Hansen’s newest overheated international warming claims are primarily based on poor science • Watts Up With That?

From Local weather And so on.

Nic Lewis

James Hansen’s newest paper “World warming within the pipeline” (Hansen et al. (2023)) has already been closely criticized in a prolonged remark by Michael Mann, writer of the unique IPCC ‘hockey stick’. Nonetheless Mann doesn’t cope with Hansen’s surprisingly excessive (4.8°C) new estimate of equilibrium local weather sensitivity (ECS)[1]. This ECS estimate is 60% above Hansen’s longstanding[2] earlier estimate of three°C. It’s Hansen’s new, very excessive ECS estimate drives, together with varied questionable subsidiary assumptions, his paper’s dire predictions of excessive international warming and its extra excessive concluding coverage suggestions, reminiscent of ‘photo voltaic radiation administration’ geoengineering.

Hansen’s new 4.8°C ECS estimate is nicely above one of the best estimate of three°C reached within the IPCC’s newest scientific evaluation report (AR6), lies exterior the AR6 seemingly (66%) vary of two.5–4°C and is nearly on the high of the AR6 90% uncertainty vary of two–5°C.[3]

Each Hansen’s new ECS estimate and his earlier estimate are primarily based totally on details about paleoclimate adjustments, significantly the extensively studied transition from the final glacial most (LGM) some 20,000 years in the past to the preindustrial Holocene. However is his new estimate (or certainly his earlier estimate) justified by the proof?

Hansen’s main LGM to Holocene primarily based ECS estimation.

For his LGM-based ECS estimate, Hansen assumes a 7.0°C rise in international imply floor temperature (GMST) between the LGM and preindustrial Holocene. This worth is 56% above the 4.5°C rise that Hansen used beforehand². Hansen cites three research in assist of his 7.0°C LGM cooling estimate (which comes from the second of those): Tierney et al. (2020), Osman et al. (2021) and Seltzer et al. (2021).

I revealed in 2021 an article that was closely essential of the Osman et al. data-assimilation (reanalysis) primarily based temperature reconstruction. Their reconstruction makes use of solely ocean sea floor temperature (SST) proxies and is predicated on a single local weather mannequin that simulates an unusually chilly LGM state[4] and, strongly influenced by the mannequin simulations, produced a 7°C estimate of LGM cooling.[5] The proxy-only primarily based reconstruction within the submitted model[6] of Osman et al. appeared extra affordable, and as proven in my article implied LGM to preindustrial GMST change of about 4.5°C.[7]

The Tierney et al. estimate of LGM cooling (6.1°C)[8] is predicated on a a lot bigger set of SST proxies, which incorporates all these employed by Osman et al. used, however equally makes use of no land temperature proxies. Tierney et al. makes use of the identical single-model data-assimilation temperature reconstruction methodology as does Osman et al.[9] Subsequently, my criticisms of Osman et al.’s reanalysis-derived LGM cooling estimate very largely apply additionally to Tierney et al.’s estimate.

The Seltzer et al. 5.8 ± 0.6 °C LGM cooling estimate is for land solely and is proscribed to 45°S–35°N, so doesn’t characterize GMST cooling. The authors use a groundwater-based proxy sort and a fancy mannequin to transform proxy values to floor temperatures. I’ve issues about some features of their strategies.[10] Furthermore, their 5.8°C land cooling estimate is predicated on the error-weighted common of solely 15 groundwater information, which present extensively various cooling[11] and is probably not adequately consultant of 45°S–35°N land areas. Seltzer et al. utterly ignore the uncertainty related to these points. Their estimate is inconsistent with most proxy-derived estimates of imply ocean cooling over 45°S–35°N, which casts additional doubt on its reliability.

I regard the latest Annan et al. (2022) LGM temperature reconstruction as producing a extra dependable estimate of GMST cooling than the Tierney et al. (2020) and Osman et al. (2021) research that Hansen et al. depends upon. The Annan et al. reconstruction makes use of an identical data-assimilation methodology to these research, however with a number of key variations. Not like them it makes use of a simulations from massive set of acceptably dissimilar local weather fashions, relatively than a single mannequin. Importantly, in contrast to these two research, Annan et al. scale the mannequin simulated temperature adjustments in order that the preliminary guess mannequin cooling estimates used are roughly centered on the info, to attenuate model-generated bias. Furthermore, Annan et al. use land temperature proxies in addition to a bigger set of ocean SST proxies to regulate the preliminary guess. They use the identical SST proxy dataset as Tierney et al., however prolong its protection with information from a extensively used earlier SST dataset, which specifically reduces Tierney et al.’s big gaps in protection of the Pacific ocean.

