As many regular readers know, I’ve pointed out several times the incident of the abrupt and sustained lowering of the Ap Index which occurred in October 2005. The abrupt step change seemed (to me) to be out of place with the data, and the fact that the sun seems so have reestablished at a lower plateau of the Ap index after that event and has not recovered is an anomaly worth investigating.
From the data provided by NOAA’s Space Weather Prediction Center (SWPC) you can see just how little Ap magnetic activity there has been since. Here’s a graph from October 2008 showing the step in october 2005:
click for a larger image
However, some have suggested that this event doesn’t merit attention, and that it is not particularly unusual. I beg to differ. Here’s why.
In mid December I started working with Paul Stanko, who has an active interest in the solar data and saw what I saw in the Ap Index. He did some research and found Ap data that goes back further, all the way to 1932. His source for the data is the SPIDR (Space Physics Interactive Data Resource) which is a division of NOAA’s National Geophysical Data Center (NGDC). He did some data import and put it all into a mult-page Excel spreadsheet which you can access here.
I had planned to do more study of it, but you know how holidays are, lot’s of things to do with that free time. I didn’t get back to looking at it until today, especially after SWPC updated their solar datasets on January 3rd, including the Ap Index. Looking at the data to 1932, it was clear to me that what we are seeing today for levels doesn’t exist in the record.
About the same time, I got an email from David Archibald, showing his graph of the Ap Index, graphed back to 1932. Having two independent sources of confirmation, I’ve decided to post this then. The solar average geomagnetic planetary index, Ap is at its lowest level in 75 years, for the entirety of the record:
Click for a larger image – I’ve added some annotation to the graph provided by Archibald to point out areas of interest and to clarify some aspects of it for the novice reader.
The last time the Ap index was this low was 1933. The December 2008 Ap value of 2, released by SWPC yesterday, has never been this low. (Note: Leif Svalgaard contends this value is erroneous, and that 4.2 is the correct value – either way, it is still lower than 1933) Further, the trend from October 2005 continues to decline after being on a fairly level plateau for two years. It has started a decline again in the last year.
This Ap index is a proxy that tells us that the sun is now quite inactive, and the other indices of sunspot index and 10.7 radio flux also confirm this. The sun is in a full blown funk, and your guess is as good as mine as to when it might pull out of it. So far, predictions by NOAA’s SWPC and NASA’s Hathway have not been near the reality that is being measured.
The starting gate for solar cycle 24 opened ayear ago today, when I announced the first ever cycle 24 sunspot. However in the year since, it has become increasingly clear that the horse hasn’t left the gate, and may very well be lame.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
The answer of what will Hathaway do, it now official, kick the can and hope. What else would you expect? http://solarcycle24.com/
The sun will awake suddenly on Tuesday night, while you sleep, and everything will be find. The hoax shall remain intact as long as possible. Afterall, the government needs the raxes.
This was mentioned on another thread, but check out the new Hathaway prediction for SC 24 (via solarcycle24.com)
http://solarscience.msfc.nasa.gov/images/ssn_predict_l.gif
Looks like a maximum of around 105 in late 2012/early 2013 and a peak value less than that of SC23. Earlier predictions were greater than that of SC23.
Lief had mentioned in an earlier thread that he though Hathaway might be coming around to his point of veiw. Looks like he was right.
“Jeff Alberts (10:37:02) :
This blog is not a peer-reviewed publication.
Oh yes it is! I daresay it gets more rigorous review that most climate science papers these days by the journals to which they’re submitted.”
Let’s face it – this is the new peer review!
I mean this as a general comment on the ongoing transition of the web into a critical global citizens’ forum – and, indeed, WUWT is a true trailblazer. We are at the threshold of monumental changes in the role of science in society.
The “science is settled/consensus” controversy will be seen as a watershed…
so, folks, make sure to vote on weblogawards!
PaulHClark (11:38:48) :
Did you get the Duhau and de Jager paper and if so do you have any comments?
I did miss it. Too much numerology for my taste. And building on data that is not all that secure. Both the sunspot numbers and the aa index were underestimated in the past. The plot you showed of days when aa was above a threshold is particularly sensitive to the underestimate of aa. As far as I can tell, there is nothing special about 1923 and the even-odd rule is likely an artifact in the first place [it is strange that when a ‘rule’ is violated by observations that people invent new mechanisms to explain it rather than to admit defeat]. At http://www.leif.org/research/Storms150.png you can find the number of magnetic storms deduced from a dedicated storm index [Dst]. The number of storms simply follows the sunspot number [the red curve] without any discontinuity around 1930.
