Abstract

Global temperature leaped more than 0.4°C (0.7°F) during the past two years, the 12-month average peaking in August 2024 at +1.6°C relative to the temperature at the beginning of last century (the 1880-1920 average). This temperature jump was spurred by one of the periodic tropical El Niño warming events, but many Earth scientists were baffled by the magnitude of the global warming, which was twice as large as expected for the weak 2023-2024 El Niño. We find that most of the other half of the warming was caused by a restriction on aerosol emissions by ships, which was imposed in 2020 by the International Maritime Organization to combat the effect of aerosol pollutants on human health. Aerosols are small particles that serve as cloud formation nuclei. Their most important effect is to increase the extent and brightness of clouds, which reflect sunlight and have a cooling effect on Earth. When aerosols – and thus clouds – are reduced, Earth is darker and absorbs more sunlight, thus enhancing global warming. Ships are the main aerosol source in the North Pacific and North Atlantic Oceans. We quantify the aerosol effect from the geographical distribution of sunlight reflected by Earth as measured by satellites, with the largest expected and observed effects in the North Pacific and North Atlantic Oceans. We find that aerosol cooling, and thus climate sensitivity, are understated in the best estimate of the United Nations Intergovernmental Panel on Climate Change (IPCC).

Global warming caused by reduced ship aerosols will not go away as tropical climate moves into its cool La Niña phase. Therefore, we expect that global temperature will not fall much below +1.5°C level, instead oscillating near or above that level for the next few years, which will help confirm our interpretation of the sudden global warming. High sea surface temperatures and increasing ocean hotspots will continue, with harmful effects on coral reefs and other ocean life. The largest practical effect on humans today is increase of the frequency and severity of climate extremes. More powerful tropical storms, tornadoes, and thunderstorms, and thus more extreme floods, are driven by high sea surface temperature and a warmer atmosphere that holds more water vapor. Higher global temperature also increases the intensity of heat waves and – at the times and places of dry weather – high temperature increases drought intensity, including “flash droughts” that develop rapidly, even in regions with adequate average rainfall.

Polar climate change has the greatest long-term effect on humanity, with impacts accelerated by the jump in global temperature. We find that polar ice melt and freshwater injection onto the North Atlantic Ocean exceed prior estimates and, because of accelerated global warming, the melt will increase. As a result, shutdown of the Atlantic Meridional Overturning Circulation (AMOC) is likely within the next 20-30 years, unless actions are taken to reduce global warming – in contradiction to conclusions of IPCC. If AMOC is allowed to shut down, it will lock in major problems including sea level rise of several meters – thus, we describe AMOC shutdown as the “point of no return.”

We suggest that an alternative perspective – a complement to the IPCC approach – is needed to assess these issues and actions that are needed to avoid handing young people a dire situation that is out of their control. This alternative approach will make more use of ongoing observations to drive modeling and more use of paleoclimate to test modeling and test our understanding. As of today, the threats of AMOC shutdown and sea level rise are poorly understood, but better observations of polar ocean and ice changes in response to the present accelerated global warming have the potential to greatly improve our understanding.

Keywords:

• Climate Change, Climate Models, Climate Sensitivity, Aerosols, Ship Emissions, Greenhouse Gases, Earth’s Albedo, Earth’s Energy Balance, Sea Surface Temperature, Storms, Ocean Circulation, Greenland Ice Sheet, Antarctic Ice Sheet, Ice Shelves, Sea Level, Point of No Return

Global warming has accelerated since 2010 by more than 50% over the 1970-2010 warming rate of 0.18 °C per decadeFootnote¹ ().Footnote² Earth is now warmer than at any time in the Holocene, the past 11,700 years of relatively stable climate in which civilization developed, and it is at least as warm as during the extreme warm Eemian interglacial period 120,000 years ago. Global temperature increased 0.4 °C during the recent moderate El Niño (a period when east-to-west equatorial trade winds weaken, allowing warm waters of the West Pacific to move toward South America), a warming much greater than during even the strongest prior El Niños. This rapid warming has baffled leading Earth scientists, who, for example, conclude that no combination of known mechanisms for warming “has been able to reconcile our theories with what has happened.”Footnote³ We conclude, on the contrary, that the known drives for climate change, principally human-made greenhouse gases and aerosols, account for observed global temperature, including a jump in sea surface temperature that amplified warming during the El Niño and has caused the widely discussed 1.5 °C temperature threshold to be breached, for all practical purposes.

Climate change burst into public attention with climate anomalies in 1988 so extreme that Time Magazine declared Earth to be “person of the year.” Rising public interest in climate change, especially the role of humanity in causing change, led to the 1992 United Nations Framework Convention on Climate ChangeFootnote⁴ and a large increase in funding for climate observations and research. The Framework Convention aimed to prevent dangerous human-caused climate change. The largest funding increase was for a NASA program initially titled Mission to Planet Earth expected to make global observations needed to understand ongoing global change.

Sidebar 1. Global surface temperature relative to 1880-1920 in is the GISS (Goddard Institute for Space Studies) analysis through October 2024.² The 1970-2010 warming rate of 0.18 °C/decade almost doubled in 2010-2023, but this higher rate is not a prediction of the future. A downturn in greenhouse gas emissions could alter projections on decadal time scales.

By 1992 it was understood that two things caused large human effects on climate: greenhouse gases (GHGs) and aerosols (tiny, generally microscopic, particles suspended in the air). GHGs cause global warming by holding in Earth’s heat radiation, acting like a blanket. The physics of this greenhouse effect is well understood and tested, for example, by comparison of Mars, Earth and Venus, with their differing amounts of atmospheric GHGs. Carbon dioxide (CO₂), produced mainly by burning of fossil fuels (coal, oil and gas) but also by deforestation, causes more than half of the human-made greenhouse warming. Human-made increases of CO₂ and the other main GHGs (Sidebar 2) have long lifetimes in the atmosphere from decades to millennia. Thus, these gases are well-mixed in the atmosphere and it is easy to measure their changing global amounts.

Human-made aerosols with greatest effect on climate are products of fossil fuel burning and biofuels (like firewood). Most aerosols increase reflection of incoming sunlight back to space and thus have a global cooling effect. Charlson and colleaguesFootnote⁵ concluded in 1992 that the climate forcing by aerosols – the cooling drive for climate change, see below – was similar in magnitude to the GHG forcing, but opposite in sign, thus tending to offset GHG warming. Aerosol offset of GHG warming is a Faustian bargain (),Footnote⁶ that is, a bargain providing present benefit without regard to future consequences. The aerosols providing a cooling benefit are also inherently dangerous particulate air pollution responsible today for several million annual deaths by respiratory, cardiovascular, and even neurological diseases worldwide;Footnote⁷ thus, as global pollution control has improved and clean energies are introduced the cooling effect of aerosols is lost: with the change of ship regulations, our first Faustian payment came due.^1,Footnote⁸

Climate sensitivity is a measure of the effect of rising levels of greenhouse gases on Earth’s temperature. It is usually defined as the eventual increase of global average temperature after a doubling of CO₂ concentration in the atmosphere compared to pre-industrial levels.

In this paper, we conclude that the estimate of aerosol climate forcing () by the United Nation’s scientific advisory body (the Intergovernmental Panel on Climate Change, IPCC) is an underestimate, and thus the Faustian bargain is worse than expected. We also show that IPCC’s emphasis on global climate models led to a marriage of aerosol forcing and climate sensitivity, such that underestimate of aerosol forcing led to underestimate of climate sensitivity. The result is a double whammy that helps explain global warming acceleration and alters projections of future climate, magnifying the danger of intergenerational injustice. The delayed response of climate still allows a potential bright future for today’s young people, but that happy result requires understanding of the factors driving climate change. These issues can be readily understood via the most basic concepts, beginning with climate forcings.

Climate Forcings

Climate forcings are imposed changes of Earth’s energy balance. If the Sun suddenly became 1% brighter, for example, that would be a forcing of +2.4 W/m² (W/m² is watts, a measure of energy transfer over time, per square meter) because Earth normally absorbs 240 W/m² of solar energy averaged over Earth’s surface. Solar variability is one of the two natural climate forcings that are important on time scales of years to centuries, the other being large volcanic eruptions that inject gases and aerosols into Earth’s stratosphere (at about 15-50 kilometers, 10-30 miles). It is helpful to compare these well-understood natural forcings with human-made climate forcings.

Our Sun is a rather quiescent star, in the family of all stars, yet the variability of dark areas (sunspots) on the solar surface has long caused suspicion that the Sun may drive climate change on Earth. Fortunately, NASA has done a good job of monitoring the solar energy received at Earth since the late 1970s. Variations of the Sun’s brightness during the solar sunspot cycle are about 0.1% (), a forcing change of about 0.25 W/m² between solar minimum and solar maximum outputs. This solar forcing, we will show, is much smaller than human-made forcings, but large enough to be a partial cause of present climate extremes.

Volcanic eruptions produce occasional large climate forcings. When Mount Pinatubo erupted in the Philippines in 1991, producing the greatest climate forcing of the 20th century, NASA had satellites in orbit that provided a precise test of aerosol forcing. SAGE (stratospheric aerosol and gas experiment) measurements yielded the average aerosol size and the dispersion of aerosol sizes.

ERBE (Earth radiation budget experiment) measured the change of Earth’s energy balance, which peaked at −3 W/m² cooling several months after the eruption, the delay due to the time for conversion of the volcanic SO₂ gas into atmospheric sulfuric acid aerosols and the time for stratospheric winds to disperse the aerosols around much of the world. Observed global cooling after the Pinatubo eruption peaked at about 0.3 °C, consistent with expectations given the ocean’s thermal inertia and the brevity of forcing (stratospheric circulation carries aerosols to polar latitudes where they descend and are washed from the atmosphere).

A huge submarine volcanic eruption on 15 January 2022 – Hunga in the Pacific Ocean, east of Australia, near the dateline – blasted about 150 million tons of water vapor and 1 million tons of SO₂ into Earth’s stratosphere. It was much less SO₂ than for Pinatubo, and any cooling effect was partly offset by warming from the added stratospheric water vapor (a GHG).¹⁶ Nevertheless, we need to estimate the possible effect of Hunga on global temperature in 2022-2023 to see if it had a significant effect on the unprecedented warming that followed. Although it is difficult to disentangle Hunga effect on Earth’s measured energy balance from natural variability (due mainly to cloud variability), analysis shows that aerosol cooling dominated over water vapor warming¹⁶ and the net volcanic forcing declined to a small fraction of its peak value by two years after the eruption. We approximate the Hunga forcing based on the Pinatubo forcing (), but with peak forcing −0.3 W/m². Like the solar forcing, the Hunga forcing is small.