Because of the significantly better proxy protection, the usage of a number of fashions and the debiasing step, the 4.5°C estimate of GMST cooling on the LGM that the Annan et al. reconstruction produces needs to be far more practical than the Tierney et al. (2020) and Osman et al. (2021) estimates that Hansen makes use of. Certainly, Annan et al. criticize Tierney et al.’s method, utilizing very chilly simulations from a single mannequin, declaring that it’ll produce a cool reconstruction no matter whether or not the info factors to a milder LGM local weather.

By coincidence[12], Annan et al.’s 4.5°C GMST LGM cooling estimate is identical worth as I adopted for the LGM derived local weather sensitivity estimation in Lewis (2023)[13]. That research reworked Sherwood et al. (2020), a really influential World Local weather Analysis Programme linked assessment of local weather sensitivity proof, which had assessed the LGM to be 5°C cooler than preindustrial.  

Hansen’s ECS estimate is elevated not solely by adoption of an unrealistically massive LGM cooling worth, but additionally by very low estimates[14] for the assorted adjustments in forcing between the LGM and the preindustrial Holocene[15]. Hansen’s central estimate for the whole Holocene to LGM forcing distinction is −5.75 W/m², a lot smaller than the −8.63 W/m² assessed by Sherwood et al.(2020) and even additional under the −8.93 W/m² greatest estimate adopted in Lewis (2023). The LGM part of Desk 1 within the Appendix offers particulars of all of the data-variable estimates used to type the Sherwood et al. (2020), Lewis (2023) and Hansen et al. (2023) LGM–preindustrial Holocene primarily based ECS estimates.

The mix of the massive GMST change and small forcing change that Hansen adopts for the LGM ends in a central ECS estimate of 4.8°C. By comparability, the Lewis (2023) LGM-based ECS estimate was 2.2°C, considerably lower than the corresponding ECS estimate from Sherwood et al. of two.8°C.[16] Each these estimates are nicely under the decrease certain of the three.6 – 6.0 °C ECS 95% uncertainty vary for ECS that Hansen estimates.

Hansen helps his LGM-based ECS estimate with one primarily based on the distinction in GMST between the LGM and the earlier, Eemian, interglacial interval. His estimate is subsequently inflated by his assumption that the LGM was very cool. Furthermore, in comparison with the LGM there may be even larger uncertainty in Eemian GMST, ice-sheet forcing, and different non-CO₂ forcing. IPCC AR6 mentions all these drawbacks and moreover factors outs that accounting for various orbital forcing is difficult for this era. Sherwood et al. (2020) didn’t try and estimate local weather sensitivity from the Eemian to LGM or another pre-LGM local weather transition as a result of the info out there are much more restricted than for the far more extensively studied LGM. I likewise didn’t accomplish that in Lewis (2023), nor do I try such an estimate right here.

Hansen’s evaluation of the PETM occasion and its implications.

Hansen additionally considers the Paleocene-Eocene Thermal Most (PETM) warming occasion some 56 million years in the past, adopting a greatest estimate for the warming concerned of 5.6°C and assuming, as did Sherwood et al. (2020) and Lewis (2023), that greenhouse gases alone produced the rise in forcing that induced the warming.

Quite than estimating ECS utilizing an estimate of the PETM forcing change, Hansen calculates what the PETM CO₂ degree would have needed to be to trigger warming of 5.6°C, primarily based on his ECS estimate of 4.8°C, assuming a pre-PETM CO₂ degree of 910 ppm and that non-CO₂ greenhouse gases contributed 25% as a lot forcing as CO₂ did. Hansen thereby calculates a PETM CO₂ degree of 1630 ppm. Actually, the right determine appears to be approximately1565 ppm; Hansen seems to have gone improper someplace in his calculations.

Though Hansen’s 910 ppm pre-PETM CO₂ focus worth is intently according to different estimates, the circa 1600 ppm worth for the PETM CO₂ focus implied by his 4.8°C ECS and his different assumptions is far decrease than the estimate of 2400 ppm, with a ±1 commonplace deviation vary of 1700–3100 ppm, that Sherwood et al. (2020) reached after assessing the out there proof. Additionally it is close to the underside of the 1400–3150 ppm uncertainty vary given in AR6.

Furthermore, Sherwood et al. assessed one of the best estimate GMST rise on the PETM to be barely decrease than did Hansen, at 5.0°C, and the non-CO₂ greenhouse fuel forcing to be larger, at 40% of CO₂ forcing. All these variations contribute to Sherwood et al.’s PETM-based ECS estimate being, at 2.5°C, barely half Hansen’s assumed 4.8°C. Furthermore, Lewis (2023), utilizing equivalent data-variable assumptions as Sherwood et al. (2020) however the AR6 method for CO₂ forcing, which is extra correct at excessive concentrations than the straightforward method Sherwood et al. used, obtains a good decrease PETM-based ECS estimate of two.2°C.

