Climate Sensitivity Uncertainty

October 3, 2014 7:20 pm3 comments

In a press conference ahead of the UN Climate Summit two week ago, UN Secretary-General Ban-Ki Moon stated: “Action on climate change is urgent.  The more we delay, the more we will pay in lives and in money.” The recently appointed UN Messenger of Peace Leonardo DiCaprio stated “The debate is over. Climate change is happening now.”

These statements reflect a misunderstanding of the state of climate science and the extent to which we can blame adverse consequences such as extreme weather events on human caused climate change. The climate has always changed and will continue to change. Humans are adding carbon dioxide to the atmosphere, and carbon dioxide and other greenhouse gases have a warming effect on the climate. However, there is enduring uncertainty beyond these basic issues, and the most consequential aspects of climate science are the subject of vigorous scientific debate: whether the warming since 1950 has been dominated by human causes, and how the climate will evolve in the 21st century due to both natural and human causes. Societal uncertainties further cloud the issues as to whether warming is ‘dangerous’ and whether we can afford to radically reduce carbon dioxide emissions.

Dr. Judith Curry

Dr. Judith Curry

At the heart of the recent scientific debate on climate change is the ‘pause’ or ‘hiatus’ in global warming – the period since 1998 during which global average surface temperatures have not increased. This observed warming hiatus contrasts with the expectation from the 2007 IPCC Fourth Assessment Report that warming would proceed at a rate of 0.2oC/per decade in the early decades of the 21st century. The warming hiatus raises serious questions as to whether the climate model projections of 21st century have much utility for decision making, given uncertainties in climate sensitivity to carbon dioxide, future volcanic eruptions and solar activity, and the multidecadal and century scale oscillations in ocean circulation patterns.

A key argument in favor of emission reductions is concern over the accelerating cost of weather disasters. The accelerating cost is associated with increasing population and wealth in vulnerable regions, and not with any increase in extreme weather events, let alone any increase that can be attributed to human caused climate change. The IPCC Special Report on Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation found little evidence that supports an increase in extreme weather events that can be attributed to humans. There seems to be a collective ‘weather amnesia’, where the more extreme weather of the 1930’s and 1950’s seems to have been forgotten.

Climate science is no more ‘settled’ than anthropogenic global warming is a ‘hoax’. I am concerned that the climate change problem and its solution have been vastly oversimplified. Deep uncertainty beyond the basics is endemic to the climate change problem, which is arguably characterized as a ‘wicked mess.’ A ‘wicked’ problem is complex with dimensions that are difficult to define and changing with time. A ‘mess’ is characterized by the complexity of interrelated issues, with suboptimal solutions that create additional problems.

Nevertheless, the premise of dangerous anthropogenic climate change is the foundation for a far-reaching plan to reduce greenhouse gas emissions. Elements of this plan may be argued as important for associated energy policy reasons, economics, and/or public health and safety. However, claiming an overwhelming scientific justification for the plan based upon anthropogenic global warming does a disservice both to climate science and to the policy process. Science doesn’t dictate to society what choices to make, but science can assess which policies won’t work and can provide information about uncertainty that is critical for the decision making process.

Can we make good decisions under conditions of deep uncertainty about climate change? Uncertainty in itself is not a reason for inaction. Research to develop low-emission energy technologies and energy efficiency measures are examples of ‘robust’ policies that have little downside, while at the same time have ancillary benefits beyond reducing greenhouse gas emissions. However, attempts to modify the climate through reducing CO2 emissions may turn out to be futile. The hiatus in warming observed over the past 16 years demonstrates that CO2 is not a control knob on climate variability on decadal time scales. Even if CO2 mitigation strategies are successful and climate model projections are correct, an impact on the climate would not be expected until the latter part of this century. Solar variability, volcanic eruptions and long-term ocean oscillations will continue to be sources of unpredictable climate surprises.

