human behavior and social dynamics), but developing new market designs and coordination mechanisms or however you want to describe the new algorithms that will manage near-future systems with very different architectures, characteristics, and information processing potentials (on that last point, the energy needs of the IoT at scale needs looking at).

I rather suspect @tom_brown is going to pull you up for over-interpreting their paper on transmission build-out (Hörsch and Brown 2017). Their results are based on perfect foresight and coordination (hence refer to point 2 above).

PS: Wirkungsketten also translates as causal relationships in the sense of a cascade or at least interconnections — is that a reference to system dynamics modeling?

References

Hörsch, Jonas and Tom Brown (June 2017). The role of spatial scale in joint optimisations of generation and transmission for European highly renewable scenarios. 14th International Conference on the European Energy Market. New Jersey, USA: IEEE. doi:10.1109/EEM.2017.7982024. arXiv: 1705.07617.


#8

Hello @berit.mueller The following summary is mostly aimed at IAM models but many of the points raised would apply to energy modeling.

References

BPIE and Climact (May 2018). Bridging the gap between modelling and new policy expectations — Briefing paper. EIT Climate-KIC.


#9

Thanks for your inputs and here are my short notes about the meeting:
EC representatives insisted that they need answers to a lot of questions (opening all tasks like uncertainty, innovation, behaviour, …) because for the next period climate and energy issues are one of the three important subjects of the EU.
That’s why they also seem to give much more importance to this new roadmap scenarios (with PRIMES; which was the reason for Bruegel to bring together this meeting because they wanted to see what else is possible). After the publication of the roadmap it should be approved in about a year.
Models should better help politicians to look at the divers points in question (also biodiversity, landscape,…
One repeated thing concerning assumptions was “be as less conservative as possible” which was especially underlined by the underestimated price development in the past.
There was a longer discussion about complexity and transparency. I think the request for transparency more and more reaches the EU – or even better is requested by the EU. New for me: IDEES now has licensed open Data (that is what the JRC representative said – but one week later at EMP-E together with Robbie we had a further discussion with them and the status is still the same as the one we discussed in the openmod (unclear if republication is possible) – we will stay in touch and push for a clear license (everybody of you should do so who is in contact with them)). Additionally they said that they opened the process of decomposition of the data to the national levels (which was not wanted when we talked in February)

One challenge we discussed was the linkage between local, national and EU models/scenarios or as someone phrased it the “consistency between the big picture and millions of small projects”. As far as I got it we didn’t come to a conclusion.
One statement that touched me was: from the macroeconomic point of view we have to decide for a technology path (e.g. e-mobility or hydrogen; etc.) whilst I thought of the “cellular approach” where every “cell” follows their approach and then we try to link them. – I cannot follow this logic I wonder if the industry market isn’t big enough for different technologies and if it wouldn’t be better from a material point of view.

Material: In a side talk someone said that there is a CGE Model with integrated material flow(from Netherlands) developed in a Project (EXIOPOLE or similar (do you know it? Otherwise I should try to get back to them)) with a free material database - this approach has been done in a static way , not structural change,…

Next discussion point GDP… index for income and for production (not for income inequalities…) ; Which indicator do we have to capture circularity? (no answer in that round)