Guest post by David Archibald
Colder is drier.
The figure above is after a figure from Maus et al 2010 “Long term solar activity influences on South American rivers”. It shows a very good correlation between solar activity, as measured by sunspot number, and the flow rate of the Parana River, the second largest river in South America. The Parana River now hosts the Itaipu Dam with installed capacity of 14,000 MW.
A recent paper in an engineering journal shows a similar strong solar control on the level of Lake Victoria in East Africa (Mason March 2010): 
As Mason notes, an interesting correlation was noticed in the early 1900s between lake level and solar activity, in the form of the sunspot number. The interest this caused waned when the correlation seemed to disappear after about 1928. The early 1960s saw a dramatic climate anomaly in East Africa. Lake levels rose significantly, including those of Lake Victoria, and flows in the Tana River in Kenya doubled. The sluice gates at the Owen Falls dam were opened to release the additional water required by the Nile waters agreement and they stayed open, almost continuously, until well into the 1990s. This surplus water also led Uganda to invest in a new hydroelectric power station at Kiira. But the lake level starting falling from 1964 with an oscillation around the falling trend. This oscillation, controlled by solar activity, is shown in the following figure from Mason:
The falling trend in the level of Lake Victoria meant that the new hydro dam at Kiira did not produce any long term, additional energy for Uganda.
Back to South America and the Itaipu Dam – it produces 90% of the electric power consumed by Paraguay and 19% of Brazil’s consumption. As Maus et al note, the relationship between smaller solar activity and low Parana’s discharge can also be found in historical records.
For example, low discharges were reported during the period known as the Little Ice Age (LIA). In particular, a traveller of that period recalls in his diary that in the year 1752 the streamflow was so small that the river could not even be navigated by the ships of that time, which were less than 5 ft draft, to be compared with ships up to 18 ft draft that can navigate it at present as far north as Asuncion in Paraguay.
Our prediction for Solar Cycle 24 in terms of F10.7 flux is shown following:
Given the link between East African and central South American rainfall and solar activity, the list of economic impacts from the current solar minimum (Solar Cycles 24 and 25) can be expanded to:
- Canadian agricultural will get a severe whacking from a shortened growing season and un-seasonal frosts.
- 24 year drought in central South America
- 24 year drought in East Africa
- Paraguay and Brazil having severe power shortages.
This list is by no means exhaustive. The last time the world witnessed mass starvation was the 1965-67 drought in India which killed 1.5 million people. Things don’t look pretty.
References:
Mauas, P.J..D., A.P.Buccino and E.Flamenco, 2010, Long-term solar activity influences on South American rivers, Journal of Atmospheric and Solar-Terrestrial Physics on Space Climate, March 2010.
Mason, P.J., 2010, Climate variability in civil infrastructure planning, Civil Engineering 163, pages 74-80.
David Archibald
July 2010
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Calling a spade a spade:
The introductory graph in this post is mislabeled.
For anyone on the detective’s trail (trying to figure out how when a person stands up in an airplane the elevation of their head increases):
Mauas, P.J.D.; Buccino, A.P.; & Flamenco, E. (2010). Long-term solar activity influences on South American rivers.
http://arxiv.org/PS_cache/arxiv/pdf/1003/1003.0414v1.pdf
The “secular trend” (which oscillates) is what will be of more interest to many folks (i.e. the vertical motion of the airplane, rather than the ripples superposed by the person standing up & sitting down).
I’ve seen this paper misrepresented too many times. There’s something to it, but it’s not what many are saying. Recommended: more care.
WOW! Stunning!
If this ain’t just another “WOW!”-signal – indicating that, in the end, it’s the Sun, which is REALLY driving the big steam engine, called Atmosphere – then what is?
Great find. Much appreciated. Thank you.
In other news, german news media – If you haven’t noticed yet: I AM a German… – are actually hyping carbonic acid in the oceans leading to mass extinction of the plancton, after hyping CO2 in the atmosphere as the bringer of hellish future temps has failed to gain traction even in my hysterically environmentally friendly homeland.
