“I do not deny there is a climate, or that it changes”
“I do not deny there is a climate, or that it changes”
This was said to me by a close family member. I called it a "standard opening line" for climate denialism. At the same time, I do not want to dismiss my family member completely. So I wanted to discuss in a medium where I can take my time to investigate different arguments in-depth to see what they entail. This also allows me to provide source material.
I'm actively drawing from claims made on the Climate page of a political party in the Netherlands. The first claim is, paraphrased, "The climate has changed in the past, therefore, change is nothing to worry about". In this post, I will take these claims at face value. Therefore, I will look into the periods, the changes involved, and their causes.
In a second post, which I hope to write after this, I will review a second claim made by the page. This claim boils down to, "extreme climate events do not happen more frequently or cause more damage than historic events of a similar nature".
The climate is always changing
According to their first claim (my translation):
The climate is always changing. The earth was warmer in the time of the Egyptians and Romans, there was a new peak in the Middle Ages, after which it was colder for a few centuries; however, since 1850, the earth warmed slightly, by about 1 degree Celsius.
One of the harder things with stories like this is that they *seem to make sense*. This claim is such a story; there is nothing wrong with it on a surface level. Climate change is the result of a tremendously complicated network of feedback, inputs and weather systems. It changes, it drifts, it changes again. A body of scientific literature studying these phenomenons has concluded that human activity has induced accelerated global warming. Unless we find otherwise, the default here is to assume “climate change” can also be used colloquially to refer to the conclusions of that field. These conclusions can be simplified to, "The global climate is warming and human activities are a primary factor behind this."
Does that mean we do not understand the changes that happened in the past? This is where the climate change discussion becomes harder, more complex, and more nuanced.
Let’s start with science. The study of previous climates is called Paleoclimatology, from “paleo”, "ancient/long ago", and "climatology". This is a complicated science because no measurements were taken. Therefore, data analysis must work by proxy methods. Among scientists in the field, there is, of course, a lot of discussion regarding these proxy methods, such as when they are misused, when they have bias, and when they are unsure.
The various proxies, used in the correct way, actually agree with each other. Therefore we can reconstruct the global climate in the past and how that came to be. And we can check if the models we use now agree with that data from the past.
Climate Reconstructions
There are quite a few re-constructions to draw from, so I will point out there is a Wikipedia page on the topic. I will focus on Ljungqvist (2010), because the PDF is easily accessible. In this paper, they present a new reconstruction of the temperature in the (extra-tropical) Northern Hemisphere, using many different paleotemperature proxies which were never included before. It shows a distinct Roman Warm Period, a Dark Age Cold Period, a medieval Warm Period, the Little Ice Age, and then a rapid temperature increase in the 20th century. This presentation agrees well with previous reconstructions and even with the quote we are discussing. One such previous reconstruction it agrees well with is the famous “Hockey Stick” model, which was reconstructed from tree-ring data. Let’s take a look at that Fig. 3 from Ljungqvist (2010):
A word about proxies – these are painstakingly documented ways of estimating direct data from indirect data. You're probably familiar with an extremely common proxy, a proxy that greatly contributed to the Industrial Revolution. I’m talking about a thermometer, a proxy for temperature which uses the expansion of Mercury as a proxy for environment temperature. Proxies are everywhere – to the point where we don’t even realize familiar ones are in fact, proxies.
So as I mentioned – there is nothing wrong with the quote, but what about its implications? If the climate changed in the past, is there certainly nothing to worry about?
The Roman Warm Period
In the previously mentioned quote, the Egyptians and Romans are mentioned in one breath. This refers to the Roman Warm period. First off, we should wonder if this was a global climate phenomenon or one local to the Roman/Egyptian region (Mediterenean). The aforementioned reconstruction is a fairly local one (Northern Hemisphere).
The Roman Warm Period affected most of the Northern Hemisphere, as can be estimated from this paper investigating temperature shifts in China, and published in the journal "Climate of the Past". It would be convenient to have reconstruction for Northern America to compare it to the world average but, alas, I could not find one.
Another paper concluded that land cover and aerosol activity in the Roman Empire had noticeable influence on some climate variables. It concludes the Romans probably had no net effect. However, their results imply that land cover might have contributed to temperature shifts and that aerosol-cloud interactions would have attenuated it.
But, as always, nuance strikes again. While I inferred the Roman Warm Period affected most of the Northern Hemisphere, it does not imply a global effect. Much like the area near my windows is a lot colder than the rest of my house –possibly because I’m not running my heating – the global climate (the overall temperature of my house in this analogy) was not noticeably hotter during the Roman Warm Period. A study published in Nature, which is generally thought of as the most impactful publisher – and as a result, the one that will use the most scrutiny both before and after publication – reviewed coherence in the global record using temperature reconstructions [PDF]. I shall quote a piece of the text:
The results shown here can explain at least two curious facts about climate epochs of the Common Era: the lack of consensus about the timing of climate epochs and the discovery of records that do not fit the standard narratives. Peak warming and cooling events appear to be regionally constrained. Anomalous globally averaged temperatures during certain periods do not imply the existence of epochs of globally coherent and synchronous climate. This global asynchronicity suggests that multi-decadal regional extremes are driven by regionally specific mechanisms, being either unforced internal climate variability or regionally varying responses to external forcings .
