The op-ed authors reference the 2017 State of the Climate Report. I helped prepare the "Lower and Mid-Tropospheric Temperature" section and it is unclear where the authors' statistic comes from ("seven times as much warming...").

The average CMIP5 model warming in the tropical troposphere does outpace observations (by a factor of 1.5 - 3.3, depending on dataset). This issue deserves (and has received) scrutiny.

The difference in warming rates between observations and models largely arises in the early 2000s. A number of assessments have concluded that this slowdown in warming in the 2000s is in part due to natural variability (the Earth's warming was slowed due to climate variability) and forcing (the real world experienced different solar and volcanic aerosol forcing than what was used in the models). Models do simulate natural "hiatus" periods like that experienced in the early 2000s, but, since they are random, they generally do not occur at the same times as in the real world (though some models happened to have a slowdown in warming in the early 2000s). Furthermore, forcing agents (greenhouse gases, aerosols, solar changes) are prescribed to models. Since we do not know the exact evolution of forcing agents for future projections, they are estimated. In this case, the estimated forcing was different than what occurred in the real world.

When these issues are taken into account, models and observations are in agreement. Over long periods, when natural variability is a smaller issue, models and observations agree on the rate of warming. Assessments of of model projections of climate change, show that models have typically been quite skillful (e.g., here).

• IPCC, AR5, Ch. 9, Box 9.2.
• Meehl et al. (2014), doi: 10.1038/NCLIMATE2357
• Gleisner et al. (2015), doi: 10.1002/2014GL062596.
• Medhaug et al. (2017), doi: 10.1038/nature22315.
• Santer et al. (2017), doi: 10.1175/JCLI-D-16-0333.1.
• Santer et al. (2017b), doi: 10.1038/NGEO2973.

Assessments of climate change include many lines of evidence. In addition to computer models, observations of many aspects of the climate (e.g., changes in land and sea ice, sea level rise, temperature, humidity, etc.) and paleoclimate evidence is considered.

Beware of cherry-picking: It is not established that satellites are the "most reliable" observations, that the tropics are the best domain with which to compare observations or models, or that we should only consider post-1979 satellite era data.

Considering a larger range of time Marotzke and Forster (2015) generally find good consistency between models and observations.

Martozke and Forster (2015), doi: 10.1038/nature14117

As suggested in the last comment this statistic is quite uncertain (1.5 - 3.3). Some climate models simulate less warming than some model satellite datasets. The reason that the average of the CMIP5 models exhibits more warming than observations is in part due to natural variability and problems with the forcing (aerosols, solar changes, etc.) in the models.

The consistency of the surface records and the well-documented, large uncertainty in weather balloon and satellite datasets suggests that this is not true.

The relative warming between the surface and the troposphere in both models and observations is in accord with our physical understanding of the tropical atmosphere.

• Po-Chedley et al (2015), doi: 10.1175/JCLI-D-13-00767.1

The authors are referring to microwave sounding instruments which provide a measure of tropospheric temperature changes. These records are useful, but have substantial uncertainty.

In the State of the Climate Report, various estimates of tropical tropospheric warming had a range of 0.12 - 0.22 K / decade across four datasets (that all use the same underlying satellite data). This is nearly a factor of two difference in the long-term rate of warming. This "structural uncertainty" results because researchers use different approaches to remove known biases that affect long-term trends. No method is perfect, which leads to widely varying estimates of atmospheric warming.

Christy, J.R., S. Po-Chedley, C. A. Mears, L. Haimberger, 2019: Tropospheric Temperature [in “State of the Climate in 2018”]. Bull. Amer. Meteor. Soc., 100 (9), S181–S185, doi:10.1175/2019BAMSStateoftheClimate.1.

This is something that is also simulated in climate models (larger warming in the minimum daily temperature than the maximum daily temperature).

The article the authors reference tends to focus on smaller spatial scales where natural variability plays a larger role. On a global scale the seasonal asymmetry in warming rates appears to be small. Taking global GISTEMP trends over 1950 - 2018 the range of trends for all seasons is 0.13 - 0.15 degrees C per decade.

Lewis and Karoly (2013), doi: 10.1175/JCLI-D-13-00032.1

This is not as universally true as suggested in this op-ed. Changes in extreme events are difficult to detect and depend on definition and region, but there is evidence for some changes in extreme weather events. For example, evidence is provided for decreases in cold nights, increases in warm days, and increases in heavy precipitation over North America.