The PETM part of Desk 1 within the Appendix offers particulars of all of the data-variable estimates used to type the Sherwood et al. (2020)and Lewis (2023) PETM primarily based ECS estimates, and the Hansen et al. (2023) estimate of CO₂ focus on the PETM. Determine 1 exhibits what the central estimate of the whole change in forcing concerned was, as a a number of of the forcing from a doubling of preindustrial CO₂ focus, for all three research’ LGM and PETM estimates.

Though there are a lot larger uncertainties concerned when estimating ECS from the PETM occasion than for the LGM – preindustrial Holocene transition, the out there PETM proof is intently in keeping with the decrease, 2.2°C and a pair of.8°C, ECS estimates derived from the LGM data-values used respectively by Lewis (2023) and Sherwood et al. (2020), however not with Hansen’s very excessive 4.8°C estimate of ECS.

Additionally it is clear that Hansen’s declare that right this moment’s human-made greenhouse fuel forcing is, at 4.6 W/m², at the least similar to the PETM forcing (which his assumptions suggest was 4.67 W/m²) is strongly at variance with the proof as assessed by Sherwood et al.[17] That proof implies a greatest estimate of PETM forcing of 1.98× that for a doubling of preindustrial CO₂ focus[18] (versus 1.17× on Hansen’s assumptions), double or extra the newest (2022) AR6-basis estimate, in Forster et al. (2023), of 0.88× for greenhouse fuel forcing (1.00× when together with that from ozone, a brief lived greenhouse fuel).

Determine 1.The forcing change between the LGM and preindustrial Holocene, and between earlier than and in the course of the PETM, implicit in every research’s assumptions. The forcing adjustments are expressed as a a number of of that from a doubling of preindustrial CO₂ focus. The related ECS estimate equals in every case the corresponding assumed GMST change divided by the forcing change proven, so a better forcing  change implies a decrease estimated ECS worth.

Conclusions

I don’t think about that Hansen’s local weather sensitivity estimation correctly assesses and pretty displays all of the out there related proof. Sadly, in contrast to Sherwood et al.(2020), IPCC AR6 and Lewis (2023), Hansen et. al. (2023) estimates ECS utilizing solely paleoclimate proxy-derived proof, which typically varies significantly in line with the proxies concerned and to the strategies used to interpret them. This opens the door for biased (cherry picked) assessments. For example, Hansen et. al. don’t even point out any research (e.g. Annan and Hargreaves (2013) and (2022)) that discover a a lot decrease LGM – preindustrial warming than their chosen worth.

Though I respect Hansen’s potential and appreciable scientific contributions, for my part papers he leads are more and more strongly biased in the direction of overheated projections and dire conclusions.[19] The “political suggestions” with nice influence on the society in Hansen et al. (2023) can’t be justified as a result of their basis could be very shaky, as proven right here for local weather sensitivity and, in relation to warming-in-the-pipeline and ocean heating, in Michael Mann’s critique.  

Nicholas Lewis                                                                                                            6 November 2023

Appendix

Desk 1. Paleoclimate proof data-variable best-estimate values^a used to estimate S and ECS

Notes:
a Information-values for Sherwood et al.(2020) and Lewis (2023) are medians which were extracted from Desk 3 of Lewis (2023), the notes to that are included right here by reference. See these papers for particulars of the sources used to derive their respective data-variable estimates.
b Utilizing the Hansen et al. (2023) Desk 1 method for CO₂ forcing

[1] ECS is outlined as the worldwide imply floor warming for a doubled CO₂ focus after so-called fast-feedbacks have been totally activated and the ocean has reached equilibrium (which it approaches inside a thousand or so years). Earth system sensitivity (ESS), which additionally consists of very sluggish feedbacks reminiscent of these from adjustments in ice sheets, represents international warming arising over for much longer timescales. ESS is related when learning paleoclimate adjustments, however not when  projecting adjustments over the following few centuries.

[2] E.g., in Hansen et al (2013) Local weather sensitivity, sea degree and atmospheric carbon dioxide. Phil Trans Roy Soc A. https://doi.org/10.1098/rsta.2012.0294

[3] A number of the newest era (CMIP6) international local weather fashions do have ECS values above 5°C, however they’re typically regarded by local weather scientists as being too delicate.

[4] Really two barely totally different variations, iCESM1.2 and iCESM1.3, of the identical underlying CESM1 GCM.

