Climate change is an issue that is not yet capturing the public opinion as it should do. One of the reasons is that the year-be-year temperature variations, at global scale, are much smaller than day-by-day variations at local scale, or seasonal variations at global or local scale. However, there is a certain parallelism between the body temperature of a person and the climate. If one records its body temperature every minute of a day, he will discover some daily variations, which could vary little less than 1 °C. Also, different persons can have different mean body temperatures, with values spanned over 1 °C or sligthly more, depending on each personal condition. An athlete immediately after a competition, or a normal person after a stressing exercise, can have a body temperature even 1°c or more superior to its normal mean. Well, these normal variations represent the “weather” of the person, and do not indicate necessarily the presence of a pathology.
However, if the temperature of a certain person start rising, more or less regularly, in the time, this variation represents a sort of “climate” of the person, with is changing, and may be a symptom of something not necessarily good for the health of the person. This is true even if such “climate” variation could be smaller than those that we have previously called “weather” variations. The scientific problem is thus to distinguish normal variations from abnormal variations, differentiating the scales. While the communication problem is how to inform in a correct but clear way the people about such dynamics.
We have now the fortune of having available more than one century of meteorological observations, carried out in several meteorological stations displaced in various places in the Earth. Several climatic centers have collected a subset of these data and have performed some analyses on such data, in order to exclude anomalous trends or data. And have calculated an estimate of the global mean temperature. Since the choice of stations and the treatment of data has been different from centre to centre, there are some differences among the datasets, but the emerging signal is univoque and incontestable: global mean temperature is increasing.
Furthermore, several climatic models (that, more properly, are now called Earth System Models) have been run by different teams to simulate the future climate of the Earth. Despite each individual model tends to give a particular answer, the current method to consider such kind of projections is to consider the ensemble of the results of a wide set of models. This has been performed, for instance, during the experiment CMIP5, whose preliminary results have constituted the core of the findings of the last IPCC report.
With Stefano Caserini, coordinator of Italian blog climalteranti.it, we have had the idea of combining the two informations, data and models, in a visual way. We have chosen as dataset the GISS and as model the CMIP5 ensembles, by selecting three different scenarios adopted: the RCP 2.6, the RCP 4.5, and the most extreme RCP 8.5, respectively corresponding to low, medium, and high emissions. In particular, RCP 8.5 scenario outlines what we would expect if the emissions will continue to change as they did until now (i.e. with a continuous increment).
This is instead the case of the milder scenario RCP 2.6:
while this is the animation of the intermediate scenario RCP 4.5:
Model simulations data are available in the period 1860-2100, while GISS observations refer to the period 1880-2016 (last datum is April). I have merged those two datasets by evaluating in each case the respectively anomaly in the common period 1880-1909 (a 30-year period, as usually it is done in climate analyses).
I have visualized the result with two different methods. I have used the spiral method of monthly anomalies, originally developed by Ed Hawkins for HadCRUT data, and I have considered the linear plot of montly anomalies, as done by myself in a recent post with HadCRUT data.
The first method is discussed in detail in this post of climalteranti.it, thus here I will describe the second, adding some short general considerations (short because I believe that these plots can talk by themselves).
These are, in my opinion, two different but impactive ways to visualize the climate change which is going on. The animations start from 1880 and, month by month, show the temperature variations up to April 2016. It is evident the initial cooling in the first decade of 1900, the warming between 1920 and 1930, the stationariety in the decade of the WWII, another weak warming immediately after, then a stasis between 1960 and 1970, and then rapid increase of warming rate since 1980, with last fifteen years able to update at least one montly record almost every year. Until the anomalous period of last nine months, which places completely out of the above range.
Future climate simulations by CMIP5 scenarios show a continuous warming, almost similar for all three scenarios up to 2030, then rapidly differentiating. Note that warming rate of models is more regular, due to the fact that these data represent an ensemble and are not just the output of a single model. The anomalous warmest records established from January to April 2016 seems to be updated around 2025-2030, when those values would become the regular climate. Then, after 2030, different scenarios start to differentiate from each other, with just the common result to show a larger anomaly in winter. At the end of this century, even in the low emissions scenario (RCP 2.6), a season like the last one will be regarded as a cool period. But, if we look at the most extreme scenario (RCP 8.5, corresponding to high emissions), it will be regarded as a sort of mini-ice age…
The scenarios begin to differentiate from about 2030… there is still a few time for trying to make occurring the mild scenario RCP 2.6 instead of the extreme RCP 8.5 one… not too much time, because greenhouse gases increases inexorably, and consequent global warming too.
We can choose… we must choose!