Comparing Maunder And Modern With Long-Term Cet

[Roger J Smith]

Comparing the Maunder and modern climate data (CET) with long-term averages

____________________________________________________________________________________

There's a lot of talk about climate change and modern trends, but sometimes this happens without much reference to actual numbers. This thread is all about the numbers.

I have a data set available with monthly CET anomalies going back to 1659. I have quality checked this carefully to make sure that it matches published data.

For the "Maunder" or low-solar-activity period that began just around the time of the CET in 1659, I used the end point 1710. So my "Maunder" data set covers 1659-1710.

For the "modern" period I used 1988 to (May) 2007 and for the "modern downturn" period I used June 2007 to present. As I am reporting this near the end of May, 2011, I have taken the liberty of inserting the provisional value for May, 2011 which can be adjusted later. This "modern downturn" period is quite arbitrary, of course, and is based on a sort of consensus of subjective analysis that I have read on various forums. How long we will remain in the "modern downturn" period remains to be seen.

This leaves "long-term" or what you might want to call "background" climate as 1711 to 1987. I will show that this is a reasonable assumption by taking 1911 to 1987 as a subset and calling that "recent." So, bear in mind, all data sets are independent except that "recent" is a sub-set of "long-term." The "recent" part of the long-term climate shows that it introduced the modern warming but in a rather irregular way.

The main point of the analysis is to show that these are four rather distinct periods by the numbers. Basically, the Maunder is about half as much colder than long-term, as the "modern" is warmer. The "recent downturn" is about half way back down from modern to long-term. So the "recent downturn" climate is about as anomalously warm as the Maunder was anomalously cold. The significance of such comparisons varies with the number of years in each period. Clearly the cold Maunder was more significant than the warm "modern" due to its length more than its anomaly.

Here are the numbers, starting with "all data" the averages for all the 353 years and five months so far. The analysis is by months as well as annual.

...................Month ....J.....F.....M.....A.....M.....J.....J.....A.....S.....O.....N.....D.......YR

ALL DATA ..............3.2..3.9..5.3..7.9..11.2..14.3..16.0..15.6..13.3..9.7..6.0..4.1....9.2 ... (1659-2011)

Maunder .................2.7..3.0..4.7..7.4..10.9..14.1..15.7..15.4..12.7..9.2..5.7..3.5....8.8 ... (1659-1710)

Long-term ...............3.3..3.9..5.3..7.9..11.2..14.4..15.9..15.6..13.4..9.7..6.1..4.1....9.2 ... (1711-1987)

(19th century) .........3.1..4.0..5.3..8.0..11.2..14.4..15.8..15.5..13.2..9.5..5.9..3.9......9.1 ... (1801-1900)

.(recent)..................3.8..3.9..5.7..8.1..11.3..14.3..16.0..15.7..13.6.10.1..6.6..4.6....9.5 ... (1901-1987)

Modern ..................5.0..5.1..6.9..8.6..11.9..14.6..16.9..16.7..14.2.10.8..7.2..5.0...10.3 ... (1988-May2007)

Modern downturn ....3.7..4.6..6.5..9.6..12.2..14.8..16.1..16.0..13.9.10.7..7.0..2.7....9.9 ... (Jun2008-May2011)

Analysis

The Maunder is uniformly colder than the long-term or all data, when monthly variability is factored in. The cold anomaly remains about half a degree to a full degree Celsius throughout the year, except in the period May to August which is closer to a quarter degree. The cold signal of the Maunder fades out somewhat around May and June when strictly compared to variability. This indicates the possible importance of easterly flow, a component which would tend to favour warmer weather in May and June.

It is no surprise that long-term almost matches all data, since long-term is about 80% of all data. The "recent" part of the long-term, basically most of the 20th century, becomes about a half degree warmer from October to March, except for February which does not move upward at all (with 1947, 1956 and 1963 in that sub-set of data, I suppose the frequency of mild may have kept pace, but not the mean temperature). This shows us that the long-term climate began to shift to warmer, but only in the lower-sun half of the year. This pre-AGW warming signal fades in April and is absent (statistically) from May to September.

Then we move into the "modern" period 1988 to mid-2007 and this is quite an upturn in all months except possibly June which shows a more modest increase. The increase against the background climate is over one degree and reaches a peak of 1.8 deg in January and 1.6 in March. But the "modern downturn" period of the past four years (June 07 to May 11) drags down this warm signal in all but April to June where it continues to move upward. December, with two very cold months, has actually outdone the Maunder on average but must be considered a short-period statistical anomaly.

The interesting comparison for me is that the "modern downturn" has basically sunk back to the "recent long-term" or 20th century pre-AGW signal except for February to June and especially March to May (spring), with a second weak retention of warming in October and November. The "modern downturn" is very similar to the "recent long-term" but remains generally almost half a degree warmer than it.

Without going into a lot of science here, you could say that we have evidence for an external signal added to natural variability that could be AGW peaking in 1988-2007 and the fading out of that signal but not its complete extinction from mid-2007 to current date. But that signal was appearing in the data from 1911 onward in the colder half of the year (I tested this in decadal comparisons not shown here which is why I chose 1911 and not 1901 or any other year ending in 1).

Is this signal human activity or solar activity, or a blend? I would have to say that solar activity looks at least as plausible as human activity, but we should also consider shift of the magnetic field allowing circulation patterns to migrate. We could expect the oceans to store up some of the warmth and release it gradually for many years after the signal (whatever blend of these factors it might be) weakenes or is eliminated. So the "modern downturn" retention of some warming signal in spring matches any hypothesis of weakened signal, it merely says that in the season most responsive to ocean temperature, the climate remains a bit warmer than long-term.

So what process is underway? This is my personal opinion and not necessarily easy to support from the data analysis:

First, there is a background warming from human activity. I realize that the CET data are managed to eliminate urban heat island encroachment, so I am leaving that factor out. But I would suspect a slow increase over time amounting to 0.5 C deg give or take 0.2, from the human activity that produces greenhouse gases and the overall modification of surface that might include the venting of warmth from urban heat islands to regional and then global climate. So whatever the actual processes, human presence probably accounts for some of this variation in the data. This signal accelerated from the "recent long-term" to the "modern" periods of data and then has either slowed or plateaued in the "modern downturn" period. It should still be imparting an upward push to temperatures so if there is a downturn, other factors must be strongly negative now.

Solar activity over the longer term seems to be about an equal factor adding 0.5 C deg to the mean temperature through the 20th century (and similar signals that are likely solar are seen in parts of the 18th and 19th centuries). This solar signal has weakened since 2003 and is therefore becoming extinguished from the data in the "modern downturn" period. Given my analysis of human factors, we may be returning to a Maunder type climate in terms of circulation, but with a warmer environment that includes retention of long-established warming in ocean heat.

A third factor identified is response of the climate to the magnetic field. This shows up more dramatically in North American data where the temperatures increased sharply from the period 1851-1890 to 1891-1910 as the magnetic field lifted northward. This signal is much weaker in Europe, where magnetic latitudes remained more static. So it gives some credence to the theory that circulation may respond in part to geomagnetic forcings. Whatever that did to hemispheric climate from 1850 to 1950, the effects likely faded out after 1950 because the NMP had only a few more degrees left to move north and started to move more west than north around 1980. But I do think that the warming of the "recent" part of the "long-term" core of the data probably indicates the importance of this factor in changing hemispheric temperature -- that warmed the oceans and the warmer Atlantic showed up mainly in the colder half-year CET temperatures. As is fairly obvious from the modern downturn period, having a slightly warmer Atlantic is no guarantee of warmer summers.

I can present any other periods of interest that might be raised in discussion. For example, I worked out the averages for the 19th century (1801-1900) which is the middle 40% of the "long-term" period ... you can see that they are very similar to the averages for 1711-1987. These are added to the table but are not otherwise discussed or referenced.

Edited May 29, 2011 by Roger J Smith

[Roger J Smith]

As a picture is worth a thousand words, here's the above in graphical form. The 19th century (not shown) would be mostly in the top of the light blue "lake of cold" that represents the Maunder minimum (blue line in graph). Then note how the bulk of the 20th century (purple) begins to warm at the winter ends of the graph in particular, and then how the modern (1988-2007) period in red shows a larger increase. The orange line is the specific average for the past four years (June 07 to May 11) and in general it lies between the peak of the warm period and the "modern" period of the 20th century. But it takes quite an excursion down through the lake of cold in December and drills well into the base of the Maunder in that one month.

[johnholmes]

A fascinating read as ever with your work Roger-thank you. I do hope it does not get 'trashed' by the pro and anti AGW sides. You have shown, from your data, that there is perhaps more to the warming of the globe than just one feature. Sad though that the majority of current climatologists will probably not be prepared to discuss its pros and cons in a constructive manner.

thanks again for a though provoking read.

[Iceberg]

Hi John I am not wishing to trash anything so hopefully this post isn't viewed in that way, Roger feels very passionate about this and puts alot of time and effort in.

A few points though.

We are nowhere near a maunder wrt SI or even sunspots so I am not sure why its mentioned tbh as it's pretty irrelavent.

All the figures are really showing is that the CET has gone from it's previous record highs in the middle part of the last decade to something a little more normal. This could be due to a number of reasons but the most likely are the La Nina effects on the jet and a preponderance to a la nina based atmospheritc set up. Whether this is due to Solar, Sunspots, PDO changes, arctic ice melt etc is certaintly open to debate. However time periods and evidence are far too short to show anything really.

I would also suggest if wait out this year and look at it again all we will really have seen is a temperary blip of CET values, at least if this current year so far and GP predicted summer are anything to go by. !

This isn't a GW/anti GW thing as certain people mostly the media were guilty of doing the same thing back when the CET was at the top of the cycle.

I maybe missing something and apologies if I am, as it's still an earlish sunday morning.

[johnholmes]

thanks for your input Ice, it will be interesting to see what Roger replies with; I'm no expert so defer to both of you.

[4wd]

Sensible well rounded analysis.

Could you upload a larger graph please?

[Roger J Smith]

I agree that the four year "modern downturn" period is too short to draw many conclusions, and it is only with December that we find any data that rival the Maunder. I selected the period because it has become a general theme of climate change discussion to point to this arresting of the previous twenty years' theme of record-breaking warmth. And in fact I draw attention to the fact that the "modern downturn" period fails its own name during the spring months and continues to show upward temperature trends in those months. Yes, I also agree that it must remain a wait and see situation no matter what one's position on the debate over causes of observed climate change. With my estimate of 0.5 C deg for the AGW component of this, I find that I am actually not as much of a "skeptic" as some although I have noted that the camp which claims all of the warming to be anthropogenic would probably argue for closer to 1.0 C deg. So basically all I'm saying is that the modern downturn gives some evidence for my perspective (which to be fair I stated as early as 2006) that observed changes are a synthesis of independent factors that I attempt to analyze here and elsewhere.

What's indisputable is that climate began to warm against the rather steady "background climate" that followed the Maunder. There is remarkably little difference between the 19th century and the long-term averages, and only a bit more if you take the post-Maunder 1711-1800 period. An upward trend begins to show on a decadal time scale around 1911 from my analysis and it seems to affect the winter half of the year much more strongly at first, hardly budging the means from April to September at all. The "modern" period of apparent AGW effects can be taken as 1988 to 2007, although in North America it is more evident from 1983 to 2006. While I have always been suspicious that this is some sort of natural warming signal that might have been forced by long-term high solar activity finally overwhelming earlier trends, or some other natural forcing, I am quite willing to accept that increased greenhouse gases must be having at least enough effect to warm up nocturnal temperatures by about a degree, hence the concession of 0.5 C overall (I didn't mention this in the first post but I have always thought of the greenhouse signal as being largely nocturnal from the evidence of ever-decreasing record low minima as compared to slowly decreasing record low maxima).

I'm sorry that the graph is a bit small (if you click on it, it will at least become readable). Working on a better version of it today. Although I have it mag x2 on my own screen working on it, the file then saves it at x1 size anyway. Will just have to fill the rectangle with the numerical range to increase the size, I guess.

However, if you want to get an overview, the graph is helpful. Just do it in these steps. The green line at 0.0 represents the average monthly temperatures through the entire period. The blue line at the base is the Maunder climate period. The blue shading depicts how far below long-term normal that was. The purple or mauve line represents the pre-AGW 20th century warming and you can easily see how that is mainly in the colder half of the year. The red line represents the 20-year modern warming period. Note that it has taken the original warming and expanded it at least twice as far upward adding significant warming in the summer half year. Then finally, compare with the orange line on the graph, showing the irregular nature of the last four years and what has been mainly a downturn except in spring, and in all cases but December, a repositioning somewhere between the former slightly warmed climate in the 20th century and the recent warming episode. December forms a major exception that is in fact colder than even the Maunder for that month.

As to the statement that this summer will be very warm and will begin to erode the significance of the modern downturn, I would say perhaps, but I am not expecting this modern downturn period to last very long into the new solar cycle, probably a return to the "modern" warming period would be the most likely outcome for 2012-20 at least. It is difficult for me to find any factors that would progressively cool off northern hemisphere SST values which would be a necessary condition to turn the "modern downturn" signal into a more robust return to long-term averages. In fact I suspect that we will see some five or ten year periods that exceed the modern warming signal at least slightly in the 21st century and that we would do so even if greenhouse gases were eliminated tomorrow.

The recent colder winters, I feel, are associated with a higher variability that is now peaking in the climate system. The cause of this higher variability may be dissonance between solar and geomagnetic factors which are working in opposite directions, aggravated perhaps by greenhouse gases and albedo changes in the arctic from pollution.

[Roger J Smith]

Here's a comparison from the 18th century when a long warm spell rapidly gave way to a colder period. Conmpare these values to the "modern warming" and "modern downturn" ...

Modern ..................................5.0..5.1..6.9..8.6..11.9..14.6..16.9..16.7..14.2.10.8..7.2..5.0...10.3 ... (1988-May2007)

Post-Maunder warming ...........3.7..4.0..5.5..8.4..11.3..14.6..16.0..15.8..13.9..9.8..6.4..4.2....9.5 ... (1711-1738)

Modern downturn ...................3.7..4.6..6.5..9.6..12.2..14.8..16.1..16.0..13.9.10.7..7.0..2.7....9.9 ... (Jun2008-May2011)

Cold interval ...........................2.1..3.1..4.5..6.7..11.1..14.3..15.7..15.7..13.7..9.0..5.6..3.3....8.7 ... (1739-1746)

What does this comparison suggest?

The natural peak of warming represented by 1711-1738 increased throughout the year by about one C degree relative to the Maunder, and came closest to the "modern (AGW) warming" period 1988-2007 in April and June. It was generally about 0.5-1.0 C deg colder than that period otherwise.

The colder period that followed (which was most noticeable for eight years 1739-1746 although it tended to continue at almost those levels for several more decades) is generally similar to the Maunder, and quite a bit colder than the "modern downturn." In other words, it was a sharper downturn than we've just seen, and returned the UK climate to about where it had been in the heart of the Little Ice Age or the Maunder minimum period. However, the exceptionally cold 1740 had a lot to do with these statistics and if you exclude 1740 the numbers look a bit more similar to the modern downturn especially in terms of how far the previous warming episode cooled off. What this suggests to me is that a similar natural progression has just occurred and that the 1740-analogue winter may have been represented more by calendar 2010 in terms of one particularly abrupt change.

However, what this 18th century downturn also suggests is that we should be cautious about the solar-climate link, because around 1738-39, "normal service" had been restored in solar activity, and a peak occurred around 1737, with another peak in 1749-50. This downturn cannot be reasonably attributed to low solar activity, and neither can cold weather around 1780-84 (between two very active solar cycles). There is apparently some other natural signal besides solar activity that can bring on a colder climate. Perhaps low solar is a reliable driver of colder climate (as with the Maunder, Dalton, and perhaps recent few years), but some other factor is also active and shows up at other times.

Final comparison for this post, how does the Dalton minimum compare to the Maunder?

Check these figures out:

Maunder ...............2.7..3.0..4.7..7.4..10.9..14.1..15.7..15.4..12.7..9.2..5.7..3.5....8.8 ... (1659-1710)

Dalton...................2.3..3.9..5.2..7.6..11.3..14.1..15.5..15.2..12.7..9.6..6.0..3.3....8.9 ... (1806-1823)

These are remarkably similar, February and March were milder in the Dalton, while December and January were somewhat colder than the Maunder, but the overall trends are very similar.

