Guest Post by Willis Eschenbach
Inspired by a claim made on WUWT that
A new study led by Professor K.M. Hiremath of the Indian Institute of Astrophysics shows the strong, possibly causative correlation between variations in solar activity (red curve) and in monsoon rainfall (blue curve) in Figure 1.
I decided to see what I could find out about the Indian monsoon. I thought I might tell the story by describing the path I walked. I started out with a huge advantage—I knew nothing about the timing, size or pattern of the Indian monsoon rains, other than that they occur during the summer. I knew that when land gets hot in summer, hot air rises, wet air is drawn in from the ocean. Result? Monsoon rainfall. Other than that I had no knowledge, a great advantage in exploration of a new dataset. Preconceptions are the enemy of science …
A Google search led to the Hockey Schtick, which fortunately reproduced two pages from the study. In turn, that identified the data source as the Indian Institute of Tropical Meteorology. Another google search located the IITM data page. I used the “Sontakke” dataset for this analysis, the page is here.
Now, on that page the rainfall data for India is divided into seven regions, three on the peninsula and four in the north. The Hiremath et al. study used all three peninsular datasets, plus the north-west region. Simply because it was higher in the list, I started by analyzing the dataset for North West India (NWI). The Sontakke dataset ends in 2006, unfortunately, but it’s what the authors used. Here are the last two decades of that record:
Figure 1. Rainfall in the North West region of India over the two decades 1987-2006. The monsoon rains come in June, July, August, and September. The top panel is the raw data. The middle panel is the average monthly component of the rainfall. The lower panel is the raw data minus the seasonal component. The trend over this period is not statistically significant. The blue line is the loess average of the data.
Yes, there is a huge difference between the four monsoon months and the dry two-thirds of the year. There is also no statistically significant trend in the two decades of data. Having seen that, I took a look at the same analysis, but for the entire period of the data.
Figure 2. As in Figure 1, but for the longer period of 1844 to 2006.
What stands out in the full dataset is the lack of much long-term variation at all. There is no significant long-term trend to the data, nor any obvious variation over the century and a half of data.
Is there a solar signal in there? Well … perhaps, but if so it’s neither large nor obvious. However, this shows the whole year, not just the monsoon months (JJAS) analyzed by Hiremath et al.
So I then looked at just the monsoon months. Why did I not start with just the monsoon months? Because I first want to see the entire signal before I start sub-setting it.
In any case, here is the total rainfall of just those four months, year by year, compared with the sunspot record.:
Figure 3. Rainfall and sunspots. Upper data is the total of the monsoon rains (JJAS) for that year. Lower data is the total sunspot count by year.
At first glance, that looks kinda hopeful … but a closer examination shows that significant correlation simply doesn’t exist. Perhaps the simplest way to demonstrate this is the cross-correlation function. This shows the correlation at a variety of lags, with positive lags showing rainfall lagging the solar changes.
Figure 4. Cross-corrlation of sunspots and NW Indian monsoon rainfall.
As you can see, the correlation at all lags is trivially small.
Now, does this show that the paper by Hiremath et al. is wrong? By no means. They looked at an average of four monsoon areas, and I’ve only looked at one of them, NW India. I haven’t examined the rest. However, it isn’t looking good for the solar theory.
I also haven’t taken a close look at their formula relating the solar activity to future rainfall. Why not? Well, I don’t have their paper.
In addition, I couldn’t verify the following from the abstract of the paper:
Those internal forcing variables are parameterized in terms of the combined effect of external forcing as measured by sunspot and coronal hole activities with several well known solar periods (9, 13 and 27 days; 1.3, 5, 11 and 22 years).
Instead, here’s what I found for the cycles inherent in the data.
Figure 5. Periodogram of the NW Indian Rainfall.
Note that there are no long-period cycles that are larger than two percent of the peak-to-peak swings of the data, so we’re way down in the noise. This is trivially small. In addition, there are no peaks at the periods mentioned of 1.3, 5, 11, or 22 years …
So, that’s the investigation to date. Still lots to do. I’ve only looked at one of the four datasets so far. Hiremath et al. looked at the total for the four areas. And I don’t have a copy of the underlying Hiremath paper, so I’m doing my own investigation.
Sadly, my time is quite short these days. I’ve taken a new job, as is my wont, but it has bizarre hours—5:30 AM until 2 PM. Now me, I’m a night owl by nature, so this has played havoc with my available time.
In addition, I’m doing strenuous physical work. We’re doing a rebuild on the lobby and facade of the local movie theater, and there has been a lot of demolition work. The lobby has 11-foot (3.3 metre) ceilings, which we’re ripping sections out of and rebuilding. Much of the work is up in the ceiling, or working up on a two level scaffolding … another part of life’s rich pageant. Now me, I’m sixty-seven, what I call my “middle youth”, so this is, well, somewhat consuming.
Not that I’m complaining, mind you. People sometimes ask me what I do for exercise … I say “I don’t pump iron … I pump wood.” So I’ve been spending my days in the gym, climbing ladders, doing the low crawl in the ceiling, and pumping wood.
However, no matter how cardiovascularly stimulating my work might be, to date it has cut heavily into the time for my climate research and investigations. So, for a while at least, I’ll be contributing less to the discussion.
My best to all,
w.
The Usual: If you disagree with someone, please have the courtesy to quote the exact words that you disagree with. That way all of us can be clear exactly what you are objecting to.