“Zu viel Kohlensäure in den Meeren: Seeigel “versauern” auch”
(Too much carbonic acid in the oceans: Sea urchins suffer from acidification, too)
http://www.n-tv.de/wissen/weltall/Seeigel-versauern-auch-article679064.html
This article is from January this year. But yesterday, the “Polarstern” – the reknowned german scientific research vessel – returned from a trip into the Arctic, claiming, that the the Oceans are actualy turning into sparkling water, which madethe headlines in Germany’s mayor network’s primetime-news.
So now it’s the carbonic acid, we have to avoid like the plague and, therefore, rather tax ourselves into oblivion?
But, well: What happened to the sealife during the Eocene then – when CO2-levels in the atmospere were 2-3 times higher than they are today?
http://www.nature.com/nature/journal/v461/n7267/full/nature08447.html
“Atmospheric carbon dioxide through the Eocene–Oligocene climate transition”
Be afraid, be very afraid – of scientists desperately looking for something to tag on the evil CO2, to justify and finance their future scientific life, I mean…
>> A Little Book of Coincidence. It is about the highly exact
>>geometrical relationships pervading the whole solar system
>>and rich in pi and phi.
And don’t forget we have added to those coincidences, by the units of measure we have chosen to use.
For instance the number of Earth days in one degree of the Precession of the Equinox, is equal to the total number of years in the whole precessionary cycle.
Why? – because we chose one degree of a circle to be equal to one day of the Earth’s orbit.
(Anyway, any spinning and precessing body must have a rhythm to it).
Such observations would be accounted for by a change in the position of the observing locations in relation to the nearest rain bearing air circulation systems.
So what matters most is the latitudinal shifts of the air circulation systems beyond normal seasonal variability.
This article provides evidence in support of my proposition that the jet stream positions are affected partly by the levels of solar activity and I have proposed a mechanism via an interplay between the solar influenced size, position and intensities of the polar oscillations in the air on the one hand and the varying rates of energy release by the oceans on the other.
While a river and a lake or two of each makes for a nice fit, but a true correlation, if there is one, ought to be between high altitude precipitation and sunspot numbers, the rivers and lakes be damned? :p
Three questions:
1) How many river systems did they test before they found a correlation this good?
2) How many different ways of manipulating the sunspot/neutron data did they try?
3) How did they account for the temporal autocorrelation in the data?
And a bonus question:
How do the answers to the above affect the risk of a Type I error?
It appears that the correlations presented are valid for only narrow geographical areas. In order to be significant it should have a wider if not global occurrence.
Parana River is in the area covered by the South Atlantic Anomaly
http://www.astro.psu.edu/~niel/astro485/lectures/lecture09-overhead02.jpg
which is highly sensitive to the solar activity, and consequently may be a connection.
Lake Victoria is in the African rift valley.
http://www.geo.arizona.edu/geo5xx/geos577/projects/mooney/images/usgs.gif
which is geologically highly unstable, and there are some tenuous indications that there may be a link between movements in the Earth crust and solar activity, hence affecting the water levels of Lake Victoria.
Both are highly speculative:
Parana river : Svensmark, solar activity-cloud formation ?
Lake Victoria: Solar activity-earthquakes, volcanoes ?
tallbloke
yes your cumulative sunspot concept is what I suggested earlier about AP index running above its long term average. I likened the earth to a cooking pot. If the temperature on the stove is ever so slightly warming the pot continues to warm even though the stove temperature is not increased further.
Ed Murphy @9:58. I decided not to renew my Nat Geo subscription after 46 years due to their pro AGW stance. So far they have been following the tribal consensus that all CC effects are bad drifting to catastrophic. The greening of the Sahel is somewhat of a surprise. Not the fact it is happening but that they reported a positive CC story.
tallbloke says:
July 22, 2010 at 10:42 pm
Andrew W says: I see sun spot numbers in Africa are different to sun spot numbers in South America.
This paper was brought to my attention on my blog a fortnight ago. The reason I haven’t done a post on it yet is I’m still trying to work out the full rationale behind the way Maus et al have processed the sunspot data. They say:
“When plotting together different quantities, the offset and the relative scales
are free parameters which are usually arbitrarily introduced. To avoid these
two artificial parameters, as a final step we have standardized the quantities
by subtracting the mean and dividing by the standard deviation of each series
shown, for the whole period 1909-2003.”
It seems on the face of it to be a reasonable approach, I think Mr Archibald should have mentioned it to avoid confusion though.