[IPCC WG1, Ch. 2, pg. 162]

This is at odds with recent work that finds a "persistent acceleration in global mean sea level rise since the 1960s" and the IPCC's attribution of sea level rise:

It is very likely that there is a substantial contribution from anthropogenic forcings to the global mean sea level rise since the 1970s. It is likely that sea level rise has an anthropogenic contribution from Greenland melt since 1990 and from glacier mass loss since 1960s. Observations since 1971 indicate with high confidence that thermal expansion and glaciers (excluding the glaciers in Antarctica) explain 75% of the observed rise. [AR5, WG1, pg. 870]

Dangendorf et al. (2019), doi: 10.1038/s41558-019-0531-8

The UN Report frequently demonstrated that the impacts of climate change (around the world, not just for island nations) are less severe and widespread at 2.7 degrees Fahrenheit compared to 3.6 degrees Fahrenheit.

This would have been a useful and easy point to make, given that it was one of the central findings of this report.

I thought these were good examples of impacts.

I couldn't find where this exact statement came from, but it seems reasonable given that:

... achieving emissions reduction targets consistent with the ambitious goal of 1.5°C of global warming under the Paris Agreement will result in the further loss of 70–90% of reef-building corals compared to today... [Box 3.4]

and that

Global warming is likely to reach 1.5°C between 2030 and 2052 if it continues to increase at the current rate. [SPM A.1]

And, regarding food security:

Increasing global temperature poses large risks to food security globally and regionally, especially in low-latitude areas (medium confidence) (Cheung et al., 2010; Rosenzweig et al., 2013; Porter et al., 2014; Rosenzweig and Hillel, 2015; Lam et al., 2016), with warming of 2°C projected to result in a greater reduction in global crop yields and global nutrition than warming of 1.5°C (high confidence) (Section 3.4.6), owing to the combined effects of changes in temperature, precipitation and extreme weather events, as well as increasing CO2 concentrations. [Ch. 3, Box 6]

This statement and similar statements would benefit from specific examples from the report or more context. Although I agree the problem is dire, I'm not sure which aspect of the problem is worse than previously thought. For example, the carbon budget (i.e., how much humans can emit and still warm less than 1.5 degrees C) has expanded from previous estimates.

It may be that some impacts will be felt earlier or will be worse than previously estimated -- this just wasn't clearly articulated.

Other statements that I thought could have used greater context or examples from the report include the quote by Bill Hare (below), or the statement that:

The new report, however, shows that many of those effects will come much sooner, at the 2.7-degree mark.

This statement deserves more scrutiny because it appears that the IEA outlook and the UN Report are considering different scenarios.

In the IEA outlook, it is true that coal remains a large component of the global energy system through 2040 under existing and expected policies ("NPS scenario"), but their projection under the "Sustainable Development Scenario," which takes international climate objectives into account, has coal accounting for 12% of global energy by 2040.

This is a useful point and comes from The Summary for Policymakers (C2.1) and is with regard to the economic transition needed to limit the world to 1.5 degrees C of warming (with limited or no "overshoot.")

The report does state that some sectors and technologies have transformed this quickly, but not on the same scale.

This should be “eastern equatorial Pacific”

This should be “central and eastern-central equatorial Pacific”

Note that the Paris target is “to limit the temperature increase to 1.5°C above pre-industrial levels” and the 0.9°C value is relative to 1951 – 1980.

I think 2017 was compared to the 1951 - 1980 average: https://earthobservatory.nasa.gov/images/91604/2017-was-the-second-hottest-year-on-record

This is reason 5 in the original statement.

This is reason 4 in the original statement.

This is reason 2 in the original statement.

This was Reason 1 in the original statement.

The original statement says:

These “natural climate solutions” could provide 18%[v] of cost-effective mitigation through 2030.

which is in reference to avoiding 2 degrees C of climate change. "Runaway climate change" is too strong here.

This information relates to reason 2 in the original statement.

This appears to be the motivation for the authors' statement. I don't know if it is true that the role of forests will be under-appreciated (in the to-be-released IPCC report) - it would have been useful to get more perspective on this - but it is certainly reasonable to highlight the importance of forests.