Data and code: The data (including the three datasets I haven’t analyzed yet, plus other indian rainfall datasets) and R code are all in a zipped folder called “Indian Rainfall Folder“. To run the code, set your R workspace to the folder, and it should be pretty turnkey.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Many studies find the link between Asian monsoons & solar activity on various time scales. I’ve previously linked some on this blog, but here are some others:
http://www.sciencemag.org/content/308/5723/854
http://www.nature.com/nature/journal/v411/n6835/full/411290a0.html
http://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-13-00482.1
Solar influence on the Indian summer monsoon during the Holocene
Anil K. Gupta,1 Moumita Das,1 and David M. Anderson2
Received 23 February 2005; accepted 27 July 2005; published 10 September 2005.
[1] The large (8%) changes in the past seasonal
insolation have a well-documented influence on the
Indian summer monsoon. However, the effect of the
small (<1%) decade to century scale solar variability is
less certain. Evidence is emerging that Earth’s climate is
sensitive to small changes in solar output on centennial
time scale during the Holocene. Comparison of a recently
published proxy record for sunspot activity with our
newly-revised higher-resolution record of the Indian
summer monsoon winds reveals multiple intervals of
weak summer monsoon during the Holocene at
multidecadal to centennial scales. Weak summer
monsoon winds correlate with reduced solar output. Our
results suggest that small changes in solar irradiance can
bring pronounced changes in the tropical monsoon. The
multidecade to century scale variations in the monsoon
winds were much larger in the early Holocene coincident
with increased sunspot numbers. Citation: Gupta, A. K.,
M. Das, and D. M. Anderson (2005), Solar influence on the
Indian summer monsoon during the Holocene, Geophys. Res.
Lett., 32, L17703, doi:10.1029/2005GL022685.
Willis interesting read thank you.
I like the description of being in your “middle youth” my Mum when in her seventies used to say she and my Dad were in late middle age as they kept seeing in the news that someone had reached 100 plus years. My Dad only gave up his allotment garden 2 years ago at 89 (still has a garden 120×55 ft) and Mum gives lifts and does errands for the old folks in her car (some of the old folk are 10/20 years her junior)
As a friend of mine says “You’re only as old as you feel and if you feel old you’re obviously feeling the wrong bit”
All the best
James Bull
Similar to exploratory factor analysis – there may be little ripple across the various measures, but the FA will find factors for you, however subtle the profile.
If after 1860 you have 14 peaks in one data set and 12 in the other, do you really expect to have much correlation? Are the usual trick going to make any difference?
Even IF there is something more in it than strong correlation. Who are the man/woman to judge if it’s A that effects B or if there is other factors on the route to B? In other words IF A -> C1 (+C2 and so on) and Cx leads to D that leads to B…..
“Figure 5. Periodogram of the NW Indian Rainfall. ”
I see you are still using the “slow Fourier Transform” code which I’ve shown distorts the signal badly and is not a valid analysis tool.
Actually, the “Slow Fourier Transform” was not invented by me. As Tamino commented on an earlier post,
So if you are unhappy with it, please go talk to Ferraz-Mello.
In future, if you think I’m wrong, you really should post up what you think is a correct analysis of the data so that we can all see the difference. I await your analysis of the NW Indian rainfall data.
w.
I have previously posted “what’s wrong” for the sunspot data. Data and code is still up on dropbox. It’s the same tool, same wrongeness. Why would I have to do it again for a different data set?
Also, you have the burden of proof wrong. When doing signal processing, the burden of proof is on the author: As previously posted, you need to sweep sine waves and make sure you are getting the correct result, make sure the windowing function is correct with low frequency sign waves and make sure aliasing isn’t happening with high frequency sign waves. Showing the impulse response is also normal.
Also, how do we know you implemented DCDFT correctly? We don’t. you have to test your code. You have to provide the tests as proof. I have 10 years of signal processing and 25 years of software engineering as a background here. I’ve shipped numerous bugs to the field and been caught out by PhDs making basic “where’s your test code” mistakes – in the medical device field, where if you get it wrong, someone dies (or stays dead in what I was working on).
Generally I would barely trust built-in Matlab and R functions, though for our purposes they are probably good enough.
Peter Sable October 9, 2014 at 10:12 pm Edit
I understand that. However, I don’t think that what you’ve posted in any way invalidates the tool. To do that, you’d have to show us how to do it right, and then show the difference between your results and mine.
And despite repeated requests, you’ve never done that … sorry, but without that you’re just flapping your gums and calling it science.
w.
> To do that, you’d have to show us how to do it right, and then show the difference between your results and mine.
?dl=0
Well, actually the source code for my method is sitting up on dropbox. It’s an early draft version so I’m still updating it. (the shape is correct but the magnitude has a bug, plus there’s anomalies near zero period I have to deal with)
output of 11 year sine wave my code:
output of 11 year sine wave your code:
source: https://www.dropbox.com/sh/5wh9dbja6x37nfa/AADhfZFr6JXWF2vCDAj407Hoa?dl=0
I believe what you are forgetting to do is window the data. Read http://www.edn.com/electronics-news/4383713/Windowing-Functions-Improve-FFT-Results-Part-I, The sidelobes I see in your code look exactly like leakage due to windowing issues in that article.
(btw the thread depth is limited I can’t reply directly to your post which is current below… )
(also if I could get some help on formatting in wordpress I’d appreciate that)
Not sure why I didn’t think of this before, but I disabled windowing on my code and got almost exactly what you got. Lots of side-lobes.
