How Data Can Help Solve Climate Change, Part 1
This post continues my climate journey as a software systems thinker seeking my place in climate.
Climate change is an incredibly complex problem, underpinned by an equally complex system that is driven by intricate interactions between a large set of forces and actors. How do we make progress in fixing this system when the success of any specific intervention is at the whim of these complex dynamics?
One option is to ensure that every solution we deploy — greener power, electrified transport, novel food production, carbon capture, etc. — has considered all system forces and aligned them for success, but this can quickly become impossible especially when many solutions are being tried, incentives run in conflicting directions, and the system is constantly changing as a result (as is the case in climate response).
System thinking gives us another option. It tells us that one of the most powerful approaches is to find levers that change the underlying dynamics of the climate system — setting up mechanisms that cause all forces and actors to intrinsically move toward alignment on the common goal of reducing carbon. In this post, I argue that data is such a lever — and a vastly-under-leveraged tool for solving climate change.
Data — in the form of metrics, standards and schemas, analytics, reporting, and common infrastructure — provides visibility and insight into the behavior of system participants. It answers questions like, how do we know whether a producer has reduced their emissions, or is merely “greenwashing”? How do we trust that a climate technology solution actually reduces emissions or reliably sequesters atmospheric carbon, and by how much? How do we manage a resource like atmospheric carbon load in the aggregate, regulating the inflows and outflows and identifying the greatest opportunities for improvement? How do we standardize reporting such that it can be trusted, independently verified, and used for accountability and novel economics?
And data doesn’t just answer those questions. It creates a grounding of common facts that helps cut through disagreements on truth or on the path forward; it creates factual visibility, that, properly deployed, can lubricate the climate system, reshape system forces, and create “guardrails” that steer system actors into new behaviors. High-quality, trustworthy data can:
- Create new forms of accountability — for example, illuminating facts about carbon footprints and enabling producers to be held to account for reductions by governments, activists, and/or shareholders, creating guardrails against bad behavior (like greenwashing) and incentives to adopt climate solutions
- Empower regulation — for example by making it possible to set resource targets and caps in a more meaningful way, and increasing “teeth” for enforcement
- Power the design of new economic incentives — for example, credits based on provable, durable removal and sequestration of atmospheric carbon; and new economic constructs — like currencies tied to carbon removal (e.g. Global Carbon Reward) — that use market forces to incentivize behavior
- Identify and increase focus on high-leverage opportunity — from providing the insight on where to optimize supply chains for the highest impact, to providing ground truth that steers investment capital to the highest-impact emission-reduction or carbon-removal solution areas as opposed to just what’s trendy
- Shape the cultural narrative by activating governments, activists, NGOs, and culture-influencers and giving them the facts they can wield to counter entrenched interests and shape new policy, advocacy, education, and cultural understanding — like An Inconvenient Truth did in the early 2000s.
Critically, data can do all these things at the same time. It’s a high-leverage intervention that can alter the “laws” of the system so it reshapes itself, a far more scalable solution than having to individually understand system conflicts and intervene to fix them for each climate solution. This means that building a robust and trustworthy data foundation is one of the most powerful tools we have to address climate change. Data is also a platform opportunity, which we need more of in the climate solution space, as I discussed in a previous post.
But how do we capitalize on this opportunity? What does a data platform look like in climate? There is, of course, plenty of climate data available today — tons of datasets are published by governmental groups and NGOs — but it’s far from a platform, with disconnected data silos, inconsistent data standards, and a lack of freshness making these datasets difficult to join and use for timely decision-making. Conversely, there are limited examples of decision-centric data platforms already in place in climate management, specifically in carbon accounting & offsetting (e.g., Persefoni, Sinai, Watershed, Patch, Gravity), though those platforms today also remain siloed and aren’t yet sufficient to drive system-wide change. I argue that we can, and need to, go farther.
I propose that we need a complete data platform for climate consisting of a combination of public and private data infrastructure and data applications, all tied together by common standards and data exchange. Having both public and private data infrastructure platforms that work together is key: private data platforms solve the need of producers and companies for “climate management” tasks like monitoring, optimization, reporting, and internal decision-making; in doing so they incentivize emission, energy, and resource-use measurement and data collection.
Conversely, in the public sphere, a public data commons is essential for governments, NGOs, activists, and the entire system, providing public transparency and accountability, enabling policy and regulatory decision-support and enforcement, and providing fodder for culturally-relevant storytelling. But a public commons can only achieve these goals if there’s high-quality, standardized, up-to-date data flowing in, which is a hard problem to solve given the cost and disincentives for transparency involved in widespread data-sharing. This is where the integration of private and public data platforms becomes critical: private data platforms provide natural value to system participants, and if those private platforms are designed to feed and interconnect with the public commons, then the public commons emerges almost as a side effect of natural activity in the private sphere.
Taken together, the integrated set of private and public data platforms provides the basis for a wide variety of climate data applications, which transform the raw data into insights that provide the transparency, analytics, decision-support, reporting, and storytelling that enable data to reshape the system.
Finally, a few more pieces are needed to complete the data picture. Data standards are critical, as they provide the glue that enables private and public platforms and applications to interoperate and share a common language — this is a huge gap today as the existing repositories of climate data vary widely in quality and coverage, and are difficult/impossible to join together on common dimensions. Designed properly, standards can also enable trust, verification, and privacy. Along with standards, a comprehensive methodology for ensuring data quality is critical — both when data enters the system, and via ongoing checks to ensure it remains accurate and relevant — since low-quality data leads to low-quality insights and bad decisions. And data sources & instrumentation are the final piece of the puzzle, the bedrock layer that collects raw ground truth on the behavior of the climate system.
System actors also play essential roles in the data picture, whether providing ground-truth data or plugging into the public data commons e.g. in the role of independent verifiers, reporters, or storytellers.
A high-level block diagram of the entire data platform might look something like this:
At this block-diagram level, this picture will look familiar to anyone who’s worked on large-scale public/private data infrastructure (it certainly is reminiscent of what I’ve seen working on data infrastructures inside Google across areas like search, healthcare, and knowledge graphs). And that gives a certain comfort that creating this kind of climate data platform is possible. But the devil is in the details, so if you’re interested in how this data platform vision might be realized for climate data, and how it could solve some of climate’s greatest systems problems, check out part 2 of this post.
In the meantime, as always, I’m eager to discuss these ideas with others thinking along similar lines — conversation and feedback welcome! You can comment here or find me at abbrown at gmail.com.