Given that we know precisely the natural climate forcings – volcanic aerosols and solar irradiance – as well as the human-made and natural greenhouse gas forcings, it is obvious that human-made aerosol forcing is the elephant in the climate forcing story that receives too little attention. Aerosol forcing occurs in part from the direct effect of human-made aerosols as they reflect and absorb incoming sunlight, but also from the indirect aerosol effect as the added aerosols modify cloud properties as discussed below. IPCC estimates the indirect aerosol forcing based largely on mathematical models.Footnote¹⁸ We suggest that this modeling fails to fully capture the fact that human-made aerosols have a larger impact on clouds when the aerosols are injected into relatively pristine air in places that are susceptible to cloud changes. Later in this paper, we use spatial and temporal changes of climate and Earth’s energy balance to explore this indirect aerosol forcing. First, however, we discuss aerosol and cloud particle microphysics.

Aerosol and Cloud Particle Microphysics

Climate forcing by aerosols depends on aerosol and cloud processes on minute scales. Aerosol composition matters, both for the direct effect of aerosols on radiation and the indirect effect on clouds. Indirect aerosol forcing arises because aerosols are condensation nuclei (tiny sites of water vapor condensation or “cloud seeds”) for cloud drops. More nuclei yield more cloud particles and brighter clouds that reflect more sunlight and cause cooling.Footnote¹⁹ More aerosols also increase cloud cover, as shown by cloud trails behind ships (“ship tracks”).Footnote²⁰ Observations to quantify these effects are challenging because human-made aerosols must be distinguished from changes of natural aerosols. Thus, there is large uncertainty in the overall net aerosol forcing.Footnote^21,Footnote²²

Simultaneous with human-caused aerosol and cloud changes, clouds also change as a climate feedback. [Climate feedbacks – response of the climate system (such as change of clouds or sea ice) to climate change – can be either amplifying or diminishing. Amplifying feedbacks increase climate change, tending to produce instability, while diminishing feedbacks decrease climate change, promoting stability.] Cloud feedback is the main cause of uncertainty in climate sensitivity, the holy grail of climate research.Footnote²³ Climate sensitivity is defined as global temperature response to a standard forcing. Observations reveal that the sizes and locations of zones with different characteristic clouds are changing – the intertropical convergence zone (encircling the Earth near the thermal equator) is shrinking, the subtropics are expanding, and the midlatitude storm zone (not near the poles or the equator) is shifting polewardFootnote²⁴ – with associated changes of Earth’s energy balance that constitute potentially powerful, but still inadequately understood, climate feedbacks. Some of the difficulties in climate modeling include cloud microphysics, such as the need to realistically simulate mixed phase (both ice and water) clouds.Footnote²⁵ As cloud modeling has become more complex and realistic, several global climate models have found higher climate sensitivity correlated with more realistic cloud distributions (Sidebar 4).

Given the importance of aerosol climate forcing and climate sensitivity,Footnote²⁸ there is a crying need for global monitoring of aerosol and cloud particle microphysics and cloud macrophysicsFootnote²⁹ to help sort out climate forcings and feedbacks.Footnote³⁰ Global monitoring of aerosol/cloud microphysical properties and cloud macrophysics from a dedicated small satellite mission has been proposed, but not implemented.Footnote³¹ The need for such data will only increase in coming decades as the world recognizes growing consequences of climate change and tries to chart a course to restore Holocene-level global climate. NASA’s 2024 PACE satellite missionFootnote³² includes polarimeters capable of measuring aerosol and cloud microphysics including aerosol and cloud droplet number concentrations, which could be a step toward a dedicated, long-term aerosol mission to monitor global aerosol and cloud properties as required to calculate climate forcings and feedbacksFootnote³³ (analogous to greenhouse gas monitoring that permits calculation of greenhouse gas forcing). In the absence of that data, we now explore less direct evidence of aerosol climate forcing.

Evidence of Aerosol Climate Forcing

Paleoclimate data suggest the important role of aerosols in global climate. In the past 6,000 years, known as the late Holocene, atmospheric CO₂ and CH₄ increased markedly, likely as a result of deforestation and methane from rice agriculture,Footnote³⁴ causing greenhouse gas (GHG) climate forcing to increase more than 0.5 W/m²,¹ yet global temperature during the late Holocene held steadyFootnote³⁵ or decreased slightly.Footnote³⁶ This divergence of GHG forcing and global temperature is a strong anomaly; CO₂ is a tight control knob on global temperature at other times in the ancient paleo record (see in Note 1 at end), as anticipated on theoretical grounds.Footnote³⁷ Aerosols, the other large human-made climate forcing, is a suggested solutionFootnote³⁸ for this “Holocene conundrum.” Aerosols increased in recent millennia as burning of wood and other biofuels provided fuel for a growing global population. Moving to recent, preindustrial, times, the required magnitude of the implied (negative) aerosol forcing from biofuel burning reached at least 0.5 W/m². Biofuel aerosol forcing has likely increased since then, as the biofuel energy source is widespread in developing countries and continues in developed countries.¹

Recent restrictions on ship emissions provide a great opportunity to investigate aerosol forcing. The International Maritime Organization (IMO) introduced limits on the sulfur content of ship fuels in stages, with the greatest global restriction effective January 2020 (Sidebar 5). Change of global aerosol forcing from this limit on ship emissions, based on IPCC’s formulation of aerosol forcing,¹⁸ is calculatedFootnote³⁹ as 0.079 W/m². Forster et al.,Footnote⁴⁰ updating IPCC’s aerosol forcing, estimate the ship aerosol forcing change as +0.09 W/m²; they also note that this ship aerosol forcing would be offset by negative aerosol forcing from increased forest fires and other biomass burning. A reviewFootnote⁴¹ of five ship aerosol modeling studies finds a range 0.07 to 0.15 W/m², with mean 0.12 ± 0.03 W/m². A recent model resultFootnote⁴² of 0.2 W/m² refers to ocean area and is thus a global forcing of 0.14 W/m². None of these modeled Ship Aerosol Forcings would have much effect on global temperature because GHG forcing currently is increasing 0.4-0.5 W/m² per decade.

However, if the aerosol effect is highly nonlinear (i.e., if aerosols emitted into polluted air have much less effect on clouds than aerosols emitted into a pristine atmosphere), decreased ship emissions may have a large effect on Earth’s albedo (reflectivity). The largest effect should be in the North Pacific and North Atlantic, where ship emissions dominate over natural sulfate aerosols (Sidebar 5). Fortunately, Earth’s albedo has been monitored for almost a quarter of a century by the CERES (Clouds and the Earth’s Radiant Energy System) satellite instrument,Footnote⁴⁴ which reveals a stunning darkening of Earth ().Footnote⁴⁵ Earth’s albedo decreased about 0.5% (of 340 W/m²), which is 1.7 W/m² additional heating of Earth since 2010! Such albedo change is equivalent to an increase of CO₂ by 138 ppm, from the 419 ppm actually measured at the beginning of 2024 to 557 ppm. However, the 1.7 W/m² increase in energy absorbed by Earth is not all climate forcing; it is partly climate feedback – cloud changes and reduced ice and snow cover caused by global warming. Our task is to apportion the 1.7 W/m² between aerosol forcing and climate feedbacks, accomplishing this in the absence of adequate aerosol and cloud measurements.

Ship-Induced Aerosol Climate Forcing

Earth’s declining albedo (darkening) is “noisy” in time and space because of the large natural variability of clouds. Earth’s albedo change () may not seem to correlate well with the 2020 change of ship emissions. However, a sharp 2020 change is clear after we consider the largest source of natural variability – the Pacific Decadal Oscillation (PDO)Footnote⁴⁶ – and additional data. The PDO is an observed natural cycle of sea surface temperature and cloud changes in the Pacific, as a large-scale manifestation of tropical El Niño/La Niña variability.Footnote⁴⁷ Absorbed Solar Radiation in the North Pacific is well correlated with the PDO from 2000 (when CERES data begins) until 2020 (), whereupon Absorbed Solar Radiation rapidly increases, when PDO cloud changes should have spurred a decrease of Absorbed Solar Radiation.

Let’s use the observed change of Absorbed Solar Radiation to estimate aerosol forcing change that occurred in 2020. The Absorbed Solar Radiation anomaly in 2020-2023 () relative to the base period (March 2000 through February 2010) reveals expected large tropical anomalies, but also increased absorption throughout the North Pacific and North Atlantic with the exception of the “global warming hole,” the region southeast of Greenland that is cooling relative to the rest of the world.Footnote⁴⁸ The increase of Absorbed Solar Radiation in the North Pacific and North Atlantic since 2020 can itself account for a global climate forcing of almost 0.5 W/m² (see Sidebar 6).

Sidebar 6. The 2020-23 anomaly of Absorbed Solar Radiation is +3.2 W/m² in the North Pacific (15-60°N, 120-240°W) and +2.8 W/m² in the North Atlantic (15-60°N, 5-80°W). These regions are 10.1% and 6.3% of global area, so, if increased Absorbed Solar Radiation in these regions were entirely an effect of decreased aerosols, it would contribute a global forcing of 3.2 × 0.101 + 2.8 × 0.063 = 0.5 W/m². Removing land areas from these two boxes reduces the Absorbed Solar Radiation anomaly to 0.42 W/m². This amount would be larger, if it were not for cooling and increased cloud cover in the “global warming hole” region southeast of Greenland. The relative cooling there is due to slowdown of North Atlantic’s overturning ocean circulation,⁴⁸ which is a northward flow of warm water in upper ocean layers with a deep southward return flow of cold water. Cooling of the “warming hole’ region induces increased cloud cover there (see and in the paper “Ice Melt”Footnote⁴⁹) that exceeds and hides the effect of decreased ship aerosols.

Increased Absorbed Solar Radiation is partly climate feedback: decreased snow and ice albedo and decreased cloud cover. However, snow/ice albedo change, apparent in near Antarctica and in the Arctic, has little role in the North Pacific and North Atlantic. Cloud feedback, including shifting climate zones,²⁴ may contribute to Absorbed Solar Radiation increase in the North Pacific and North Atlantic, but the largest cloud feedback is expected to be in the Southern Hemisphere south of 30°S.Footnote⁵⁰ An illuminating picture of where and when the global darkening (of ) exists is provided by zonal-mean (i.e., average around the world at each latitude) Absorbed Solar Radiation (). Latitudes are compressed toward the poles in so that an increment of latitude in the graph is a true measure of area on the globe. Northern Hemisphere midlatitudes are the dominant region of increased Absorbed Solar Radiation, with a large increase beginning in 2020. Some increase of Absorbed Solar Radiation also began in early 2015, at the time a severe restriction on fuel sulfur was imposed in coastal regions around northern Europe, North America and Hawaii (Sidebar 5). If ships only used low-sulfur fuel while in port and switched to high-sulfur fuel on the open sea, then the coastal restrictions would have little effect,Footnote⁵¹ but it is possible that some ship operators switched to low-sulfur fuel more generally.

