Long before FCDI came into being, CSIRO was leading the way in climate science and projection modelling. FCDI has always had CSIRO as a key partner and right from the start, well before we finally got funded. Together, we work to improve climate decisions using better climate science.
Without any exaggeration, CSIRO is one of FCDI’s most important partners in furthering Australia’s contribution to global climate science and answering climate questions.
CSIRO spearhead much of the research in this space and FDCI is proud to fuel the engine of climate forecasting and modelling they use to peer into the future. In this edition, we’ll be exploring the origins and purpose of FCDI’s relationship to CSIRO and how through our partnership, public and private sector organisations are able to forecast the climate with more trustworthy tools and greater confidence than ever.
Read our interview with John Clarke (Research Team Leader, CSIRO Climate Science Centre) and discover how FCDI and CSIRO work together to empower Australians with better climate science.
– Dr Tomas Remenyi, Chief Investigator
Iron Sharpens Iron – CSIRO’s partnership with FCDI
With CSIRO (Climate Science Centre) Research Team Leader, John Clarke.
Long before joining CSIRO’s Climate Science Centre, John was working in the television industry, lugging hefty, pre-digital (one imagines Beta-Max) video recording equipment around the wilderness of northern Queensland and NT.
While working in a conservation management role in the national parks service as part of his television career, John became discouraged about how the traditional conservation paradigms were struggling to adapt to the new climate change narrative.
And in a most extreme example of a divergent career change, John took off the Merrell’s and turned in the TV gear for a stack of books on data modelling. Specifically, climate data.
John now leads a team of researchers specialising in developing climate change projections. John and his team’s work focuses on the near future of the climate: decadal and multidecadal timescales. And their objective is to draw on multiple lines of the best quality evidence in the form of data, to generate answers to some of the most urgent climate questions being posed today.
Scaling from Continent-Sized to A Bit Bigger Than Vatican City
When hearing the nomenclature surrounding John’s wheelhouse, Regional Projections, one might be fooled into thinking his focus was on areas the size of a local government area (LGA), or a district like Riverina, for instance. This is only partly right.
"In that context, regional means anything smaller than the whole planet”, shared John. “So, a region could mean a whole country, like Australia, or at the opposite end of the scale, a small Pacific island like Niue, which is tiny”.
John’s work, he explained, is projecting into the future a variety of plausible climate scenarios expected to occur within a specific region.
“What we’re projecting is how the main state of the climate might change in the future”, John continued, “so when we’re trying to gain an understanding of that future, we use very sophisticated computer models, which are in their essence, very similar to weather forecast models. However, they are tuned in different ways to look at larger-scale systems over longer periods”.
Take greenhouse gas emissions, for example. “So we have to use a set of scenarios that isolate fossil fuel burning,” John says, “and we can then ask what happens if they continue at the current level, or if there is a reduction in greenhouse gas emissions?”
We quickly find that the weather metaphor runs dry for comparison when scaled up to continent level climate projections several decades into the future. Unlike the weather forecasts we’re all used to, explained John, the possible futures being investigated are both more complex, integrating many different types of data, and offering few opportunities for a feedback loop:
“To look into the future, one might predict it’s 30 degrees and sunny, but it might be 29 with patchy cloud when tomorrow comes. We can then look at those differences and use them to inform the model and tweak it accordingly. We’re not afforded that because we’re projecting decades into the future when it comes to climate projection.”
Asking complex climate questions using FAIR data
The ability to perform these kinds of feats of data science inevitably rests on the availability of evidence. That evidence is the growing pool of real-time (or close to real-time) data constantly uploading and refreshing itself from myriad data sources harvested from millions, perhaps even trillions, of observational instruments that generate petabytes of data.
When observations match with projections, particularly those from the 90s, we can be confident that the models are doing a good job.
“Our view on the best way of doing climate projections is to draw on as many sources of information as you can. That increases your confidence that collectively, all of the information used together has a much better chance of covering the range of possible futures, including the one that is going to happen”.
What makes this almost science-fiction sounding endeavour possible is the marriage of CSIRO’s climate modelling expertise with sources of FAIR Data. What is FAIR DATA?
“Findable, Accessible, Interoperable, and Reproducible, that's what the acronym stands for. Where we expend public money to produce data, it should be easy to find it should be easy to use, and we should set it up so that multiple people across multiple platforms can use it. And you ought to be able to find enough information that you could then yourself go off and make your version of it and get the same results. That's reproducible. And systems like the Federated Climate Data Initiative are right in that space. They're ticking all those boxes”.
“Where the Federated Climate Data Initiative comes in,” John explains, “Is by making a system that allows the open sharing of data that’s been produced by all the data modelling people around the world. It makes it much easier for everyone to draw on that data, wherever they are. Then, if users want to drill down deeper and look at a particular part of that plausible future, they can find the best data that best fits the kinds of questions they want to ask”.
Connecting the Climate Science Dots - a podcast from the FCDI
Get the inside scoop and hear how our amazing team is overcoming obstacles to create the FCDI with Connecting the Climate Science Dots.
Hosted by former SBS and ABC journalist, Sam Ikin, Connecting the Climate Science Dots is available wherever you get podcasts.
AUTHORS Dr Tomas Remenyi
COVER IMAGE Jake Allison
Through these types of initiatives, Eratos enables federated access to data, facilitating the building of sophisticated AI models and workflows to transform the data into actionable insights across research, government, smart cities, industry and beyond.
The Eratos Platform utilises Amazon Web Services (AWS) to leverage best in class cloud scaling architecture, including Amazon Elastic Container Service (Amazon ECS), Amazon Elastic Kubernetes Service (AWS EKS), Amazon Elastic File System (Amazon EFS) and Amazon Simple Storage Service (S3).
Comments