I couldn't find the statement link. I believe this is the statement.

Although background on the underlying research is provided further down, it is important to put this information into the context of the original study to give a sense of timescale and uncertainty.

The underlying study reviews the paleoclimate record and finds periods in which the Earth's climate may serve as an analogue for the future. The authors note that the Earth, with increased atmospheric carbon dioxide and a warmer global temperatures, exhibited reductions in ice at the poles, savanna expanded into the region where the Sahara desert currently exists, and sea levels were higher. It's important to keep in mind a few points:

• drastic changes (e.g., 20-foot sea-level rise) are expected to occur over long timescales (i.e., not in our lifetime)
• these past climate analogues don't exactly match our current situation (Earth's orbit, for example); so we should not expect that the past is a perfect predictor of the future
• the magnitude of these changes does depend on human actions over the next century (i.e.., how much carbon dioxide will we emit?)

This is a good point to make: research indicates that there is a non-negligible chance we have already emitted sufficient greenhouse gases to commit the Earth to 1.5 degrees C of warming.

This is an important point: that the potential discrepancies between models and paleoclimate data are mainly for large increases in carbon dioxide (and warming).

Somewhere in here, it should be noted that the past is not expected to be a perfect predictor of the future (or at least give some sense of the uncertainties involved in this research). The periods considered in the original study had important differences relative to the current Earth: different Earth-sun orbits, continental configurations, and land ice, for example. These differences affect the Earth's climate response to greenhouse gases.

This paragraph is more appropriately caveated (using the word "could" and "large areas" and "edges") in comparison to the leading sentence, "collapsing polar ice caps," which is a little less clear. A useful way to frame this information would be that: past climate have seen these profound changes and it is plausible that the future may look similar.

The study suggests this is an upper bound. This is important, since this seems to be a factor motivating the attention-grabbing headline.

From the study:

model-based climate projections may underestimate long-term warming in response to future radiative forcing by as much as a factor of two

and

models may underestimate observed polar amplification and global mean temperatures of past warm climate states by up to a factor of two on millennial timescales

This "factor of two" is also derived using model simulations of large increases of CO2 (~4x pre-industrial levels) and comparing the model simulated warming to paleoclimate data from the early Eocene climatic optimum, which was roughly 50 million years ago. During this time, the Earth's continental configuration (land mass locations and elevation) differed from the present and these continental configuration differences were not reflected in this model/paleo-data comparison. When the authors account for this, the differences between the models and the paleoclimate record appear to be substantially reduced. This "factor of two" seems to arise in part due to an apples-to-oranges comparison.

This paragraph is well-qualified by the author quotes in the subsequent paragraphs and the conclusion from the original study:

...we can conclude that even for a 2\(^\circ\) C (and potentially 1.5\(^\circ\) C) global warming - as targeted in the Paris Agreement - significant impacts on the Earth sustem are to be expected.

The next few paragraphs represent a good summary of the original research.

An example of this (over the last 65 million years) is noted in this study (press release).

It might have been useful to cite some scientists and make a stronger connection between sea level rise and the rest of the article here. It is not clear what timescale this magnitude of sea level rise would happen on.

This is another important point to make -- there will be periods where the Earth warms less quickly (not every year will be a record year), but the heating of the Earth continues.

This is an excellent and important point.

This is a small, but useful illustration that there is uncertainty in these datasets and their representation of record warmth.

Just to be clear, the central estimate is 5.86. You get 'more than seven' by taking the upper end of the uncertainty range. Not necessarily wrong, but a little misleading.

I think this is a little misleading, because the study does not really look at 'how bad' will climate change be, it instead looks at 'how much' climate change do we have in store. The apocalyptic stuff seems to be made via analogies with Venus.

It would be helpful to describe what this means. Michael Mann defines this later on in this article (greater than 2 degrees C warming). I think this context (e.g., 'game over for existing climate targets') would be useful (or using different language here).

Some important points follow later on, which do a great job putting these values in context. Including:

• this range of warming is compatible with existing model estimates
• the uncertainties are large and this study does not purport to give the authoritative warming estimate
• instead, this is another line of evidence that the climate could be a bit more sensitive to greenhouse gases than models predict
• the implication is that this makes it a bit harder to stay under IPCC targets of 1.5 and 2 degrees C

Some have also pointed out that momentum is growing for renewables on the business side (in addition to purely climate considerations).