So, how much of the 1.7 W/m² darkening of the Earth () is from feedbacks and how much is likely aerosol forcing? The high latitude snow/ice feedback can be estimated from the data in : the latitude ranges 60-90°N and 60-90°S contribute +0.07 W/m² and +0.08 W/m² to the 2020-23 global Absorbed Solar Radiation anomaly, respectively. In contrast, the latitude range 30-60°N contributes +0.53 W/m² in 2020-23 and +0.67 W/m² in 2023. Part of this increase of Absorbed Solar Radiation may be cloud feedback, but that should tend to be offset by reduced ship aerosol forcing in ocean areas other than the North Pacific and North Atlantic. Half of the world is covered by these other ocean areas, where ship aerosols have an effect on cloud albedo even in regions without visible ship tracks.Footnote⁵² We conclude only that the aerosol forcing induced by International Maritime Organization restrictions on ships could be of the order of 0.5 W/m², thus much larger than the aerosol forcing (0.079 W/m²) estimated in the IPCC formulation (see above).

Sea Surface Temperature

Sea surface temperature is indicative of heat stored in the ocean’s upper “mixed layer,” which is characterized by a single temperature because of continual turbulent stirring by wind and waves. Thus, sea surface temperature change is a good diagnostic to assess the effect of climate forcings over the ocean. The zonal-mean (i.e., average around the world at each latitude) Sea Surface Temperature anomaly relative to 1951-1980 climatology ()Footnote⁵³ contains a strong imprint of the Absorbed Solar Radiation anomaly at 30-60°N and reveals a clear global picture. The large global warming in 2023 is a combination of a moderateFootnote⁵⁴ tropical El Niño and additional warming that is largest at middle latitudes in the Northern Hemisphere. We interpret the additional warming as mainly the effect of reduced human-made aerosols, especially aerosols produced by ships. The ship aerosol effect is greatest in the North Pacific and North Atlantic because that is where ship-produced sulfate aerosols exceeded natural aerosols (Sidebar 5), but ship aerosols have some effect over most of the world ocean. We show below that natural climate forcings in 2022-2023 also made a contribution to the appearance of an unprecedented spike in global warming.

Sea surface temperatures will decline as the tropics moves into its La Niña phase, but we expect cooling to be limited, as it was after the 2015-16 El Niño (). Temperature will not decline to pre-El Niño levels because Earth is far out of energy balance,Footnote⁵⁵ with more energy coming in than going out and with the vast majority of the excess energy being stored as heat in the ocean. Earth’s energy imbalance is measured by the combination of CERES satellite instruments⁴⁴ (which measure change of the imbalance) and 4,000 deep-diving Argo floats⁵⁵ distributed around the ocean (which calibrate the satellite data by measuring change of ocean heat content over a decade). A revealing picture is provided by the zonal-mean energy balance at the top of the atmosphere over the ocean (). Most of the increased energy uptake occurs in the extra-tropics of the Northern Hemisphere. Earth’s global energy imbalance was 0.6, 1.11, and 1.36 W/m² in 2001-2014, 2015-2019, and 2020-2023, respectively. The imbalance over 30-60°N ocean was 0.67, 1.41, and 2.56 W/m² in the same periods.

The location and timing of changes in Earth’ energy balance and sea surface temperature support our interpretation that decreasing ship emissions are a major cause of the recent increase of Earth’s energy imbalance and accelerated global warming. Our interpretation is at odds with that of the United Nations Intergovernmental Panel on Climate Change (IPCC). This issue must be illuminated because of implications for climate sensitivity, climate forcings, and policies that will be needed to maintain a climate similar to that in which civilization developed and thrives.

Marriage of Aerosol Forcing and Climate Sensitivity

IPCC’sFootnote⁵⁶ estimate of aerosol forcing () is updated through 2023 by Forster et al.Footnote⁵⁷ (). Aerosol precursor emissions increased rapidly after World War II, as fossil fuel use grew. Global sulfur emissions reached 120 megatons of SO₂ by 1970 and stayed near that level until 2005 (Sidebar 8). Sulfur emissions in the United States began to decline in the 1970s due to the Clean Air ActFootnote⁵⁸ based on concern about acid rain, but decreasing aerosols in developed countries were largely compensated by growing emissions around the world until early in the 21st century. We expect, contrary to the IPCC, that aerosol forcing is more nonlinear, i.e., small emissions in relatively pristine air have an outsized effect. This expectation is supported by our conclusion that the moderate emission reduction (about 10 megatons of SO₂) due to ship fuel regulations (Sidebar 8) altered aerosol forcing by as much as 0.5 W/m².

Sidebar 7. IPCC’s formulation for aerosol forcing¹⁸ is close to linear in global precursor emissions, i.e., proportionate to global emissions (Figure SM1 in Supplementary Material). Thus, IPCC’s aerosol forcing is near its maximum value of about −1.3 W/m² by 1970 and stays near that value until 2005 ().

As a contrast to IPCC’s aerosol forcing scenario, we consider aerosol scenarios, A and B (), which approximate the Matrix and OMA aerosol models of Bauer et al.Footnote⁶² Matrix (Aerosols A) explicitly models aerosol microphysical processes, so we might hope that it is more realistic than OMA (Aerosols B), which is simply based on aerosol mass. The Bauer aerosol models use the same CEDS (Community Emissions Data Systems)⁵⁹ emission data employed by IPCC, but we believe the Bauer models more realistically capture the nonlinearity of the aerosol effect on clouds, i.e., the fact that aerosols emitted into a pristine environment have a greater effect on clouds than aerosols emitted into air that is already heavily polluted.Footnote⁶³ The models agree that IPCC understates aerosol forcing: aerosol forcing increases until 2005, when a “turning point”Footnote⁶⁴ is reached mainly due to emission reductions in China during 2006-2014.Footnote⁶⁵ The continued increase of aerosol forcing in 1970-2005 has major ramifications for understanding of climate sensitivity.

Aerosol forcing and climate sensitivity are each important and should be independent issues, but, due to the absence of global aerosol and cloud measurements needed to calculate the aerosol forcing accurately, aerosol forcing and climate sensitivity were wedded in an inappropriate shotgun marriage. We now seek to disentangle and expose their relationship with simple computations understandable to a broad audience. That goal requires that we first take a fresh look at the classic climate problem: how much will Earth warm if atmospheric CO₂ is doubled?

Global Temperature Response to Doubled CO₂

Global climate models (GCMs) are complex, requiring supercomputers for climate simulations. Modeling activity is well organized – Climate Model Intercomparison Projects (CMIPs) are conducted prior to the IPCC reports – but young researchers note that heavy emphasis on GCMs tends to crowd out a well-balanced focus on underpinning science issues, critical thinking, theoretical comprehension, and communication.Footnote⁶⁷ In addition, the complexity of GCMs places lead climate modelers in a position of expert gatekeepers to knowledge about climate, and their focus on complex models limits the public’s understanding of climate change.

As a partial antidote, we suggest a simple calculation that is more amenable to understanding. It involves two steps. First, we need the Global Temperature Response to an instant doubling of atmospheric CO₂ for a range of plausible climate sensitivities. Second, we use these response curves for simple calculation of global temperature change due to known climate forcings of the past century. Results provide insights about climate sensitivity, aerosol forcing, and causes of the unusual global warming in 2023-2024.

Global Temperature Response to doubled CO₂ was chosen by Prof. Jule Charney, chair of a pioneering study of climate sensitivity,²³ as a tool to study climate change.Footnote⁶⁸ It was an astute choice, allowing primitive global models – that were just budding half a century ago – to be used for computations that helped illuminate issues in climate physics. The doubled CO₂ studies were conducted for an idealized case in which the world’s ice sheets were imagined to be unchanging. The effect of climate change on ice sheets is a crucial issue, but the complications of ice sheet change needed to wait until there was a better understanding of the atmosphere and ocean.

Climate response to doubled CO₂ forcing is now routinely calculated for GCMs to calibrate a model’s sensitivity – and the results reveal important climate physics. The doubled CO₂ Global Temperature Response of the most recent fully-documented GCMFootnote^69,Footnote⁷⁰ of the Goddard Institute for Space Studies, dubbed GISS (2020), has sensitivity ∼3.4 °C, as shown in (in blue) along with an earlier model of sensitivity 2.6 °C (in green), and a response estimated for 4.5 °C sensitivity (in orange). The long timescale of the temperature responses – representative of today’s climate models and presumably of the real world – are a consequence of the ocean’s great thermal inertia.

About 40% of the eventual (equilibrium) warming is achieved in 10 years, 60% in 100 years, and 90 percent in 1,000 years.Footnote⁷¹ The early response is largest over continents, where the response is not held down tightly by the ocean’s great thermal inertia.

The long delay of climate in achieving its equilibrium response is both a curse and a blessing. The problem is that the public responds to threats that it sees and feels today, not to future threats perceived by scientists; thus, a great amount of future warming may be built up before actions required to stem climate change are undertaken. On the other hand, delayed response provides humanity time to alter climate forcing so that the equilibrium warming – or even the 100-year warming – and the most consequential consequences may never occur. The delayed climate response, actions to address it, and the effect of these actions are a prodigious topic that we can only introduce in our discussion below.

Climate Forcing Scenarios

To calculate global temperature, we must first specify the forcing scenario. On the century time scale, greenhouse gases (GHGs) and aerosols are dominant forcings, as other forcings are either negligible in magnitude or oscillatory with little long-term effect. The growth rate of GHG forcing () reached 0.4 W/m² per decade by 1970 and neared 0.6 W/m² per decade by 1980, when the growth rate of methane slowed and the Montreal Protocol began to reduce the growth of trace gases that threatened the stratospheric ozone layer.Footnote⁷³ IPCC projections of future GHG growth rates (yellow region in ) are discussed below under Policy Implications. RCP = Representative Concentration Pathway, these are IPCC scenarios; RCP2.6 requires CO₂ emissions start declining by 2020 and go to zero by 2100; RCP 4.5 is a “moderate” scenario in which emissions peak around 2040 and then decline; RCP 8.5 is a worst-case scenario in which emissions continue to rise throughout the twenty-first century.

IPCC data has GHGs as the sole cause of rapid global warming during 1970-2005, with aerosol forcing nearly constant over that period. Climate forcing with IPCC aerosols thus grew during 1970-2005 about 0.5 W/m² per decade () based on GHG growth (). In contrast, with the Aerosol A and B scenarios () the GHG + Aerosol forcing grows only 0.2-0.3 W/m² per decade during 1970-2005 and then accelerates (). The 1970-2005 period with contrasting forcings thus allows us to examine the issue of climate sensitivity.

