A Close Look at Australia's New Climate Assessments
Fun 'n games with climate scenarios
Today’s is the first in a series of posts for THB Australia Week. Coming soon, an update on trends in Australian normalized disaster losses and an analysis of historical and projected decarbonization of the Australian economy. Please share THB with your Aussie friends, family, and colleagues!
Last week, the Australian government released a suite of reports on climate change. Among them were:
The first National Climate Risk Assessment (NCRA) — which projects the future risks and impacts of changes in climate for Australia;
Modelling Sectoral Technology and Emissions Pathways to 2035 and Net Zero Emissions - Final Report (NZ35) — which projects technologies and costs associated with implementing Australia’s proposed path to net zero.
Today, I take a look at the scenarios that underpin these assessments and find some significant methodological issues — which compromise the utility of the assessments for informing policy.
Specifically, the good news — that the NCRA leaves behind SSP5-8.5 — is canceled out by some bad news: The NCRA emphasizes an equally implausible scenario in SSP3-7.0.
In start contrast, the NZ35 report starts with the optimistic IEA STEPS baseline (well below SSP2-4.5 for those scoring at home), which makes the costs and challenge of net zero much more manageable than they would be had the NZ35 report used the same extreme scenario of the NCRA.
The use of an extreme scenario to project impacts and an optimistic scenario for projecting costs is, charitably, a methodologically unsound approach to assessing science to inform policy.
Less charitably, exaggerated impacts and low-balling costs is exactly the sort of approach you’d take if the goal was political advocacy rather than clear-eyed assessment. Let’s take a closer look.
The NCRA based its projections on Australian government climate modeling performed over the past few years. A report released in July detailed the approach, explaining that the SSP5-8.5 scenario had been left behind in favor of SSP3-7.0:
We use this scenario as representative of higher levels of climate change. . . This scenario is selected as a fallback from the extremely high SSP5-8.5 (‘Fossil Fuelled Development’), which is seen as implausible in some respects by many . . .
As one of these “many,” it is rewarding to see that awareness of the implausibility of SSP5-8.5 is getting through to the climate assessment community. However, the adoption of SSP3-7.0 totally negates any of the positives, as SSP3-7.0 is in every respect just as implausible as SSP5-8.5.
Just two features of this scenario are sufficient to demonstrate its implausibility.
SSP3-7.0 has a global population of more than 12 billion in 2100 which is coupled with a massive increase in global coal consumption, together resulting in massive emissions to 2100. SSP3-7.0 has less coal consumption than SSP5-8.5, but a much greater population, so it maintains a very high radiative forcing to 2100. Thus, when used as the basis for projecting future climate, it leads to extreme results.
In 2025, plausible projections of future global population have seen peak population moving closer and closer to the present. Projected population in 2100 is now well below 10 billion (United Nations) and others see less than 8 or 9 billion as more likely. I’ve recommended taking the under.
Most importantly, no population expert sees a population of 12 billion in 2100 as plausible.
For coal consumption, SSP3-7.0 projects that global coal consumption will triple from today by 2100, reaching 50% of total global energy consumption (from about 25% today).
Javier Blas, energy columnist at Bloomberg, just published (on X) a sneak peek of the draft 2025 IEA World Energy Outlook projection of coal consumption over coming decades in its forthcoming scenarios, shown below.
The IEA appears to have rediscovered energy sobriety, projecting that coal will plateau and then slowly decline to less than 150 exajoules (EJ) in 2050 under “current policies.”1 The IEA STEPS (“stated policies) scenario goes even further, projecting coal to drop to near 100 EJ.
In contrast, SSP3-7.0 projects coal consumption to increase to 2050, approaching 300 EJ, and to nearly double from there by 2100. From today’s perspective, a world choosing to go all in on coal is implausible.
Climate projections using SSP3-7.0 are of no use for informing policy. They are not useful as a baseline, nor as a stress test, nor as a worst-case scenario. SSP3-7.0 should go the way of SSP5-8.5, and none too soon.
Remarkably, the NCRA completely ignores where most research suggests where the world is more plausibly headed — somewhere between 2 and 3 Celsius increase in global temperatures from a pre-industrial baseline. The figure below, from the modeling that supports the NCRA, shows that gap with a red oval that I added.
The NCRA provides policy makers with no information whatsoever on projected climate impacts based on the most plausible scenarios from recent research. It would not be unfair to call this climate assessment malpractice. Close observers will see in the figure above that SSP3-7.0 has a central value of 4.0 C in 2100 (which is consistent with the IPCC), whereas the NCRA instead calls it a 3C scenario. That is simply wrong and a bit too clever.
If I were to advise an Australian decision maker on how best to interpret the NCRA as an assessment of plausible climate futures,2 I’d recommend looking at the results of the SSP1-2.6 scenario — used as a lower bound — but more accurately interpreted as a plausible future. Clearly, 2C in 2100 is closer to a 2-3C range than is 4C. But the choices of SSP1-26 and SSP3-7.0 were far from ideal, compromising the value of the NCRA.
The choice of scenarios underpinning the NRCA is unfortunate because no doubt many intelligent, hard working, and well meaning people put a lot of time into the assessment. I do wonder how many of them were aware of the issues with the underlying scenarios and if anyone raised concerns. How these poor scenario choices were made would be worth understanding so that such a mistake does not happen again
The use of SSP3-7.0 as a key scenario to project changes in climate stands in stark contrast to how the NZ35 assessment uses an optimistic scenario as a baseline from which to project the costs of mitigation.
That scenario is IEA STEPS (“stated policies”), which is described as follows:
The CSIRO Stated Policies (CSP) scenario is the ultimate baseline for all scenarios modelled here. The CSP scenario was developed in Brinsmead et al. (2023); it is based on stated policies internationally and within Australia and projects a 2.6°C temperature increase by 2100. It was developed to translate the IEA’s Stated Policies Scenario (STEPS) (IEA, 2021a) to an Australian context.3
The projections for 2100 for the scenarios used to project impacts and costs are summarized in the slide below, from my talk tomorrow at the 2025 Aon Australia Hazards Conference.
Assessing science to inform policy makers is challenging in the best of cases. In the case of Australia’s new climate assessments decisions made about what scenarios to use have led to results that are at the same time too pessimistic for climate change and too optimistic for the prospects for mitigation.
We might expect such a perspective from climate advocacy groups. From government agencies responsible for calling things straight, we should expect better.
If you think that everyone should understand climate scenarios and why they matter, please click that “❤️ Like.” More likes mean that THB rises in the Substack algorithm and gets in front of more readers. Thanks!
Comments welcomed!
Please consider subscribing or upgrading to a paid subscription. Paid subscribers make THB go and also have access to THB Pro, with PDFs of some of my books, THB Insider, Five Figures, and paywalled THB posts. Thank you!
The IEA had previously projected a rapid phaseout of coal.
Because there is not a plausible socio-economic scenario underneath the climate projections, NCRA projections of future impacts are of little utility.
The 2021 IEA STEPS scenario is already out of date — The 2024 version projected 2.4C by 2100.
When will climate scientists admit the obvious that government agencies have been captured by climate change advocacy groups for almost two decades?
It seems to me that the only reason bodies like CSIRO persist in using implausible emissions scenarios like SSP 3.0-7.0 is because the perceived urgency of climate change dwindles significantly in scope and time when more realistic scenarios are employed. Thus it becomes more and more difficult to maintain that climate change is THE existential crisis of our time. Justifying revolution in control of energy becomes untenable. Rather, with realistic emissions, climate change becomes one of a number of issues to be ranked and rated for priority and resources. This, in my opinion, is what advocates find intolerable.