[5] Supporting my critique of their reanalysis, I confirmed that its co-located estimates have been uncorrelated with values from the cave speleothem proxies it used. Furthermore, in contrast to proxy-based reconstructions (e.g., Caufman et al. (2020)), which present early Holocene GMST 5,000–9,000 years in the past being circa 0.5°C  larger than late preindustrial (1750) GMST, their reanalysis confirmed that interval as being vital cooler than late preindustrial GMST.

[6] The ultimate, revealed model of Osman et al. was barely totally different, with no point out of the change within the peer assessment recordsdata, and cooled marginally extra close to and on the LGM.

[7] Utilizing a more moderen mannequin era than the older ones Osman et al. used to derive the ratio used to transform their ex-polar sea floor temperature (SST) proxy-based estimates to GMST.

[8] No rationalization is obtainable of why, disconcertingly, this differs from the 5.9°C of their submitted manuscript.

[9] The Osman et al. cooling estimate could also be larger as a result of they use solely a subset of Tierney et al.’s proxies, and haven’t any proxy protection in any respect within the central Pacific ocean.

[10] Whereas Seltzer et al. validated their mannequin on fashionable proxy information, it is probably not legitimate for circumstances on the LGM, when components reminiscent of vegetation cowl, rainfall patterns, and seasonal temperature and precipitation cycles have been totally different. A further concern is that Seltzer use two various strategies to pick out LGM samples, which give noticeably totally different LGM cooling estimates (5.8°C and 4.8°C).  They like the tactic giving the upper estimate, nonetheless it is going to have a tendency to select significantly chilly temperatures from datasets, and so could overestimate LGM cooling. Curiously, the typical of the Annan et al. LGM reconstruction’s cooling estimates for grid cells co-located with the 15 proxies that Seltzer et al. use is near the 4.8°C estimate that Seltzer et al. receive utilizing their second methodology.

[11] Thus, excluding two significantly cool, low assessed uncertainty proxy estimates would cut back the estimate by 0.4°C.

[12] Lewis (2023) was submitted sooner than was Annan et al (2022).

[13] Lewis, N., 2023. Objectively combining local weather sensitivity proof. Local weather Dynamics, 60(9), pp.3139-3165.

[14] Hansen’s estimate of the greenhouse fuel forcing change, relative to that from a doubling of CO₂ focus, is 27% decrease than the worth assessed by Sherwood et al. and in addition adopted in Lewis (2023). Furthermore, Hansen ignores the numerous mud aerosol forcing change assessed by Sherwood et al., likewise additionally adopted in Lewis (2023).

[15] Within the method for estimating ECS, the change in forcing (ΔF) types the denominator: ECS = F_2×CO2 ×  ΔT/ΔF, the place ΔT is the GMST change in equilibrium and F_2×CO2 is the forcing from a doubling of atmospheric CO₂ focus. All forcing adjustments given on this article are for efficient radiative forcing (ERF), the principal forcing metric utilized in AR6.

[16] Sherwood et al. (2020) and Lewis (2023) centered on a local weather sensitivity measure, S, a neater to estimate approximation to ECS that’s usually derived for international local weather fashions as an alternative of ECS. They accordingly transformed their underlying paleoclimate ECS estimates into estimates of S.

[17] Hansen’s declare right here that right this moment’s human-made GHG forcing is 4.6 W/m² is furthermore inconsistent with the assertion earlier in his paper that it’s 4.13 W/m² for 2022 (1.03× his worth for a doubling of CO₂ focus, near the AR6-basis ratio), and rising by 0.5 W/m² per decade.

[18] Based mostly on the AR6 method for CO₂ forcing

[19] Furthermore, I’m unsure that Hansen is totally up with latest developments in local weather science. For example, when discussing totally different measures of radiative forcing, a subject on which Hansen wrote a seminal paper in 2005, he wrongly claims that (in contrast to for the formulae in his equation (3)) AR6 makes use of the biased-low F_o measure for its long-lived (non-ozone) greenhouse fuel efficient radiative forcing (ERF) estimates, and on that foundation adjusts the AR6 forcing values. The 2013 (AR5) IPCC evaluation report did use F_o, however AR6 makes use of a measure that, like Hansen’s most popular F_s, excludes the impact of floor temperature change.

[20] Hansen estimates ECS from adjustments in greenhouse gases (GHG) between the LGM and the mid-Holocene. The GMST estimate he makes use of is identical for that interval as for the instantly preindustrial interval (circa 1750). Nonetheless, adjustments in non-GHG forcing brokers are extra unsure for the mid-Holocene than for the preindustrial interval  (which most forcing change estimates relate to). Specifically, aerosols and land floor albedo might have been considerably affected by deforestation beginning nicely earlier than the preindustrial interval, a risk talked about in Hansen’s paper. This makes ECS estimates primarily based on LGM to mid-Holocene adjustments extra unsure than these primarily based on LGM to preindustrial adjustments.