In addition, although solar and volcanic forcings have little effect on long-term trends, we want to know if they play a role in the unusual 2023-24 global warming. Thus, we include recent change of these natural forcings (lower right corners of ) based on satellite data. Effects of the January 2022 Hunga volcanic eruption were too small to stand out above other sources of variability, but a comprehensive analysis¹⁶ concluded that cooling by Hunga aerosols exceeded warming by water vapor injected into the stratosphere by the volcano as well as induced ozone changes. We use the well-observed 1991 Pinatubo eruption to define the time scale for aerosol formation and removal from the stratosphere (Sidebar 3), with the forcing for Hunga an order of magnitude smaller, based on SO₂ injection amounts.

We approximate Ship Aerosol Forcing – for clarity and due to lack of information – as an instant +0.5 W/m² forcing added on 1 January 2020. Part of the forcing may have been added in January 2015 by emission limits in coastal North America and Northern Europe, if those led to changes of fuel use on the open ocean. Limits imposed in July 2010 and January 2012 may have had a small effect (Sidebar 5), but the main Ship Aerosol Forcing change occurred in January 2020.

Global Temperature Scenarios

Global temperature change can be calculated in seconds because climate responds mainly to the planetary energy imbalance, with less dependence on forcing mechanisms. Thus, knowledge of the forcing magnitude and the doubled CO₂ temperature response suffices for the calculation. Further, dependence of the response on a specific forcing can be accounted for via an “efficacy” factor.Footnote⁷⁴ Temperature change is obtained with a single equation – the most elementary in Isaac Newton’s calculus – in an intuitive calculation that requires no advanced mathematics training to understand (see Note Footnote⁷⁵ at end). Validity of this temperature calculation rests on the assumption that the ocean general circulation is not altered by the climate forcing. Fixed ocean circulation should be accurate enough for the past century, but there are signs that overturning circulations (see Notes 48, 95, 96 at end) in the North Atlantic and Southern Oceans are now both on the verge of major disruptions that, if allowed to proceed, will dramatically alter future climate, as discussed below under “The Point of No Return.”

Climate forcings of and the calculation (see Note 75 at end) yield global temperature change. GHGs and global temperature have been accurately measured since the 1950s, making post-1950 temperature relative to 1951-1980 () best-suitedFootnote⁷⁶ for testing the ability of alternative aerosol histories to capture the 1970-2005 global warming rate. The conclusion is that all three aerosol forcing scenarios of can fit observed 1970-2005 warming and produce at least moderate warming acceleration, but they require successively higher doubled CO₂ sensitivities: about 3- 3.5 °C, 4.5 °C and 6 °C, for the IPCC, Aerosols A, and Aerosols B scenarios, respectively.Footnote⁷⁷ Aerosol scenarios could be tweaked in each case to obtain arbitrarily close fit to observed temperature, but there is no point to do that. The main conclusion is that modern temperature change does not provide a tight constraint on climate sensitivity because aerosol forcing is not measured; however, if aerosol forcing is nonlinear (as in Aerosols A and B), IPCC (and thus most of the scientific community) has underestimated climate sensitivity.

Another test is the magnitude of global warming since preindustrial time, including the unique 2023 warming. Calculated temperatures for 1850-2024 are provided in Figure SM4 (in Supplementary Material), but for clarity expands 21st century temperature. All three aerosol scenarios can reach 1.6 °C warming in 2023, but IPCC aerosols require a high sensitivity that then could not match warming of the past 50 years.Footnote⁷⁸ The IPCC aerosol scenario and IPCC best estimate of climate sensitivity 3 °C do not produce warming to +1.6 °C in 2023; most decidedly, they cannot produce +0.4 °C warming in 2023, even with the help of the modest, observed, El Niño that can only add a temporary +0.2 °C. This difficulty has led to consternation that “something is wrong.” In contrast, Aerosols A and B scenarios, with their associated climate sensitivities (∼4.5 and 6 °C for 2 × CO₂), the ship aerosol forcing, and a +0.2 °C El Niño, readily reach +1.6 °C current warming.

The warming detail needing explanation is the +0.4 °C leap in 2023. We interpret temperature change in the 2020s as being affected by the strong decline of the Pacific Decadal Oscillation (PDO) in 2020-22, which temporarily hid the ship aerosol effect on global temperature. The PDO is a natural cycle of sea surface temperature patterns in the Pacific with associated cloud changes. Positive PDO has cloud cover that yields increased absorption of solar radiation by the ocean surface (). The PDO moved rapidly into negative values in 2020, which normally results in a negative anomaly of absorbed solar radiation, but the opposite occurred. In contradiction to the PDO, absorbed solar radiation reached +6 W/m² averaged over the North Pacific (), this being, we suggest, at least in part the expected result of reduced ship aerosols. Ocean surface mixed layer and sea surface temperature in the North Pacific and North Atlantic rose steadily during the four years 2020-23 ( and ) in the regions of maximum aerosol effect (Sidebar 5). By 2023, PDO had reached bottom () and no longer added to global cooling, so the effect of decreased aerosols began to show up in global temperature.

Conclusions from the global temperature calculations are substantial. IPCC’s calculated aerosol forcing and best estimate for climate sensitivity are not consistent with observed warming. Aerosols A and climate sensitivity 4.5 °C for doubled CO₂ are consistent with observed warming, which is encouraging because aerosols A is based on the newer Bauer model that simulates aerosol microphysics and 4.5 °C sensitivity agrees well with glacial-interglacial climate oscillation in the past 800,000 years,¹ the only paleoclimate case with accurate knowledge of climate forcings in equilibrium climates, as required for empirical assessment of equilibrium climate sensitivity.Footnote⁷⁹

Now we must look in more detail at global temperature change in the 2020s, specifically asking whether the unprecedented global warming of 2023 contains information that can help confirm or refute the large ship aerosol forcing that we inferred from Absorbed Solar Radiation. The 2023 global warming was nominally the result of an El Niño, but can the El Niño alone explain the magnitude of the warming?

Fingerprinting the Climate Acceleration Mechanisms

Interpretations of the 2023 warming are bookended by Raghuraman et al.Footnote⁸⁰ and Schmidt.³ Raghuraman et al. conclude that the 2023 warming is explained by the El Niño, while Schmidt concludes that the extreme warming cannot be explained by even the full array of mechanisms in global models. Raghuraman et al. note that the 2023 El Niño rose from a deep La Niña, so, despite the El Niño being modest, the Niño3.4 (equatorial Pacific temperature used to characterize El Niño status) change from 2022 to 2023 was about as large as the Niño3.4 changes that drove the Super El Niños of 1997-1998 and 2015-2016. They suggest that the change of annual mean global temperature between 2022 and 2023 (0.28 °C, ) can be accounted for by El Niño warming. However, change of annual mean temperature (black squares in ) does not capture the magnitude of the 2023 warming, which is exposed by the 12-month running mean temperature (which includes the annual mean every December). The La Niña held down global temperature in 2020-2022, but the modest following El Niño cannot account for the remarkable 0.4 °C global warming. We conclude that Schmidt is partially right: something else important is occurring.

Here, like a detective who dusts a doorknob and lifts a fingerprint with clear adhesive tape, we extract fingerprints of the mechanisms that caused the acceleration of global warming in two simple steps. The first step is to remove the long-term trend of global temperature (0.18 °C per decade) by subtracting it from the global temperature record since 1970. (The long-term trend is caused by the net greenhouse gas plus aerosol forcing.) What remains is the blue curve in , which is global temperature change due to other forcings and natural variability. The main source of natural variability is the tropical El Niño cycle, shown by the temperature anomaly in the tropical Niño3.4 region (red curve). Thus, as a second step, we subtract the El Niño variability from the blue curve,Footnote⁸¹ obtaining the green curve in .

Fingerprints in the green curve are apparent. Most obvious is the 0.3 °C global cooling caused by the Pinatubo volcanic eruption, but even the maxima of solar irradiance (a forcing of only ± 0.12 W/m², ) cause detectable warmings consistent with prior analyses.Footnote⁸² The portion of the fingerprint of present interest is the decade-long anomaly that began in 2015 () and grew to an astounding +0.3 °C in 2023, which we will associate mainly with ship aerosol forcing. This anomaly does not coincide with reduction of aerosol emissions in China, which began in the first decade of the century but left a still highly polluted atmosphere in China.

How much global warming is expected today from a ship aerosol forcing added five years ago (January 2020), for our estimate of a 0.5 W/m² forcing? A 0.5 W/m² CO₂ forcing yields 0.2 °C warming ().Footnote⁸⁴ But what is the efficacy of an ocean-only forcing relative to global CO₂ forcing? To evaluate that, we ran GCM climate simulations with uniform forcing over the ocean (Figure SM3, Supplementary Material), finding an efficacy 77% the first year, 93% the second year, and 100% by the third year.Footnote⁸⁵ Thus, global warming for 0.5 W/m² ship forcing today is 0.2 °C. Raghuraman et al. are correct that the 2023 El Niño rise from a deep La Niña makes the 2023 El Niño effect similar to the 1997 and 2015 El Niños (), but that leaves an enormous 0.3 °C global warming to explain. The Sun is now near maximum irradiance, a min-to-max forcing of 0.24 W/m² (). Doubled CO₂ forcing of 4 W/m² yields warming of 1-1.5 °C in five years (), which we must reduce by the ratio of solar and CO₂ forcings (0.24/4), yielding a solar cycle warming of 0.06-0.09 °C and leaving just over +0.2 °C warming to explain. Our estimated 0.5 W/m² ship aerosol forcing yields +0.2 °C warming in 2024. Thus, the global warming anomaly in 2023-2024 is accounted for well and supports our estimated ship aerosol forcing.

Reconciling Our Analysis and Aerosol Models

How can we reconcile our estimate of 0.5 W/m² for ship aerosol forcing with the six aerosol modeling studies mentioned above,^41,42 which are in mutual agreement that the global ship aerosol forcing is small, in the range 0.07-0.15 W/m²? Let’s first summarize our alternative analysis of the aerosol forcing and then suggest an approach to resolve the large difference.

Our initial estimate of the ship aerosol forcing was based on the precise CERES satellite data, calibrated absolutely with Argo float data.^44,55 The CERES data show that Earth’s albedo (reflectivity) decreased 0.5% since 2010, corresponding to a 1.7 W/m² global average increase of Absorbed Solar Radiation. Based on the spatial and temporal coincidence of the increased absorption with regions where the effect of ship aerosols should be largest – the North Pacific and North Atlantic – we infer a Ship Aerosol Forcing of ∼0.5 W/m², an order of magnitude larger than follows from the IPCC aerosol formulation. We also show that the albedo feedbacks due to reduced high latitude snow and ice constitute no more than 0.15 W/m², so there is plenty of room in the 1.7 W/m² to also accommodate the cloud feedback implied in the shifting of climate zones identified by Tselioudis et al.^24,Footnote⁸⁶ Sea surface temperatures (SSTs) support our interpretation. SSTs are rising fastest where the aerosol forcing is largest, in the North Pacific and North Atlantic, and they are rising at low latitudes where aerosol forcing is widespread, even though smaller.

