Climate Sensitivity is Influenced by Surface Warming Patterns
One of the main difficulties in climate science is to decrease the amount of uncertainty in determining climate sensitivity. Climate sensitivity measures the amount of warming on Earth’s surface when carbon dioxide levels double compared to preindustrial levels. This uncertainty is significant because it combines many different processes, ranging from small-scale aerosol-cloud interactions to large-scale atmospheric and ocean circulations, into a single value. The main contributor to this uncertainty is the challenge of accurately measuring and simulating clouds.
Multiple pieces of data are utilized to approximate climate sensitivity, such as simulations by climate models with different levels of complexity, observations from the previous century, indicators that track changes in climate from distant history, and scientific theory. The potential range of estimates for climate sensitivity has remained consistently uncertain at 1.5–4.5 K for many years, but recent efforts from the research community have begun to narrow this range (see Figure 1).
In this figure, you can see the best estimates of climate sensitivity, which is the range of likely values at 66% confidence, from various assessments that have been conducted using multiple studies and evidence. The minimum to maximum range shown by climate models is represented by thin bars. The data used in this figure were obtained from studies by Charney et al. (1979), Houghton et al. (1990, 1995, 2001), Masson-Delmotte et al. (2021), Meehl et al. (2020), Sherwood et al. (2020), Solomon et al. (2007), Stocker et al. (2013), and Zelinka et al. (2020). The abbreviations CMIP and IPCC AR stand for Coupled Model Intercomparison Project and Intergovernmental Panel on Climate Change Assessment Report, respectively.
In the beginning of the 2010s, a significant difference was found between climate sensitivity estimates from climate models and those based on observed warming and radiative balance. The latter refers to the balance between incoming solar radiation and reflected radiation, as well as outgoing terrestrial radiation. Estimates based on observed warming indicated much lower values compared to those from models. The discovery of the pattern effect, which links climate sensitivity to the geographic pattern of surface warming, has been a crucial development in resolving this puzzle. This advancement has been recognized as one of the most promising ways to further narrow down estimates of climate sensitivity in the future [Forster et al., 2021].
“Manipulation, Response Loop, and Optimal Climate Conditions”
Increasing the amount of greenhouse gases in the Earth’s atmosphere results in an excess of energy (less energy is able to escape into space), known as forcing. In order to regain a balance of energy, the planet must warm. However, this warming causes alterations in the climate system: The level of water vapor, a greenhouse gas, rises in the atmosphere, the area covered by reflective snow and sea ice decreases, and the properties of clouds change. These and other effects related to energy absorption can either increase or decrease the amount that the planet warms in response to the initial forcing. Therefore, the feedbacks play a crucial role in determining the sensitivity of the climate to a given forcing.
If the increase in global temperature is distributed evenly, it will have a different effect on radiation than if it is concentrated in an optimal climate zone.
Researchers previously believed that the impact of global temperature on radiative feedback was solely based on the overall global temperature [Gregory et al., 2004]. However, it has been discovered that the spatial distribution of surface warming also plays a significant role. Just as the effect of applying pressure to someone’s entire body differs from tickling their feet, a uniform increase in global temperature will have a different radiative response compared to concentrated warming in a specific area with high radiative damping capabilities.
There are many factors that impact changes in surface temperature, such as greenhouse gas levels, regional aerosol patterns, and natural oscillations involving the ocean and atmosphere. Other influences include continental boundary conditions and the presence of ice sheets and sea ice. Over the last 40 years, the changes in surface temperature have shown a clear spatial pattern, with certain areas even experiencing cooling despite an overall global warming trend of approximately 1 K (see Figure 2, bottom left).