Obama, B., 2017: "The irreversible momentum of clean energy," Science, DOI: 10.1126/science.aam6284.

This is interesting. It might also be helpful to point out that the Earth is currently in a relatively warm phase (as the graph below illustrates).

A number of studies point to this, using varied methodologies. Although no study is particularly convincing in isolation, it is interesting that this sort of analysis frequently points toward a climate that is a bit more sensitive to greenhouse gases than the CMIP5 model ensemble.

Some examples include:

Fasullo, J. T. and K. E. Trenberth, 2012: "A Less Cloudy Future: The Role of Subtropical Subsidence in Climate Sensitivity," Science, 338, 792 - 794.

Sherwood, S. C., S. Bony, J.-L. Dufresne, 2014: "Spread in model climate sensitivity traced to atmospheric convective mixing," Nature, doi:10.1038/nature12829.

Tan, I., T. Storelvmo, M. Zelinka, 2016: "Observational constraints on mixed-phase clouds imply higher climate sensitivity," Science, 352, 224 - 227.

Siler, N., S. Po-Chedley, C. Bretherton, 2016: "ariability in modeled cloud feedback tied to differences in the climatological spatial pattern of clouds," Clim. Dyn., DOI 10.1007/s00382-017-3673-2.

Volodin, E. M., 2007: "Relation between Temperature Sensitivity to Doubled Carbon Dioxide and the Distribution of Clouds in Current Climate Models," Atmos. and Oceanic Physics, 44 (3), 288 - 299.

Implicit in this estimate (which relies on RCP8.5) is that society continues to rapidly emit (business as usual throughout the next century). But not knowing future emissions also leads to a lot of uncertainty in future warming.

I wasn't totally sure where these numbers came from. I think this should be 3.4 - 6.4?

In the original article it says: "Reaching atmospheric CO2 concentrations of up to ~900 parts per million (ppm) by 2100 CE, the ensemble of Representative Concentration Pathway 8.5 (RCP8.5) scenario Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations (6) projects a global mean SAT rise (relative to PI conditions) of 4.84 K with an ensemble range of 3.42 to 6.40 K."

This has some implications for the interpretation of some of the warming estimates given later.

While others have documented differences in the rate of warming between models and observations, this study is different in that the authors suggest that these differences are because models are too sensitive to increases in atmospheric carbon dioxide. Several studies disagree with this assessment in both the surface and atmospheric record.

Other factors contribute to this apparent discrepancy (and some are noted in the original study). These include the potential for observational biases, errors in external forcing prescribed to the models, and real world multidecadal climate variability. So the attribution of model-observational differences predominantly to models is not supported.

M. Richardson, K. Cowtan, E. Hawkins, M. B. Stolpe (2016): Reconciled climate response estimates from climate models and the energy budget of Earth, Nat. Clim. Change, 6, 931 – 935, doi: 10.1038/nclimate3066.

Santer, B. D., et al. (2017): Causes of differences in model and satellite tropospheric warming rates, Nat. Geo. 10, 478 – 485, doi: 10.1038/NGEO2973.

Note that although there is nothing wrong with attempting to remove these effects from the temperature time series, it can be quite uncertain and subject to methodology, especially over short timescales.

As noted in the research paper, multidecadal variability may also effect the results. Several studies have noted that multidecadal variations in the rate of ocean heat uptake have affected observed warming rates.

Santer, B. D. et al. (2001): Accounting for the effects of volcanoes and ENSO in comparisons of modeled and observed temperature trends, J. Geophys. Res., 106, D22, 28,033 – 28,059.

Chen, X. and K.-K. Tung (2014): Varying planetary heat sink led to global-warming slowdown and acceleration, Science, 345 (6199), 897 – 903.

This study does not include the effect of smaller volcanoes late in the record that would have reduced real-world warming. Since these volcanoes were not included in model simulations, the model simulations would also have too much warming, since they did not include accurate representation of the effect of volcanic activity. This would have increased the model-observational discrepancy.

Santer, B.D. et al. (2014): “Observed multivariable signals of late 20th and early 21st century volcanic activity,” Geophys. Res. Lett., 500 – 509, doi: 10.1002/2014GL062366.