How can we explain the small aerosol forcing produced by aerosol-cloud models? Aerosol-cloud modeling involves complex microphysical interactions and is still at a primitive stage. Perhaps there is pressure to get “the right answer:”⁶⁶ ship aerosol emissions are a small fraction of total human-made emissions, so the ship aerosol forcing must be small to avoid contradicting the authoritative IPCC assessment (see Sidebar 5). Scientific reticenceFootnote^87,1 may also come into play, a preference to move cautiously toward an answer that differs from that of recognized authority.

Priority in physics is given to observations, which here is global monitoring of aerosol and cloud particle microphysics and cloud macrophysics. As noted above (in Aerosol and Cloud Particle Microphysics), a relevant instrument is being tested on a current NASA mission, but adequate monitoring requires long-term observations of specific data by several instruments.³¹ Aerosol and cloud changes are needed to evaluate climate forcings and climate sensitivity, which thus warrants a dedicated satellite mission, so that needed information will be available in the future as climate change rises toward the pinnacle of public interest. Radiation balance (CERES) observations must be continued with a new satellite and Argo data need to be expanded, especially around Antarctica and Greenland (see below).

There is a danger of “being too late” with policy-relevant information. Thus, we also make an effort to define a near-term ship aerosol footprint that will help verify, or disprove, our inference of a large ship aerosol forcing as soon as possible.

Ship Aerosol Footprint: Near-Term Climate Prediction

Ship aerosol forcing should have an indelible footprint that will be obvious soon: SSTs (sea surface temperature) and global temperature will remain abnormally high. The ship aerosol effect is largest in the North Pacific and North Atlantic where human-made sulfate aerosols dominate over natural aerosols, but ship emissions are substantial at low latitudes in both hemispheres. Global SSTs, and thus global surface temperature, should remain high even during the next La Niña. Global warming of 0.2 °C from ship aerosol reduction will grow slowly beyond year 5 of forcing initiation (), but it will prevent global temperature from falling much below +1.5 °C relative to preindustrial (late 19th century) time. Thus, our prediction is that global temperature averaged over the El Niño/La Niña cycle has already reached the +1.5 °C threshold.

These high SSTs constitute a heavy footprint for people. Increased SSTs are indicative of rising heat content of the ocean’s wind-mixed surface layer, which provides energy for stronger storms with more extreme rainfall amounts. The rote explanation that “warming of 1 °C allows the air to hold 7% more water vapor” is not a full explanation of storm intensification and climate impact. Water vapor’s fueling of storms – thunderstorms, tornadoes, and tropical storms – is a main factor causing more extreme storms and floods.Footnote⁸⁸ A paperFootnote⁸⁹ attached to the first author’s congressional testimony in 1989, describing research by a team of eight scientists, concluded that global warming increases “moist static energy”Footnote⁹⁰ near Earth’s surface and causes a larger portion of rainfall to be in more powerful thunderstorms that rise to greater heights, as opposed to gentler rainfall from stratiform clouds. Also, increased heat content in the ocean’s surface layer provides energy for rapid tropical storm intensification and drives stronger, wetter, storms.

Our main conclusions – that climate sensitivity is higher and aerosol forcing is greater than in IPCC’s best estimates – add to the climate threat. Nevertheless, the climate system’s slow response allows the possibility to avoid the “point of no return,” the point when disastrous climate change would run out of humanity’s control. A happy ending – with restoration of a propitious climate – requires an understanding of this greatest climate threat.

The Point of No Return

Tipping points are a big concern in popular and scientific discussion of climate change. The most dire belief is that today’s accelerated warming is a sign of runaway feedbacks that are pushing climate beyond multiple tipping points, thus causing global warming acceleration that threatens eventual collapse of civilization. Our analysis does not support such beliefs. Instead, we find that observed acceleration of global warming is caused by a human-made climate forcing: reduction of atmospheric aerosols, especially aerosols produced by commercial shipping.

Climate feedbacks are real; paleoclimate evidence shows that “fast” feedbacks (water vapor, clouds, and sea ice) amplify climate sensitivity from 1.2 °CFootnote⁹¹ for doubled CO₂ with no feedbacks to as much as 4-5 °C, i.e., these well-known feedbacks more than triple the equilibrium climate response. However, equilibrium climate response is slowed by the ocean’s thermal inertia. For example, warming from the 2020 reduction of ship aerosols is one-third complete after five years; the next third requires a century and the final third requires millennia. The mechanism that causes continued slow warming is Earth’s energy imbalance – thus the additional warming will never occur, if we reduce net climate forcing to restore Earth’s energy balance.

Tipping pointsFootnote⁹² are also real. Some feedbacks can pass a point such that the process accelerates and causes amplifying climate feedback. For example, global warming may melt Arctic permafrost, releasing large amounts of greenhouse gases to the atmosphere. Or heating and drying of the Amazon rainforest may reach a point that the rainforest is not self-sustaining, with fires releasing much of the carbon stored in the forest. Many tipping point processes are reversible if Earth cools, but the recovery time varies and may be long for some feedbacks.

The most threatening tipping point – the Point of No Return – will be passed when it becomes impossible to avoid catastrophic loss of the West Antarctic ice sheet with sea level rise of several meters. Large areas in China, the United States, Bangladesh, the Netherlands, island nations, and at least half of the world’s largest cities would be substantially submerged, an irreversible result on any time scale that people care about. Rising seas would be accompanied by increasing climate extremes that are already emerging at global temperature of only +1-1.5 °C.Footnote⁹³ Emigration from populous coastal areas and other vulnerable/disaster-prone regions would add to emigration from increasingly inhospitable low latitudes. Sea level would not stabilize after West Antarctica collapses: there is at least 15-25 m (50-80 feet) of sea level in Antarctic and Greenland ice in direct contact with the ocean. The last time Earth was at +2 °C relative to preindustrial time – in the early Pliocene – sea level was 15-25 m (50-80 feet) higher than today. Sea level change takes time, so coastlines would be continually retreating.

Clearly, we must avoid passing the Point of No Return. Learning how we can do that requires understanding how the ice sheets, ocean, and atmosphere work together.

Ice Sheet, Ocean, Atmosphere Interactions

The IPCC reports dismiss shutdown of the overturning ocean circulation and large sea level rise on the century time scale as low probability, even for high emission scenarios. How did they reach that conclusion? Models with a specific modeling approach. Climate models are an essential tool because there is no natural precedent for rapid human-made climate forcing. A complete global climate model includes the ice sheets, ocean, and atmosphere. Dynamic ice sheets are the most recent of these components to be modeled in detail and are the most challenging due to the wide range of spatial scales: from small-scale action of freeze-thaw cycles in breaking up ice to large-scale movement of ice sheets over land surface and sea floor terrain. Global climate models are supposed to allow realistic interactions among the ice sheets, ocean and atmosphere, but if one of these components is not simulated well, it affects the others.

Twenty years ago, the first author (JEH) had discussions with field glaciologistsFootnote⁹⁴ who were frustrated with IPCC reports and models that they believed portrayed ice sheets as unrealistically lethargic.Footnote⁹⁵ Their concerns were based mainly on observed effects of water – on, within, under, and at the edges of the ice sheets – that could speed the movement and disintegration of the ice. Concern that ice sheet models were too “stiff” led to an alternative perspective on ice sheet stability⁹⁵ based on Earth’s energy balance and feedbacks among the ocean, ice, and atmosphere; this perspective suggested that ice sheets are more mobile in the real world than in ice sheet models. Support for this perspective was provided by paleoclimate data, which revealed oscillations of ice sheet size that could not be produced by existing ice sheet models.⁴⁹

Integrated modeling – with ice sheets, ocean, and atmosphere all included in one model – is one approach that should be, and is, being pursued. But if it is the only approach, there is a danger that it will be slow to achieve real-world dynamic realism. A complementary approach is to use well-tested atmosphere-ocean climate models with testable assumptions for ice sheet behavior. The objective is to compare the model results with reality in hopes of learning things about ice sheet behavior and future climate impacts. This latter modeling approach was pursued with the Goddard Institute for Space Studies climate model, but first it was necessary to address fundamental issues about ocean models, which have their own uncertainties.

Stefan Rahmstorf, the world-leading expert on the ocean’s overturning circulations,Footnote⁹⁶ described a tendency of ocean model development to produce models that are unrealistically stable.Footnote⁹⁷ A related concern about ocean models was their widespread tendency to produce excessive small-scale mixing of ocean properties. As we have already discussed, the excessive mixing of surface heat anomalies caused global models to under­estimate (negative) aerosol forcing.Footnote⁹⁸ Another effect of excessive mixing is to increase stability of the ocean model against possible shutdown of the overturning circulation. When freshwater from melting ice sheets is injected into the ocean surface layer it reduces the density of the salty surface mixed layer. The density reduction tends to decrease the amount of cold, salty, dense water that sinks toward the ocean floor in polar regions in winter; if the density reduction is sufficient, it can even shut down the overturning circulation. Ocean models with excessive, unrealistic, mixing tend to homogenize the water and prevent shutdown. Special effort was made to eliminate unphysical mixing in the Goddard Institute for Space Studies atmosphere-ocean climate model.Footnote⁹⁹ This model was used for climate simulations for the 20th and 21st centuries, and a paper was submitted for publication in 2015.

Ice Melt, Sea Level Rise, and Superstorms

The full title of the submitted paper,Footnote¹⁰⁰ “Ice melt, sea level rise, and superstorms: evidence from paleo­climate data, climate modeling, and modern observations that 2 °C global warming is highly dangerous,” summarized our strategy to assess the danger of passing the Point of No Return. Insight based on combining information from paleoclimate studies, climate models, and ongoing climate change is essential to obtain early, reliable, assessment of climate change. The paper passed extensive peer review and was published in 2016.Footnote¹⁰¹

This “Ice Melt” paper paints a picture of Eemian climate (120,000 years ago) of relevance to climate change today. Eemian global temperature was about +1 °C relative to the preindustrial Holocene.Footnote¹⁰² Mid-Eemian sea level was about the same as today and the Antarctic and Greenland ice sheets were similar to their present sizes. Late in the Eemian period,Footnote¹⁰³ sea level rose several meters within a century, the rapid rise being recorded in the rate that coral reef-building “backstepped” toward the shore in response to the rising seas.Footnote¹⁰⁴ It is likely that the added sea level was from collapse of the West Antarctic ice sheet because that ice sheet sits on bedrock hundreds of meters below sea level, making it vulnerable to ocean warming and rapid disintegration.