The left side of Figure 2 shows the warming pattern observed in the tropical Pacific over the last few decades. This warming is strongest in regions with deep convection, such as the western Pacific, and spreads throughout the troposphere, causing significant warming in the upper atmosphere. This increased warmth also contributes to a stronger release of heat into space. However, there are also areas in the eastern Pacific that have experienced cooling. This cooling in the upper atmosphere and warming at the surface creates a strong inversion layer in the lower troposphere and promotes the formation of shallow cloud cover, which reflects solar radiation and helps keep the Earth cooler. On the right side of the figure, a typical representation of climate change from coupled climate models is shown. These models create their own warming pattern, which differs from the observed one. In these simulations, there is less warming in the upper atmosphere, a decrease in the strength of the lower tropospheric inversion, and a reduction in cloud cover over time. As a result, these models are less effective at cooling the Earth’s surface compared to what is observed in nature.
The response to clouds and atmospheric temperature is most influenced by variations in warming across space. In the warmest tropical areas, deep convection easily transmits surface conditions upwards throughout the troposphere (up to approximately 10-15 kilometers) and then horizontally across much of the world, making the western Pacific a favorable location for climate conditions. The warmer air above the comparatively cooler waters in the eastern Pacific or Southern Ocean serves to stabilize the lowermost portion of the troposphere, which allows for the development of widespread low-lying clouds such as stratus and stratocumulus. These low clouds are situated in such a way that they effectively cool the planet and counteract some of the initial warming effects (see Figure 2, top left).
Looking at the Issue from Various Perspectives
In recent years, three areas of study have come together to emphasize the importance of including input from all climate research fields in order to fully comprehend the effects of patterns.
In the past, three different areas of study have demonstrated the link between radiative feedbacks and the pattern of surface warming. The first area involved analyzing climate feedbacks and sensitivity in simulations of transient climate change. Using a basic energy balance model that extrapolated the relationship between global temperature change and radiative imbalance, researchers could estimate the equilibrium temperature change (climate sensitivity) assuming feedbacks remained constant [e.g., Gregory et al., 2004]. However, when longer, fully equilibrated simulations were conducted, it became clear that this simple approach underestimated the actual equilibrium climate sensitivity. This is due to the changing surface warming pattern, which initially emphasizes more stabilizing radiative feedbacks but later, during equilibration, highlights less stabilizing radiative feedbacks [e.g., Senior and Mitchell, 2000; Rugenstein et al., 2020].
The second part of the study connected the concept of continuous feedback to the attempts of determining equilibrium climate sensitivity using past data. This was mentioned earlier. When feedback is calculated based on observations or simulations of only the atmosphere, using the observed warming patterns from the last few decades, it suggests a lower level of warming compared to simulations that include an interactive ocean. These simulations have the ability to generate their own surface warming patterns [for example, Gregory et al., 2020].
The third area of study was inspired by oceanography and revealed that the cooling impact of ocean heat absorption varies depending on its location: One unit of heat absorbed in higher latitudes has a greater cooling effect on Earth compared to the same unit absorbed in low-latitude oceans. This discrepancy is significant because the majority of heat absorption by the ocean happens at higher latitudes. This phenomenon, known as ocean heat absorption efficacy, is another example of how radiative feedbacks are affected by surface temperature patterns [Winton et al., 2010; Lin et al., 2021].
Over the past few years, the three areas of research have come together to show that comprehending the impact of patterns is enhanced by, and maybe even dependent on, the input from all climate researchers studying how the ocean and atmosphere interact on a large scale and the processes that determine how regions and the entire globe respond to external influences.
In order to promote collaboration between different communities, a group of 140 scientists from various disciplines and countries came together for a US CLIVAR workshop in May 2022 in Boulder, Colorado. We will now discuss the conclusions reached during this workshop and highlight important areas that require immediate attention from the scientific community.
The Past Is a Poor Analogue for the Future
Studies on the pattern effect have highlighted restrictions in using equilibrium climate sensitivity as a means of predicting future warming. Similarly, using current warming trends to determine equilibrium climate sensitivity also has its limitations. The main takeaway from the pattern effect is that past observations of climate change do not provide a definitive limit for climate sensitivity [Sherwood et al., 2020; Forster et al., 2021].