They use the term "Climate Epochs" to refer to periods such as the Roman Warm Period, the Little Ice Age, and similar incidents I discuss in this blogpost. Their verdict is exceedingly clear: There is no spatial consistency and no global effect, in any of these pre-industrial variations. All are driven by regional mechanisms, be they natural or a response to non-natural forcings.
This paper also comes to another important conclusion:
In contrast to the spatial heterogeneity of the pre-industrial era, the highest probability for peak warming over the entire CE (Fig. 3c) is found in the late 20th century almost everywhere (98% of global surface area) except for Antarctica, where contemporary warming has not yet been observed over the entire continent. Thus even though the recent warming rates are not entirely homogeneous over the globe with isolated areas showing little warming or even cooling, the climate system is now in a state of global temperature coherence unprecedented over the Common Era.
Spatially heterogeneity, or “basically this is happening everywhere” is a particularly clear indication of a global phenomena, that of warming. The warming is also unprecedented over the Common Era, which is the time for which they look into the data.
Note that this is a conclusion reached by evaluating both direct measurement and proxy data.
So, now that we know the Roman Warm Period was not a global epoch (therefore, the Roman Warm Period isn't an argument against climate change science), it is interesting to wonder which regional factors determined the local effect on the Northern Hemisphere.
This study in Nature Geoscience looked at reconstructions and simulations (in which we take our understanding of climate physics and test it against data) [article]. Here, they took the (often local) reconstructions to construct reliable global mean temperatures and compared those to 23 CMIP5 climate model simulations (standardized global climate simulations). It suggests, but does not conclude, that most pre-industrial climate epochs are related to volcanic activity (not necessarily more, but also, possibly, less, activity) and the resulting aerosols. In this study they primarily focus on results for 1300-1800, which are more firmly inside of the simulated timescale. I would also hazard a guess that volcanic activity is better understood for this era.
Data shows that when volcanic activity went back to normal levels, when the warming ceased, the climate returned back to its “baseline”. Therefore,we have the “Dark Age Cold Period”. Note that “baseline” is a rather sloppy term here; mostly, I mean that strong extra factors (volcanic activity) are no longer relevant, so the normal system of sun, sea, wind, air, etc., takes over.
Medieval Warm Period
The Medieval Warm Period (800-1400 CE) is a part of the study discussed earlier, and there are well known causes for the Medieval Warm Period. I think this one sums it up best [PDF]:
The persistent positive phase reconstructed for the MCA appears to be associated with prevailing La Niña–like conditions possibly initiated by enhanced solar irradiance and/or reduced volcanic activity (21) (Fig. 4D) and amplified and prolonged by enhanced AMOC.
This paper primarily talks about how atmospheric and sea temperatures are coupled, which leads to various effects such as the NOA (North Atlantic Oscillation) and La Nina (a cooling interaction between atmosphere/wind and the oceans). They reference a paper on Volcanic and Solar influences on climate. But their results are clear:
There was more sunlight
There was less volcanic activity
The atmosphere/sea system prolonged the effects
Little Ice Age
So, from the previous paper, we know that several factors influenced a warm period. After that, the warm period came to an end. With more volcanic activity and less solar irradiance,we entered a “cool” period. Overall, there are numerous possible causes [Wikipedia] that likely contributed.
Strikingly, solar activity was low. Both the Maunder and Spörer Minima happen in this timespan. Solar activity recovered back to more normal levels during the 19th century. However, the last 50 years have not seen increased solar activity compared to the last century. We cannot explain recent warming with solar activity.
Meanwhile, volcanic activity was high, particularly at the onset,of the Little Ice Age, leading to cooling effects. As we previously discussed regarding the Medieval Warm Period, these effects are prolonged due to atmosphere/sea system interaction. According to the study, changes in solar activity aren’t even required to reproduce the effects.
This doesn’t mean that other effects can be neglected, or that I have just magically found the answer they’re still searching for. They haven’t found a definitive answer yet, but these factors are all contributing.
Modern age
Finally, we come to the Modern age, the last century or so. I think it’s important that we first talk about it. As the political party wrote on their page:
After 1850, the earth becomes slightly warmer: by about one degree celsius."