This study alone does not support this kind of broad statement, especially given that there are already several studies that disagree with the authors results.

It’s important to note that climate scientists do work on both understanding model-observational discrepancies and improving models. Another aspect to consider is that observations also have biases that can affect such comparisons.

This is right – models have the wrong ‘external forcings’ (such as volcanic forcing), which is likely an important factor in model-observational differences.

It is also possible that errors in removing volcanic forcing and natural variability could also affect model-observational comparisons.

Solomon, S., J. S. Daniel, R. R. Neely III, H. –P. Vernier (2011): The Persistently Variable “Background” Stratospheric Aerosol Layer and Global Climate Change, Science, 333 (6044), 866 – 870, doi: 10.1126/science.12066027.

Santer, B. D., et al. (2014): Volcanic contribution to decadal changes in tropospheric temperature, Nat. Geo., 7, 185 – 189, doi: 10.1038/NGEO2098.

One issue with such comparisons is that the observational record of atmospheric warming is quite uncertain as evidenced by the large trend differences between different groups that create satellite temperature datasets and even between versions of each group’s datasets.

Mears, C. A., F. J. Wentz, P. Thorne, and D. Bernie (2011): Assessing the uncertainty in estimates of atmospheric temperature changes from MSU and AMSU using a Monte-Carlo estimation technique, J. Geophys. Res., 116, D08112, doi: 10.1029/2010JD014954.

Again, the authors would presumably have a different trend over 1979 – 1993 if they repeated their calculation using their new dataset, which has fixed several non-climatic artifacts.

One curious aspect of this statement is that the trend in over the first 15 years (1979 – 1993) is seemingly taken from a much older version of the UAH satellite lower tropospheric temperature dataset. This dataset has since been improved as inhomogeneities have been discovered and corrected, including satellite orbital decay, an error in the diurnal correction, and a miscalibration of the NOAA-9 satellite. Each of these issues would have spuriously reduced the trend in the UAH dataset – presumably if this specific result were revisited models and satellites would agree much better over 1979 – 1993.

F. J. Wentz and M. Schabel (1998): Effects of orbital decay on satellite-derived lower tropospheric temperature trends, Nature, 394, 661 – 664.

Mears, C. A. and F. J. Wentz (2005): The Effect of Diurnal Correction on Satellite-Derived Lower Tropospheric Temperature, Science, 309, 1548 – 1550.

Po-Chedley, S. and Q. Fu (2012): A Bias in the Midtropospheric Channel Warm Target Factor on the NOAA-9 Microwave Sounding Unit, J. Atmos. Ocean. Technol., 29, 646 – 652, doi: 10.1175/JTECH-D-11-00147.1.

Spencer, R. W., J. R. Christy, and W. D. Braswell (2017): UAH Version 6 Global Satellite Temperature Products: Methodology and Results, Asia-Pac. J. Atmos. Sci., 53(1), 121 – 130, doi: 10.1007/s13143-017-0010-y.

The periods compared by the authors are somewhat arbitrary (1979 – 1993 and 1979 – 2017) since they are comparing their new result with their older study (in 1994). This makes it a little difficult to compare with other studies on this topic, though, again, this result is not necessarily surprising.

This is not a particularly new result. It has long been known that warming has not accelerated in the observations in the last decade. Although this is well-documented in the surface record, it has also been noted in the satellite record as well.

Foster, G. and S. Rahmstorf (2011): Global temperature evolution 1979 – 2010, Env. Res. Lett., 044022, doi: 10.1088/1748-9326/6/4/044022

Gleisner, H., P. Thejill, B. Christiansen, J. K. Nielsen, (2016): Recent global warming hiatus dominated by low-latitude temperature trends in surface and troposphere data, Geophys. Res. Lett., doi: 10.1002/2014GL062596.

Medhaug, I., M. B. Stolpe, E. M. Fischer, and R. Knutti (2017): Reconciling controversies about the ‘global warming hiatus’, Nature, 545, 41 – 47, doi: 10.1038/nature22315.

Santer, B. D., et al. (2017): Causes of differences in model and satellite tropospheric warming rates, Nat. Geo. 10, 478 – 485, doi: 10.1038/NGEO2973.