So I think your periodogram code just needs a window function added.
I’m still flummoxed as to why you would implement an FFT in a different manner though… FFT is built in to scilab, matlab, Octave, and probably R.
Peter Sable October 10, 2014 at 5:26 pm
Peter, I’ve explained this to you before. It’s because my method 1) does not require that the data be regularly spaced, and 2) is quite tolerant of missing data. The FFT routines built into R are neither … which was the exact reason that Ferraz-Mello implemented it in 1981.
So for example I have a dataset which is the IntCal98 14C dataset. The space between the samples is variously 10, 20, 30 , 40 50, 415, and 1000 years … my method for FFT handles it without comment.
How about yours?
w.
Okay, now I understand the missing data part, though typically interpolation is what a signal processing person would use as a separate step.
You are still not windowing your data and that’s creating plenty of artifacts
This shows what happens when you don’t window a sine wave. It doesn’t look like the single period you would expect, it looks like energy scattered all over the spectrum.
?dl=0
source (lots of other fun stuff like period resolution limits):
https://www.dropbox.com/sh/xsu1mr76bqhlgz1/AABJTKn4Pj96YVjTyfbkhA_ba?dl=0
Here’s what happens when you window the data before feeding it to your slow Fourier Transform” code. I believe this proves that you need to window your data. Nice clean spike at the correct period of the sine wave:
?dl=0
Simple correlations with the sun will never be statistically detectable unless combined with ocean oscillations.
The strength of monsoons have been clearly linked to pattern changes in ocean surface temperatures. When the Pacific is in a LA NIna like state monsoons are stronger and there are more floods in Asia and droughts in the Americas. When in an EL Nino like state monsoons are weaker and there are more droughts in Asia and more floods in the Americas.
During the Medieval Warm period there was high solar activity and the ocean were in a La Nina like condition with the appropriate pattern of droughts and floods. During the Little Ice Age when solar activity was reduced with the appropriate effect on monsoons and droughts.
The position of the ITCZ is also a factor and it is well established that the ITCZ moved southward during the LIttle Ice Age and low solar activity.
jim Steele October 8, 2014 at 11:13 pm Edit
Thanks, Jim, always good to hear from you.
However, that claim of yours assumes that 1) there are “simple correlations with the sun” [and the weather], and 2) they are being hidden by the ocean oscillations.
I don’t notice any evidence for either of those claims. Do you have any?
The monsoon is, as are many emergent phenomena, temperature related. It is driven by the difference in temperature between the land and the ocean. So anything that affects that difference might well affect the monsoon. Obviously, that could include the El Nino/La Nina oscillations, as well as the PDO. I’ll have to take a look at those.
Note also that the monsoons have a huge cooling effect on the surface, so obviously they are another one of the emergent thermoregulatory mechanisms that control the earth’s temperature.
However, the path from that to a solar connection is murky and indistinct at best, and non-existent at worst.
All the best,
w.
“Note also that the monsoons have a huge cooling effect on the surface…”
Another fact that gives the lie to the positive feedback of water vapor assumed in all the GCM’s.
Willis says “that claim of yours assumes that 1) there are “simple correlations with the sun” [and the weather], and 2) they are being hidden by the ocean oscillations.
I don’t notice any evidence for either of those claims. Do you have any?”
There are numerous papers showing the oceans absorb heat during La Nina like conditions, the lack of clouds and clearer skies in the eastern Pacific during La Nina can allow 5 to 200 more W/m2.
I don’t argue that there is a simple correlation with the sun, I argue attempts to determine one will never be detected in the short term of decades to centuries because absorption and ventilation of heat and changing sea surface patterns will obscure the sun’s contribution. But a stronger correlation emerges over longer timescales and the paleoclimate is littered with studies showing changes in cosmogenic nuclides like Be10 correlate with climate change.
A most pertinent paper is Asmerom et al (2007) Solar forcing of Holocene climate: New insights from a speleothem record, southwestern United States. Geology
They claim “We show that periods of increased solar radiation correlate with decreased rainfall, the opposite to that observed in the Asian monsoon, and suggest that a solar link to Holocene climate is through changes in the Walker circulation and the Pacific Decadal Oscillation and Kl Nino-Southern Oscillation systems of the tropical Pacific Ocean. Given the link hetween increased warniing and aridity in the southwest, additional
warming due to greenhouse forcing could potentially lead to persistent hyperarid conditions, similar to those seen in our record during periods of high solar activity.
Willis,
If you want the Hiremath in pdf, my email is: joel(dot)obryan(at)gmail(dot)com
I will reply with the pdf of the Hiremath doc.
Joel (in tropical soggy Tucson Arizona)
Thanks, Joel, I’ve emailed you.
w.
At first glance that figure 3 looks more than “kinda hopeful”.
There are parts of that time period that look very hopeful.
Perhaps it would make sense to look at the periods that seem to work (1850-1900ish and a century later) and see if those periods seem to work in the other regional data as well?
It is possible that there is a real solar effect on the Indian Monsoon but that it is swamped by other effects at some times.
Thanks, M.
So your plan is, ignore the data that doesn’t agree with the theory, and just analyze the data that does agree with the theory? You sure you want to follow that plan? Because to me, it seems not just unscientific but anti-scientific.
Regards,
w.
No,
I was saying that the data that does agree with the theory may actually indicate that that the theory only applies at certain times. Other factors are more significant at other times.
So I described a test that would check that… if it applies in one region it should apply in all regions when the theory is applicable.