Whether or not anthropogenic climate change is exacerbating extreme weather events, vulnerability to extreme weather events will continue owing to increasing population and wealth in vulnerable regions. Climate change (regardless of whether the primary cause is natural or anthropogenic) may be less important in driving vulnerability in most regions than increasing population, land use practices, and ecosystem degradation. Regions that find solutions to current problems of climate variability and extreme weather events and address challenges associated with an increasing population are likely to be well prepared to cope with any additional stresses from climate change.

Oversimplification, claiming ‘settled science’ and ignoring uncertainties not only undercuts the political process and dialogue necessary for real solutions in a highly complex world, but acts to retards scientific progress. It’s time to recognize the complexity and wicked nature of the climate problem, so that we can have a more meaningful dialogue on how to address the complex challenges of climate variability and change.

IPCC AR5 lowers the bottom of ‘likely’ climate sensitivity range

The sensitivity of our climate to increasing concentrations of carbon dioxide is at the heart of the scientific debate on anthropogenic climate change, and also the public debate on the appropriate policy response to increasing carbon dioxide in the atmosphere. Climate sensitivity and estimates of its uncertainty is a key input into the economic models that drive cost-benefit analyses and estimates of the social cost of carbon.

One of the most significant outcomes of the recent IPCC Fifth Assessment Report is the change in conclusions relative to the Fourth Assessment Report. The equilibrium climate sensitivity (ECS) is defined as the change in global mean surface temperature at equilibrium that is caused by a doubling of the atmospheric CO2 concentration. The IPCC AR4 conclusion on climate sensitivity is stated as:

“The equilibrium climate sensitivity. . . is likely to be in the range 2.oC to 4.5C with a best estimate of about 3.oC and is very unlikely to be less than 1.5C. Values higher than 4.5oC cannot be excluded.” (AR4 SPM)

The IPCC AR5 conclusion on climate sensitivity is stated as:

Equilibrium climate sensitivity is likely in the range 1.5°C to 4.5°C (high confidence), extremely unlikely less than 1°C (high confidence), and very unlikely greater than 6°C (medium confidence) (AR5 SPM)

The bottom of the ‘likely’ range has been lowered from 2 to 1.5C in the AR5, whereas the AR4 stated that ECS is very unlikely to be less than 1.5C. It is also significant that the AR5 does not cite a best estimate, whereas the AR4 cites a best estimate of 3C. Further the AR5 finds values of ECS exceeding 6C to be very unlikely, whereas the AR4 did not have sufficient confidence to identify an upper bound at this confidence level. The stated reason for not citing a best estimate in the AR5 is the substantial discrepancy between observation-based estimates of ECS (lower), versus estimates from climate models (higher).

My recent finding showing best estimate of 1.64 degrees C per doubling of preinsdustrial level of CO2eq

Nic Lewis and I recently authored a paper, published two weeks ago in the journal Climate Dynamics (2014), which arrived at a best estimate of 1.64 degrees C. This is much lower than the best estimates expressed by the other participating scientists in Climate Change National Forum (e.g., Scott Denning’s best estimate of 3 degrees C [link]; Andy Dessler’s best estimate of 2.7 degrees C [link]). Nic and I also managed to constrain a ‘likely’ range (17–83% probability) of 1.25-2.45 degrees Celsius, while still taking into account the substantial uncertainties in external forcings, particularly aerosols. This lower bound extends below the IPCC AR5’s ‘high confidence’ range (5–95% probability) of 1.5-4.5 degrees Celsius.  Our findings for that probability range were 1.05-4.05 degrees Celsius.

Here are excerpt from Nic Lewis’ blog post at ClimateAudit on the paper:

Our paper derives ECS and TCR estimates using the AR5 forcing and heat uptake estimates and uncertainty ranges. The analysis uses a global energy budget model that links equilibrium/effective climate sensitivity (ECS) and transient climate response (TCR) to changes in global mean surface temperature (GMST), radiative forcing and the rate of ocean heat uptake between a base and a final period. The resulting estimates are less dependent on global climate models and allow more realistically for forcing uncertainties than similar estimates such as those from the Otto et al (2013) paper.

