Guest post by David Archibald
The first prediction of the current climatic minimum was made by Hubbert Lamb in 1970 in a report (Weiss and Lamb) for the German Navy. He did it by making a reconstructed record of the average frequency of south-westerly surface winds in England since 1340. Quoting Lamb “We sense a cycle or periodicity of close to 200 years in length.” and “There may be a valuable indication of the origin of this apparent 200 year recurrence tendency, in that the sharp declines of the south-westerly wind indicated in the late 1300s, 1560s, 1740s-1770s and now, in each case fell at about the end of a sequence of sunspot cycles which built up to periods of exceptionally great solar disturbance (around 1360-80, the 1570s, the 1770s, the 1950s and more recently). The frequency maxima of the south-westerly wind, and evidence of warm climate periods in Europe sustained over several decades, all bear a similar relationship to these variations of the Sun’s activity.”
Following is Figure 11.6 from Lamb’s 1988 book “Weather, Climate and Human Affairs”:
The frequency of the southwest wind at London is shown by the solid line. A tentative forecast (broken line) is made simply by moving the whole curve 200 years to the right, i.e. the forecast implied by accepting the apparent 200 year recurring oscillation shown by the series.
The most successful prediction of the current minimum, in terms of lead time and detail, was made by two researchers in the US later that decade. Using tree ring data from redwoods in Kings Canyon in California, in 1979 Libby and Pandolfi forecast that, “by running this function into the future we have made a prediction of the climate to be expected in King’s Canyon; the prediction is that the climate will continue to deteriorate on the average, but that after our present cooling-off of more than the average decay in climate, there will be a temporary warming up followed by a greater rate of cooling-off.”
In a Los Angeles Times interview, Libby and Pandolfi gave a more detailed forecast:
“The forecast is for continued cool weather all over the Earth through the mid-1980s, with a global warming trend setting in thereafter for the rest of the century – followed by a severe cold snap that might well last through the first half of the 21st century.”
“Both the isotope record and the thermometer record show neat agreement for the cold decades at the ends of the 17th and 18th centuries, when temperatures fell by 1-10th to 2-10ths of a degree.”
“More recently, the world has enjoyed an agricultural boom during the past 70 years or so. The Earth’s annual average temperature has risen by about 1 to 1½ degrees, about as much of an increase as the decrease during the Little Ice Ages, during this interval.
When she and Pandolfi project their curves into the future, they show lower average temperatures from now thorugh the mid-1980s. “Then,” Dr. Libby added, “we see a warming trend (by about a quarter of 1 degree Fahrenheit) globally to around the year 2000. And then it will get really cold – if we believe our projections. This has to be tested.”
How cold? “Easily one or two degrees,” she replied, “and maybe even three or four degrees.”
The remarkable thing about the Libby and Pandolfi prediction is that they got the fine detail right, up to the current day, which gives a lot of credence to their projection for the next fifty years.
In 2003, two solar physicists, Schatten and Tobiska, published a paper which included the following prediction: “The surprising result of these long-range predictions is a rapid decline in solar activity, starting with cycle #24. If this trend continues, we may see the Sun heading towards a “Maunder” type of solar activity minimum – an extensive period of reduced levels of solar activity.”
The next prediction of the current minimum was made by Clilverd et al in 2006 using low-frequency solar oscillations:
Clilverd predicted that Solar Cycles 24 and 25 would have amplitudes similar to that of Solar Cycles 5 and 6 of the Dalton Minimum before a return to more normal levels mid-century.
A Finnish tree ring study (http://lustiag.pp.fi/holocene_trends1000_INQUA.pdf) followed in 2007 – Timonen et al. This is a portion of a figure from that study showing a forecast cold period starting about 2015 that is deeper and broader than any cold period in the previous 500 years:
Summary
Libby and Pandolfi provided timely warning of the current cooling more than thirty years ago, through the proper use of tree ring data. Given the enormous societal and financial consequences of that cooling, it would be good application of climate research funds to have a number of groups replicate and update the Libby and Pandolfi work.