_________________________________________________________
It is a reasonable approach, but it couldn’t change the shape of the curves. Yet the sunspot numbers in the two graphs have noticeably different shapes, not only in the timing and relative heights of their maxima, but also in the little wiggles than amazingly match those in the river flow curve. Of course I’m not suggesting that anyone would deliberately fudge the data to make his theory more plausible … Oh, wait…
alan says:
July 22, 2010 at 10:13 pm
“[…]richly timbred musical tone. To understand long-term climate change it is necessary to break down this complex wave into its major components, a kind of Fourier analysis of the climate signal.[…]”
It might be a resonance. Lake Victoria is BIG so it influences the local climate in a big way. Consider:
-Sunspots influence cloud coverage
-Cloud Coverage influence precipitation and evaporation
-The lake level influences evaporation (through water temp and size of the lake area)
and via evaporation, also cloud coverage and precipitation.
It is a system that could resonate locally, and such a resonance can amplify itself, or it can break down and dissipate quickly.
Other river/lake topologies do not have to show the same resonant frequency or strength!
You can say i’m just pulling this out of my arse or you can look at the variables involved and see that there is a possible circular dependancy which would be necessary for a resonance. Turning the data into an audible sound or Fourier-analysing it would surely be revealing. The human ear is very sensitive for detecting repetetive patterns.
Sunspot activity creates the extra warmth the earth needs to separate it from at least a mini-ice age.
Best estimates:
Less than> 100 total sunspot mean for a cycle for a 9 cycles in a row = mini-ice age
Greater than 200, but less than 300 total sunspot mean for a cycle = solar minimum.
Severe winters and droughts
Normally 2 of 9 cycles per 100 years
400 total sunspot mean allows the earth to warm up or cool down quickly at the end and beginning of cycles. We had 4 of these from 1878 to 1933.
3 to 6 – 600 to 1000 total sunspot mean gives us global warming, melting glaciers and terrifying hurricane seasons.
DirkH says:
July 23, 2010 at 5:36 am
“[…]way. Consider:
-Sunspots influence cloud coverage[…]”
Hypothetically, i should add. Also applies to the other dependencies, but to a lesser degree.
Odd that in figure 3 that sunspot count is a trailing indicator of lake level rather than a leading indicator.
Andrew W says:
July 23, 2010 at 12:42 am
If this sort of vague half arsed manipulated data was being used to promote a theory that didn’t suit the people who follow this site you’d all be quite rightly condemning it as nonsense.
By “a theory that didn’t suit the people who follow this site” you must be referring to the Alarmists’ idea that our C02 is causing all manner of global disasters, and will, if not curtailed lead to global catastrophe, meaning we must immediately change our sources of energy to far more expensive ones, costing trillions. See the difference? For thinking people, the idea of a strong solar connection to climate is at least an interesting one.
Correlation (or in this case half a correlation) is not causation.
Paul Birch says:
July 23, 2010 at 5:16 am
“[…]It is a reasonable approach, but it couldn’t change the shape of the curves. […]”
An exact shape match might be what the researcher wanted to achieve to improve his correlation coefficient. For a resonant mechanism it would not be very important though; frequency components are more important in that realm. It would be of interest to compare the Fourier transforms of Maus’ processed sunspot data and the standard data series to see whether Maus’ processing introduces strong new frequency components; this does not necessarily have to be the case even if the wave shape looks slightly different.
ShrNfr says:
July 23, 2010 at 5:41 am
“[…]Odd that in figure 3 that sunspot count is a trailing indicator of lake level rather than a leading indicator.”
This can happen in a resonance (because there is not a one-way causation even if one of the causative directions is far stronger than the other; in this case phase shifts might lead to the “energy receiving” side seemingly “anticipating” the next period of the “energy delivering” side).
Oh, now i see. Yes, the sunspot number in Mauas’ paper looks strange. I understand tallblokes reservations.
Found Mauas’ paper
http://arxiv.org/abs/1003.0414
Found the reason for the different shape of the sunspot data in Mauas’ paper; tallbloke has left out an important sentence:
“The data were
smoothed with an 11-year running mean,
detrended by substracting the long term
component. All series were standardized
by subtracting the mean and dividing
by the standard deviation of each series
shown for the period 1971-2000.”