More on the underestimate of aa can be found here:
http://www.leif.org/research/Analysis%20of%20K=0%20and%201%20for%20aa%20and%20NGK.pdf
If Livingston and Penn theory is right then the Hathaway curves will change into straight lines (though they are already appearing on the x axis) and we´ll have another “lost cycle”. Could it be possible?
Jeff Alberts (10:37:02) :
This blog is not a peer-reviewed publication.
Oh yes it is! I daresay it gets more rigorous review that most climate science papers these days by the journals to which they’re submitted.
Touche’
crosspatch (15:18:31)
I would point out though that any deep sea venting of CO2 is not immediately available to the atmosphere as it is released under pressure and into low temperatures.
Leif
Thanks – as ever very helpful – I am familiar with your paper on the underestimate of the aa index – hence my comment that “I note you have concerns over the aa index”.
It is also interesting to note the different thresholds that scientists use in this area having read numerous papers.
As my learning increases I can now understand how frustrating it must be for you and all your colleagues to not have good accurate data over a ‘climatic timescale’.
All that said I still found the chart in Fig 2 in the Duhau and de Jager paper to be potentially very telling. Even if we accept that there must be increasing doubt as one goes back in history there is a marked trend which very broadly fits with the climate reconstructions of Moberg and Zhang.
Moreover I think there is a suggestion that the aa index reflects a much higher level of activity since 1923. I do not think we can discount geomagnetic activity as a key element in the climate warming of the last 80 years. How such enhanced activity impacts the Earth’s climate though is a complete mystery to me at least.
Thanks again for taking the time to respond.
“MartinGAtkins (13:58:39) :
I would point out though that any deep sea venting of CO2 is not immediately available to the atmosphere as it is released under pressure and into low temperatures.”
CO2 venting at mid-ocean spreading centers is an important and non-trivial issue. I haven’t been active in the field for almost thirty years, but would be happy to delve into it if there is an interest (under another thread?).
To illustrate some of the complexities: vast temperature gradients, pressure of the water column, internal pressure of the ejected gas; and there are the properties of the various components of the CO2/carbonate system. Add to this the fact that there is dissolved organic carbon in the seawater, and that, depending on physical criteria, bursting bubbles can form new organic particulate matter…
in short, submarine volcanoes play a hugely important role in the the global carbon flux and thus Earth’s metabolism and heat budget. A great deal of deep-sea observational work is required to reach an understanding beyond the simplistic models that are built on spot-checks using invasive robot vehicles…
(and the climate change warriors want to combat, fight and control all that before having the slightest idea of the dynamics of the presumed enemy…)
PaulHClark (15:06:47) :
I do not think we can discount geomagnetic activity as a key element in the climate warming of the last 80 years.
Not discounting is not the same as having evidence for. Solar activity and geomagnetic activity in the middle of the 19th century were on par with those of the later half of the 20th. Solar cycles 9, 10, and 11 were just like cycles 23, 22, and 21. I believe temperatures were rather different.
—-
a preview would be nice.
REPLY: Yes and so would a dozen other things, but beggars can’t be choosy, and I use the WordPress.com free hosting service. Complaining to them might help, as it if often requested- Anthony
Leif Svalgaard (07:49:35) :
The 0.1% is the variation from max to min for the time where we have actual data [1978-2009]. It scales pretty well with the size of the cycle, so for the small cycles early in the 20th century it should be something like 0.05%. The cycle average would be about half of the max swing, so 0.05% for recent cycles and 0.025% for the small cycles.
I smell a rat in here somewhere, the fact that the IPCC accept this figure for one and we also have examples where the sunspot activity has been higher than cycles between 1978-2009…..perhaps some re jigging is required.
Also on the subject of the solar floor, there seems no reliable record to prove this beyond doubt. Proxy records are doubtful and aa records starting in 1868 dont go back far enough. What is required is an accurate measurement taken during the lowest point of the Maunder.
It seems very possible to me the max variation could be a lot higher than .01% or 1.3Wm2.
nobwainer (Geoff Sharp) (16:01:34) :
What is required is an accurate measurement taken during the lowest point of the Maunder.
First you tell me which year you consider the deepest part of the Maunder Minimum, then we’ll see…
It seems very possible to me the max variation could be a lot higher than .01% or 1.3Wm2.