Late Eemian climate also featured shutdown of the North Atlantic overturning circulation, as revealed by ocean cores of seafloor sediments.Footnote¹⁰⁵ Shutdown of this ocean circulation short-circuits interhemispheric transport of heat by the global ocean conveyor,Footnote^106,Footnote¹⁰⁷ which normally transports a huge amount of heat – 1,000 trillion watts – from the Southern Hemisphere into the Northern Hemisphere. That heat amounts to 4 W/m² of energy averaged over the Northern Hemisphere, but it is mostly concentrated in the North Atlantic region, which is thus warmer than expected for its latitude. When the ocean conveyor shut down, that heat stayed in the Southern Ocean, where it may have contributed to collapse of the West Antarctic ice sheet. Meanwhile, in the North Atlantic region, there was evidence of powerful storms. This picture of the Eemian, if filled out in finer detailFootnote¹⁰⁸ including the sequencing of events, may help us anticipate where our present climate is headed, if effective actions are not taken to halt and reverse human-made climate change, restoring relatively stable Holocene climate.

Climate simulations in “Ice Melt” were carried out with a climate model that passed crucial tests such as having deepwater formation at several locations close to the Antarctic coast, a test that many other models failed. In the climate projections, it was assumed that growth of ice sheet melt would be nonlinear, based on paleoclimate data showing that sea level on occasion rose several meters in a century. Freshwater fluxes into the ocean were estimated as 360 Gt/year (a gigaton, Gt, is one billion tons) in the Northern Hemisphere and 720 Gt/year in the Southern Hemisphere in 2011 with doubling times for these rates being either 10 years or 20 years. The largest freshwater source is melting of ice shelves, the tongues of ice that extend from the ice sheets into the ocean.Footnote¹⁰⁹ The range of doubling times for freshwater injection – 10 years to 20 years – was based on limited observations, but still seems to be an appropriate estimate. Observations that help improve this estimate are needed.Footnote¹¹⁰

Our climate simulations led to the staggering conclusion that continued growth of ice melt will cause shutdown of the North Atlantic and Southern Ocean overturning circulations as early as midcentury and “nonlinearly growing sea level rise, reaching several meters in 50-150 years.”Footnote¹¹¹ These results contrast sharply with IPCC conclusions based on global climate models. Growing freshwater injection in the Ice Melt model⁴⁹ already limits warming in the Southern Ocean by the 2020s with cooling in that region by midcentury. In contrast, models that IPCC relies on have strong warming in the Southern Ocean. Observed sea surface temperature is consistent with results from the Ice Melt model,⁴⁹ but inconsistent with the models that IPCC relies on ().Footnote¹¹²

Earth’s temperature only reached the Eemian level, +1 °C, about a decade ago and is now already at +1.5 °C. It’s crucial that we understand the implications of this warming for today’s young people and for their children and grandchildren. We must understand it well enough, soon enough, that we can avoid handing them a planet headed irrevocably to the Point of No Return, with ice sheets headed for collapse and sea level out of humanity’s control.

Long-Term Climate Change

Are the United Nations and public well-informed about the status of long-term climate change? The Secretary General of the UN in the past few years has made increasingly frantic statements about the urgency of actions to stem global warming, but in the context of unrealistic appraisal of the possibility of achieving the goal of the Framework Convention on Climate Change. Frank admission of the status of climate change and the implausibility of limiting global warming to a level below 2 °C with the present policy approach is needed. Realistic assessment is needed to help evaluate the actions that are needed to provide the best chance to attain and preserve a propitious climate and environment for today’s young people and their descendants.

Global temperature leaped up in the past two years, passing the +1.5 °C level, and it will continue to rise for at least the next few decades, with natural oscillations about the human-made long-term change. The recent acceleration of the global warming rate should not last long. That acceleration is driven mainly by a unique forcing, the forcing of about 0.5 W/m² caused by reduction of sulfur emissions from commercial ships, not by runaway feedbacks or climate tipping points.Footnote¹¹³ Our Faustian debt is not paid off by any means; the warming due to reduction of ship aerosols is only one-third complete, but the second and third portions will occur over a century and a millennium, which gives humanity time to take action. Potential additional reduction of aerosols, mainly of continental sources, is about 1 W/m² according to IPCC, but more likely in the range 1.5-2 W/m² for all aerosol sources, including wood and other biomass burning.¹ Although that is large forcing and large potential warming, if the world were to return to a pristine pre-human atmosphere, nothing approaching complete return is plausible with a human population of billions that is still growing. Burning of wood and other biofuels will continue for the foreseeable future and human-caused forest and grass fires are likely to grow as climate extremes increase. We need to measure aerosol changes better, but additional aerosol changes seem unlikely to be a major drive for climate change in the next few decades.

Thus, continued global warming will depend mainly on fossil fuel emissions ().Footnote^114,Footnote¹¹⁵ The resulting greenhouse gas climate forcing () is now increasing almost 0.5 W/m² per decade, an amount that dwarfs changes of other climate forcings, including aerosols. Ever since the Kyoto Protocol was achieved in 1997, it has been hoped that voluntary goals for emission reductions would slow the growth of global emissions as needed to avoid dangerous climate change. In fact, global emissions accelerated, demonstrating that short-term economic self-interest trumps concern about long-term degradation of the global commons. The gravity of the situation is shown by , which compares reality with the greenhouse gas scenario (RCP2.6) designed by IPCC to limit global warming to less than +2 °C. Annual growth of greenhouse climate forcing is now more than double the amount in IPCC’s target scenario, which was never realistic because it relied on an assumption of massive carbon capture at powerplants with permanent burial of the captured CO₂. Carbon capture at the gigaton scale does not exist; the estimated annual cost of CO₂ extraction is now $2.2-4.5 trillion dollars per year,Footnote¹¹⁶ and the gap between the IPCC scenario and reality is rising rapidly (). Such hypothetical large-scale carbon capture will not happen in anything near the required timeframe.

Imaginary, implausible, scenarios are harmful. Misleading plans for “net zero” emissions by midcentury – while present policies guarantee that high fossil fuel emissions will continue – disguise failure to face reality. How is it that the United Nations advisory structure appears to be so oblivious of real-world energy needs and the time scale on which fossil fuel emissions will realistically be brought down? Contrary to hype of some environmental organizations, fossil fuels are not a narcotic pushed on the public by an evil industry; they are a convenient condensed form of energy that has helped raise the standard of living in much of the world. The realpolitik is: as long as the global commons are available as a free dumping ground for pollution, most nations with fossil fuel reserves will exploit those reserves. A radical change of global climate policy is needed, as discussed in our final section below.

Given this grim picture, what is our basis for optimism? Why do we believe that it is realistic to avoid passing the Point of No Return? Our optimism is based on the growing interest of young people in the condition of the world that they and their descendants will live in, and in their conviction that they should follow the science. The scientific approach, as we will explain, has potential to lead to a radical change of policy.

Global Justice: Policy Implications

Global emissions will remain high and climate will pass the Point of No Return, if the atmosphere continues to be a free dumping ground for fossil fuel emissions. Current emissions are coming more and more from nations with emerging economies, such as China and India, but climate change is caused by cumulative (total historical) emissions Footnote^117,Footnote¹¹⁸ (), for which the United States and Europe are the largest contributors. This responsibility becomes even more apparent in per capita contributions to cumulative emissions (, based on 2020 populations). The per capita cost of removing prior emissions, as needed to restore Holocene climate, is shown on the right-hand scale of , based on the most optimistic (low end) cost estimate.¹¹⁶ This large cost provides one measure of the scale of the climate problem.

Global injustice of the present political approach to climate change is obvious. Intergenerational injustice is clear: young people and their descendants will suffer consequences of climate change that was initiated and left unchecked by older generations. International injustice is also manifest, as many nations – especially those at low latitudes and low elevation – will be hit hardest by climate change, despite having little responsibility for climate change. The Framework Convention on Climate Change – overseen by the United Nations with annual COP (Conference of the Parties) meetings and supported by the Inter­governmental Panel on Climate Change (IPCC) – was supposed to stem climate change so as to minimize these global injustices, but it has been ineffectual. Why? We assert that this political approach has not followed the path dictated by science. There is evidence that young people are fed up with this ineffectual political approach and wish to follow the science, as we can illustrate with important examples.

The most fundamental need is for a rising price on carbon emissions, which is the essential underlying policy needed to guide the world to a prosperous clean-energy future. Economic scientists overwhelmingly agreeFootnote¹¹⁹ that a simple rising carbon fee (tax), collected at domestic fossil fuel mines and ports of entry, with 100% of the funds distributedFootnote¹²⁰ uniformly to the public as “dividends,” is the most effective and socially just way to implement a carbon fee. Low-income and most middle-income people would gain financially, with the dividend exceeding their increased energy costs. Student body presidents at colleges and universities in all 50 states in the U.S. agreed to “follow the science” and support carbon fee and dividend.Footnote¹²¹ Later 700 high school student leaders from all 50 states endorsed this approach.Footnote¹²²

A second example of following the science is also informative. Although a rising carbon fee is the underlying requirement to phase out carbon emissions, it is not sufficient. Governments also must assure that adequate carbon-free technology is available. Yet, rather than supporting competition among alternative energies, most governments chose to support innovation and development only of “renewable” energies, a political “solution” that serves to hamstring future generations by slowing the transition away from fossil fuels.Footnote¹²³ Buried deep in IPCC reports is information that nuclear power has the smallest environmental footprint of major energy sources, but politics caused a failure to develop modern nuclear power (Sidebar 10). It takes time to drive down the costs of new technology – as demonstrated by solar and wind power – but there is still, if barely, time for additional nuclear power to be brought on-line to provide the firm (available 24/7) energy needed to complement renewables, including the ability to provide high-temperature energy required by heavy industry. It may be just in time to help us avoid passing the Point of No Return.

