Climate change "for dummies"

According to relatively simple physics the average temperature of the Earth should be about 33 degrees lower than it is now. Thankfully we have an atmosphere consisting for a small fraction of greenhouse gases such as water vapor, carbon dioxide (CO2), methane (CH4), and other gases that keep the Earth at the current temperature level. Because we we learned to manage the landscape for agriculture and other uses and invented the combustion engine we developed to a pretty smart society, but as a side effect we are increasing the amount of greenhouse gases in the atmosphere. Simple logic tells us that the Earth will get warmer and that is the story you have heard for the past decades from climate scientists.

For lazy folks:

But maybe it is a conspiracy!

Then came climategate showing that a number of climate scientists was not fully respecting all scientific rules. And mistakes in the "climate science bible" IPCC fourth assessment report were found. Now, that is not too strange for such a large report but what is worrying is that all mistakes portrayed things worse than they actually were. In addition, the average temperature of the Earth has not risen much over the past decade or so, something that was not foreseen by any climate model and is often used as an argument that we do not have an influence on the climate ("CO2 goes up, temperature not, so CO2 does not impact temperature"). Add to that the stories about scientists making up or manipulating data, and I can fully understand that people do not take the words of a scientist for granted anymore.

So it is best to judge for yourself

And that is what I hope you will do with the graphs below. We will focus on the past 100 years or so because that is where we have the best data and before that we probably had relatively little impact compared to natural fluctuations. Obviously we start with the most important and easiest to understand graph, global temperature: Temperature

This is the mean annual temperature of the Earth (Python code), taken from the Met Office Hadley Centre and the Climatic Research Unit at the University of East Anglia (HADCRUT4). The data is plotted as an anomaly, so it shows the difference in temperature, not the absoslute value. Now, you see an overall increase in temperature but a lot of variability, both from year to year and also over longer time scales. For example, there was no temperature increase from 1940 to 1970. There are other datasets that give more or less the same picture with regard to variability, but the trend varies a bit between those datasets: the one I plotted above gives a trend of 0.16 degrees C per decade over 1979-2011, GISS also gives 0.16, and UAH (satellite data, only available from 1979 onwards) yields 0.14.

The next thing to do is to look at the cause of this warming. CO2 is obviously an important candidate, because that is a greenhouse gas which warms the planet and its concentration is increasing. So we will plot the CO2 concentration derived from ice cores and direct measurements (from 1958 onwards) on top of the temperature graph: Temperature and CO2

So CO2 and temperature do line up, but far from perfect, especially in the first 50 years of the record (Python code). Nobody would argue based on this graph that CO2 is the sole cause of changes in climate.

On short time scales we do understand the variability pretty well. For example, when in June 1991 the volcano Mt.Pinatubo erupted the temperature dropped for two years because the massive amount of aerosols that reached high altitudes shielded some of the sunlight. And when we had a strong El Nino year in 1997-1998 the global temperature jumped by a few tens of a degree. But hey, we were trying to figure it out for ourselves, so here are plots of natural drivers of climate change and top left again the temperature anomaly data: some text

It is a mixed bag of data (Python code). Solar radiation increased up to 1960 or so, volcanoes cooled more in the second half of the 20th century than before but it is usually not a long lasting signal. El Nino's were more prevalent when the temperature increased mostly in the 1970-2000 period, but seems to be cooling the planet now. The same holds to some degree for the Pacific Decadal Oscillation (PDO), while the atlantic meridional overturning circulation (AMO) has a rather similar shape as the temperature graph but without a trend (or at least not a natural one).

A relatively simple way to disentangle the role of those influences is to perform a multiple linear regression where these natural factors are combined in such a way that best describes the measured temperure data. That is the final graph I will show where the blue line are the temperature measurements, the green line the best fit of temperature to CO2 and all the natural drivers shown above, and the red line the same exercis but without CO2, so trying to explain the pattern with only natural factors: some text

Clearly, with CO2 you get a pretty good match explaining 92% of the variability in temperature (Python code). Most of the warming is due to CO2 but the AMO also plays an important role. Without CO2 the statistical method has to boost the variables with a positive trend: the sun, AMO, and volcanoes. This way you get some of the trend but the pattern looks far from good, and does not make physical sense with for example more volcanoes leading to more warming.

Interestingly (and now it gets a bit more technical), this approach can also give a rough indication about the climate sensitivity. If we convert CO2 concentration to radiative forcing (using 5.35 * ln(CO2 / CO2 pre-industrial)) and make the -admittedly- questionable assumption that all other anthropogenic forcings cancel eachother, the best fit says the warming is 0.48 degrees C per W/2 increase. In other words it points towards a transient climate sensitivity of about 1.8 degrees C per doubling of CO2.

Now, you can ask yourself whether this is bullet-proof evidence. The answer is no; there may always be a natural cause that we do not understand yet, or something else going on in the complex climate system we have yet to discover. But to me and almost all fellow scientists, these three things:

  • According to well understood physics an increase in greenhouse gases should warm the planet
  • The concentration of greenhouse gases is increasing and the planet is warming
  • There are no natural variations in the climate system that could have caused this warming

provides a whole lot of evidence that humans are warming the planet. Actually, this is also not where most discussion is about in the community, even most climate skeptics agree that the planet warms due to CO2. The discussion is about the rate of warming, which depends on to what degree nature amplifies or dampens a certain amount of warming. But in the public debate it would help a lot if we first would agree on the relatively simple things as pointed out above.