Bonkers in Yonkers
How outdated climate science became formalized in policy through regulatory capture
Earlier this week a reader shared with me the following news story in the New York Post:
The reader wrote: “You’ll never guess the scenario!” We can all guess the scenario. It is RCP8.5. As I looked into the New York state government’s report that was the basis for the news article, I discovered a cautionary tale for how flawed science becomes formalized in policy via regulatory capture by a handful of experts.
Before proceeding, let’s review two things.
First, climate change is real and poses significant risks for our collective futures. Sea level rise is among those risks, and is going to continue inexorably into the future with some considerable uncertainties. Policy makers should take sea level rise seriously.
Second, policy making will be more effective if grounded in the best-available science. In 2024, a climate scenario with radiative forcing of 8.5 watts per meter-squared (W/m^2) in 2100 is not the best-available science for projecting the future. The figure below shows how extreme such a scenario is compared to plausible scenarios of the future. The figure shows radiative forcing levels projected to 2100 and beyond under three different scenarios — (left) a 3.4 W/m^2 scenario, which we have identified as being most consistent with past emissions and their near-term projections to 2050, (center) a 4.5 W/m^2 scenario, which sits above a “current policies” trajectory according to the UNFCCC, and (right) the very extreme SSP5-8.5 scenario (an update to RCP8.5).
You don’t have to be a climate scientist to see that SSP5-8.5 is far more extreme than the more plausible scenarios consistent with history and the projected near-term future to 2050. Extreme, implausible scenarios have their purposes, but are misused if prioritized as the basis for policy and planning.1
In 2016, the state of New York mandated via regulation not only that RCP8.5 be used in climate policy planning, but that an extreme version of RCP8.5 must be used — the 95th percentile outcomes of RCP8.5 model runs — that is, the most extreme values from the most extreme scenario.
The fact that RCP8.5 is implausible has come to be accepted in the climate scenarios community. The real world is not headed for a RCP8.5 future, which means that planning for a RCP8.5 world is not just misguided, but bad science, poor policy, and a waste of taxpayer money.
Let’s take a detailed look at how this bad science became codified in New York law and regulations and, even worse, how New York policy makers are right now doubling down on the bad science.
Almost a decade ago, the NY governor signed into law the Community Risk and Resiliency Act (CRRA) requiring “science-based state sea level rise projections”:
SECTION 3-0319
Sea level rise projections
Environmental Conservation (ENV) CHAPTER 43-B, ARTICLE 3, TITLE 3
§ 3-0319. Sea level rise projections.
The department shall, no later than January first, two thousand
sixteen, adopt regulations establishing science-based state sea level
rise projections. In adopting such regulations, the department shall
consider information including, but not limited to, reports of the
Intergovernmental Panel on Climate Change, the National Oceanic
Atmospheric Administration Climate Assessment, the Sea Level Rise Task
Force report created pursuant to chapter six hundred thirteen of the
laws of two thousand seven, projections prepared by the New York City
Panel on Climate Change and any other relevant regional, state and local
reports. The department shall update such regulations no less than every
five years.
Subsequently, in 2016 the NY Department of Environmental Conservation (DEC) adopted regulations — Part 490 —that detailed the content of these “science-based” sea level rise projections, explaining:
Sea level rise is one of the most direct and observable effects of climate change in New York and DEC is required by law to develop science-based sea level rise projections to guide decision making and permitting in the areas most at risk.
The 2016 regulations mandated the use of two scenarios, RCP4.5 and RCP8.5, and the DEC justified the use of these two scenarios based on a 2014 report — Responding to Climate Change in New York State (ClimAID) — as follows:
The Department has considered numerous factors in proposing to base Part 490 on the ClimAID projections rather than on more conservative, less protective projections based primarily on process modeling.2
First, adoption of projections based on the ClimAID report ensures that regulators, planners and others have access to projections developed specifically for New York State and accounting for regional and local factors not considered in development of global sea-level rise projections.
Second, the ClimAID research was conducted by the same research team that provided the NPCC projections,3 using the same methodologies, which have been peer reviewed and published in established scientific journals.