Sidebar 10. Based on construction materials (steel, concrete, etc.) for a nuclear power plant and cost of nuclear fuel, nuclear energy could be among our least expensive energies, but it is not at this time. Many governments, especially states in the U.S., required utilities to have “renewable portfolio standards” rather than “clean energy portfolio standards,” thus providing an unlimited subsidy to renewable energy for decades, stunting investment in nuclear power. Disinformation played a role in opposition to nuclear power, e.g., in the emphasis of danger in “nuclear waste.” Nuclear waste is contained and has caused little problem, especially in comparison to waste from other energy sources; even old technology nuclear power demonstrably saved millions of lives.Footnote¹²⁴ In addition, exaggerated danger of tiny amounts of nuclear radiation are used by nuclear power opponents to require regulations that slow nuclear construction and increase costs.Footnote¹²⁵

Nuclear power warrants a further comment because it stands as a warning about the next big issue in climate change policy, which we will discuss next. Opposition to nuclear power was “successful” in blocking development of nuclear power for several decades, thus excluding modern nuclear power from the toolbox to deal with climate change. Given the absence of low-cost, ultrasafe, modern nuclear power in the 21st century, fossil fuels were the practical option for firm electric power as the complement to intermittent renewable energy. Thus, there has been a long delay in phasedown of fossil fuel emissions in developed nations and a vast infrastructure of fossil fuel powerplants and high-carbon industry was built, especially in emerging economies. In turn, as so vividly illustrates, global temperature in excess of +2 °C was locked in, absent purposeful actions to affect Earth’s energy imbalance.

Purposeful Global Cooling

Today’s older generations – despite having adequate information – failed to stem climate change or set the planet on a course to avoid growing climate disasters. And they tied one arm of young people behind their back by supporting only renewable energies as an alternative to fossil fuels. Now, as it has become clear that climate is driving hard toward the Point of No Return, there are efforts to tie the other arm of young people behind their back. We refer to efforts to prohibit actions that may be needed to affect Earth’s energy balance, temporarily, while the difficult task of reducing greenhouse gases is pursued as rapidly as practical – namely Solar Radiation Modification (SRM). Purposeful global cooling with such climate interventions is falsely described as “geoengineering,” while, in fact, it is action to reduce geoengineering. Humanmade climate forcings are already geoengineering the planet at an unprecedented, dangerous, rate.

We, the authors – who range in experience from young people just beginning our careers to older scientists who have spent half a century in research aimed at better understanding of Earth’s climate – are concerned about the danger of again “being too late” in informing the public about actions that may be needed to preserve the marvelous world we inherited from our parents. We do not recommend implementing climate interventions, but we suggest that young people not be prohibited from having knowledge of the potential and limitations of purposeful global cooling in their toolbox. We do not subscribe to the opinion that such knowledge will necessarily decrease public desire to slow and reverse growth of atmospheric greenhouse gases; on the contrary, knowledge of such research may increase public pressure to reduce greenhouse gas amounts.

Given that global temperature is already +1.5 °C, given Earth’s present energy imbalance of about +1 W/m² (see below), given the evidence that climate sensitivity is high, given the expectation of at least moderate additional reduced-aerosol warming, and given the prospect of additional greenhouse gas emissions (), we conclude that the world is headed to temperatures of at least +2-3 °C. If such global warming occurs and persists, it will push the climate system beyond the Point of No Return, locking in sea level rise of many meters and worldwide climate change, including more powerful storms and more extreme floods, heat waves, and droughts. Given the difficulty of achieving consensus on policy actions, research is needed during the next decade to define the climate situation better and the efficacy of potential actions to minimize undesirable climate change. For that purpose, the United Nations IPCC approach, heavily emphasizing global climate modeling, is insufficient. Observations and research are needed to better understand effects of the ocean and atmosphere on ice sheets, as is a focused effort to understand rapid sea level rise during the Eemian period. Research on purposeful global cooling should be pursued, as recommended by the U.S. National Academy of Sciences.Footnote¹²⁶ Solar Radiation Modification to counter global warming was suggested by Mikhail BudykoFootnote¹²⁷ in 1974 and later by Paul Crutzen.Footnote¹²⁸ Their idea is to mimic the cooling effect of a volcano by injecting sulfates into the stratosphere. A benefit of such aerosol cooling was revealed in climate simulationsFootnote¹²⁹ in which aerosols equivalent to the Pinatubo volcanic injection were added over the (1) entire globe, (2) Southern Hemisphere, (3) Southern Ocean and Antarctica, or (4) Antarctica (Figure SM5, Supplementary Material). The aerosols cool the Southern Ocean at depth () in mirror image of ocean warming caused by greenhouse gases. The importance of this finding is the implied effect on processes that determine ice sheet stability (Sidebar 11).

Sidebar 11. Ice shelves adhered to the Antarctic continent extend down the side of the continent to depths as great as 2 km in the Southern Ocean, where they provide the strongest buttressing forceFootnote¹³⁰ holding the ice sheet in place. Ice shelves are the “cork” that prevents rapid expulsion of Antarctic ice into the Southern Ocean – especially the vulnerable West Antarctic ice, which rests on bedrock below sea level.Footnote¹³¹ The rapid Eemian sea level rise likely was preceded by melting of Antarctic ice shelves. Today, ice shelves around Antarctica are again melting, with the melting accelerated by slowdown of the ocean overturning circulation. The overturning is driven by cold, salty water near the Antarctic coast that sinks to the ocean floor, compensated by rising, warmer water; this circulation is an escape valve for deep ocean heat. Global warming today is increasing ice melt around Antarctica, freshening and reducing the density of the upper ocean, thus reducing the overturning circulationFootnote⁴⁹ and escape of ocean heat to space during the cold Antarctic winter. Based on a conservative estimateFootnote¹¹⁰ of observed ice melt in 2011 and a 10-year doubling time for the melt rate, a global climate model yields a 30% slowdown of the overturning circulation in 2025,Footnote¹³² consistent with observational data.Footnote¹³³ Thus, today the ocean surface layer around Antarctica is freshening and cooling (, Cheng et al.),Footnote¹³⁴ but the ocean below is warming. Purposeful aerosol cooling recharges this overturning Antarctic circulation, allowing deep ocean heat to escape to the atmosphere and space and cooling the ocean at depth while warming much of the thin surface layer as the upwelling deep-ocean heat melts sea ice ().

There are numerous recent modeling studies^, on the effect of stratospheric aerosols, including a strong reminderFootnote¹³⁵ that a high greenhouse gas scenario such as RCP8.5 creates such great warming and melting that aerosol intervention will almost surely be fruitless in the end. Our is a shocking revelation that real-world greenhouse gases are increasing at nearly the RCP8.5 rate. Policy must focus on reducing actual greenhouse gas emissions to a steeply declining growth rate relative to RCP8.5 (). Solar Radiation Modification (SRM) – whether via stratospheric aerosols or otherwise – should be considered only as a possibility to address temporary overshoot of safe global temperature while atmospheric greenhouse gases are reduced as rapidly as practical. With that caveat, numerous studies, e.g., Footnote^136,Footnote¹³⁷ suggest that stratospheric aerosols have potential to reduce the risks of Antarctic ice loss and sea level rise. However, it must be borne in mind that the greatest uncertainty is in ice sheet response to changing climate. Ice sheet modeling is still so primitive that it is difficult to have confidence in these modeling studies, per se.

Modeling limitations are why we suggest comparable emphasis on paleoclimate studies, climate modeling, and modern observations of ongoing changes. In the latter category, there is the global, natural experiment of cooling by stratospheric aerosols provided by the 1991 Pinatubo volcanic eruption, which spread aerosols into both hemispheres. The maximum negative forcing was about −3 W/m², more than enough to offset Earth’s present energy imbalance of 1-1.5 W/m² and cause global cooling. Such negative forcing, if maintained for years, would cause reversal of fast feedbacks, including regrowth of sea ice area. Major effects of the brief Pinatubo forcing included global cooling in the next two years that peaked at 0.3 °C and a 50% reductionFootnote¹³⁸ in the growth rate of atmospheric CO₂ that lasted about three years. Negative effects included a temporary reduction of stratospheric ozoneFootnote¹³⁹ in the tropics and adverse changes of precipitation patterns.Footnote¹⁴⁰

Tropospheric aerosols are a suggested alternative cooling mechanism.Footnote¹⁴¹ The inadvertent global experiment arising from the sudden restriction on sulfur content of ship fuels is analogous to the Pinatubo experiment. Based on our analysis, this ship experiment indicates the potential for a large cooling effect via tropospheric aerosols. The environmental impact of spraying salty sea water into the air with the intention of seeding clouds may generate less concern than some other cooling mechanisms, but much more scientific and engineering research is needed to explore the topic.Footnote¹⁴²

Investigations of purposeful global cooling occasionally raise a concern of a potential threat it might pose to ambition to reduce emissions. This hypothesis is often called moral hazard. This concern is largely contested in research on individuals,Footnote^143,Footnote^144,Footnote¹⁴⁵ but we take it seriously. Importantly, whether moral hazard plays out should depend on how SRM is framed, e.g. as a panacea or get-out-of-jail card vs. a complementary mea­sure. SRM must be presented as an auxiliary tool that could help reverse some of the damage already set in motion by the fossil fuel industry and irresponsible politics. The environmental movement and academia have a huge responsibility in steering public debate on SRM, which they have largely shunned to date.

However, even in the worst case, if SRM would in some degree distract from GHG cuts, it still may be a risk worth taking, given the limited potential that greenhouse gas reductions alone now have for avoiding some catastrophic climate impacts. If, as for us, the main concern is with limiting climate disasters and subsequent human suffering, then the mere possibility of moral hazard is not per se a strong/valid reason for rejecting SRM research.

The main ethical issues here are (1) whether the climate impacts that are already unavoidable even with the most stringent emission scenario (not to mention with worse and more likely scenarios) are acceptable for those who are doomed to bear them; (2) whether SRM might help avoid some of these impacts, and how this benefit would compare to the possible side effects of the given SRM; and (3) how much SRM may distract from the main challenge of greenhouse gas reduction. There is a need for analysis that compares the risks and benefits of purposeful global cooling scenarios against scenarios with no such cooling. This comparative risk analysis is typically absent in objections to SRM research; in a similar vein, proponents of SRM research should appreciate valid concerns about the moral hazard hypothesis and deal with it in a comparative “risks vs. risks” framework. Although public perception research is nascent, it must be noted that it shows stronger support for SRM in the Global South than in the Global North, probably because of younger average age and greater exposure to climate hazards.Footnote^146,Footnote^147,Footnote¹⁴⁸

There is no expectation of purposeful global cooling in the near-term. For now, what is needed is a strengthened resolve to transition away from fossil fuels and adequate funding to assist nations presently suffering climate disasters. Countries most responsible for climate change will be expected to provide funding. Recognition of a growing obligation may encourage phasedown of relevant emissions. These issues are complex, but now unavoidable.

In any event, based on the discussion in this article, we believe it is likely that purposeful global cooling would be more helpful than not for limiting disastrous climate impacts. However, international agreement on such actions is undesirable until the risks and benefits of SRM are better established, and unlikely before there is better understanding of the science as well as evidence of extreme, undeniable, climate change that persuades the public of the common sense and desirability of action. That requires time, probably decades. Thus, it is important to be aware of likely near-term climate change, and to have the data needed to interpret the climate change.

