by Tony Brown
Over the last 15 years interesting things have been happening at CET-the world’s longest instrumental record, dating to 1660.
It was partly as a result of watching its evolution that I decided to try to reconstruct it further back than Manley-creator of the 1660 record- reaching 1538 as described in my article ‘The Long Slow Thaw?’
Here is CET from 1772 to the current day, which is the preferred Met Office data
Figure 1
Parker, D.E., T.P. Legg, and C.K. Folland. 1992. A new daily Central England Temperature Series, 1772-1991. Int. J. Clim., Vol 12, pp 317-342 (PDF)
“The graph above shows annual anomalies relative to the 1961-1990 average. The red line is a 21-point binomial filter, which is roughly equivalent to a 10-year running mean.”
Some two and a half years ago I graphed the 2013 version of CET against CO2 concentrations and posed several questions. [link] Do the results demonstrate that once you get to around the 300ppm level that the law of diminishing returns sets in as the logarithmic curve of CO2 versus temperatures takes effect? Does it illustrate nothing at all and the current downward CET slope is merely a blip from the 1990’s ‘hump’ that will increase sharply again as more CO2 is added?
Figure 2 below is an update of the 2013 chart using a 5 year running mean;
As some 30 months have elapsed since the original, several things can now be better discerned. Firstly, the hump has in effect been erased (but not withstanding that 2014 was the warmest year in the 1772 record).
Secondly, there has been a number of rather cold years-compared to the recent norm and thirdly, that the upwards trend line from the start of the record can be seen. This erasure was perplexing and whilst the 15 year upwards trend was widely discussed, the subsequent 15 year decline has received little attention and although the current lower plateau is still at a fairly high level it is matched by periods in the past. To attempt to discover what is happening we graphed the seasons back to 1660 using the Manley Met office data. The result was this;
Figure 3
As can be seen, the graph uses actual temperatures rather than anomalies and shows the evolution of the seasons with a gradual warming over the centuries, although of course with fits and starts. It can be seen that whilst the Summer warming overall since 1660 has been somewhat small , Spring Autumn and Winter are much more pronounced. Note; this is a linear trend
The warming of Winter is perhaps the most notable, but least surprising feature, as the record covers the various periods of the Intermittent Little Ice Age
Clearly, not all seasons in the same year are in sync, and a run of (say) warm Winters might be counter balanced by a run of (say) cool Summers. Where they are all in sync, that will likely cause a substantially warmer or cooler year than ‘normal’. However it only needs one season to be substantially different to the others for it to materially affect that years mean temperature. Indeed one month might do it, for example this has been a very mild November in the UK that has impacted strongly on the entire years CET which, until then, had been rather cooler than usual.
Whilst the relationship of one year’s season in relation to other seasons in the same year can not be observed in the following graph, by including all the data in one graphic from 1660 this enables us to see the generality of the trends better.
Figure 4; 5 year running mean
Eyeballing the hump (figure 1) which reached its peak around 2000 and setting the subsequent trend to coincide with the decline of some half a degree centigrade since then, reveals this;
Figure 5
Clearly all seasons, other than Autumn have shown a noticeable decline, with that season appearing, for some reason, to have remained roughly static. Drilling down to each Autumn month shows this;
Figure 6
So the static nature of autumn can be shown to be comprised of a noticeable decline in September, counter balanced by rises in October and November which, overall, were not sufficient to counter the overall substantial CET decline from 2000. What is the explanation for the erasure of the ‘hump’ which certainly elicited a lot of scientific comment on its way up but somewhat less on the way down?
I asked the following of the Met Office earlier this month:
“ Is the hump and apparent decline we can observe in Hadley CET 1772 in recent decades correct and a true representation of our evolving climate, or did the change of stations and methodology with the advent of the Hadley 1772 data set in 1992 correctly reflect the weighting they should have been accorded, in particular Ringway? The apparent incongruities seem to date from around the time the new data base would have come into use.
In this respect I note the subsequent amendments to the data arising from David Parker and Briony Horton’s paper of 2005 ‘ UNCERTAINTIES IN CENTRAL ENGLAND TEMPERATURE 1878–2003 AND SOME IMPROVEMENTS TO THE MAXIMUM AND
MINIMUM SERIES’.
I received the following reply from the Met Office;
“I note your thoughts on possible inconsistencies/inhomogeneities/’incongruities’ also; of course with best scientific interests at heart, we do ensure as much as possible that any changes in the stations used (and currently we use Stonyhurst, Pershore College and Rothamsted) are allowed for by doing an overlap comparison, so that not just biases but changes in standard deviations of these stations are allowed for to get a series that is as closely homogeneous as possible.”
