Chill in the air: record low temps in 10 states

Ken Fabos,
You’ve forgotten to post the obligatory “Gore 08” slogan at the bottom of your comment.

alexjc38 (10:58:11) :
“I’m wondering whether the people who believe we can somehow “stabilise” the world’s climate by adjusting CO2 levels, like selecting values on a thermostat, have a clear idea of what would be the perfect state of worldwide heat and cold?”
This was raised by George Smith in another thread, when he said “…and what temperature would you set it to if you could?”
Which prompted me to think about WHO gets to decide where to set it?
Talk about a political mess. I do believe it’s an excellent question to ask someone engaged in this debate tho. It at least makes them think about something they haven’t been programmed to respond to…yet.
Jim

But, but, but the arctic’s melting:
http://www.timesonline.co.uk/tol/news/environment/article5014744.ece

” DAV (13:37:17) :
>> deleted <<
That may be true but the values changes from the previous record will become less and less over time. After a while, the record can only be broken by a value that has more precision than the measuring equipment. IOW: you will asymptotically approach the “real” extreme. ”
Well DAV, that presupposes that what is being measured is supposed to be some value that never changes. Also many physical processes exhibit something like 1/f noise; that are actual variation in observed value, (so they are real data) but the variations from any sort of main trajectory can grow indefinitely but at ever reduced frequency.
Now if you average every single data point (of a single variable) that has ever been measured, then I agree that the next data point is going to change the mean by a diminishiong amount (on average)
I have been known to claim that the “big bang” was just the bottom end of the 1/f noise spectrum.
And for hungry Brute out there; Nyquist is quite simple and you are very familiar with it unless you just never watched a horse opera movie or TV show. TV and movies are sampled data systems, in that you are presented with discrete pictures taken at certain intervals of a scene that is continuously changing; for example the wagon wheels on the runaway wagon with the maiden in distress, are rotating at some speed so the spokes move around at a certain rate, and unless they are non uniformly spaced, each spoke eventually replaces its neighbor at some frequency. If you sample at exactly that frequency, the spokes appear to be stationary; whcih is a false reconstruction of what is really happening.
A fairly simple statement of the Nyquist sampling theorem, would be:- “Any BAND LIMITED continuous function; f(x,y,z) can be completely reconstructed from a set of discrete samples of the function , so long as the sampling rate is at least twice the highest signal frequency contained in the band limited signal. So for example an audio music signal that contained no signal frequencies higher than say 20 KHz, must be sampled at least as fast as 40 KHz, in order to accurately reconstruct the original music, without distortion. if the sampling rate was 40 KHz, but the original music included signals as high as 25 KHz, which would be entirely inaudible to most people, upon reconstruction, would yield a noise component that was 5 KHz BELOW the 20 KHz band limit, or 15 KHz, so it now appears as an in band signal that is erroneous, and in an audio system would be eminently hearable. signal at twice the band limt or 40 KHz, same as the sample rate, would appear at zero frequency, which is the average value of the signal, and so even the average is not recoverable for just a 2:1 undersampling.
No central limit theorem, or other mathematical prestidigitation can rescue you from Nyquist violations; the noises are in band and cannot be filtered out (unless some how you know exactly where they are and could devise a narrow band filter to exclude that band, whcih would also exclude any real signal that was there.
Now in Global temperature determination, you ahve spatial variations over the globe, covering a total temperature variation range that could go from +60 C or hotter, to -90 C, all on the same day. A 6PM weather map of just the SF Bay area for example last night, showed over a 10 deg F range of temperatures occurring over a spatial extent of no more than 50 x 50 miles.
As for time variations; I understand that most of these weather stations report daily min/max temperatures. If the daily temperature excursion was simply sinusoidal, thent eh min, and max would occur 12 hours apart, and you would in fact record two samples per complete daily cycle, which is OK. But the daily temperature record is not sinusoidal, and even if you ignore clouds moving around during the day, it is not sinusoidal, since heating under a 6000 Kelvin solar source, and cooling by IR radiation to outer space, are not comparable processes. So the daily min/max strategy is already in violation of Nyquist even without cloud changes during the day, and cloud changes do change the amount of heating or cooling that aoccurs in that area.
Heat island measurements observe local ofsets due to the heat island, but they are often used to represent a much larger areal extent, than just where the temperature hots pot offset occurts; and that is the crux of the heat island problem; not that heat islands report abnormal temperatures due to very local effects; but that those abnormal numbers are then used to represent remoter areas; that is a Nyquist violation. the heat island is aphenomenon well outside the “band limit” of the GISStemp sampling process..
But in any case; suppose you could actually measure the true average surface temperature of the earth (you can’t afford to); there is no scientific validity or meaning to the result; because the heat flows, into or out of planet earth; which is what you are really wanting to know (are we heating or cooling), are not simply related to local temperatures. Some effects are related more to the fourth power of the absolute temperature; not to the average temperature. (such as Infra-red radiation from the surface). Other effects are more dependent on the fifth power of the local temperature. The peak radiant emittance of the emitted IR varies as the fifth power of absolute temperature; and the wavelength at which that peak occurs varies inversely as the absolute temperature, so the amount of IR radiation that falls within say the resonance absorption band of the CO2 molecule, varies in a very complex way with both the surface (emitting temperature) and the receiving (atmospheric) tempertaure, which determines the kinetic statistics of the CO2 molecules, and hence the Doppler broadening extent of the intrinsic CO2 line width.
The thermal processes over water are quite unlike those over a rocky desert, or an alpine tundra, ar a tropical rain forest. So even an accurate knowledge of the true earth average temperature; tells you nothing meaningful about the net heat balance of the planet. It’s what i call a phone book average.
You can average all the telephone numbers in your local phone directory, and come up with an average telephone number for your region. It might even be a real telephone number that would ring if you dialled it; but unless it is your telephone numebr, it is of no earthly interest or use to anyone, and even more useless, is what happens to the average telephone number when they print a new edition of your phone book with some new numbers.
You can apply statistical analysis to any set of numbers you want to; even the calculated output values of F(x); but that doiesn’t mean there is any meaning to the results. Some numbers are jsut meant to be different.