If it only applies at sometimes in some places, sometimes, then we have no support for the theory. The evidence is picked to support the theory not the theory picked as it is supported by the evidence.
But saying that a theory is not the whole answer so it can’t be any part of the answer? That is a illogical.
Thanks for the clarification, M, the misunderstanding was mine.
However, I fear that I know of no tests or other methods that would falsify your idea that a “theory only applies at certain times”. It’s possible, to be sure, but how would you test that?
You say we could look to see if the same periods seem to work with “other regional data” … which data, and how would we know?
In addition, how would we identify the time periods in question? What thresholds need to be exceeded to say “this is a relevant time period and that one is not”?
Finally, suppose (as is almost sure to be the case) that phenomenon A, say solar vs. rainfall, agrees with phenomenon B, solar vs. temperature … but only part of the time. Are we then to assume that whatever links the two phenomena “only applies at certain times” as well, just as you recommend assuming for the solar-rainfall relationship itself? And how far does this recursion continue?
I’m sure you can see the practical problems.
w.
I agree with Jim Steele.
Weather originates in the oceans with nearly all rain arising from and descending back into it. Land masses are merely interruptions between oceans. It is why rain forest removal etc will not alter climate, which more than anything else is a function of latitude, i.e. angle of sun’s rays. The ocean is controlled by the moon, which directs currents according to its declination cycles. The SOI is the result. Current reversal is called El Nino and the weakened winds that result together with slacker currents retard monsoonal activity.and also cyclone development
For a very rough rule of thumb there are two approx El Ninos per sun cycle, with each return averaging about half a decade. But it is not regular. EN may vary between 2-7 years and one solar cycle may vary between 9-14 years. Longer term, 10 lunar cycles equate approximately to 17 sun cycles.
Ken Ring
http://www.predictweather.com
“However, it isn’t looking good for the solar theory.”
Well, how’s it looking for CO2?
I’d say the BS detector needle is pegged at 11.
What say you Willis?
Looks abysmally bad for CO2, say I …
w.
Another anti-cycle-clone.
You didn’t read my post. I said lunar cycles would be a more reliable guide to monsoon periodicity
Willis:
I agree, taking a job sure sucks (up your time 😉
Since I’ve been chasing money, my time for thought and posting is way off. But, gotta eat… (STILL waiting for all that Big Oil money that is supposedly sloshing around…. /sarc)
On the monsoons vs solar: An ‘eyeball’ of the rain pattern looks like a (poor) about 80 year cycle with peaks about 1875 – 1960 (yes, I know that is 85 years… I said “about’ 😉 so maybe longer cycle time has some promise, but the data only has one cycle in it (and that one is lousy and could be a random excursion).
On Fig.5 it looks like about a 16 year lag from the drop of sunspots to the plunge of rain in about 1910. The larger envelope of the two both seem to ’round over’ in similar fashion on each side of that drop (with the lag). Perhaps “period” isn’t quite so important as overall ‘envelope’ ( plot lagged rain vs sunspots and check R value?) When you have multiple not quite regular thing interacting, looking for a regular period will fail. Sunspots have an average 11.1 (or so) period, but actually studiously avoid having that average in any one cycle. Clusters both sides of it ( 9 ish and up to about 12-14 ish). Add in quasi-cycle lunar movements (it wanders over long terms) and you want a wiggle match more than a periodicity peak. Will it find anything? While I suspect so, it is just speculative on my part.
Jevons (of Jevons Paradox) studied the British Empire grain production/price data from India and found a significant coorelation between solar cycles and grain production. Yes, more than just Monsoons involved, but the data are what they are. There is some kind of connection, just a murky one. It has been ‘found’ several times since the 1800s and Jevons. ( IIRC he was follwing up on Herschel. )
Does any of that have enough strength to make any real difference? I doubt it. There have been some grain traders who claimed to use sunspots to make predictions and thus money. Then again, some of them claim to use all sorts of crazy stuff…
IMHO, the reality will likely end up back at Lunar / Tidal and the solar connection just being due to orbital resonance causing a slight coorelation between spots and the moon cycles; but even that is entirely speculative. Still, doing a comparison of tide patters (long term) in the area to monsoons and ocean cycles might be useful.
Thanks for the story and the analysis. Hope the theatre turns out well!
kenmoonman October 9, 2014 at 1:54 am
I agree with Jim Steele.
And I agree with both.
The Temperature of the Ocean off India’s coast is going to be what determines the Monsoon.
You heard it hear ‘third’ folks : )
In Australia, we have folk at the Bureau of Meteorology who are capable of homogenising this data to ensure there is a strong trend – up or down – it doesn’t matter – whatever pleases you.
Why then are we seeing this essay at all?
Good question. Hiremath’s 2004 paper on the same topic is available in its entirety in .pdf simply by Googling “Hiremath monsoon”. Can’t recall if I’ve linked to it on this blog before or not, among the many instances of the influence of the solar cycle on weather & climate I’ve cited in the past.