References
Clilverd. M.A., Clarke, E., Ulich, T., Rishbeth, H. and Jarvis, M.J., 2006 “predicting Solar Cycle 24 and beyond” Space Weather, Vol. 4, So9005, doi:10.1029/2005SW000207
Libby, L.M. and Pandolfi, L.J. 1979, Tree Thermometers and Commodities: Historic Climate Indicators, Environment International Vol 2, pp 317-333
Schatten, K.H. and W.K.Tobiska 2003, Solar Activity Heading for a Maunder Minimum?, Bulletin of the American Astronomical Society, 35 (3), 6.03
Timonen, M., Helema, S., Holopainen, J., Ogurtsov, M., Eronen, M., Lindholm, M., Merilainen, J and Mielikainen, K. 2007 “Climate patterns in Northern Fennoscandinavia during the Last Millenium” Xvii INQUA Congress
Weiss, I. and Lamb, H.H. 1970 ‘Die Zunahme der Wellenhohen in jungster Ziet in den Operationsgebieten der Bundesmarine, ihre vermutliche Ursachen and ihre voraussichtliche weitere Entwicklung, Fachlich Mitteilungen, Nr. 160, Porz-Wahn, Geophysikalisher Bertungsdiesnt der Bundeswehr.
Thanks to David once again for the warning. We, in central MN, are in our second year of serious gardening, and putting food by, purchased a superior light-table to get a jump on our abbreviated season. Fortunately the 2012 winter drought ended with normal spring rains and McCabe, 2004 seems to forecast acceptable precipitation whatever befalls the high plains, south and south west.
Pity the poor Left Coast Luddites.
A brief reconstruction of key charts in Archibald’s “Solar Cycle 24: Implications for the United States” (2008).
http://rhinohide.wordpress.com/2012/05/20/archibald-2008-redux/
This leads me to speculate that ice ages could be linked to a combination of oceanic circulation cycles and solar activity – if the cold ocean water surfaces at the same time as a solar grand minimum there is a perfect scenario for dramatic atmospheric cooling which in turn leads to precipitation and if that precipitation freezes on land we have a negative feedback from albedo to add to the mix of the cold ocean cycle an and possible solar influence via Svensmark GCR cloud seeding aerosols in the extended solar minimum – we live in interesting times and have a wonderul opportunity to learn about how our climate works ahead of us. I hope i’m gone before the next ice age though!
Lets see.
Lamb took a past record and spliced it onto a current record by wiggle matching the humps.
People like that method because they like the answer.
In mann used that method, folks would pitch a fit.
Libby, L. M. & L. J. Pandolfi. (1974) Temperature Dependence of Isotope Ratios in Tree Rings
http://www.real-science.com/libby-and-pandolfi-were-correct
Steven Mosher says:
May 20, 2012 at 11:02 pm
In mann used that method, folks would pitch a fit.
======
Splicing two different proxies/instruments, with two inherently different frequency (bandpass) responses is generally regarded as a big no-no in fields outside of climate science. Especially when the splice is made in the temperature domain and temperature is the dimension of interest.
Lamb’s hasn’t done this. He displaced a single data set in the time domain, not the temperature domain, to see if the data was cyclic and thus suitable for forecasting. There is nothing controversial about this. It is routinely done in many areas outside of climate science.
The important point is that Lamb’s technique is not intended to say the earth has warmed by a fraction of a degree over X years. The technique simply says that if the past is an indication of the future, then we can expect future climate to mimic the past.
What I’ve always found interesting about the tree ring data, is that it peaked around the same time as the solar maximum, and temperatures leveled off some 10+ year after. Had mann et al done their splice in the time domain rather than the temperature domain they may well have caught this and successfully predicted this.
The divergence problem is only a problem if you assume that temperatures post 2000 were going to continue to increase and accelerate. However, if trees are a proxy for solar activity, not temperature, then the divergence makes perfect sense after you allow for lag due to climate inertia.
The tree ring divergence successfully predicted that temperatures were going to level off. However, experimenter bias prevented this message from being correctly interpreted. Instead the assumption was made that something had gone wrong with tree rings as a prediction tool.
Tree rings are not a proxy for temperature. They (plants) are a proxy for solar activity, as Herschel noted some 200 years ago. Photosynthesis is driven by sunlight, not heat.
Temperature is also a proxy for solar activity, so at best trees can provide an indirect proxy for temperature.
Steven Mosher says:
May 20, 2012 at 11:02 pm
That’s a rather ridiculous and silly statement Steve – as the term splicing together isn’t really applicable to Lamb? Just replicating the past data and projecting it forward is a pretty standard way of looking at forecasting all kinds of stuff !
By direct contrast, Mann chopped and changed using selected data and grafted on a ‘more suitable’ dataset (i.e. instruments) to try and make a point – hardly the same at all….
and as to your statement ”people like that because they like the answer” – that’s even more crass, as there is no real ‘answer’ in projections/forecasts but Mann was trying to scare the world with his ‘method’…..ideological thinking at it’s worst, deliberate misinformation at best….
I see your point (I guess?) – that such a method is hardly deeply ‘scientific’ – but as you are so fond of reminding us all – thats what models are all about – in that sense, the best model ‘representation’ is of course, the past observations as they are the ‘real thing’ (assuming they are valid observations of course! – you know, not like, just one tree sample or anything! LOL). For cyclical type things, such a simple method is perfectly reasonable as a forecasting tool……especially as in this case he used ONE dataset!
Mods – a post to Pamela has gone missing, posted just before my last post to Gail – please would you have a look for it?
DirkH (May 20, 2012 at 4:28 pm) commented/asked about Solar-Terrestrial-Climate Weave http://i49.tinypic.com/2jg5tvr.png “Fascinating, Paul – the checkerboard pattern repeats after 22 years (a full magnetic reversal cycle). What variable do you show? Simple temperature anomaly?”
Not temperature anomaly. Basically you’re looking at normalized lunisolar-neutralized Schwabe-extent semi-annual-grain global-WIND anomalies, which inform about the equator-pole temperature GRADIENTS which drive them.
The weave can be isolated from BOTH daily atmospheric angular momentum (AAM) AND daily length of day (LOD) records using DOZENS of methods. IT’S ROBUST. You can be SURE that anyone objecting to its existence is operating from a base of ignorance &/or deception.
Piers Corbyn works with the event series which underpin the weave. NASA has had opportunity to be aware of the existence of the weave since at least 1997.
There are clean, beautiful cross-scale connections to both interannual & multidecadal timescales.
—
@Ulric Lyons (May 20, 2012 at 5:18 pm)
The solar wind records you use — where are the measurements taken?
kevinUk
Mosh has told me he does not really have much time for Lamb and Parker (who gave further detail to the CET record) and is somewhat sceptical about the value of contemporary observations. He likes maths/computers and replicable data (although I often try to point out that the basic data is suspect) . That is all fair enough and I admire Mosh’s work but it is only one line of enquiry and I think that knowing about the past can tell us things about the future. Ideally anecdotal material needs to be married up to scientfic studies, when you then have a very potent combination. For example I just spent the day at Exeter Cathedral going through their records. for the period 1279 to 1353.
Of note during that period is that;
‘‘the work force was much reduced in number on account of the weather (during the winter) though work does not seem to have ever altogether ceased on this account.’
That sort of information is interesting in itself but needs to be tied in with other information before we can say that during this period there were no severe winters.
However the anecdotal directly complements other evidence in this entry;
‘ 1783- extra poor relief in extreme cold. ‘
A quick google will show the reasons this was needed.
So the approach of people like Lamb was extremely useful but has become unfashionable, but I in my very modest way believe that the approach of merging sciece and anecdotal evidence is still worthwhile and try to continue it in my own modest work.
tonyb
This is just crap and it does nothing to our credibility to reproduce this kind of rubbish.
The key thing to do is look at the frequency graph and see whether the size of any of “dominant” frequencies is statistically significant. The answer is that not only are they not statistically significant but they are completely normal. In other words, draw a smooth line from the left to the right of fig 2 (a hump) … then plot a line through the lowest peaks and through the highest peaks (you’ll get a crescent) and then see if anything at all pokes through.
ANSWER: nothing at all is abnormal.
It looks like pure noise. Completely natural & random variation. This is a lot of hullabaloo about noise.
In other words: this graph has zero predictive power. Or to turn it around these researchers haven’t a clue what they are talking about. It is the same standard statistical nonsense as Mann’s Hockey stick. More apt, it is the same statistical nonsense as the Boeker paper that invented “global warming” … again this was a frequency analysis by statistical incompetents who thought a bit of noise was some kind of cycle.
What is it about Fourier transforms that turns scientists minds into jellies and make them statistical morons?
A graph showing frequencies is no more difficult to analyse statistically than any other graph, but as soon as it gets transformed “scientists” think there is some kind of magic that allows them to throw out the rules of statistics.
Before these idiots report this kind of nonsense, they have to answer the question: are any frequencies statistically unusual. If as this case, they are entirely within the range of all other frequencies, then they are either total incompetents for reporting this or fraudulent.
Steven Mosher says: May 20, 2012 at 11:02 pm
In mann used that method, folks would pitch a fit.
There are two usual indicators of poor work in climate science:
1. Splicing results
2. Using Fourier transforms to “discover” cycles.
No folks. Fourier transforms do not discover cycles, they create cycles. The technique is simple: take random pink noise, convert it to the frequency domain, throw away every bit of signal that gives it a natural look (high frequencies) and then convert it back to show you have discovered that … when you haven’t a clue what you are doing anyone can make data have cycles.
I am almost tempted to go and look at the papers suggesting they have discovered cycles and show that all these cycles cluster around 10% of the sample length. No … on looking it is closer to 20%
I am really annoyed at this nonsense paper. People who have no clue how to interpret frequency graphs … have about as much knowledge of frequency as they have of Tocharian … shouldn’t be allowed to publish their childish gibberish nonsense. They certainly should be forced to do a basic course like electronics where this kind of analysis is (was) taught as a matter of course.
It is about as useless as taking a sample of boulder clay and saying that the rocks in the boulder prove it is unusual because most of the sample is clay. Yes most of it is clay (by number), yes the graph of clay size will increase steadily only to descend into a mess when sampling boulder samples, because there are so few large grains (boulders) there is more statistical noise and/or bigger boxes (hence the way the graph spreads out on the right).
Likewise, sampling frequencies. There are a lot more high frequencies in this kind of sample, so the graph starts smoothly from the left and then starts to get “grainy” to the right with wide-spaced highs and lows, but all you need to do is imagine the right side squeezed up as tight as the left, and you will see that it is a smooth continuum of noise.
“Tree rings are not a proxy for temperature. They (plants) are a proxy for solar activity, as Herschel noted some 200 years ago. Photosynthesis is driven by sunlight, not heat.”
That could make tree rings a proxy for global cloudiness too.
So if the level of solar activity affects the global air circulation and consequent cloudiness then that would turn up in the tree ring data.
Pretty much squares the circle in my view.
It would have been increased global cloudiness during weaker cycle 20 that reduced tree growth for a while and caused the apparent divergence problem.
What does tree growth show since then ?
Does anyone know of any paper which graphs the N-S movement of the timberline (which is likely to be a climate proxy) rather than tree rings (which is likely to be a proxy for tree-tree competition response to climate …. which is next to useless for long period changes)
“This leads me to speculate that ice ages could be linked to a combination of oceanic circulation cycles and solar activity – if the cold ocean water surfaces at the same time as a solar grand minimum there is a perfect scenario for dramatic atmospheric cooling .”
I’ve been saying as much in my articles for several years. Climate swings could well be a consequence of solar and oceanic variability moving in and out of phase.