The reason for the poor predictive power of the climate record from past decades is that the observed surface warming pattern caused feedbacks that were more stabilizing than the ones projected for the future. The warming was particularly pronounced in the sweet spots of the western Pacific and the subtropical eastern Pacific—a perfect combination for enhanced radiative damping (Figure 2, left). In particular, low-lying cumulus clouds in the eastern Pacific covered a larger area and reflected more sunlight back to space than studies suggest they will in the future.
The challenge in measuring the pattern effect arises from the uncertainties in surface temperature trends, ocean heat absorption, and the limited length of time that Earth’s energy balance has been observed.
The discussion and size of the observed pattern effect are still being discussed and uncertain. The current most accurate estimate suggests that the radiative feedback or damping from long-term carbon dioxide forcing is 0.5 ± 0.5 watt per square meter per kelvin weaker compared to the radiative feedback seen since the late 1800s [Andrews et al., 2022]. This value is similar in magnitude to the individual radiative feedbacks that are affected by the pattern effect. The wide range of uncertainty allows for two contrasting possibilities: The pattern effect may have significantly slowed down global warming – meaning that the average global temperature would be much higher by now if the recent warming had occurred in a different spatial pattern. This suggests that the pattern effect has the ability to strongly influence future warming rates as the surface warming pattern changes. On the other hand, the other end of the uncertainty range suggests that the pattern effect may have had little impact over the past century and may only be relevant over shorter periods of time or in drastically different climate conditions.
It is challenging to accurately measure the impact of patterns due to uncertainties in surface temperature trends, ocean heat absorption, and the limited data on Earth’s energy budget. Our incomplete understanding of how clouds respond to their surroundings and the limitations of climate models also hinder our ability to fully comprehend the extent of the pattern effect.
Key questions in the study of climate dynamics revolve around the origins of surface temperature patterns, the relationship between Earth’s radiation budget and these patterns, and the interdependence between the two. This requires a deeper understanding of factors driving sea surface temperature patterns, such as decadal coupled variability, ocean heat uptake rates, and the impact of aerosols, greenhouse gases, and volcanoes. A crucial objective is to accurately explain past patterns and make predictions about their future evolution. Additionally, it is necessary to determine how changes in surface temperature affect local and distant top-of-the-atmosphere radiative fluxes in terms of magnitude, spatial scale, and direction (positive or negative). Finally, the connection between ocean heat uptake and radiative feedback must be examined on a global and regional level over different time periods.
Cautious consideration is necessary when interpreting paleorecords.
The way in which past changes in surface warming and the planet’s radiation balance impact the future of Earth is limited. Similarly, analyses of paleorecords for both recent and ancient time periods must consider the pattern effect in order to accurately apply them to future climate projections. For instance, the cooling pattern observed during the Last Glacial Maximum (LGM) differed from current warming patterns and those expected for the next few decades, as well as the equilibrium following a doubling of carbon dioxide. This difference not only includes the direction of change, but also the spatial distribution of magnitudes. The LGM is currently the most useful constraint for determining the upper limit of climate sensitivity [Sherwood et al., 2020], making the details of the pattern effect crucial.
Outstanding questions concern how representative pattern changes in the past century, the past millennium, or quasi-equilibrated times millions of years ago are of expected future changes.
We have gained knowledge about the pattern effect, and we have also discovered that radiative feedbacks vary depending on the global mean temperature. For instance, a 1 K increase from the Last Glacial Maximum (LGM) results in different feedbacks (such as the sea ice albedo feedback) compared to a 1 K increase from a Miocene hothouse world, or a 4 to 5 K increase in a high-emission scenario within a century or two from now (such as the water vapor feedback [Bloch-Johnson et al., 2021]). These uncertainties from the pattern effect and feedback temperature dependence make it challenging to accurately estimate climate sensitivity based on paleorecords. However, understanding and quantifying these effects can help improve the relevance of these records in predicting climate sensitivity and future warming.
Outstanding questions concern how representative pattern changes in the past century, the past millennium, or quasi-equilibrated times millions of years ago are of expected future changes. The paleorecord could further be crucial for understanding the timescales and relative importance of internal variability, forced response, and extratropical forcing to rates of warming or cooling in the equatorial Pacific in nature and in climate models.
Can we rely on model patterns in predicting climate change scenarios?
Can climate models accurately replicate observed patterns of warming? This question arises from our comprehension of the pattern effect. Coupled climate models have their own expression of internal variability due to a freely evolving ocean surface. Therefore, it is unlikely that a coupled model would simulate El Niño events at the exact same time as they occur in nature, unless by coincidence. This makes it challenging to compare models and observations, particularly on longer timescales where our understanding of the spectrum of internal variability in the real world is limited.
Unfortunately, coupled climate models seem unable to simulate observed surface warming patterns across some key regions in which surface sensitivity modulates clouds, even when they account for internal variability. Most important, models do not re-create observed cooling in the equatorial and subtropical eastern Pacific and Southern Ocean (Figure 2, right), and it is concerning that all models have the same sign of error in the trend patterns [e.g., Wills et al., 2022]. In addition, models strongly disagree about the timescales over which future warming may occur in these regions.
The development of worldwide integrated models that can accurately simulate storms and clouds, although limited to a few months or years, offers a promising prospect.
Our current understanding of the factors that cause inaccuracies in climate models’ surface warming patterns limits our ability to assess these inaccuracies. It is unclear whether these models’ failure to replicate observed trends is a result of inaccuracies in how they account for internal variability, aerosol forcing, interactions between the ocean and atmosphere, atmospheric convection at different levels, radiative properties of clouds over the Southern Ocean, depth of the ocean’s thermocline, and ocean circulation, or if there are other contributing factors.
One of the most important concerns is whether predictions made by climate models will continue to deviate from actual observations in the future, as they have in relation to recent conditions (see Figure 2, left and right). If this does occur, how will these model discrepancies impact radiative feedbacks and global warming rates? It is necessary to measure whether coupled models attempt to offset their inaccuracies in surface temperature trends and radiative feedbacks by overcompensating for ocean heat absorption or aerosol influences.
The emergence of global coupled models that can accurately simulate storm and cloud behavior is a promising development, even though they can currently only simulate short time periods. How can we effectively evaluate surface warming patterns and radiative feedback in these simulations and compare them to observations and lower-resolution models, considering that even decades of data may not be enough to confidently identify forced trends in certain regions? And for those relying on the climate change projections of these models, how much do the remaining carbon budgets [Zhou et al., 2021] depend on the pattern effect in regards to global warming targets, near-term warming rates, and efforts to detect and attribute changes?
A Collaborative Outlook
Collaborating across disciplines and incorporating ideas from various research communities can lead to solutions for the aforementioned issues. This could also have a positive impact on related research topics such as the impact of surface warming patterns on tropical cyclone intensity and future rainfall patterns in the southwestern United States and South America.
New and promising methods and findings are starting to surface. These include focused experiments and comparisons, expanded satellite observations of clouds and radiation around the world, usage of Argo floats to measure heat absorption in the deep ocean, estimates of radiative imbalances before 2000, and new observations on how clouds respond to environmental factors. These advancements give hope for our capability to accurately measure the impact of patterns and its consequences in the coming years.
Acknowledgments
The authors express their gratitude to all participants of the US CLIVAR workshop on pattern effects. M.Z.’s research was funded by the U.S. Department of Energy (DOE) Regional and Global Model Analysis program and conducted at Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. P.C. received support from the UK Natural Environment Research Council through grants NE/T006250/1 and NE/V012045/1. T.A.’s work was supported by the Met Office Hadley Centre Climate Programme, funded by the Department of Business, Energy and Industrial Strategy, and also received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement 820829. M.R.’s research was funded by NASA under grant 80NSSC21K1042.