The Next Decade or Two

Are the public and United Nations well-informed? Not if judged by assertions that global warming can be kept “well below 2 °C,” the goal of the Paris Agreement,Footnote¹⁴⁹ without purposeful global cooling (in addition to phasedown of greenhouse gas emissions). Intergovernmental Panel on Climate Change (IPCC) scenarios that achieve that target, such as RCP2.6 in , are implausible. We also conclude that IPCC underestimated cooling by human-made aerosols, and, largely as a result of that, IPCC’s best estimate of climate sensitivity (3 °C for doubled CO₂) is also an underestimate. More realistic assessment of the climate situation will be needed during the next decade or two, if the world is to finally comes to grips with climate change reality.

Epilogue

Young people feel anxiety about climate change and their future. A surveyFootnote¹⁵² of 10,000 16-to-25-year-olds in ten nations found that 60% were “very worried” or “extremely worried.” Two-thirds of them felt that governments are failing them, and, specifically, that governments are not acting according to science. Are they on to something? How could they get that impression? They see shootings in their schools. They see growing wars in the world. They see climate changing. In all cases, they see innocent people suffering with ineffectual government response. Yet they have faith in science. They ask: what is the truth? They do not want a sugar-coated answer: “Oh, don’t worry, you will all be wealthy soon, so you can take care of the problems.” They see that their parents are struggling, not becoming wealthy. Instead, governments borrow money from young people, leaving them with the obligation to pay off the debt. Yet young people want to work for a bright future and they ask of science: what is the big picture, the long story?

Failure to be realistic in climate assessment and failure to call out the fecklessness of current policies to stem global warming is not helpful to young people. The life of the first author (JEH) covers the period in which policy constraints developed. With the permission of the coauthors, the rest of this epilogue describes his perception of why policies do not serve the best interests of the public. The developments refer to the United States, but they are relevant to many nations.

The United States was ill-prepared for war when it entered World War II in 1941, but before the war was over the country had built a powerful, successful, military. The nation used its influence to help establish the United Nations and rules-based international bodies that promoted free trade and raised living standards in much of the world. The U.S. maintained a strong military, given the perceived threat of the Soviet Union, but President Dwight Eisenhower, in his 1961 Farewell Address, warned of danger in “the military-industrial complex.” The draft of that speechFootnote¹⁵³ – with input from his brother, Milton, then President of Johns Hopkins University – referred to the military-industrial-congressional complex, but the President deleted “congressional” before delivering his address on national television. When Milton asked about the omission, Eisenhower explained “It was more than enough to take on the military and private industry. I couldn’t take on the Congress as well.” The public wishes he had. The public knows that Washington is a swamp of special interests, with huge negative effect on the public’s best interest.

John F. Kennedy, in campaigning for the Presidency of the U.S. in 1960, received rousing support on university campuses when he proposed a Peace Corps to promote world friendship, and he gave the Peace Corps high priority when he assumed office in 1961. Kennedy was promptly introduced to the “deep state” when he let the Central Intelligence Agency orchestrate an invasion of Cuba by Cuban exiles, which ended in fiasco at the Bay of Pigs; Kennedy was angry at the CIA, but blamed himself for accepting their plan. In 1963 President Kennedy gave a surpassing “Peace Speech”Footnote¹⁵⁴ that led to a nuclear test ban treaty with the Soviet Union, and, shortly before his assassination in November 1963, decided on a specific plan to withdraw from Vietnam regardless of the military situation there.Footnote¹⁵⁵ If Kennedy had served eight years instead of 2 years and 10 months, perhaps America would have followed a different path, but, instead, the United States now has a military presence in almost too many nations to count.Footnote¹⁵⁶

The path followed by the United States after JFK’s assassination was not chosen by the American people, who, in fact, have a distaste for meddling in the internal political affairs of other nations. As a NASA post-doc in 1967 and 1968, I worked in a Columbia University building a short distance from where students protested the Vietnam war and Columbia involvement in war-related research. Most Americans accept the need for a strong military, but not the continuous pursuit of global military hegemony with interference in the internal affairs of other nations. The origin and continuation of a militaristic approach with frequent, often secret, support of “regime changes” in nations deemed unfriendly to our interests – as opposed to Kennedy’s greater emphasis on being “as a city upon a hill,”Footnote¹⁵⁷ a positive example with the eyes of all people on us – is important for understanding why effective actions to preserve climate are not being taken and how this can be changed.

Now let’s summarize emergence of climate change science and the world’s political response.

Charles David KeelingFootnote¹⁵⁸ initiated precise measurements of atmospheric CO₂ in 1958, confirming that humanity was changing our atmospheric composition. During the 1960s and 1970s concern grew about possible impacts on climate, culminating in the 1979 Charney report²³ that concluded climate sensitivity was likely in the range 1.5-4.5 °C for doubled atmospheric CO₂, thus implying large potential climate change. Paleoclimate data supported high climate sensitivity, favoring a sensitivity in the upper half of Charney’s range.Footnote¹⁵⁹ Observed, ongoing, global warming added to concerns, leading to adoption of the UN Framework Conven­tion on Climate Change⁴ at the Rio Earth Summit in 1992 and the Kyoto ProtocolFootnote¹⁶⁰ in 1997, with the objective of limiting changes of atmospheric composition, so as to avoid dangerous human-made interference with climate.

Subsequently, every year for three decades, the nations of the world have gathered for the annual Conference of the Parties (COP), duly noting the growing climate threat. Nations duly promise to take action to reduce their emissions, and each year (barring a pandemic or global recession) global emissions actually grow (). Why? Fossil fuels are a marvelous, condensed energy source that raises living standards. As long as their waste can be dumped into the global commons, the atmosphere, without paying a fee for the cost to society, they will continue to be used and the climate problem will remain unsolvable. There are still plenty of fossil fuels in the ground. Individuals and nations will not readily give up the benefits that fossil fuels can bestow.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

28. Estimates for equilibrium climate sensitivity (ECS) of 3 °C or less for doubled CO₂ were common for decades not only because GCMs with simple cloud schemes yielded such sensitivities, but because the main paleoclimate study of climate sensitivity supported low sensitivity. The large CLIMAP project (CLIMAP project members, “Seasonal reconstruction of the Earth’s surface at the last glacial maximum,” Geol Soc Amer, Map and Chart Series, No. 36, 1981), which reconstructed Earth’s surface conditions during the last ice age, found SSTs not much colder than today. SSTs were based on an assumption that tiny shelled marine species would migrate to stay in the temperature zone where they live today. However, if, instead, species partly adapt to changing temperature over millennia, a larger SST change would be inferred. Recent studies (M. B. Osman et al., “Globally resolved surface temperatures since the Last Glacial Maximum, Nature 599 (2021): 239-44) that exclude any assumption about migrating species, instead relying on chemical proxies for temperature change, yield ice age cooling of 6-7 °C, implying ECS of 4.8 °C ± 1.2 °C for doubled CO₂. Narrowing this range requires improved evaluation of the “efficacy” of ice sheet climate forcing, which requires realistic simulation of clouds in GCMs.

IPCC’s AR6 climate analysis, unlike earlier IPCC reports, makes good use of paleoclimate data, especially the climate sensitivity study of S.C. Sherwood et al., “An assessment of Earth’s climate sensitivity using multiple lines of evidence,” Rev Geophys 58 (2020): e2019RG000678. That study is exceptionally comprehensive, but its estimate of ECS (2.2-4.9 °C, 95% probability) is marred by an estimate that glacial-interglacial temperature change (the only paleo case with accurate greenhouse gas amounts) was only 5 °C and by an assumption that paleo aerosol changes are a climate forcing. Natural aerosol changes are a climate feedback, like cloud changes; indeed, aerosols and clouds form a continuum and distinction is arbitrary as humidity nears 100 percent. There are many aerosol types, including VOCs (volatile organic compounds) produced by trees, sea salt produced by wind and waves, black and organic carbon produced by forest and grass fires, dust produced by wind and drought, and marine biogenic dimethyl sulfide and its secondary aerosol products, all varying geographically and in response to climate change. We do not know, or need to know, paleo aerosol changes because those changes are feedbacks included in the climate response. The choice of Sherwood et al.to count estimated paleo aerosol changes as a forcing caused an underestimate of ECS.

57 P.M. Forster et al., Supplement of Indicators of Global Climate Change 2023: annual update of key

indicators of the state of the climate system and human influence Earth Syst. Sci. Data 16 (2024): 2625–58 <a href="https://doi.org/10.5194/essd-16-2625-2024-supplement" rel="nofollow">https://doi.org/10.5194/essd-16-2625-2024-supplement</a>.

75 Let T_C(t) be global temperature change calculated by a GCM at time t after an instant CO₂ doubling, i.e., one of the curves in . T_C(t) can be used to calculate the expected temperature for any climate forcing scenario. For example, assume that climate was in equilibrium in 1850. The temperature in 1851 is the product of the forcing added in 1851 (expressed as a fraction of 2 × CO₂ forcing) x T_C (year 1). Global temperature in 1852 is the sum of two terms, the first term being the forcing added in 1851 × T_C (year 2) and the second term being the forcing added in 1852 × T_C (year 1) – and so on for successive years. An equation for this is .

T_G is our “Green’s function” estimate of global temperature and dF_e is the forcing change per unit time divided by the doubled CO₂ forcing of 4 W/m². Integration begins when Earth is in near energy balance, e.g., in preindustrial time. The 5000-year run of the GISS (2020) model used in the Pipeline paper¹ for was a bit of an outlier for T_C(t) in the first year, e.g., Earth’s energy imbalance (EEI), which was initially 4 W/m², decreased rapidly to 2.7 W/m² averaged over year 1. For our present paper, we made 5 more 2 × CO₂ runs and used the ensemble-mean to define a smooth T_C(t). This “ultrafast” response is still present in the ensemble mean, but for year 1 the ensemble average EEI is 3.0 W/m². Also, the ensemble-mean warming in years 10-50 is less than the average warming in those years in the 5000-year run, as the GISS (2020) single model runs had multi-decadal variability that is believed to be unrealistic. Our estimate for 4.5 °C Global Temperature Response to 2 × CO₂ is obtained by multiplying the 3.4 °C Global Temperature Response by a scale factor that allows the 4.5 and 3.4 responses to begin to increase similarly at time t = 0, but diverge on a decadal time scale toward their equilibrium responses. The scale factor is S(t) = S_f – exp[−(t − 1)/r] × (S_f − 1), where S_f = 4.5/3.4 and r = 13 years.

114 Hefner M, Marland G, Boden T et al. Global, Regional, and National Fossil-Fuel CO₂ Emissions, Research Institute for Environment, Energy, and Economics, Appalachian State University, Boone, NC, USA. https://energy.appstate.edu/cdiac-appstate/data-products (20 August 2023, date last accessed).