From New Astronomy, via Science Direct:
Influence of the solar activity on the Indian Monsoon rainfall
K.M. Hiremath a,*, P.I. Mandi b
a Indian Institute of Astrophysics, Bangalore 560034, India
b Basaweshwara Science College, Bagalkot, Karnataka, India
Received 5 March 2004; received in revised form 21 March 2004; accepted 1 April 2004
Available online 22 April 2004
Communicated by W. Soon
Abstract
We use 130 years data for studying correlative effects due to solar cycle and activity phenomena on the occurrence of
the Indian Monsoon rainfall. We compute the correlation coefficients and significance of correlation coefficients for the
seasonal and the annual data. We find that: (i) for the whole years 1871–2000, the spring and southwest monsoon rainfall
variabilities have significant positive correlations with the sunspot activity during the corresponding period, (ii) the FFT
and the wavelet analyses of the southwest monsoon rainfall variability show the periods 2.7, 16 and 22 year, respectively
(similar to the periods found in sunspot occurrence data) and, (iii) there is a long-term trend indicating a gradual
decrease of occurrence of rainfall variability by nearly 2.31.3 mm/year and increase of sunspot activity by nearly
3.91.5 sunspots/year compared to the activity of previous solar cycle.
We speculate in this study a possible physical connection between the occurrence of the rainfall variability and the
sunspot activity, and the flux of galactic cosmic rays. Owing to long-term positive and significant correlation of the
spring and southwest monsoon rainfall variabilities with the sunspot activity, it is suggested that solar activity may be
included as one of the crucial parameter in modeling and predicting the Indian monsoon rainfall.
2004 Elsevier B.V. All rights reserved.
Keywords: Solar cycle and activity phenomena; Indian Monsoon rainfall; Flux of the galactic cosmic rays
Kip Hansen October 9, 2014 at 6:44 am Edit
Seriously?
Look, Kip. I use what I have. I took the data used by the authors, and I took an independent look at whether there is a perceptible solar signal in the NW Indian monsoon rains. I found nothing. I wrote up my findings. I note that you have found nothing wrong with my work.
Come back when you have something equivalent to contribute, and perhaps you’ll understand this crazy process called “science”. Heck, there are still three more sections of their data to analyze … but it seems you’d rather ask foolish questions than actually do some work.
w.
Reply to Willis ==> To me, this type of premature, superficial “hack-and-slash” criticism of papers you haven’t read adds nothing to our greater knowledge, says nothing about the paper being discussed, and serves only to muddy the scientific waters.
It might be better for to wait until you know what they are saying before you start in.
Kip Hansen October 9, 2014 at 5:08 pm
Kip, I thought I’d made it clear that I was not analyzing the Hiremath et al. paper. I said I was INSPIRED by their paper to look at the monsoon rains, and I also said that my results did NOT invalidate their paper, viz:
Far from “hack-and-slash” criticism, I barely discussed their paper at all. Now, at some future date I may do an analysis of their paper. But as I clearly stated, all I’ve done so far is look at a part of the data they used.
It might be better for to wait until you notice what what I am saying before you start in.
Finally, why is this analysis of the monsoon rains “superficial” in your not-so-humble opinion? I’ve done a variety of analyses of the data, and you’ve done nothing but complain, cavil, and carp … which of us is being “superficial” here? Do you have anything scientific to add to this conversation?
w.
Personally investigating questions that inspire you is perfectly valid — I do it all the time, often on points and across fields of knowledge a bit esoteric — approaching bizarre — so that my wife and friends think I’ve lost it. They may be right.
Whatever you call the thing you do when you write this type of piece, it is not science — science requires understanding what knowledge we have already gained in a field; taking advantage of the years of work other colleagues have put in on a question; reading, understanding, questioning, probing and testing to see what part of their work answers valid questions about the world around us with probably valid answers; and attempts to replicate their work. Spending an otherwise idle evening or two playing with one’s “Maths and Graphs Toolkit” and using easily accessed vaguely related datasets found under your local lamppost where the light is good is not the same as years of hard work on a narrow question — and writing up the “results” you obtain from your interesting play is not science — it is sophomoric pretension.
I enjoy your travel travelogues — but not your pseudo-science travelogues.
Willis, is there a simple algorithm to back convert your cycle-length/amplitudes (i.e. Figure 5) into a signal vs time format?
Not sure what you mean, Doc. It sounds like you want to recreate the original signal, but we already have that, so I’m not clear about your request.
Thanks,
w.
Excellent, Willis. I wondered about India because the Indian Ocean seems to have a mind of its own and doesn’t change temperature the way the others do (I am basing that on Tisdale’s work).
I also appreciate the straightforward methods that ‘finds things’ but which are in the mud. The usefulness of that is of someone finds a powerful multiplier the initiating impulse is already demonstrated.
Is there a cloudiness data set v.s. Temp or rainfall?
I subscribe all or in part to the hypothesis advanced by Happ, et al, that the 100% swing in the UV component of TSI over solar cycles, combined with associated solar magnetic flux variation (as indicated by sunspots), is a major driver of both weather & climate change on earth, via these solar effects on ozone, clouds, air pressure, SST & other atmospheric & oceanic parameters.
http://climatechange1.wordpress.com/2008/11/21/the-enso-driver/
These influences are brought home to me every year during the Pineapple Express & Chinook winds in Oregon (& on longer timescales by PDO switches, most memorably in 1977), & by ENSO swings in Pacific coastal South America.
Monsoons in China, SE Asia, the Indian Subcontinent & the Arabian Peninsula however are also strongly correlated with these fluctuations, as are droughts & floods on opposite sides of the Pacific.
CO2, not so much.
Experience of the North American monsoon in AZ & NV also reinforces this view, IMO.
milodonharlani October 9, 2014 at 8:20 am Edit
milodon, I just took a look at your link and found nothing about either sunspots or solar magnetic flux variation. There were two mentions of “UV”, both in terms of what happens to it after it hits the earth, nothing about the “100% swing”.
