A Minimum of 3.5 Degrees C by 2100?
The Depressing Results of a New Simulation Study
Last week George Backus, formerly of Sandia National Labs, sent me a copy of his new research study. George is one of the smartest risk experts around so I looked forward to reading his latest analysis. Much of his work at Sandia focused on the potential consequences of climate change for macroeconomic threats and potential military conflicts at the local and international levels, but his 2010 report Assessing the Near-Term Risk of Climate Uncertainty: Interdependences among the U.S. States remains one of the most insightful risk-based looks at the implications of climate change for the United States. His new study, Climate Risk and Response: Too Much and Too Little, is much broader in scope, and challenges much of today’s conventional wisdom when it comes to mitigating climate change.
Weighing in at more than 300 pages and more than two years in the making, Backus’ study uses a simulation model to assess our collective ability to rapidly decarbonize the global energy system. In doing so it explores the implications of likely barriers to rapid decarbonization, and assesses what those barriers means for global temperature targets. A simulation model is well suited to this task because the causal linkages and feedback dynamics yield a far more policy-relevant picture of the functioning of the overall system than do normative economic analyses based on least-cost optimization.
I spent three days going through the study, and I see it as a major contribution to evidence-based climate change decision-making. That’s why I’m bringing it to your attention. I’ll briefly review the study, point you to a dedicated website where you can explore a subset of key graphics from the study, as well as to where in the Climate Web you can dig deeper (and download the report if desired).
A few points regarding the simulation model:
Extending out to 2100, the model focuses on rapidly decarbonizing the global energy system by as close to 2050 as possible.
The model does not recreate the climate models and economic models that form the basis for most mitigation analysis today, but the algorithms used to specify the simulation model’s “Referent Case” replicate the results of those models.
While the results of the simulation are likely to come across as depressing, the model is arguably optimistic in not including the risks associated with climate tipping points, or the costs of the kinds of climate events (or disasters) already being witnessed in the West, e.g. 115+ degrees F leading to hundreds of deaths in Oregon and Washington, and the huge fires currently burning in several Western states.
The simulation model goes on to assess, at a systems level, our ability to rapidly decarbonize the global energy system:
“100% electrification with renewable energy” is selected as the basis for the study’s policy package, in part because Backus consider it implausible that global decision-makers can universally agree upon and implement a complicated strategy relying on numerous initiatives from decarbonizing the electric system to restoring natural systems and transitioning to a circular economy.
The study avoids an optimization-based modeling approach, given the inevitably imperfect ambition and/or mobilization levels that the study assumes (quite rightly) will continue to characterize global efforts to tackle climate change.
The study arrives at several over-arching conclusions:
Because today’s climate change mitigation conversations are largely based on “most likely climate change” on the impacts side, and “response optimization” on the mitigation side, an enormous amount of climate risk (Consequence x Probability) ends up being overlooked.
Both socioeconomic impact uncertainty and climate uncertainty have to be accounted for. Socioeconomic impact uncertainty is more important than climate uncertainty for deciding how to trade off lives and economics costs.
Taking into account predictable barriers and choke points when it comes to rapid decarbonization of the energy system, it is unlikely that the global average temperature increase can be kept below 3.5 degrees C by 2100.
The earliest a carbon-fuel transition could be completed is 2057, and a net-zero transition can’t happen before 2065.
It is plausible to talk about bringing the average global temperature back down by continuing to deploy direct air capture after the energy system has been full decarbonized, but climate change tipping points may prevent that.
The economic impacts of climate change, combined with the costs of trying to mitigate climate change, will be ruinous for “disadvantaged” countries as defined in the study. “Advantaged” countries will (in their own self-interest) need to step up to substantially fund decarbonization of the global energy system.
Using renewable energy to build new renewable energy capacity while rapidly transitioning to a fully decarbonized system will be difficult and costly. The relatively long energy payback periods associated with renewable energy sources, as compared to fossil fuels like natural gas, complicate the challenge considerably.
Nature-based solutions are a risky way to tackle climate change, given that their effectiveness could be transitory at best in the face more than a 2 degrees C increase in average global temperature.
The bottom line? It’s critical that we do as much as we can to mitigate climate change, but thinking we can hit today’s 1.5 or 2 degree C targets is #greenwishing. And while geoengineering is not part of the simulation study per se, the study concludes that the situation is likely to get bad enough that a lot of geoengineering efforts will be deployed.
The study incorporates a number of assumptions of likely interest:
Reliance on solar, wind, energy storage, and direct air capture to get to 100% renewable energy and to “net zero.”
Because biomass energy remains an overall net source of CO2 as long as biomass-based generating capacity is increasing, it’s not the answer if rapid decarbonization is the objective.
The study does not rely on nuclear energy due to its costs and lengthy ramp-up times, which would worsen scenario outcomes. Nuclear energy is also a proliferation risk in a warming world with all the associated economic and political upheaval that may occur.
Hydropower is not expanded as part of the response strategy because of its lengthy energy payback period, and because of the challenges dams will be facing in helping grapple with the implications of climate change for agriculture, fish, etc.
The study does not rely on hydrogen because it is inherently less efficient than direct electrification and would result in higher costs and a longer transition period than the scenarios evaluated in the study.
While a truly integrated global or national power grid would save quite a bit of money, the study assumes that individual regions are likely to have to deal with many of the challenges of a rapid decarbonization without the benefit of such a grid.
When it comes to these assumptions and many other aspects of the report, Backus provides extensive references (more than 1,000 in total).
As you might surmise, this study has potentially big implications for:
1.5 and 2 degree C target conversations
Business risk and risk management conversations
Carbon offset conversations
Natural climate solutions conversations
Scenario planning conversations
Indeed, almost everyone is likely to find something NOT to like about this new study, and I asked the author about this problem. He responded:
“I think everybody is a specialist and nobody is an integrator. The world is an integrated whole and climate needs to be addressed in that context. When feedback is included, the results are very different from those assumed by looking at a particular element in isolation (as is almost universally done.) I did the study because I think its unexpected conclusions and viewpoints are really important.”
So do I, and I hope you’ll explore it further!
Some Links for Digging Deeper: