165 years of Climate Change

Here is a full animation of all Hadcrut4.3 monthly spatial anomaly data. In the early years you can see individual ship voyages! Note the strong variations throughout, especially in northern continents during winter months, and try to spot the El Niño events.

The video uses my new colour table which I think is better than the Met Office scheme. Blue is for DT < 0 and red for DT> 0. The table is made using a linear interpolation  over a ± 10 degree range. Anything just below zero looks slightly cool and anything above zero looks slightly warm. For example this is the plot for October 2015 using the new colour scheme.

H4 Temperature anomalies for October 2015 using new temperature colour scale.

H4 Temperature anomalies for October 2015 using new temperature colour scale.

In my opinion this better reflects the overall warming pattern last month, compared to the Met Office plot which exaggerates it.

This entry was posted in AGW, Climate Change, climate science, Institiutions, Science, UK Met Office and tagged , . Bookmark the permalink.

16 Responses to 165 years of Climate Change

  1. Bryce Payne says:

    A question, a suggestion, a comment.

    Question: Are the anomalies for each month in the animation differences from the overall annual temperature 1961-1990? I assume the anomalies are with respect to average annual temperature 1961-1990 since the seasonal anomalies are so sharp.

    Suggestion: Would it not be more informative with respect to the apparent underlying question, or at least a major portion of that question, to present the animation, or an accompanying animation of anomalies that compare a monthly mean in each year to the mean of the same month for the selected reference time period (1961-1990)? This would, presumably, take out much of the cyclical seasonal variations, to reveal whether or not there is a clear underlying trend.


  2. Bryce Payne says:

    Sorry, hit “post comment” button too soon.
    Comment: In the end, for October 2015 at least, the world appears to be considerably more red than blue……

  3. clivebest says:

    Hi Bryce,

    Answer: The anomalies for say January 2013 are equal to the difference from the average of all Januaries between 1961 and 1990. Likewise February is the difference from all average Februaries etc. These anomalies are calculated for each meteo station on land, and for each 5 degree cell on ocean surfaces. On land the anomalies of all stations within a 5 degree cell are averaged together.

    Suggestion: Yes that is a good idea. I could make 12 animations of variations for each month or perhaps seasonal.

    I have also made an animation of the annual variation…..




  4. Bryce Payne says:

    Is your annual variations animations accessible? If so, where/how?

  5. Bryce Payne says:

    Your annual variations animation popped up after my last comment, which led me to another question. I noted that, at least as colors rendered on my computer screen, your “around zero anomaly” color (green) according to the scale under the image is centered at about +0.3 deg C, and seems to run from about -0.6 to about +1.0 deg C. So your scale is biased to the high side. That is, cooler periods will change visible shift from green toward blue giving the appearance of cooling, with substantially less temperature change than the amount of warming needed to cause the color to shift toward red. In other words, your color scale is biased to not show warming……???

    • Clive Best says:

      It’s simply that I first used a default rainbow colour scheme. Then I later myself made a new blue to red transition centered on zero. I just haven’t had time yet to redo the original annual animation using the new scheme. You have to be careful with some of the other colour schemes you see including those in th IPCC reports. They tend to exagerrate warming by using a logarithmic colour scale.

      Perhaps I should redo the annual version.

  6. Bryce Payne says:

    Please do best you can to revise to a reasonably balanced, zero-anomaly centered color. Also, if I may make another suggestion, perhaps it would work better to render only two colors and their respective intensities instead of three. That is, eliminate any color for near zero anomalies. If you can then get the scale/scheme reasonably zero-anomaly-centered, it will be a useful and informative piece. This will leave open the issue of missing data in early years, but that problem is tacit, or could be addressed with another animation aimed specifically at data availability over time. Appreciate your efforts.

  7. Clive Best says:

    Here is an updated annual H4 anomaly animation using the new colour table. The range is now extended to ± 4C.


  8. Bryce Payne says:

    Thanks for most recent effort. To my eye much clearer. I do not know how much trouble it would be, or if it even matters given how white to red the world gets in the last twenty years or so, but could you nudge the lower end of the red scale to just a bit stronger red, or the lower (nearer zero anomaly) of the blue scale to just a bit weaker blue (nearer white). There is still a contrast imbalance in the near-zero anomaly part of the continuum (blue side is clearly blue all the way to zero, but red side is so weakly pink that it appears white up to about 0.2 to 0.3 deg C). Like I said works for me as is, just have to recognize that white is red just less so, and not neutral.

  9. Nick Stokes says:

    I have experimented a lot with animation and color schemes too. I think you do need a non-linear scale like the Met Office uses. I use atan or tanh. The reason is that otherwise you convey virtually no information about oceans. In fact I have tried schemes that ensure each color has an equal number of cells. That maximises visual information. If you adopt not “exaggerating warming” as a priority, you end up not showing much at all.

    • Clive Best says:


      Actually I am still working on it ! There is no correct answer here. The problem with the non-linear scale is that it shows small variations as being large variations. The problem with the linear scale is (perhaps) that it washes out long term variations. However what becomes clearer is just how strong natural variability is in northern continents, and just how small it is in southern continents. For example there really is not much strong variability in Australia as compared to say North America. The answer clearly is due to the effect of oceans. Any colour scheme has to at least show el Nino’s

      Another interesting observation is just how seasonal these variations (weather) really are. It is much stronger during northern winter months.

      So I haven’t finished yet. I dissagree about non-linear scales. They may accentuate global warming, but they are not an honest presentation of reality. They are just helpful to scare the public ! There are some good examples of this in AR5.


  10. Frank says:

    I posted this comment at Judith’s.

    Clive Best had two posts lately that beautifully illustrate the dramatic difference between natural variability and global warming. The first is a global movie of 150 years monthly temperature anomalies (HadCruT 4.3) for grid cells on a scale from 0 to +10 degC in increasingly intense shades of red and 0 to -10 in increasingly intense shades of blue – both in increments of 1 degC. Intense swathes of both colors regularly sweep across continents. Recent highly-publicized events like the cold on the East Coast of the US last two winters or the heat wave in Europe in the summer of 2003 are relatively routine events that get lots of publicity because the involve regions with lots of people and press coverage. Similar extreme events occurred before the rise in GHGs: extremely winter 1941/2 when the German army was stopped outside Moscow, July/Aug 1936 in the Central US (Dust Bowl era) or the intense cold in the Northern Rockies in Feb 1936. (The movie has a slower speed that enables the user to stop at these dates.)


    By contrast, global warming amounts to only 1 degC, the smallest step change visible in this movie! So why do maps of global warming contain so man bright red areas? Clive explains that these maps showing global use a logarithm scale, with the smallest change being step changes being 0-0.2, 0.2-0.5, 0.5-1.0, 1.0-3.0, and 3.0-5.0 degC. Trivial changes are amplified compared with


    I like to say that no individual has “experienced” global warming, because that change is trivial compared with the typical monthly variation in weather. Even more extreme events occur on the multi-day/weekly time scale.

    I doubt you have time to show a movie as part of your testimony, it would be a dramatic way to illustrate how the importance of a couple tenths of a degree of warming has been exaggerated.

  11. Bryce Payne says:

    Seems to me some of us miss the point of climate change. It actually lies exactly, or at least mostely, in variations that cannot be readily observed within recent historical variation. So, maybe use both scales as each of them aids visualization of different things. Just be specific (and clear, honest) about what the purpose and weaknesses of a method of presentation is.

    Along the same line, I commented early on about Clive’s animated time series, and, speaking for myself only, Clive’s scaling put a different (and proper and helpful) light on the data, but it does not change an apparent, and seemingly strong underlying warming trend.

  12. Pingback: Clive Best: 165 years of Climate Change | Climate Collections

Leave a Reply to Nick Stokes Cancel reply