New York City has already adopted the NPCC/ClimAID projections for its planning purposes; a State regulation based on alternative projections could create confusion among the public, planners and regulated community.
Third, ClimAID provides projections for the entire tidal coast of the state including the Hudson River upstream to the federal dam in Troy, rather than just Long Island and New York City.
Fourth, the proposed projection distribution (low, low-medium, medium, high-medium and high) constitutes a range suitable for risk-based planning and review of projects of varying projected life times and criticality. By having a full range of projections that include a high-medium and high value, decision makers will be able to consider the possibility of more rapid sea-level rise as it relates to particular projects in the context of relevant programs under CRRA.
Fifth, this projection distribution has been useful to communities conducting adaptation planning along the Hudson River.
The ClimAID report, originally published in 2011 and updated in 2014, explained that the use of RCP4.5 and RCP8.5 was intended to bracket possible future outcomes:
RCP 4.5 and RCP 8.5 were selected to bound the range of anticipated greenhouse gas forcing at the global scale.
We now know in 2024 that the world is currently undershooting a RCP4.5 trajectory and the climate science community is creating updated scenarios that sit well below the range of forcing bracketed by RCP4.5 and RCP8.5. The assumptions made in 2014 of the long-term future are no longer valid in 2024. When an evidence base evolves, policy should keep pace — this should be uncontroversial.
The 2016 regulations acknowledged that the most extreme scenarios are “unlikely” but argued that they were “plausible”:
Inclusion of unlikely but plausible projections provides benchmarks against which long-term decisions, e.g., those regarding critical infrastructure and land-use change, can be evaluated for low-probability but high-consequence events. If Part 490 did not include higher projections of sea-level rise, then decision makers would not be able to even consider the possibility of such levels occurring.
The regulations also acknowledge that when the extreme scenarios are included, that local decision makers tend to focus their decision making on these “unlikely” scenarios:
Communities and stakeholders in New York State that have been presented with the ClimAID projections have tended to adopt and plan for high levels of sea-level rise rather than more moderate levels. These stakeholders have placed a high degree of importance on ensuring the viability of proposed infrastructure investments and the social and economic fabric of their communities from even unlikely eventualities.
Interestingly, among the public comments submitted when the regulations were initially proposed was a request to “explain why certain conclusions of the Intergovernmental Panel on Climate Change (IPCC) were "ignored."“ Indeed, the NY regulations are explicitly at odds with the conclusions of the IPCC on sea level rise, and the DEC makes the case that the IPCC was incomplete or in error:
The ClimAID projections incorporate expert judgment of ice loss based on accelerating rates of melt and seaward movement of ice, positive feedbacks and non-linearities that are not necessarily accounted for in the process-based and statistical modeling approaches described by the IPCC.
For more than a decade, the choice of climate scenarios at the focus NY policy making was (and continues to be) made by a handful of experts whose views expressed in various reports are well outside the scientific consensus of the IPCC on sea level rise — their projections are far more extreme. These individuals — participants in the NPCC, ClimAID, and New York State Climate Assessment — have an enormous influence on NY policy making related to climate, because they decide what scenarios of the future should be considered and what scenarios are not considered.
Whatever the understandings and motivations of these experts, the continued misuse of climate scenarios in New York is an example of regulatory capture by a small group of like-minded people who sit largely out-of-sight, protected by their authority and the inscrutable complexity of climate scenarios. There is little oversight, accountability, or review of decisions made by these experts — decisions that have profound implications in policy.
Right now, the New York DEC is updating its choice of scenarios for use in policy and planning. But rather than correcting their unwarranted emphasis on the most extreme outputs from RCP8.5, the DEC has proposed to adopt an even more extreme scenario — SSP5-8.5, alongside continued use of RCP8.5.