Base and final periods were selected that have well matched volcanic activity and influence from internal variability, and reasonable agreement between ocean heat content datasets. The preferred pairing is 1859–1882 with 1995–2011, the longest early and late periods free of significant volcanic activity, which provide the largest change in forcing and hence the narrowest uncertainty ranges.

Table 1 gives the ECS and TCR estimates for the four base period – final period combinations used.

Table 1: Best estimates are medians (50% probability points). Ranges are to the nearest 0.05°C. Lewis and Curry, Climate Dynamics, September 2014.

Table 1: Best estimates are medians (50% probability points). Ranges are to the nearest 0.05°C. Lewis and Curry, Climate Dynamics, September 2014.

 AR5 does not give a 95% bound for ECS, but its 90% bound of 6°C is double that of 3.0°C for our study, based on the preferred 1859–1882 and 1995–2011 periods.

Considerable care was taken to allow for all relevant uncertainties. One reviewer applauded “the very thorough analysis that has been done and the attempt at clearly and carefully accounting for uncertainties”, whilst another commented that the paper provides “a state of the art update of the energy balance estimates including a comprehensive treatment of the AR5 data and assessments”.

There is thus now solid peer-reviewed evidence showing that the underlying forcing and heat uptake estimates in AR5 support narrower ‘likely’ ranges for ECS and TCR with far lower upper limits than per the AR5 observationally-based ‘likely’ ranges of: 2.45°C vs 4.5°C for ECS and 1.8°C vs 2.5°C for TCR. The new energy budget estimates incorporate the extremely wide AR5 aerosol forcing uncertainty range – the dominant contribution to uncertainty in the ECS and TCR estimates – as well as thorough allowance for uncertainty in other forcing components, in heat uptake and surface temperature, and for internal variability. The ‘likely’ ranges they give for ECS and TCR can properly be compared with the AR5 Chapter 10 ‘likely’ ranges that reflect only observationally-based studies, shown in Table 1. The AR5 overall assessment ranges are the same.

The CMP5 GCMs used for AR5 all have ECS values exceeding 2°C, whereas 70% of our preferred main results ECS the probability lies below that level, and over 90% lies below the 3.2°C mean ECS of CMIP5 models. The 33 CMIP5 models with suitable archived data show TCR values exceeding our preferred best estimate of 1.33°C in all but one case, with an average TCR exceeding the top of our 1.8°C ‘likely’ range.

The study does not assume any possible contribution to the increase in GMST from indirect solar influences not allowed for in the AR5 forcing estimates, or from natural internal climate variability affecting ocean heat uptake and/or forcing.

One of the most important contributions of this paper is the assessment of uncertainties in external forcing on estimates of climate sensitivity, which is something that has hitherto been only partially allowed for in most climate sensitivity estimates. As pointed out by the AR5 and this paper, the uncertainties in external forcing are substantial, particularly for aerosols.

Is this paper the last word on climate sensitivity estimates? No. The uncertainty analysis in the Lewis and Curry paper relates only to the uncertainty in external forcing, surface temperature and ocean heat uptake. There remains considerable meta uncertainty in the determination of climate sensitivity, including how the problem is even framed.

In particular, the energy balance approach does not account for factors that do not directly relate to the energy balance, e.g. solar indirect effects and natural internal variability that affects forcing (although an attempt has been made in the Lewis and Curry paper to make some allowance for uncertainty associated with these factors) . Further, there was ‘something else’ going on in the latter 19th and early-mid 20th century that was causing warming, that does not seem to relate directly to external forcing. The paper does attempt to factor out the impact of the Atlantic Multidecadal Oscillation through the selection of base and final periods, but this is by no means a complete account for the effects of multi-decadal and century scale internal variability, and how this confounds the energy balance estimate of climate sensitivity.

Resolving the reasons for differences between observational/energy balance estimates and GCM estimates of climate sensitivity is an issue of substantial importance. At this point, I find the estimates in the Lewis and Curry paper to be the most convincing estimates available to date.