Let’s take a look at the temperature reconstruction in the Ljungqvist (2010) study. To reiterate, this studied temperatures in the non-tropical part of the Northern Hemisphere. We discussed this study in previous sections and saw that it is a representation of local temperatures, not global temperatures. Let’s take a look at Figure 3 from Ljungqvist:
As you can see, while temperatures vary, they usually do so by, at most, 0.4 degrees – over a period of hundreds of years. The extreme minimum, around 1700, followed a 700 year cooling trend. Looking at (only) this presentation, you wouldn’t necessarily say there is a problem – the climate recovered from that minimum, right?
To answer that, let’s take the last 20 years into account as well. Remember, the chart above is comparing temperatures to the average of 1961-1990. The chart below, however, has 1850 as it’s base value. Because Ljungqvist’s chart has 1850 at about -0.3, you would have to take whatever we find in recent data and compare it to -0.3 to see the total difference. Let’s take a look:
Global Temperature, near surface. Taken from European Environment Agency website.
Hold up. This chart goes to a maximum of 1.5 degrees Celsius? Plus, the current warming of “About 1 degree” is added on top of the -0.3 that we saw in the Ljungqvist chart. This means its value in that chart would be a 1.3 degree anomaly, or more than 300% of what we saw as a normal range of fluctuation for the last 2000 years?
That’s the problem here. When climate science started predicting climate in the 1970s, they forecast warming (which was mostly speculative before then). At that time, the annual anomaly versus 1850 wasn’t even that large, just a few tenths of a degree. In the 1980s – consensus formed – and the anomaly grew to about 0.5 degrees Celsius. By 1988, the Intergovernmental Panel on Climate Change (IPCC) formed, with an anomaly of 0.5 degrees Celsius. While I’m writing this, the anomaly, ever growing, is at 1.0 degrees Celsius.
Our forecasts from the past have already come true for recent years but the upcoming forecast is worse. Modelling is very complicated, even though the results are spectacular. Let’s take a look at how models perform:
How models perform. Yellow: 58 different simulations produced by 14 different models. Black: The observations of near-surface temperatures. Red: The mean of the models. Also added are horizontal lines for major volcanic eruptions. FAQ 8.1, Figure 1 of IPCC AR4
Those simulations are really, really good. In this case, I suspect they were provided the volcanic data (when eruptions actually happen). If you were to run many simulations of volcanoes erupting at different times, you would just average out the effect of volcanic eruptions and wouldn’t see any direct responses to actual volcanic eruptions. But, we can see a rather clear verdict.
I can hear you’re thinking – and I’m thinking it with you – there’s a catch! If those simulations were fed what actually happened, then naturally they will do well at predicting what actually happened - they knew all along! Yes, you’re right, they have a much higher likelihood of performing well.
So, let’s look at how the original forecasts held up. There’s various papers and articles on this topic(example 1, example 2) but I will focus on one with a “plain language summary” and I will quote from it:
We find that climate models published over the past five decades were generally quite accurate in predicting global warming in the years after publication, particularly when accounting for differences between modeled and actual changes in atmospheric CO2 and other climate drivers. This research should help resolve public confusion around the performance of past climate modeling efforts and increases our confidence that models are accurately projecting global warming.
This seems rather clear – models are performing really well. Recent model improvements are in the details but the trend of global warming and underlying causes have been clear for a long time. A (now, somewhat old but favourite) infographic can be found here.
When we know model results are trustworthy, we can use them to predict data. So, if we set factors like solar radiance to be constant, we can see how that influences mean near-surface temperature.
And only one factor actually follows the temperature increase:
Not surprisingly, it’s Greenhouse Gases.
Summary
The climate always changes. It does so in natural cycles and in response to volcanic eruptions, solar flares, and everything else. But these do not cause global changes in as rapid or extreme fashion as the current observed changes.
The Roman Warm Period wasn’t a global period of high temperatures, and when it ceased locally we mostly see a return to very average conditions. When solar irradiance and low volcanic activity turned up, we enter the Medieval Warm Period. When this changed, particularly volcanic activity and solar minima, we entered the Little Ice Age. But rather than recovering back to an average level, we see a massive increase in the modern Age. Investigating the mechanism, we find that only human activities, in particular greenhouse gas emission, can account for the massive and unprecedented rise in temperature found.
The effect of extreme greenhouse gas output resulting from human activities is over 100x that of volcanic activity [source], and it is growing:
CO2 Emissions taken from Our World in Data
Humans are having a major effect on the global climate and that effect is accelerating. That’s a problem, and we need to address it.
We’re in a climate emergency — we are spending our carbon budgets, and they only give us a chance of limiting the warming effect to 1.5 degrees Celsius — and we are not treating it as such. We really should be. There are likely to be more effects than just warming - shifts of biodiversity, collapse of ecosystems, major changes to agricultural output. All of these I hope to address in a next post that addresses the claim about extreme events.
In the past, I’ve written about things you can do to reduce your carbon footprint. The most important thing is to vote — only that way you can affect 90% of the causes. Signing petitions might seem silly, but it is the only way we can show politicians that we care outside of voting years.