However I think that on longer timescales the Milankovitch cycles are primarily responsible for the ice ages themselves whilst the sun/ ocean phasing variations would account for the climate variations within both ice ages and interglacials.
I have previously suggested that relative climate stability during interglacials could be a consequence of a short period when sun and ocean cycles were largely offsetting one another to reduce climate variability.
During glacial epochs the sun / ocean variations more often supplement one another to produce much greater variability.
Due to current landmass distribution the sun / ocean phasing only results in relative stability for only about 10% of the time hence the approximate 9 : 1 dominance of ice ages for the past severqal million years.
I have had that proposition in the public domain for several years now.
“The dependence is complicated, but the bottom line is the jet stream is weaker when there’s less ozone (it has to do with latitude-dependent temperature gradients across the upper atmosphere; those gradients are strong in winter and weak in summer). Ozone creation depends on UV from the Sun, which is weaker during a solar minimum. See where this is going? Weaker magnetic activity on the Sun means less ozone which means a weaker jet stream which means it meanders more, bringing cold air south in some places.”
Very much a reflection of my previous work but they have the sign of the solar effect wrong.
Ozone above 45km decreased to give cooling aloft when the sun was more active and is increasing now that the sun is less active.
See work by Joanna Haigh and others.
Furthermore I have pointed out that the weaker more variable jetstream MUST be accompanied by stratospheric warming and not cooling because that is what pushes the jets more equatorward as witness the events observed in shorter term ‘sudden stratospheric warming’ events.
So, the only scenario that fits observations is contrary to established climatology. The stratosphere must cool naturally when the sun is active and warm naturally when the sun is less active.
It is because established assumptioins were wrong (and still are) that it was necessary to have recourse to the theories about the CFC effects on ozone and the CO2 effects on stratospheric temperatures. Only with those theories could the observations be accounted for.
But if one simply reverses the sign of the solar effect above the tropopause it all becomes clear as to what is going on.
Ron Broberg says:
May 20, 2012 at 10:37 pm
It is very flattering for someone to replicate work you have done four years ago, and I thank you even though you are a warmer.
Scottish Sceptic:
I agree all that you say at May 21, 2012 at 2:18 am. However, with respect, I point out that you state the issue in so extreme a form that you risk ‘throwing out the baby with the bathwater’.
For example, you say;
“Likewise, sampling frequencies. There are a lot more high frequencies in this kind of sample, so the graph starts smoothly from the left and then starts to get “grainy” to the right with wide-spaced highs and lows, but all you need to do is imagine the right side squeezed up as tight as the left, and you will see that it is a smooth continuum of noise.”
True, and the human mind is very good at discerning patterns. Indeed, people can ‘see’ patterns (e.g. cycles) which do not exist. However, observation of apparent cycles can lead to hypotheses which induce study to determine mechanisms. Often the study reveals that the apparent cycles are merely random fluctuations, but sometimes such study leads to important knowledge (e.g. the mechanisms which create day and night, and the seasons).
So, discernment of patterns has often led to fundamental scientific discoveries. And discernment of apparent cycles in climate is observation of a possible pattern in climate behaviour.
Two such apparent cycles have obvious importance; viz. an apparent ~1,000-year cycle and an apparent ~60-year cycle.
These cycles are apparent because there is data which indicates there was
• the Roman Warm Period (RWP),
• then the Dark Age Cold Period (DACP), then the Medieval Warm Period (MWP),
• then the Little Ice Age (LIA), and
• then the Present Warm Period (PWP),
and also there is data which indicates global climate
• cooled from ~1890 to ~1910,
• then warmed from ~1910 to ~1940,
• then cooled from ~1940 to ~1970,
• then warmed from ~1970 to ~2000, and
• then cooled from ~2000 to the present.
As you say, these cycles are merely apparent so may be misleading. However, they may be a result of real climate mechanisms. Assuming they are not misleading, then they indicate that either
• global temperature will not rise until ~2030 before rising towards temperatures of the MWP
or
• global temperature will start to cool towards temperatures of the LIA.