Cartoon originally published in Climate Etc.

Cartoon originally published in Climate Etc.

 

Note by JC: This blog post was an adaptation from two recent posts on Climate Etc. See the original posts here and here.

THE FORUM'S COMMENT THREAD

  • While Judith Curry brings up some interesting notions in her post, I don’t agree with many of her arguments and conclusions.

    I’m pleased that Dr. Curry acknowledges that “uncertainty in itself is not a reason for inaction” (see also https://theconversation.com/why-climate-uncertainty-is-no-excuse-for-doing-nothing). I do find that conclusion slightly at odds with her frequent calls to put less effort in mitigation. Curry says that “deep uncertainties remain”, while at the same time apparently basing her anti-mitigation viewpoint on the assumption that climate sensitivity (ECS) is low. If this deep uncertainty however extends to ECS, one would think that the risk of substantial warming entails a substantial risk that is worth hedging against. Is she so sure that ECS is low and impacts benign? In short, I sense some inconsistencies in her approach to uncertainty.

    There is no science without uncertainty…

    [Read full response: ‘Uncertainty doesn’t imply nothing is known or nothing should be done‘ by Dr. Bart Verheggen on CCNF.]

  • Robert Way below comments on the Lewis-Curry paper using a global temperature data set with poor Arctic coverage and thus underestimated Arctic temperatures. This raises an important point (that may have been discussed in the literature but I haven’t seen it): climate sensitivity can be defined with respect to any global temperature data set with reasonable coverage. Since coverages and assumptions vary, so to will the climate sensitivity vary depending on the data set.

    Thus, the sensitivity of the HadCRUT4 global mean to CO2 doubling will in general be slightly different from the sensitivity of the GISTEMP global mean to CO2 doubling, which in turn will be different from the sensitivity of the UAH MSU TLT global mean to CO2 doubling, which in turn will be different from the sensitivity of reanalysis-based global mean surface temperatures to CO2 doubling. None of these are truth, and it’s generally not possible to compute conversion factors so as to translate one sensitivity into another.

    Likewise, estimates of climate sensitivity from paleoclimate data will depend upon the global temperature data set used to estimate the change in global temperature from the past to the present.

    This is all in the noise level, though, compared to the remaining sensitivity uncertainty, both within a study (e.g., Lewis-Curry) and between studies.

  • A comment on Dr. Curry’s last sentence: “At this point, I find the estimates in the Lewis and Curry paper to be the most convincing estimates available to date.”

    It is quite normal for the author of a study to find that study’s results to be more convincing than those of any other study. Presumably, the study’s author has been able to directly address all of her concerns while doing her own study, while other studies will in general leave one or more of her personal concerns unaddressed. I certainly tend to find the results of the studies I conduct much more convincing than the results of studies that other people perform on similar topics.

    So how are the rest of us supposed to interpret the Lewis-Curry result? Bart Verheggen summarizes the results of other types of investigations into climate sensitivity and how the Lewis-Curry results compare. There are basically three types: estimates based on the current well-observed change, estimates based on computer models designed to simulate the current climate accurately, and estimates based on changes in climate in the geologic past. The latter two types show different, higher sensitivities than the first types.

    If one knew nothing more, one would conclude that the higher values are more likely because they arise from independent estimates. Scientists can dig deeper, though, and address the relative strengths and weaknesses of each method. The IPCC concluded that there is no clear winner: all have different weaknesses, and none is obviously superior. The methods using the current well-observed changes are attractive because they are based on better data, but then why do all other methods give answers that are systematically higher?

    It’s tempting to conclude that the correct answer is somewhere in the middle, but the discrepancies suggest the possibility that there might be some major systematic problem with one or more techniques. In the parlance of election projections, this one is too early to call.

Leave a Reply

You must be logged in to post a comment.