Sorry … not convincing in the slightest. It may be wonderful that you have a subscription to the hypothesis, but I fear I prefer, you know … evidence.
w.
The evidence for solar influence on climate is overwhelming to self-evident.
Did you read the link? UV affects ocean T directly & via its effect on ozone, which also affects air pressure & winds. The correlation is highly statistically significant, but if you don’t think so, no worries.
Thanks, milodon, but when someone is asked for evidence, not coming forth with a scrap of evidence while claiming that the mystery evidence is is “overwhelming” is … well … underwhelming.
It appears that you do not understand that we’re talking about the variations due to the sunspot cycle, and not simply “solar influence on climate”. Everyone knows the day warms when the sun comes up.
If you do understand that we’re talking sunspot cycles, please show us the ~11-year cycles in your climate variables such as ocean temperatures, Milodon, the cycles that relate to the variations due to sunspots. Yes, it is evident that the sun affects the climate … but it is far from evident that sunspots or any of the many aspects of the ~11 year solar cycle (e.g. solar wind, heliomagnetism, cosmic rays, etc) affect the climate.
And please … don’t send me links to junk papers like the one Chilean tree paper you or someone tried to slip by last time this was discussed.
Finally, you say:
I’m sorry, but I did read the link. I saw no analysis of the UV effects, in fact you only mention UV twice, and and only UVB, and it’s in the same paragraph, viz
All true, but that paragraph says nothing about your claims of how “UV affects ocean T directly & via its effect on ozone, which also affects air pressure & winds”. And those are the only mentions of UV in your article.
Is there a solar-cycle related effect on climate? I don’t know. All I know is that I have looked long and hard for convincing evidence and found none. I have looked in a variety of manners, I’ve used a variety of mathematical analyses (as in this example) and I simply haven’t found it.
This is just the latest in a long string of failed attempts to find a solar cycle signal in a wide variety of climate datasets. I now get to add “no sign of solar cycles in NW Indian monsoon data” to my list of the many places I’ve looked for and failed to find a trace of the solar cycle. I can see you don’t like that … but if the signal is there as you claim, how about you show us how to detect it? I’ve provided the data, so show us the solar cycle signal, demonstrate how it’s done for us.
Regards,
w.
The ‘100% swing’ you refer to is only in the EUV which is a tiny fraction of the total UV emitted by the sun. The EUV isn’t involved in stratospheric O3 production and in fact barely makes it into the thermosphere.
Thanks for the preliminary analysis, Willis.
And good luck with the theater lobby.
Al Gore shares a Nobel Prize while Willis Eschenbach works construction. Maybe climate change really is more dangerous than Ebola. Maybe climate changed caused it. Maybe Ebola will go extinct and it will be all our fault. And maybe it won’t. –AGF
I’m sure that Willis is in better physical condition than Fat Albert.
And unlike Prince Albert, Willis isn’t dependent on Big Oil money.
Willis, this topic is almost as old as both statistics and meteorology. The Yule-Walker equations, well known in statistics, combine the names of Udny Yule, a statistician remembered today in respect to spurious regression, and Gilbert Walker of the Walker circulation. Both sunspots and Indian monsoons crop up in early studies. You’ll find some interesting articles if you combine the topics and author names.
Something good out of NCAR:
Entire .pdf open access.
Abstract
The eponym “Walker Circulation” refers to a concept used by atmospheric scientists and oceanographers in providing a physical explanation for the El Niño–Southern Oscillation phenomenon, whereas the eponym “Yule–Walker equations” refers to properties satisfied by the autocorrelations of an autoregressive process. But how many statisticians (or, for that matter, atmospheric scientists) are aware that the “Walker” in both terms refers to the same individual, Sir Gilbert Thomas Walker, and that these two appellations arose in conjunction with the same research on the statistical prediction of climate? Like George Udny Yule (the “Yule” in Yule– Walker), Walker’s motivation was to devise a statistical model that exhibited quasiperiodic behavior. The original assessments of Walker’s work, both in the meteorology and in statistics, were somewhat negative. With hindsight, it is argued that his research should be viewed as quite successful.
http://projecteuclid.org/euclid.ss/1023799000
Steve McIntyre October 9, 2014 at 9:28 am Edit
Indeed. The first conjecture, to my knowledge, was that of Herschel, who claimed (incorrectly) that wheat prices followed the sunspots.
I just did that, and I couldn’t find an article by either author on the subject. Do you have any further hints?
All the best, and thanks for both your excellent blog site and your excellent work.
w.
Herschel was right.
Milodon, thanks for your science by assertion. No citations, no support, just you saying so.
However, before you start believing Hershel, here’s an interesting read.
I’ve also run the numbers myself. Turns out that his conjecture was based on stunningly little data. I dug up the wheat prices for the period following Hershel’s claim, and it all fell apart … as these claimed solar correlations tend to do whenever we get additional data. I suppose I should publish it, another piece of analysis I’ve done but never published.
Regards,
w.
IIRC, I provided you with the grain price data to which you refer. By all means, please publish.
My statements have been repeatedly supported by the studies which I keep linking but which you keep ignoring, to include in this very comment section. Before posting on incomplete data, you still need to do a thorough literature search before making false assertions about the influence of solar cycles. Your failure to read what others have done means that you keep reinventing the wheel, as Dr. Spencer showed with respect to tropical ocean updrafts.
milodonharlani October 9, 2014 at 8:15 pm
Dr. Spencer incorrectly claimed that my theory of the thermal control of the planetary temperature by a host of emergent phenomena, in particular tropical thunderstorms, was the same as the “supergreenhouse effect” theory of Ramanathan. Nothing could be further from the truth, as both you and he should know had you read both Ramanathan’s and my work.