Indeed it is possible, it even happened as the variation is 0.1%.
To say that “it seems very possible to me” is not science and is not based on anything [it seems].
Jeff Alberts :
Neither is the Royal Meteorological Society. They claim to be. But how can they “review” when there is no data or methodology provided in the submissions they publish? click
Oh, and don’t forget to vote. We need your help! You can vote once every 24 hours here: click
[The voting page may take a little time to load due to the high traffic, but once it loads, voting only takes about three seconds.]
You might be interested to know that terrible cold descended on Europe in the 1590’s ( just about the entire decade) before the onset of what is now known as the Little Ice Age and the Maunder Minimum.
Emmanuel Le Roy Ladurie , Times of Feast, Times of Famine (1967, New York, 1971).
A small new Cycle 24 sunspot has popped up today in the southern hemisphere.
Will it make it as a new sunspot is the question?
http://www.solarcycle24.com/
My back-of-the-envelope calculation says that it will take about a decade of very cold weather before we reach a tipping point and all of the global warmers become global coolers.
Leif Svalgaard (07:29:30) :
tallbloke (00:27:52) :
By the way, thanks for your Ap graph. By my ever so accurate cursory eyeballing, it does seem that the C20th was generally more active than the C19th, which is only to be expected given the sunspot numbers.
Some comments:
1) the average before 1900 is 13.9, after 1900 14.9, so the difference is not great.
2) before 1900, the observatories often lost the strongest storms [trace went off the record]. You can see that in the diminished number of spikes in the curve. I have not yet ‘made up data’ to compensate for that.
3) the Earth’s magnetic field has decreased by 10% since the 1840s. This makes the Earth a bit more sensitive to the solar wind and slowly inflates Ap [and aa] as time goes by. I have also not yet compensated for that effect.
So it is quite possible [likely in my opinion] that there is no real difference.
Paul S (02:29:00) :
the Aa index was lower in at least the first 6 months of 1912 than it was at any time in 2008, at least in the spidr data.
The ‘official’ Aa-index from SPIDR [and elsewhere] is too low [by some 3 units] before 1957.
Thanks again for all the time you are spending brainstorming this stuff with us Leif.
A couple more thoughts in response:
From the graphs it is clear that there is an upper threshold beyond which increased Rmax in the sunspot count doesn’t translate into higher Ap values. Maybe there is also a lower threshold below which the electromagnetic activity translates to less heating. The universe is full of non-linear dynamics after all. This is why I think the graph Paul Clark posted is interesting. It does no harm to look at the data in a variety of ways to see what stands out. (By the way Paul, when will we get a smorgasbord of e/m indices to play with on woodfortrees?? 🙂
If the post 1957 readings are reliable, Ap may be slightly inflated by the fact that the earth’s magnetic field has diminished, but if that lowering has been reasonably linear, we are only talking 3% or so for 1/3 of the period of record from 1840. During that period since 1957 until the current sudden dropoff, Ap has remained at averagely high levels with an increasing differentiation from TSI and sunspot numbers which far exceeds that 3%.
If the earth is “a bit more sensetive to solar wind” due to it’s lowered magnetic field, there’s a possibility it’s climatic responses to increased input and fluctuation may have been magnified in a non-linear way. If you do decide to “compensate” for that, I would hope you would present the resulting data alongside the existing graph and keep both data series available. We’ve had bad experiences of scientists who ‘adjust’ the data series in line with their theories and hide the originals. 😉
So many potential variables….
Leif (15:54:22)
The observation I would make (and it is only an observation) is that from 1870 to ~1912 the aa index was in decline and this corresponds with temperatures declining over the same period (HadCRUT3 Global Monthly Mean). see this chart from ncdc
http://www.ngdc.noaa.gov/stp/GEOMAG/image/aassn07.jpg
From 1923 to the end of the C20th there is a marked rise in aa index (especially evidenced by rising aamin) and temperatures rise. The rise is most pronounced from 1985 when the aa index depicts very active levels.
I guess we’ll have to wait and see how this pans out – if geomagnetic activity falls off and temperatures continue to rise then I will assign this notion to the bin – until then my antennae are raised.
MartinGAtkins;
“I would point out though that any deep sea venting of CO2 is not immediately available to the atmosphere as it is released under pressure and into low temperatures.”