For Ken Fabos,
Could you please point us to the scientific evidence that has proven that AGW is true.
Please note Ken, I’m asking for scientific evidence, not PR announcements.

Basil:”New records, given how long a history we have now of temperature data, are going to be, by their nature, extremes that are outside any reasonable range of what’s considered “normal.”
I would agree only if the records are well outside and isolated from a consistant range. If the record is simply the lowest of a lot of close by low temps, then that could and should be considered part of the normal range of temps. Same with high temps. For example, if there are a bunch of low temps over time for a certain date between 47 and 50 degrees with 47 being the lowest on record, then you could say that it is still part of a normal range. But if there were no other low temps under , let’s say 54 degrees, then the 47 would be an extreme outside the normal range. The point I was trying to make was that I think it misleading to focus in on a single temperature and call it normal or average. No one could possibly think that a low temp of 63 degrees would be “abnormal” when the cited normal temp is 65 degrees. Yet isn’t that what all these charts, graphs and models try to do? They try to pin down a specific data point as the benchmark for every other data point. I think using ranges is a better way to go, especially when dealing in generalities with the public.

Have you seen paper Dr Vincent Gray released? It talks about average temps.
I case any of you want to read it the link is here.
http://nzclimatescience.net/index.php?option=com_content&task=view&id=369&Itemid=1

George E. Smith (17:24:07) : Well DAV, that presupposes that what is being measured is supposed to be some value that never changes. Also many physical processes exhibit something like 1/f noise; that are actual variation in observed value, (so they are real data) but the variations from any sort of main trajectory can grow indefinitely but at ever reduced frequency.
Now if you average every single data point (of a single variable) that has ever been measured, then I agree that the next data point is going to change the mean by a diminishiong amount (on average)
I have been known to claim that the “big bang” was just the bottom end of the 1/f noise spectrum.
Not at all — at least not in the near future. It is highly unlikely that the Earth’s temperature will rise indefinitely if one is to take the lessons of thermodynamics to heart.
Don’t confuse mathematical model with reality.
Brute (15:56:10) : If the “average” temperature is rising, shouldn’t the extreme high temperature rise also?…… Somewhere?…. At sometime during the course of the 100 year period? Is it simply because we are discussing such infinitesimally small increases? Are you guys suggesting that there is a ceiling or limit on extreme high temperatures?
Well, yes and no. How’s that for simple ? 😉
Temperature (large scale) is connected to heat energy. The availability of energy affects the maximum temperature. The max/min records can only get wider by definition. There’s only so much energy to go around.
Another max/min example: Take an auditorium of 1000 people.. There’s an obvious limit to how loud or quiet any given set of occupants can be because there is a limit to how loud or quiet any given person can be without electronic or mechanical aid. The audience can’t only be at maximum loudness if it consists of the 1000 loudest individuals. Now if there are more loud people than quiet people, the audiences will become louder on the average. However, just because there are more people who are louder doesn’t mean the loudest possible persons have gotten louder.
Kohl Piersen (14:28:11) : … er, I’m no going to quote all of that. Let me just say there are usages of averages that are meaningful — center of mass, e.g. I have to admit that locating the center of population might not have much use to most people.
Since you’ve a lot of questions, the ones that should be uppermost in your mind is “can the Earth really have an average temperature?” and “If so, what would it mean (no pun intended)?”
Here’s someone pondering just that.

Perhaps we should add this to the discussion.
http://www.signonsandiego.com/news/state/20081026-2032-wst-maunakea-snow.html

Apparently, AGW has had a significant impact on Walden Pond…time for another concert?
http://blog.wired.com/wiredscience/2008/10/climate-change.html
Jim

When I read an article that tells me that there’s a) been a 4deg increase in temperatures at Walden Pond in the past 100yrs, and b) lilacs, along with %36 of the other plants at Walden Pond are in “imminent danger” because of this increase…well…I guess I really am a skeptic.
Jim