So the Met office-who have a lot of experience in this, believe the stations used are not showing a bias. I met compiler of the 1772 CET data base, David Parker, at the Met office a couple of years ago and it does indeed seem highly unlikely he would have created a methodology that would allow stations to be selected that would show a notable bias.
So it appears that temperatures have been declining rather than the methodology being at fault . The change point in the direction of temperatures roughly coincide with the last big El Nino. Will the current El Nino add another twist to the decline?
November has shown how susceptible our climate is to the wind direction. Winds from different points of the compass bring very different conditions, ranging from warm wet Westerlies to dry cold Easterlies (in the Winter) generally always warm Southerlies and generally always cold Northerlies. November 2015 has been typified by an extended run of South Westerly winds and has been notably mild, except on the two occasions, separated by some weeks, when the winds came from a Northerly direction and the temperatures briefly plummeted.
A future area of my research relates to updating work by Hubert Lamb who created an interesting graphic of the number of days in a year that the UK (and generally Europe) has South Westerly winds. The data begins in 1340 and ends in around 1978. I have asked the Met Office for the source of the data whereby Lambs’ graphic can be updated on a like for like basis, as the wind direction in recent decades may have a bearing on the current temperature decline.
Will it continue its decline? It has been rising for some 400 years by some measures, so taking a linear historic view I would expect it to continue to rise. However, it depends on what has caused the rise. Its start clearly predates increased CO2, but whether it is now a mix of natural variability and CO2, or whether our natural variability remains the overwhelming factor, perhaps influenced by wind direction and jet streams, amongst many other factors, is difficult to determine.
The final graphic attempts to provide a wider context to recent temperature changes. It illustrates the fluctuating nature of glaciers over the last 1000 years and superimposed is extended CET.
Figure 7
It is drawn from the numerous glacier studies made by E Roy Ladurie and published in the book ‘Times of Feast, Times of Famine’ and also encompasses more recent research by such as C Pfister. It should be seen in its generality, as the nuances of short term advances and retreats cannot be accurately captured. It covers primarily Alpine Glaciers and some from North America. The glacier recession continues to the present day. The general considerable glacial advances from around 1550 during the Little Intermittent Ice Age can be seen, although the numerous warm years, even during glacial advances, nor the cold years during glacier recession, do not show up.
It must be pointed out that the observations that are the subject of this article illustrate a short period that doesn’t really represent a significant trend, although they certainly come under the category of ‘interesting.’
During the last 15 or so years there has, by some measures, been a ‘pause’ in global warming. Taking a global average temperature may disguise the nuances of the temperature changes locally in individual countries and CET may therefore be representative of a group of countries that have shown a decline that has counter balanced continued warming elsewhere.
Notes and comments
- Statistics can often be used to prove whatever point the author wants to make, so in including 1 year linear, 5 year running mean and the 10 year running mean of the Met office itself, hopefully the reader can examine the data from a variety of angles.
- Temperatures can be seen to be rising in the earliest period of my reconstruction-(prior to the official 1772 or 1660 data) The period around 1540 included probably several of the warmest years in the record. Whether it coincides with a longer extended warm period at that time is as yet unknown to this author although in a paper ‘The year-long unprecedented European heat and drought of 1540 – a worst case’ in the journal Climatic Change an international group of 32 scientists shows that in 1540 Western Europe suffered a heat wave and ‘Megadrought’ that were broadly similar to the modern European heat waves of 2003 [link]
The compilation book ‘Climate since AD 1500’ edited by Phil Jones and incorporating work by a number of scientists, notes the warm periods around 1550 and 1630 and the cold intervals that separated them; Growing warmth is confirmed by Professor Dr C. Pfister the noted historian and geographer who identified Heat waves in 1525 and 1616 (authors note; roughly comparable to Europe 2003/2010)
- From a graphing viewpoint it is debatable as to where the CO2 line should be placed. I chose to place figure 2 around the black trend line as the CO2 and temperature trend line probably needs to start together at the same place. This also provides clarity and context to the graphic, although others might feel the CO2 line should be placed elsewhere.
- The Met office, KNMI, Mike Hulme, Hubert Lamb and Mike Lockwood were amongst those believing CET had a wider significance as a reasonable (but not infallible) temperature proxy that might reflect European Wide, Northern Hemisphere or even some sort of Global proxy. However, it is not the intention of this paper to claim that its findings and observations are relevant to any region other than Britain, although this is obviously a topic that warrants further examination at some point.
Tony Brown; November 25th 2015
Charts kindly provided by Ed Hoskins
JC note: As with all guest posts, please keep your comments relevant and civil.Filed under: Data and observations