References
Andrews and colleagues (2022) examined the impact of past sea surface temperature patterns on radiative feedback in their study published in the Journal of Geophysical Research: Atmospheres. The paper, with the reference e2022JD036675, is available online at https://doi.org/10.1029/2022JD036675.
2 concentrations
According to Bloch-Johnson and colleagues (2021), there is an increase in climate sensitivity with higher levels of CO2 in the atmosphere.2
The varying levels due to changes in temperature have been studied in the article “Feedback Temperature Dependence” published in Geophysical Research Letters. The article can be found at https://doi.org/10.1029/2020GL089074 with the reference number e2020GL089074.
Forster, P., et al. (2021), The Earth’s energy budget, climate feedbacks, and climate sensitivity, in Climate Change 2021: The Physical Science Basis. Working Group I Contribution to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, edited by V. Masson-Delmotte et al., pp. 923–1,054, Cambridge Univ. Press, New York, https://doi.org/10.1017/9781009157896.009.
Gregory, J. M., et al. (2004), A new method for diagnosing radiative forcing and climate sensitivity, Geophys. Res. Lett., 31, L03205, https://doi.org/10.1029/2003GL018747.
2 be estimated from historical climate change?
Can we accurately estimate the climate sensitivity to CO2 based on past climate changes? – Gregory, J. M., et al. (2020)2
Can the effects of past climate change be determined? This was explored in a study published in Climatic Dynamics, with a reference of “54, 129-157” and a DOI of “https://doi.org/10.1007/s00382-019-04991-y”.
In a recent study by Lin et al. (2021), it was found that Southern Ocean heat absorption plays a significant role in the changing radiative feedback in CESM. This study was published in Geophysical Research Letters and can be accessed at https://doi.org/10.1029/2021GL093302.
Rugenstein, M., et al. (2020), Equilibrium climate sensitivity estimated by equilibrating climate models, Geophys. Res. Lett., 47, e2019GL083898, https://doi.org/10.1029/2019GL083898.
C. A. Senior and J. F. B. Mitchell (2000) studied how climate sensitivity changes over time in their article “The time-dependence of climate sensitivity” published in Geophysical Research Letters. The study found that climate sensitivity varies between 2,685 and 2,688 in different periods, with a DOI of 10.1029/2000GL011373.
In their recent study, Sherwood et al. (2020) evaluated Earth’s climate sensitivity by utilizing various pieces of evidence. The results were published in Reviews of Geophysics and can be found at https://doi.org/10.1029/2019RG000678.
The authors Winton, Takahashi, and Held (2010) discuss the significance of ocean heat uptake efficiency in relation to short-term climate change. Their findings are published in the Journal of Climate and can be accessed through the following link: https://doi.org/10.1175/2009JCLI3139.1.
Wills, R. C. J., et al. (2022), Systematic climate model biases in the large-scale patterns of recent sea-surface temperature and sea-level pressure change, Geophys. Res. Lett., 49, e2022GL100011, https://doi.org/10.1029/2022GL100011.
Zhou, C., et al. (2021), Greater committed warming after accounting for the pattern effect, Nat. Clim. Change, 11, 132–136, https://doi.org/10.1038/s41558-020-00955-x.
Author Information
Maria Rugenstein of Colorado State University in Fort Collins, Mark Zelinka of Lawrence Livermore National Laboratory in Livermore, California, Kristopher B. Karnauskas of University of Colorado Boulder, Paulo Ceppi of Imperial College London, and Timothy Andrews of Met Office Hadley Centre in Exeter, United Kingdom.
Reference: Rugenstein, M., Zelinka, M., Karnauskas, K. B., Ceppi, P., & Andrews, T. (2023). The importance of surface warming patterns in determining climate sensitivity. Eos, 104. Retrieved from https://doi.org/10.1029/2023EO230411 on October 31, 2023.
Text © 2023. The authors. CC BY-NC-ND 3.0
Images are protected by copyright unless otherwise stated. Any unauthorized use is strictly prohibited without the explicit consent of the copyright holder.