Importantly, either of those outcomes would falsify the AGW hypothesis and would strongly suggest the existence of climate mechanisms which generate the cycles and, therefore, warrant investigation.
Richard
Ulric Lyons says:
May 20, 2012 at 7:32 pm
Well from my analogues 2015 to 2024 looks like it will the longest cold period this century, and several poor seasons are definitely going to impact agricultural production. Some decent grain reserves would be very much appreciated in 4+ years time.
– can you elaberate on the theory behind your “analogues”?.
I have no idea what is meant here.
Paul Vaughan says:
May 21, 2012 at 12:45 am
“Piers Corbyn works with the event series which underpin the weave. NASA has had opportunity to be aware of the existence of the weave since at least 1997.”
Thanks a lot, Paul. I was looking for possible correlations to explain Piers’ work (which would be impossible if, as Pamela assumes, the weather oscillations were just oscillating freely without an outside controlling influence). This looks enormously important.
Stephen Wilde says:
May 21, 2012 at 3:05 am
“Furthermore I have pointed out that the weaker more variable jetstream MUST be accompanied by stratospheric warming and not cooling because that is what pushes the jets more equatorward as witness the events observed in shorter term ‘sudden stratospheric warming’ events.”
A SSW is accompanied by simultaneous cooling in the equatorial stratosphere:
http://www.cpc.ncep.noaa.gov/products/stratosphere/temperature/archive/30mb9065_2010.gif
http://www.cpc.ncep.noaa.gov/products/stratosphere/temperature/archive/02mb6590_2010.gif
http://www.cpc.ncep.noaa.gov/products/stratosphere/temperature/archive/30mb2525_2010.gif
http://www.cpc.ncep.noaa.gov/products/stratosphere/temperature/archive/02mb2525_2010.gif
This temperature differential between polar and equatorial stratosphere seems to be apparent in only one hemisphere at a time.
@- “When she and Pandolfi project their curves into the future, they show lower average temperatures from now thorugh the mid-1980s. “Then,” Dr. Libby added, “we see a warming trend (by about a quarter of 1 degree Fahrenheit) globally to around the year 2000. And then it will get really cold – if we believe our projections. This has to be tested.”
I feel a bit like the boy in the Hans Christian Anderson story of the boy who pointed out the lack of apparel…
The prediction in 1979 that it would cool until the mid/late 80s, then warm by ~0.3degC, the cool from 2000 is … Well, rubbish.
That is not what happened. Last year, and this year are vying to be the hottest La Nina years in the instrumental record. Each decade since 1979 has been warmer than the last by at least 0.13degC
Another prediction was made around the end of the 70s, start of the 80s. This one was not based on past pattern matching or speculative projections of possible solar activity. It was based on the measured rise in CO2 and the basic physics of the GHG effect. It provides a graph to show, from its perspective in 1980, when the predicted warming from CO2 will emerge from the ‘noise’ of natural variation by one and two sigma of statistical significance.
It is MUCH more accurate than the prediction by Libby and Pandolfi.
Of course I recognise that it directly contradicts many peoples beliefs about the climate, and provides no support for a prediction that ‘any moment now’ it will start cooling perhaps almost as fast as it has warmed for the last four decades.
http://thedgw.org/definitionsOut/..%5Cdocs%5CHansen_climate_impact_of_increasing_co2.pdf
Have a close look at the graph in fig7. If you can overlay the actual land temperature data since 1950 and you can see just how good a prediction it is.
If anything something of an underestimate, but not to the extent that Libby and Pandolfi got it wrong!
“A SSW is accompanied by simultaneous cooling in the equatorial stratosphere:”
I know. To be more specific what is required is warming of the stratosphere towards the poles relative to the temperature at the same height over the equator.
Anyway, overall, there was stratospheric cooling when the sun was active and now with the less active sun that cooling has stopped and there may now be some warming.
Fabulous post.