It’s the 11 year running mean that changes the shapes! tallbloke, this is entirely legitimate IMHO. It just dampens high frequencies a little and, yes, changes the waveform severely, but entirely legit.
[reply] I was looking at the 2008 version of the paper. RT-mod
LightRain says:
July 22, 2010 at 10:36 pm
“It’s funny, well not really, but Lake Superior is supposed to be down significantly compared to normal.”
Ah but if it were true, I was sat on the shore of Lake Superior less than 3 weeks ago busily grilling hot dogs and cooking S’Mores and the Lake level is “normal”
In 2007, in the midst of a four year drought, it was trending lower versus recorded levels but it did not passed the record low level which was set in 1926. Then in October 2007 the lake level increased almost overnight by about 4″ after one major rain event, I haven’t calculated the amount of water involved there but given the Lake’s surface area of 31,700 sq miles it was substantial.
There are many reasons why the lake level might fluctuate year to year. Precip is of course a key factor but it’s by no means the only one. The dredging of a new deeper shipping channel on the St Mary’s River near Sault Ste Marie has also reputedly increased the amount of water leaving the lake which would also impact lake levels.
I have compared the Oxfordshire monthly 150 year long rain record to the sunspot number.
It is clear that if there is any correlation, it is at best sporadic and inconclusive.
http://www.vukcevic.talktalk.net/Oxf-rain.htm
Correlation is not causation.
But in this case I’ll make an exception!
That’s impressive evidence that solar magnetic activity has some profound unexpected influences on the earth. I wonder what other correlations are out there waiting to be discovered.
Nice to see the inclusive nature of this blog site. Where humans with open minds can discuss the “possibles” that can have an effect upon the climate rather than the singular approach that C02 is the only driving force, (and wonder of wonders that is lucky coincidence!! -because the patent cure is trading the stuff to make millionaires Billionaires and beyond and allow control the masses of humanity to the will of said Billionaires even if it turns out that it is not the culprit!!) to a changing climate.
I am sure that there are many different influences that have an effect upon our environment and planet. Inclusive discussion is one way of identifying ways to unravel the mystery of it all.
The only “trick” that should be insisted upon is that before any theory becomes fact it is subjected to scientific validation/falsification.
Publication these days is really only a means of recording research and data archiving as theories are explored, in the hope that great minds might find that elusive idea to advance our understanding of the complexity of our climate.
No one scientific discipline has the right to ignore the existence of another competing theory until the science is proven one way or the other.
Keep the discussion alive, we all learn something. Thanks Anthony!
ShrNfr says:
July 23, 2010 at 5:41 am
Odd that in figure 3 that sunspot count is a trailing indicator of lake level rather than a leading indicator.
More than odd unless one can come up with an hypothesis explaining how Parana River flow could effect the number of sunspots.
The lag must certainly be an artifact of the methods used to generate the plot.
Funny what you find when you are trawling about isn’t it ?
Here’s a 2007 paper concerning water level fluctuation in the Great Lakes published by the US Geological Survey
http://pubs.usgs.gov/circ/2007/1311/
While it’s a complex subject here’s a paragraph from the Summary page that caught my eye :
“The reconstructed water-level history of Lake Michigan-Huron over the past 4,700 years shows three major high phases from 2,300 to 3,300, 1,100 to 2,000, and 0 to 800 years ago. Within that record is a quasi-periodic rise and fall of about 160 ± 40 years in duration and a shorter fluctuation of 32 ± 6 years that is superimposed on the 160-year fluctuation. Recorded lake-level history from 1860 to the present falls within the longer-term pattern and appears to be a single 160-year quasi-periodic fluctuation. Independent investigations of past climate change in the basin over the long-term period of record confirm that most of these changes in lake level were responses to climatically driven changes in water balance, including lake-level highstands commonly associated with cooler climatic conditions and lows with warm climate periods. The mechanisms underlying these large hydroclimatic anomalies are not clear, but they may be related to internal dynamics of the ocean-atmosphere system or dynamical responses of the ocean-atmosphere system to variability in solar radiation or volcanic activity.”
I actually saw an article on this by the BBC of all people. It was part of a natural history broadcast and may therefore have slipped past the NGOs at the BBC. The presenter was on the river with a local scientist as he explained that the flow was decreasing again having been high for several years.