[Optimus Prime]

I've generally considered the current warming trend to be associated with, by the majority, human activity. Human activity may not drive the climate too much (on a scale of all variables I'd say the human influence is 3.0% as a total) but that human activity influences parts of the ecosystem that are most prone to the change of the climate. While human activity might in itself only contribute a small share of what's going on, it's manipulation on all the climate factors, particulary the 13.0 gigatons of CO2 we release per year and the deforestation/deserification is magnifiying and possibly driving the natural cycle. Almost like powering an engine beyond its capacity.

Does that make any sense?

And a great thread by the way.

[Weather-history]

I'm baffled why December in the last 20 years has shown a distinct lack of warming and if anything seems to be bending the other way.

[Iceni]

I'm baffled why December in the last 20 years has shown a distinct lack of warming and if anything seems to be bending the other way.

Thought I'd throw this into the mix… make of it what you will.

http://www.climatelogic.com/forecasts/review-winter-2010-europe.html

[Thundery wintry showers]

I'm baffled why December in the last 20 years has shown a distinct lack of warming and if anything seems to be bending the other way.

I'm almost certain that it's due to atmospheric circulation. We went through a spell of unusually cold Decembers in the 1960s which were characterised by an anomalously high frequency of northerly winds. This trend reversed during the 1970s and 1980s which contained a large number of warm Decembers characterised by frequent south-westerlies. The Decembers of the 1990s and 2000s then reverted back to the 1960s type pattern. The contrast between the anomalous prevalence of warm synoptics in December between 1971 and 1988, and cold synoptics since then, has resulted in no warming trend.

Conversely the Januarys and Februarys of the 1990s saw an unusually high frequency of westerly and south-westerly winds, and then merely reverted to nearer-normal synoptics during the 2000s leaving the Decembers standing out as anomalously cool.

[North-Easterly Blast]

I'm almost certain that it's due to atmospheric circulation. We went through a spell of unusually cold Decembers in the 1960s which were characterised by an anomalously high frequency of northerly winds. This trend reversed during the 1970s and 1980s which contained a large number of warm Decembers characterised by frequent south-westerlies. The Decembers of the 1990s and 2000s then reverted back to the 1960s type pattern. The contrast between the anomalous prevalence of warm synoptics in December between 1971 and 1988, and cold synoptics since then, has resulted in no warming trend.

Conversely the Januarys and Februarys of the 1990s saw an unusually high frequency of westerly and south-westerly winds, and then merely reverted to nearer-normal synoptics during the 2000s leaving the Decembers standing out as anomalously cool.

After the cold 1960s Decembers, the switch around 1970 to a run of mild Decembers that lasted through the next two decades looks so dramatic in the CET series. I wouldn't say that the 1990s and 2000s Decembers reverted back to the 1960s type pattern - the last two decades have seen rather more of an "average" run of Decembers after the mild run through the 1970s and 1980s. The Decembers of the 90s and 2000s were nothing like as cold overall as the 1960s Decembers. Basically for December we were cold in the 1960s, mild during the 1970s and 80s, then average in the 1990s and 2000s.

January and February have been a bit like December in a delayed fashion. Between 1977 and 1987 we went through a spell of often unusually cold Januarys and Februarys, containing an anomalously high frequency of northerlies and easterlies (similar to the 1960s Decembers). This trend for January and February completely reversed during the 1990s and 2000s with both decades containing a large number of warm Januarys and Februarys dominated by frequent south-westerlies. (it was similar for Jan / Feb in the 90s / 00s as it was for December in the 70s / 80s).

No, January and February have not reverted to nearer normal synoptics in the 2000s - January, all the way up to 2008, contained often persistent south-westerlies. February, after 2002, became a bit nearer normal compared to 1988-2002, although often still with warm synoptics - even Feb 2011 was a very mild south-westerly month, which shows that mild SW'ly Febs have not been far away to this present day.

Edited June 14, 2011 by North-Easterly Blast