PUBLIC COMMENT THREAD

  • Pingback: What Climate Scientists are Saying About the Global Warming Hiatus

  • http://ourchangingclimate.wordpress.com/ Bart Verheggen

    While Judith Curry brings up some interesting notions in her post, I don’t agree with many of her arguments and conclusions.

    I’m pleased that Dr. Curry acknowledges that “uncertainty in itself is not a reason for inaction” (see also https://theconversation.com/why-climate-uncertainty-is-no-excuse-for-doing-nothing). I do find that conclusion slightly at odds with her frequent calls to put less effort in mitigation. Curry says that “deep uncertainties remain”, while at the same time apparently basing her anti-mitigation viewpoint on the assumption that climate sensitivity (ECS) is low. If this deep uncertainty however extends to ECS, one would think that the risk of substantial warming entails a substantial risk that is worth hedging against. Is she so sure that ECS is low and impacts benign? In short, I sense some inconsistencies in her approach to uncertainty.

    There is no science without uncertainty…

    [Read full response: ‘Uncertainty doesn’t imply nothing is known or nothing should be done‘ by Dr. Bart Verheggen on CCNF.]

  • Pingback: Uncertainty doesn’t imply nothing is known or nothing should be done

  • Pingback: Back from the twitter twilight zone: Responses to my WSJ op-ed | Climate Etc.

  • http://atmo.tamu.edu/profile/JNielsen-Gammon John Nielsen-Gammon

    Robert Way below comments on the Lewis-Curry paper using a global temperature data set with poor Arctic coverage and thus underestimated Arctic temperatures. This raises an important point (that may have been discussed in the literature but I haven’t seen it): climate sensitivity can be defined with respect to any global temperature data set with reasonable coverage. Since coverages and assumptions vary, so to will the climate sensitivity vary depending on the data set.

    Thus, the sensitivity of the HadCRUT4 global mean to CO2 doubling will in general be slightly different from the sensitivity of the GISTEMP global mean to CO2 doubling, which in turn will be different from the sensitivity of the UAH MSU TLT global mean to CO2 doubling, which in turn will be different from the sensitivity of reanalysis-based global mean surface temperatures to CO2 doubling. None of these are truth, and it’s generally not possible to compute conversion factors so as to translate one sensitivity into another.

    Likewise, estimates of climate sensitivity from paleoclimate data will depend upon the global temperature data set used to estimate the change in global temperature from the past to the present.

    This is all in the noise level, though, compared to the remaining sensitivity uncertainty, both within a study (e.g., Lewis-Curry) and between studies.

  • http://atmo.tamu.edu/profile/JNielsen-Gammon John Nielsen-Gammon

    A comment on Dr. Curry’s last sentence: “At this point, I find the estimates in the Lewis and Curry paper to be the most convincing estimates available to date.”

    It is quite normal for the author of a study to find that study’s results to be more convincing than those of any other study. Presumably, the study’s author has been able to directly address all of her concerns while doing her own study, while other studies will in general leave one or more of her personal concerns unaddressed. I certainly tend to find the results of the studies I conduct much more convincing than the results of studies that other people perform on similar topics.

    So how are the rest of us supposed to interpret the Lewis-Curry result? Bart Verheggen summarizes the results of other types of investigations into climate sensitivity and how the Lewis-Curry results compare. There are basically three types: estimates based on the current well-observed change, estimates based on computer models designed to simulate the current climate accurately, and estimates based on changes in climate in the geologic past. The latter two types show different, higher sensitivities than the first types.

    If one knew nothing more, one would conclude that the higher values are more likely because they arise from independent estimates. Scientists can dig deeper, though, and address the relative strengths and weaknesses of each method. The IPCC concluded that there is no clear winner: all have different weaknesses, and none is obviously superior. The methods using the current well-observed changes are attractive because they are based on better data, but then why do all other methods give answers that are systematically higher?

    It’s tempting to conclude that the correct answer is somewhere in the middle, but the discrepancies suggest the possibility that there might be some major systematic problem with one or more techniques. In the parlance of election projections, this one is too early to call.