Modeled vs. observed global temperature: with and without ‘makeup’

modeled and observed global temperature with and without makeupGlobal average surface air temperature is one of the most well-recognized metrics of contemporary climate change – hence the term ‘global warming’. One reason for this is that many impacts of climate change are expected to be proportional to the amount of global average warming that occurs over the next several decades to centuries. This is why, for example, the Paris Accord explicitly states climate change mitigation goals in terms of global average temperature.

Projections of global temperature are often based on the output from physical global climate model simulations and thus there is great interest in the agreement (or lack thereof) between modeled and historically observed global temperature.

Official reports (like the IPCC report), tend to present the comparison of modeled and observed global temperature in a format like that shown in ‘panel a’ above. This plot shows the model-mean and the model-spread (+/- 2 standard deviations) of global average temperature since 1861 (black) compared to observations (yellow). Various possible future scenarios are also shown (red, magenta, blue, cyan) which differ due to different assumptions about how much greenhouse gasses humanity might emit.

In ‘panel a’ there appears to be quite a bit of agreement between modeled and observed global temperature from 1861 to the present and thus this seems to provide compelling visual support for climate models’ ability to simulate/project global average temperature in the future.

However, I think that it is important to point out that part of this visual support comes from some nontrivial ‘makeup’ being applied to the comparison. Firstly, these temperature time series are all expressed as anomalies relative to a 1986-2005 baseline period (and then re-zeroed to be relative to preindustrial temperatures). This has the visual effect of forcing the models to essentially agree with each other and to essentially agree with observations over this 1986-2005 time period. Secondly, the spread around the model-mean value is calculated after the anomalies are taken which has the visual effect of minimizing the range of modeled temperatures. Overall, this results in an impressively small model spread around observations over the historical record and a relatively constrained spread for each of the individual future projections.

The raw model output, without this ‘makeup’ applied, is shown in ‘panel b’ above. In ‘panel b’, the y-axis is the absolute value of simulated and observed global average temperature in Kelvin. It is still the case that observations are more-or-less in the middle of the model simulations, but it can now be seen that the range of simulated values for absolute global average temperature is pretty large (~2.5C). In fact, this range is approximately as large as the amount of warming that we might expect to see over the remainder of the 21st century.

Does this matter? from a visual perspective, ‘panel b’ seems to inspire less confidence in our projections of future warming than ‘panel a’ does. However, the relevant question is: do model biases in the absolute value of temperature have a strong relationship with potential model biases in the projection of temperature change?

It seems as though the magnitude of the model biases in global average temperature do have some relationship with the magnitude of modeled future warming. However, these biases do not matter so much that they would seriously undermine the model projections over the next century or so (see discussion around Fig. 9.42a In Ch9 of Working Group I in the 5th IPCC Report; and discussion around Fig. 2 and Appendix B in Hawkins and Sutton, 2016). Therefore, I think it is reasonable to compare modeled and observed temperature change the way it is done in ‘panel a’ as long as we don’t completely forget about ‘panel b’.


Posted in Climate Change | 1 Comment

Defending science at the People’s Climate March

I accepted an invitation to speak at the San Jose People’s Climate March on April 29th, 2017 and I have reproduced what I said below. *Note that I do not necessarily endorse the People’s Climate March policy platform and I would not wish to defend any statements other than my own*.

Good afternoon. My name is Patrick Brown and I am a climate scientist.

I would like to say a few words today in the name of celebrating science – and in defense of science’s ability to inform society on issues of critical importance – like climate change.

Now as most of you know, the primary agent of current climate change is increasing levels of carbon dioxide from the burning of fossil fuels.

I want you to think about whether or not you would know this without modern science. Could you surmise this on an intuitive level?

Carbon dioxide is a colorless, odorless, tasteless gas. But despite its invisibility to our senses, the methods of science have allowed us to identify its existence and eventually measure it in the atmosphere.