So your attempt at using him as a reference just shows that you haven’t done your homework. If you think the two theories are the same please point out how and where. I made the same request of Dr. Spencer … he became suddenly reticent. I was greatly saddened, because he was (and is) one of my heroes, but in this instance, he was way wrong.
How about you? You gonna show us all how Ramanathan’s work is the same as mine, as Dr. Spencer fatuously claimed but could not demonstrate? You gonna try to back up your empty false claims?
Please try. I look forward to it.
w.
NOTE: My post at the time in response to Dr. Roy’s character attack is here. I leave it to the reader to decide who is blowing smoke …
Willis, may I suggest you update your definition of a monsoon because this “I knew that when land gets hot in summer, hot air rises, wet air is drawn in from the ocean. Result? Monsoon rainfall. ” is truly outdated.
Here is a primer regarding atmospheric circulation: http://ddata.over-blog.com/xxxyyy/2/32/25/79/Leroux-Global-and-Planetary-Change-1993.pdf
I invite you to read Leroux Dynamic Analysis of Weather and Climate 2010, Springer-Praxis, chapter 4 on Tropical Circulation:
“• Monsoon circulation
The intervention of the continental thermal factor alters this symmetrical distribution. The Sun’s yearly motion displaces maximum warming from one tropic to the other, the theoretical maximum displacement being 46° 54′) (cf. Fig. 3-B). The thermal factor can certainly affect the pressure field, though this is not possible over the oceans, where Anticyclonic Agglutinations dominate and warming is never excessive. Over continents, however, the ITL corridor is locally very deep, and displaced by moving thermal lows, which distort the meteorological equator, lagging more or less behind the Sun’s zenithal motion. The trade wind coming from an AA is thus caused to cross the geographical equator by way of a transequatorial pressure declivity between high pressure in one hemisphere and low pressure in the other. The Coriolis or geostrophic force, a function of the sine of the latitude, is nil at the Equator, and the initial trade wind can cross the Equator under the effects of inertia and of the pressure gradient. Progressive penetration into the opposite hemisphere, towards the axis of the thermal low which is moving away from the equator, increases the geostrophic force, now of opposite sign: the flux is diverted eastwards as a monsoon (photo 19), and the lines of flux are now perpendicular to the isobars (Fig.22).
Genetically, a monsoon is therefore the extension into one hemisphere of a trade wind originating from an AA in the opposite hemisphere, or directly from an MPH, drawn across by a thermal low in the summer hemisphere. The term summer monsoon is thus tautological. The monsoon meets, along the ME, a (continental) trade wind from the opposite hemisphere, and the area of extension of this trade is considerably reduced.”
ITL: Intertropical Low Pressure
ME: Meteorological Equator
MPH: Mobile Polar High
Please correct me if wrong.
Indeed thanks for the precision…
I’m sorry, Tom, but I see nothing in my description that disagrees with your description. Yours goes through all of the details, to be sure … but if the rising air from the heat over the land isn’t drawing in the moist monsoon air, then what driving it? Sure, they say it is “drawn across by a thermal low in the summer hemisphere” … but that’s just what I said. Hot air rises and draws in moist air.
w.
The thermal low help draw the monsoon air deeper inside continents but the key point is the strength of of the circulation from the opposite hemisphere, driving MPHs passed the ME, transforming trades into monsoon winds. Thus it ties to the balance of the two hemisphere’s circulation modes. As long as there is no understanding of what triggers at short, medium term the switch between slow and rapid modes, we do not get to the bottom of it. For instance regarding a short term switch, why would we have during a winter, a sudden strengthening of all expulsed MPHs from the Arctic or Antarctic for a 3 weeks period then a reprieve then again another increase? Same if we look at a few years were indeed the Milankovitch explanation cannot explain them. So I think there is much more to consider and that has yet to be explained before ruling solar influence in or out based on teleconnections. The merit of Leroux was to describe the geometry of the circulation especially critical in the lower troposphere.
There are advantages and disadvantages to being an autodidact. On the one hand, it perhaps frees you to think outside the box, but on the other it raises the likelihood that your idea may already have been thought of by others and written about extensively without your notice, and that you may not have learned enough about the topic to comment on it knowledgeably. That goes double for subjects as complex as climatology and meteorology.
In the case of how monsoons work, the description in your post lacks certain important observations which Tom has pointed out. What he said is not what you entirely said, not that your description was necessarily wrong as far as it went, but merely incomplete. IMO yours describes better the weak, subtropical American SW Monsoon than the powerful, more tropical monsoons of Asia and Africa.
Tom and Sturgis, thank you both for your comments. I do understand the value of greater understanding and more detailed descriptions.
Was my description incomplete? Of course. Every description, including Tom’s and Leroux’s, is incomplete.
Was my description wrong? Not anywhere. Monsoons don’t happen in the winter. Why?
I said:
Leroux said, inter alia, that the moist monsoon air is “drawn across by a thermal low in the summer hemisphere”.
So I knew little about monsoon rains, but nothing that I knew was wrong.
Leroux’s point was that when land gets hot in summer, and the hot air rises, and wet air is drawn in from the ocean, that the air is even drawn from the other hemisphere.
