Why are we climate scientists? For us, there is a number of reasons: we feel a strong bond to nature, we like to solve puzzles and we want to understand the mechanisms of what we see every day. And – even if it only manifests at the end of a causal chain – we want to contribute to a just and livable world via working in climate science.
Thus, due to distant and abstract state funding or because we subscribe to it on a personal level, most of us contribute to reaching the UN Sustainable Development Goals. At its heart, this collection of global challenges wants to achieve a better and more sustainable future. But this can only be accomplished by one thing: working together.
Today’s financial and technological infrastructure makes international collaborations easy, but we also need a philosophical, moral framework to do so – a “Hippocratic oath” for scientists if you will, becoming the guardians of the ga..science. We already have a name for that, it is called Open Science and FAIR data. The benefits of sharing climate data openly are very clear:
- The climate crisis CANNOT be addressed if our knowledge is hidden and we know that data availability upon request simply does not work. The Paris Agreement states that its implementation “requires […] transformation based on the best available science”. The keyword here is “available”. What good can the best science do, if it is not available?
- Open data is also needed for grassroot social and environmental justice ideas. Data is key “[…] as a tactic in environmental justice advocacy”. Data can tell a story and can help underprivileged communities taking a stance against much more powerful adversaries. But how can these communities use the data, if it is not out there?
- From a climate science perspective, observations from a single location usually cannot tell much about the entire climate system. It needs to be put in context, to be compared to other data and different regions to generate knowledge and create an impact.
- And finally, open data makes sense financially. Think about the return of investment we had from sharing weather observations between nation-states to create the best possible weather forecast for all humankind – the benefits were and are immeasurable.
These examples demonstrate the advantages of openly sharing climate data. But that also means we have to bid farewell to the idea of data “belonging” to the scientist. To be blunt, researchers who hoard data or raise obstacles act as barriers to discovery. Or to put it in the words of Mick Watson: “Open science isn’t a movement, it is just (good) science. It is also the future”.
Unfortunately, we are not quite there yet: A recent study found that 33% of contacted scientists declined data sharing, with most stating lack of time to search for data or simply loss of data as reasons. That said, we can see amazing examples already coming out of the climate community. The World Climate Research Programme made sure that all the CMIP6 model runs are openly available – a huge data basis for not only the IPCC, but thousands of papers and countless use cases in the private sector. The Linked Paleo Data effort curates the often rather chaotic distribution of paleo-climate data and makes it much easier to connect proxies with model output, so we can generate new knowledge about the climate.
The immense impact open data can have for society could also be seen in the recent stir-up paper by Google, who managed to create a data-driven weather forecast that can beat the best-of-the-best process-driven weather forecasts in the world. Even though this is a result of immense human and computing resources at Google, fueled of course by their financial capabilities, the Acknowledgements at the end of the paper tell you the key ingredient of this recipe, namely open-access training data from our very own community, from Copernicus: “It is important to emphasize that data-driven MLWP relies critically on large quantities of data and their quality […]. Therefore, rich high-quality data sources […] are invaluable.” Thanks to that open data, weather forecasts will be more affordable for a wider range of nations, stakeholders, or scientists, and more human and animal lives can be saved in the future. And that is how we come a tiny bit closer to reaching the UN Sustainable Development Goals.
These large-scale examples are all nice and dandy, but how do we apply these ideas on an individual level? From a technical perspective, you can find high-level guides, e.g. (ironically behind paywalls) in Nature and Science describing best practices on how to publish your data. And for more hands-on advice, we refer you to the very competent open science teams at your institution.
But for us, all this comes second to one key aspect: Being so open sounds vulnerable, and being vulnerable sounds extremely scary… And you are right, publishing data and code means being vulnerable, and yes, that is scary. We know. We are scared too, we feel you.
But what exactly are we afraid of? Are we scared of the fetish for perfectionism in our community? The fear of finally being exposed as an imposter, to not belong here? The notion that mistakes are signs of failure?
Well, we truly believe: That’s bull****. Forget perfectionism – it is the mortal enemy of any progress. And you absolutely belong and we love to have you.
In fact, what society needs from you right now is not to be perfect, but to be brave. You are amongst the few that really understand the complexity of the climate system, the limitations of your data, and the potential of what can be done with it. The urgency of the climate crisis does not allow us to wait any longer. If we want to be part of the solution, part of change, we need to be brave – for minorities, Indigenous communities, future generations, and ourselves.
And yes, being brave never was and never will be easy, that one is for sure. But maybe we can actually learn from the efforts and lessons in climate change mitigation: We don’t need to be perfect, we just need to try a tiny bit more. So let’s make climate science a tiny bit better, let’s make it open. With our simple code, our messy annotations and our weird data formats. That would be brave.
This is a heavily edited summary of the recently published Viewpoint “Open Climate Science is brave Climate Science” by Lena M Thöle and Martin Wegmann 2024 Environ. Res. Lett. 19 122001. DOI:10.1088/1748-9326/ad893f