The methods of science also allowed us to figure out that carbon dioxide contributes to what would become known as the greenhouse effect, and thus it can affect global temperatures.

When we began burning fossil fuels to power our societies, there was no intuitive reason to think that emissions of this invisible gas might be able to affect the global climate.

There was no intuitive reason to think that burning coal in Pennsylvania could contribute to sea level rise in Sydney….Yet we now know that this is the case.

In order to discover truths about our world that are beyond our intuition, we need to allow science to flourish freely.

let’s imagine for a moment that across the world, free scientific inquiry had been overtly suppressed over the last several centuries.

If free scientific inquiry had been suppressed, we wouldn’t be able to measure the temperature of the planet:

We wouldn’t know that the atmosphere is warming (as measured by a global network of weather stations, satellites, and weather balloons).

We wouldn’t know that the oceans are warming (as measured by ships, buoys, and satellites)

We wouldn’t know that the Earth’s ice is melting: sea ice, alpine glaciers and the Greenland and Antarctic ice sheets

And we wouldn’t know that global sea levels are rising

If we had accepted the stifling of free scientific inquiry we wouldn’t be able to ask questions about the future.

We would have no idea, for example, that if we were to burn all available fossil fuels, that would be sufficient to melt the entire Antarctic ice sheet over the next several thousand years – 6 million cubic miles of ice.

This would be enough melting to dramatically reshape the world’s coastlines.

This location where we stand would be under water, it would turn the California Central Valley into an inland sea, and it would It would essentially remove Florida from the map.

If we accepted the stifling of science we wouldn’t know that the rates of global warming under a ‘business as usual’ future are at least 10X faster than any global climate change experienced over the past 65 million years.

Finally, if we had stifled free scientific inquiry, we wouldn’t know how difficult it will be to limit global warming: We wouldn’t know that in order to simply stabilize global temperatures, we need to reduce carbon dioxide emissions by 80%.

Now, fossil fuels are not all bad – there is a reason that we have been using them. Everything that we materially value requires energy and historically, fossil fuels have provided the most affordable way to produce that energy.

This has led to successes that should not be ignored: large reductions in infant mortality, reduced rates of many horrible diseases, reduced poverty at a global scale, and robust increases in average lifespan.

But If science had been stifled, we would only know about the salient benefits of fossil fuel use and we would be totally ignorant to the dangers of burning fossil fuels.

Now, science has no “unquestionable truths” and as a scientist, I spend a great deal of my time questioning what other scientists think.

This skeptical aspect of science is precisely why we have so much confidence in scientific conclusions: Our confidence in the conclusions comes from the conclusions surviving challenge after challenge.

Thus, I have no problem with anyone challenging any scientific conclusion, including the findings that I just laid out.

BUT…but…If you want to question a scientific conclusion, you need to have good evidence (Preferably written up in a scientific paper and submitted to a scientific journal for peer review).

So I DO have a problem, when politicians or anyone else in power, have the hubris to cavalierly dismiss scientific conclusions out of hand – essentially dismissing hundreds of millions of cumulative hours of careful scientific work – simply on intuition.

and yes, referring to global warming a “Chinese hoax” fits into this category.

To conclude, I want to emphasize that free scientific inquiry, unencumbered by ideological opposition of those in power, is what has made it possible for us to understand the non-intuitive dangers of fossil fuel burning.

Namely that an invisible gas called carbon dioxide, that is produced from burning fossil fuels, is the primary driver of current climate change and that without curbing our emissions of this gas, we will experience rates of global change that are truly exceptional in the geologic record.

Therefore, any politician who engages in the suppression of inconvenient scientific conclusions or who has the hubris to dismiss those conclusions on intuition is being woefully irresponsible.

And that irresponsibility should be punished by damage to their reputations and a price paid at the ballot box.