While this is certainly interesting and valuable information (thanks, Tom), I’m not sure how it relates to the purported solar cycle in the monsoon rains.
My best to you both,
w
PS—Tom, you discuss “ruling solar influence in or out “. While we could rule solar influences in (if we could find any evidence), we can never rule them out. As I said above, all I can say is I’ve looked hard and haven’t found them. Doesn’t mean they’re not there …
Tom Rude, Sturgis Hooper,
Reading your comments (and Willis’) I have to conclude that the monsoon is initiated and maintained by continental thermal low pressure, without which there is no monsoon, and the other natural effects that you describe all depend on that. Those effects may intensify the monsoon but they do not initiate it nor maintain it.
mpainter, the thermal lows over continents do help drawing the monsoon air deeper YET genetically a monsoon is a a trade wind that has crossed the equator. Hence its origin depends on the dynamics of the hemisphere circulation from which the trade wind initiated. Thus you’ve got it backward! In fact the intensity and aerial extension of regions affected by the Monsoons every year-widespread or narrow band of highly concentrated rains- depends on the circulation dynamics.
TomRude October 13, 2014 at 11:31 am says
Umm … no. Here’s an excellent definition
Note that:
• They say the “primary cause is the much greater annual variation of temperature over large land areas compared with neighboring ocean surfaces”, just as I said.
• They say nothing about trade winds crossing the equator.
• They occur in a number of regions far from the equator, such as Chile
So your claim is simply not upheld by the actual usage of the term. The “primary cause” of the monsoons, as I said and as per the definition above, is that in summer the land gets hotter than the ocean, so the air over the land rises and draws in moist air from the ocean.
Regards,
w.
PS: What I gave above is no random dictionary definition. It is from the official glossary of the AMS, the American Meteorological Society. I’m sorry, but I’ll take their word over that of some anonymous internet commenter.
Willis Eschenbach
October 13, 2014 at 12:35 pm
The Atacama Desert of Chile is closer to the equator than the US Southwest, which also experiences monsoonal flows. Once across the ME, the phenomenon can carry to pretty high latitudes. The Pineapple Express, which brings heavy rain from around Hawaii to the Pacific NW, results from the Madden-Julian Oscillation, so might in effect be considered a mid-latitude monsoon, yet occurs typically when the land is cold.
milodonharlani October 13, 2014 at 12:55 pm
Not sure what your point is here, milodon. The driving force for all of these is the land-sea temperature difference. When this gets large enough, it MAY draw winds across the meteorological equator (which is the latitude line at 5°N) … or it may not. For example, you get the Asian monsoon as far north as Korea, about 40°N … and I’m not buying that those are somehow misplaced southern hemisphere winds. Might be so, but I’d need to see the data to believe it.
w.
Willis, I’ll take Leroux’s work versus the AMS glossary especially because 1) he had studied and explained monsoon’s influence on the meteorology and climate of tropical Africa for his PhD the publishing of which was sponsored by the OMM, 2) because after years of following satellite images and animations, Leroux’s description is so obviously accurate and 3) because you claim asking for respectful discussion but the “anonymous internet commenter” is quite demeaning, especially since I provided a more complete reference (his 2010 book) that you should bother to read. It would in fact help you understand that the definition you just gave did not disprove anything and alluded to the balance of the circulation’s dynamics between the two hemispheres: “The monsoons are strongest on the southern and eastern sides of Asia, the largest landmass, but monsoons also occur on the coasts of tropical regions wherever the planetary circulation is not strong enough to inhibit them.”
Sadly I wont be around to see Monsoons return to the Sahara.
http://www.bbc.co.uk/blogs/legacy/23degrees/2011/06/will_monsoons_once_again_retur.html
“Strange as it may seem 8000 years ago when the cave paintings in Wadi Sora were made the Sahara was getting more sunlight than it is now. And that extra heat helped bring the monsoon rains to this desert. But how did the Sahara get more solar energy?
But over a 41,000 year period it changes, wobbling between 22.1 and 24.5 degrees. Back when the Sahara was green, the tilt was close to its largest possible angle, 24.2 degrees. Which meant that 8000 years ago the Sun shone more directly, more intensely over the Northern hemisphere.
“This precessional wobble takes 23000 years to complete one cycle, so it will be 23000 years before Polaris will come back round to be our northern star again”
Tilt & precession are parameters in the Milankovitch Cycle of orbital & rotational mechanics ruling glacial & interglacial cycles.
I confess that it never ocurred to me that the Sahara is a desert because it is not getting enough sunlight.
I guess that I am pretty behind ward on the latest findings in climate.;)
http://www.bbc.co.uk/iplayer/episode/b04kn99x/jungle-atlantis-2-death-of-angkor-wats-megacity#group=p025tw6r
“Buried in the Cambodian jungle lie the lost remains of the great medieval city of Angkor, once the capital of one of the world’s greatest civilisations. Today, only the great stone temples like Angkor Wat survive. But Angkor was once a teeming metropolis, full of life – the biggest city on Earth. ”
It would appear that during the 14th century the Monsoons became erratic and the rains failed.
Coincides with the start of the LIA????
Yes. North American cultures like the Mississippian (eg. Cahokia) & Anasazi however suffered severe, prolonged droughts during the Medieval Warm Period, leading to their decline or disappearance before the onset LIA.
http://www.jstor.org/discover/10.2307/20622439?uid=3739856&uid=2129&uid=2&uid=70&uid=4&uid=3739256&sid=21104781560007