Their justification for using the (even more) extreme SSP5-8.5 scenario is based on a new report from the same experts that recommended RCP8.5 in 2014. They acknowledge that the scenarios may be “unrealistic” but charge ahead regardless:
This report uses two SSPs: SSP2-4.5 and SSP5-8.5. There have been some arguments made among scientists and policy makers that the greenhouse gas forcing/concentrations associated with RCP8.5 may be unrealistic. Here, we enumerate some of the reasons we chose to include these two RCPs and RCP 8.5 especially. First, these scenarios have the same end-of-century radiative forcing as the two RCPs used in previous New York State projections, providing continuity with that work. Second, they match the pathways of greatest interest to stakeholders based on the Needs Assessment responses (NYSERDA, 2020).4 Third, relative to many other scenarios, there were also a large number of climate models available for these two SSPs. Fourth, they span a broad range of what we consider plausible climate outcomes, including the high-consequence outcomes that are critical for risk management (Wuebbles 2023).5 The IPCC has never assigned likelihoods to any of the RCPs, which should instead be thought of as scenarios. Fifth, while most would probably argue that the balance of evidence may suggest that lower greenhouse gas (GHG) forcing/concentrations are more likely than those associated with RCP8.5, that does not mean that the changes in extreme-event statistics, or climate impacts, associated with RCP8.5 are unrealistically high. As it becomes clearer that current climate impact models underestimate plausible high tails even when driven by only moderate GHG levels, it follows that using projections of only moderate GHG forcing would unrealistically narrow the upper bounds for the model-based distributions for impacts like heat waves and extreme precipitation.
The amendments to Part 490, the NY regulation establishing what climate scenarios must be considered in planning and policy, are currently open for public comment until April 29, and a public hearing will be held April 22. New York is on the verge of locking in SSP5-8.5 and RCP8.5 for the next five years, just as the climate community is starting to move away from these implausible scenarios.
Grounding policy in the best-available science sounds great, but pulling that off in practice can be quite difficult.
❤️Click the heart if you think 8.5 scenarios are past their sell-by date!
Thanks for reading! In the comments, I’d welcome links to other examples of how extreme scenarios are formalized in law and regulation. I am aware that California does so, as does New Zealand and several regulatory agencies of the U.S. government. There is surely an important paper to be written on how RCP8.5 has become institutionalized, making moving past it that much more difficult. THB is reader engaged and reader supported. If you are a free subscriber, please consider upgrading to join the community and to support independent writing and research!
I’ll soon have a post on how to think about unlikely futures in policy and planning.
Here the phrase: “more conservative, less protective projections based primarily on process modeling” means the projections of the IPCC AR5. The regulations are thus explicitly rejecting the science of the IPCC.
The NPCC is the New York City Panel on Climate Change.
The “Needs Assessment” referenced here surveyed local decision makers views of what scenarios should be used. It is clear that the respondents simply parroted what they were told by the experts. For instance, the report highlights the following quote: “The low range scenarios are not useful at all...They are not even a good projection of what we think is happening...and keeping it in the mix is very confusing for people. It creates a real communication issue. You know I think if 90% of scientists think we’re on the high end of the range, stop showing the low end of the range.” The report states, “There is a sense that the higher-end scenarios are more realistic and therefore should be used in planning.” There is no evidence in the report that either of these false claims were corrected or even identified as false.
Oddly, the reference here to “plausible climate outcomes” goes to an obscure paper on geoengineering that performs no assessment of plausibility.
When I used AI in my 1969 master's thesis, "AI" meant computational techniques that would solve problems that are difficult or impossible for a human. It still does. For your information, this is how "AI" itself answers the question I posed to Professor Pielke, the question being "what are applications of ai in climate science" see: https://www.perplexity.ai/search/what-are-applications-B61XpWSHS3usMWIOEQgoCQ My concern was, how much of this is hype and how much can we expect the new computational techniques and associated hardware used in "AI" to impact the field of climate science. The problem seems to require an "Honest Broker" since the general public has no idea how climate models work. If you are concerned about bias, this is a general issue when it comes to climate models, which include a tremendous amount of guesswork and arbitrary decisions.
Does Piltdown Mann live in NY?
Would explain a lot.