Ocean volcanism has been pumping CO2 and other material into the oceans for as long as there have been oceans. I don’t see what difference it makes that there is a delay before it could possibly reach the atmosphere. Whatever delay factor there is, CO2 was also being emitted before it.
The current rise in CO2 COULD be the result of under sea vulcanism in the past!!
tallbloke (23:35:05)
I may be being presumptive but I am not the same person as Paul Clark (woodfortrees) so I am afraid I can’t help with your request re the e/m indices.
@tallbloke (05:07:37) who said “Chris, I’ve tweaked your graph to give you a better idea of what’s happening. Smoothing the temperature date at 1/3 of the solar cycle length brings out the solar signal in the temperature data better than smothing over the whole cycle length.”
The reason I used the same averaging (smoothing) on both solar & temperature data is that otherwise comparisons of lag are incorrect. This is because averaging introduces it’s own lead/lag effect, so the only way to be sure you are not creating spurious lead/lag is to use the same averaging.
And 11 years of averaging is the minimum needed for sun-spot data to (mostly) loose it’s cyclic nature. Hence my previous graph:
http://www.woodfortrees.org/plot/hadcrut3vgl/from:1850/mean:132/plot/sidc-ssn/from:1850/mean:132/scale:0.01/offset:-0.8
As some people have pointed out, there is a large discrepancy after 1988. However, there have been previous discrepancies, such as around 1960, which eventually disappeared. To me this simply indicates that there are other factors at work than ONLY solar, but this should hardly be a surprise.
These other factors also make it hard to be sure that solar leads temperature, which would be one requirement for solar to be the cause & temperature to be the effect.
However, the 1988 discrepancy has been growing larger for ~15 years to the present (2003.5 on the graph, as 11/2 years are ‘lost’ due to averaging):
http://www.woodfortrees.org/plot/hadcrut3vgl/from:1980/mean:132/plot/sidc-ssn/from:1980/mean:132/scale:0.01/offset:-0.8
If I were to use AGW arguments, I would say that 15 years is only “weather”, and that we need to at least 30 years to be sure!
@Stephen Fisher Wilde
I am uncomfortable using PDO, AO, etc temperature trends to predict global temperature, because PDO/etc are temperatures themselves. It is not clear to me that this is not simply circular arguments, where the result (temperature) is being used to “predict” the result (temperature)…
Chris H
These other factors also make it hard to be sure that solar leads temperature, which would be one requirement for solar to be the cause & temperature to be the effect.
Since we agree that over the long term, there is a good correlation between solar activity and global temperature (notwithstanding lags, leads and divergences introduced by 60 year oceanic oscillations, accentuated by the Phil Jones effect, James Hansen effect etc), we can safely conclude that it solar which is the cause and earth temperature which is the effect. This is because if the line of cause and effect were to be the other way round, we would be living in a very strange universe indeed where the temperature change of a small planet affected the output of a G type star from 93M miles distance.
The contention of AGW is that the divergence is caused by increased atmospheric CO2, but I contend it could equally be many other factors. If the temperature continues to fall, while CO2 continues to rise, their hypothesis is looking shaky. If the proponents of AGW say that the fall in temperature is the temporary effect of soon to be reversing cycles, they need to assess how much of the late C20th warming was due to the positive phases of these currently downturning cycles rather than CO2. But I expect we’ll have to do that for them. 😉
PaulHClark (00:57:01) :
tallbloke (23:35:05)
I may be being presumptive but I am not the same person as Paul Clark (woodfortrees) so I am afraid I can’t help with your request re the e/m indices.
Paul, apologies, it was me who was being presumptive.
If the Paul Clark I thought you were is reading, my request stands. :o)
Actually, Paul and I have exchanged a few emails, and in my last I suggested he create a facility for uploading a background image for the graph output. This would enable us to fiddle with wiggle matching to our hearts content. Or insert our favourite picture of a polar bear, frozen mammoth etc. 🙂
Leif,
“If you add 1 W/m2 for a year, the temperature for that year will be 0.05K warmer, if you add it for 10 years, the temperature for those ten years will be 0.05K warmer, if you add it for 100 years, the temperature for those 100 years will be 0.05K warmer.”
Temperature where? Is your control in a vacuum?
RICH (07:47:03) :
“If you add 1 W/m2 for a year, the temperature for that year will be 0.05K warmer”
Temperature where? Is your control in a vacuum?
Temperature on the Earth averaged globally. And on the Moon, too, and anywhere else in the Universe, for that matter.