Posted in Uncategorized | Leave a comment

Climate Science for People Really in a Hurry

Posted in Climate Change | 3 Comments

2016 Update to our ’empirical unforced noise’ analysis

This is an update to our 2015 Scientific Reports paper: Comparing the model-simulated global warming signal to observations using empirical estimates of unforced noise. The paper used a novel statistical estimate of unforced variability that was derived from reconstructed and instrumental surface temperature records. We used our statistical estimate of unforced variability to aid in our interpretation of recently observed temperature variability (more info here).

Our paper used global temperature data through 2013 since that was the most recent year in the major global temperature datasets at the time that the paper was submitted. Below I update Figures 2 and 3 from the paper, incorporating the data from 2014-2016.


Figure 2 updated through 2016.


Figure 3 updated through 2016.

Posted in Uncategorized | Leave a comment

Do ‘propagation of error’ calculations invalidate climate model projections of global warming?

My thoughts on claims made by Dr. Patrick Frank (SLAC) on the validity of climate model projections of global warming:


Posted in Uncategorized | 49 Comments

2016 Global Temperature Update to Hansen’s 1981 Projection

It is always useful to check past predictions against eventual observations. Below is the NASA GISTEMP observed global temperature (updated through 2016) overlain on top of various projections of CO2-induced warming from calculations published in 1981 (Hansen et al. 1981). 2015 and 2016 are literally off of the chart. This does not imply higher equilibrium climate sensitivity than that represented by the dashed line (5.6C) because these calculations did not include the effects of anthropogenic increases in non-CO2 greenhouse gasses. There are a number of other important caveats to this juxtaposition like Hansen’s model not allowing for unforced/internal variability as well as differences between the assumed and actual growth rate of atmospheric CO2 ect. Nevertheless, it is an interesting comparison.


Posted in Uncategorized | 4 Comments

2016 update of modeled vs. observed global temperature

NASA released their 2016 global mean surface temperature data today. With this datapoint in, observations are now above the average climate model value for this point in time (using 1986-2005 as the baseline):


This graphic uses the RCP 4.5 emissions scenario for the models but the divergence between RCP 4.5 and steeper emissions scenarios is not appreciable until the mid-21st century (see e.g. Figure 1 here).

Posted in Uncategorized | 4 Comments

Why do climate models disagree on the size of global temperature variability?

We have published a new paper titled “Spread in the magnitude of climate model interdecadal global temperature variability traced to disagreements over high-latitude oceans“. Here is a brief summary:

Natural unforced variability in global mean surface air temperature (GMST) is of the same order of magnitude as current externally forced changes in GMST on decadal timescales. Thus, understanding the precise magnitude of unforced GMST variability is relevant for both the attribution of past climate changes to human causes as well to the prediction of climate change on policy-relevant timescales.

Climate models could be useful for estimating the true magnitude of unforced GMST variability provided that they more-or-less converge on the same answer. Unfortunately, current models show substantial disagreement on the magnitude of natural GMST variability, highlighting a key uncertainty in contemporary climate science. This large model spread must be narrowed in the future if we are to have confidence that models can be trusted to give useful insights on natural variability.

Since it is known that unforced GMST variability is heavily influenced by tropical Pacific surface temperatures, it might be tempting to suppose that the large inter-model spread in the simulated magnitude of GMST variability is due to model disagreement in the amount of simulated tropical Pacific variability. Perhaps surprisingly, our study shows that this is not the case and that the spread in the magnitude of model-simulated GMST variability is linked much more strongly to model disagreements over high-latitude oceans. Our findings suggesting that improving the simulation of air-sea interaction in these high-latitude ocean regions could narrow the range of simulated GMST variability, advance our fundamental understanding of natural variability, and appreciably improve our ability to forecast global warming on policy-relevant timescales.

Posted in Climate Change | 7 Comments

Video Summary of my PhD Dissertation


Posted in Climate Change | Leave a comment

What do historical temperature records tell us about natural variability in global temperature?

I have published an article, written for a general audience, summarizing the results of our 2015 Scientific Reports study.

Posted in Uncategorized | Leave a comment