Why the future of energy data is open source

Data is the most valuable resource of the 21st century economy. Yet it is unlike any traditional resource in a very fundamental way: it can be copied and shared infinitely. This post explores what the future of energy data looks like, and how an open and collaborative data framework can help improve information flows to expedite the transition to a zero-carbon economy. With collaboration being one of the four pillars of COP26, the importance of open data should be higher up the climate agenda.

Data is recognised universally as crucial to supporting decarbonisation efforts. Despite this, open data, as defined by the Open Source Initiative, is still limited. There are instances where open source data is problematic in the energy sector; for instance, hacking of electricity grids is seen as a growing national security risk. There are, however, often more mundane reasons why open source data is the exception rather than the norm in the energy sector. These reasons include a lack of standardisation and shared repositories, as well as a culture of hoarding to preserve business models.

Indeed, data hoarding is commonplace even amongst not-for-profit organisations. With the exception of a limited number of progressive organisations, such as Open Climate Fix and CarbonPlan, it is largely unheard of for not-for-profit organisations to open source their models for fear of credibility and replication challenges. There are two reasons why we believe these fears are likely exaggerated.

First, the more not-for-profit organisations share and collaborate, the more data gaps they will identify. Since most data and the associated assumptions are not shared, users of this data cannot present their own analyses, and thus cannot build on the work of others. As a consequence, hours of work and funding could be spent on work that has already been done, rather than on genuine collaborations that focus on areas of genuine value.

Second, open data can create open markets, which have the potential to reduce costs for the end consumer and spur a Cambrian explosion of use cases. According to analysis by the Energy Systems Catapult, a UK thinktank, data visibility helps support a modern energy system through greater levels of operational optimisation. This is particularly evident with production or utilisation data from energy assets. For instance, analysis by TransitionZero, showed how plant-level production estimates of Chinese coal plants could be used to back-calculate other important data points, such as fuel consumption, carbon emissions, emissions intensity, water use, air pollution, operating costs, net profitability and abatement costs. These estimates can then be used to inform carbon trading and net-zero alignment analysis.

While hoarding behaviour can preserve not-for-profit business models and employee salaries, it is likely slowing the transition to a zero carbon economy. This is particularly the case in the electricity sector, where governments will need to incentivise and regulate an increasingly complex system, due increased levels of variable renewable energy and the electrification of other sectors. For this reason, TransitionZero is embarking on an ambitious project, Future Energy Outlook (FEO).

Published twice a year, FEO intends to provide critical insights on energy markets, policy, and the future of the global energy system. It will be based on objective data, transparent assumptions, open-source modelling, and dispassionate analysis. FEO will use a scenario-based approach and rigorous scientific methods to illustrate how the energy system might evolve during the coming decades and how it is affected by key variables, chief among them the policies adopted by governments around the world. Crucially, FEO will be guided by three high-level principles:

• Transparent model methodology, assumptions, and code base available for users to replicate
• Informed by a geographically diverse and internationally regarded steering group to ensure local buy-in
• Based on least cost principles with no policy adjustment, so this pathway can clarify the cost of decarbonisation

Moreover, FEO will be supported by a steering committee of geographically diverse and internationally regarded institutions. The provisional plan is to productionise the model and publish the first report in Q4 of 2022.

If information precedes action, energy data needs to be open and collaborative to support decarbonisation efforts. In doing so, data-centric organisations would save time, be more effective and support business models previously unimaginable. We believe data science and open data is a vastly untapped resource to fuel the collaboration necessary to power the transition to zero carbon.