“I’ll bet they are measurably higher. Average global temps for a decade around 1995 and 2005 and see what you get. Then try it for a few more decades back.”

Except since 2005 CO2 has continued to rise and temps have fallen precipitously. Doesn’t appear that CO2 has much of anything to do with weather OR climate!!

HAHAHAHAHAHAHAHA

If you are a properly-brought-up-environmentally-conscious-global-citizen it’s a reasonable statistical inference that you will likely end up as grandfather to a child who has four grandparents who have no other grandchild, or perhaps you will have no grandchildren at all. It’s quite comical to hear the zero population enthusiasts try to justify their aims by appeals to future generations. Those who believe that the best thing we can do for our children is to not have them really don’t have any stake in the future.

REPLY: I sent him a personal email, he’s in a state of denial and refuses to look. Turns out he’s the curator of a prominent museum of science, and you’d think he’d have a better handle on dealing with the public aka “us” – Anthony

‘Lord of the Rings’ ‘An inconvenient truth’ “IPCC Assessments 1-4″ -six of my favourite science fantasy novels.

Incidentally I meant the Romans from 600BC (not AD) so that gives us 2000 years of ‘anecdotal’ records.

Tonyb

Hank, I’m not sure what you ment, but what I did use were the 1st 13 freq.
in the freq. domain. The freq. increments are given by 1/(# sample [512] times the time increment [1 year] ). So freq. from 0.002 to 0.025 cycles per year were used to re-construct the filtered input in time. Does that help?

Tonyb – The historical-weather climate observation was only that, and would be worth many threads, which I would love to get into. However “given the time alloted to us”, [Gandolf, Lord of the Rings], I guess I’d better stick to what I started out, namely apply signal conditioning methods to get better insight into long term weather/climate patterns.

As you say History tells us a great deal-there are many papers on all the events you mention, many posted by my self and others in this forum over the months.

The trouble is that these events are called ‘anecdotal’. A complete Roman army destroyed when their enemies crossed the frozen Rhine is anecdotal! Yet cherry picking highly dubious bristlecone pine proxy is the epitome of science! The Romans were great ones for recording weather so we can reconstruct the climate of much of the Known world from around 600AD to the fall of the Byzantine empire in 1453AD.

The Vikings of course are dismissed as being a very localised anomaly.

George E Smith

Did you read my link to the Hansen paper? This describes the number of thermometers he used from 1880 to construct his global temperature. The numbers worldwide in 1850 were about 20 (that could be considered reliable).

There is no doubt the climate irrationalists are scared of history. The story hinges on limited climate variation within natural variability during constant co2 levels which only in recent years goes beyond natural limits due to increased co2 concentrations.

Tonyb

E.M. Smith

Words fail me. The thermometer readings are used to determine a temperature for each point of the Earth’s surface, then you integrate over the surface. If you have two streches of land of equal size, one with 10000 thermometers and one with 10 thermometers, the former having an average temperature of 20°C, the other of 22°C, then the average temperature over both is 21°C and not 20.002°C, as you imply. “”"

Well you are forgetting that there are approximately zero thermometers over 70 percent of the earth’s surface. It is my understanding that around 1850 there were precisely 12 thermometers in the Arctic (north of +60 Lat), and that number increased over the years to around 86 or so, and then in recent years decreased to avout 72 or so; my guess being that cionicided with the collapse of the Soviet Union.

It is laughable to suggest that this hodgepodge of sampling methodology, in any way conforms to the basic laws of sampled data theory. Having 10,000 thermometers on one area and ten on another, doesn’t improve the situation over having ten on each.

Nyquist does not reqire that the sampling intervals be equal; but the maximum sample spacing (anywhere) sets the band limit of the recoverable signal; not the minimum spacing. Ergo, GISStemp, and HADcrut are both just a set of numbers bearing no relationship to the temperature of planet earth.

All of the wonderful manipulations of statistical mathematics can be appled to any arbitrary set of numbers; those numbers don’t actually have to have any linkage to each other; they are simply a set; yet they have an average or mean, a median, a standard deviation; any high falutin extraction you want to make; you just can’t extract any information from that set; there isn’t any to be had.
It can be argued that the maximum information density is carried by white Gaussian noise; because no matter how long a sample sequence you want to study, at no point can you predict the value of the next sample; you can’t even predict if it will be higher or lower than the latest value. So in that sense the signal is 100% information, having zero redundancy.

Conversely, GISStemp, or HADcrut have close to zero information content; they violate Nyquist both temporally, and in spades in spatial sampling.

You only need a violation by a factor of two, to make the average unrecoverable (without aliassing noise).

And the Central Limit theorem cannot buy you a reprieve from a Nyquist violation.

2 – Hank, your comment on the famine in Ireland, correlates to research I have been doing on our various families. The result was how many population migrations were caused by the weather/climate? There seems to be a pattern from looking at the various family backgrounds. These are just pieces of info, without any research, but a pattern might emerge.
Unfortunately we have very little Chronicles to go on, but a history of Pomerania, Germany, etc. give glimpses.

~500 B.C. Goth and other Scandinavian tribes head south into Pomerania (northern Poland now). These included the Goths, Vandals etc. Cause of the migration, seems to famine (poor growing weather).

~100 A.D. Roman occupy England, making wine from local grapes. (warmer weather?)

~400 A.D. What is called by the Goths “the great migration” move out of Pomerania, south, into the Roman empire. About that time the Rhine freezes over, allowing the tribes to cross into Roman land. About that time the Huns burst out Eurasia.

~900-1000 A.D. Vikings settle in Iceland and Greenland. Settlement in North America.

~ 1350 – Last Viking settlement in Greenland, beginning of long cold period.

Anyway there seems to be some correlation to weather and mass migrations of the past, and seem to follow a 800-1000 period. That whole area would be worth a few papers.

Sorry for the spelling mistakes-it ‘escaped’ before its time!

tonyb

As long as you’re going to check with Anthony, let’s back up a step, and ask him to direct his attention not just to SHAPS and TOBS, but rather at the validity of the sum of these adjustments, as related by the graph which started this discussion;

http://www.ncdc.noaa.gov/img/climate/research/ushcn/ts.ushcn_anom25_diffs_urb-raw_pg.gif

which YOU claimed were attributable mostly to SHAPS and TOBS, were incontrovertible, entirely justified and necessary.

I indicated the notion that anyone would adjust the most recent readings of objective instrumentation up by .6 C and then claim there’s been .6C warming in the last 100 years and the science is settled is rather a strange way to conduct sciencere. That’s like starting with a hypothesis, and then adding fudge factors to the data until you get what you want.

The alternative, of course, would be to take the objective measurements, and after realizing there’s been a century of no change in the data, start to look at those measurements and see if there’s some change that’s been camoflaged by sighting issues or time of observation issues. We’ve clearly gotten the cart before the horse on this.

Sincerely,
Bob Kutz

Excellent work I might add. I was intrigued to see how the FFT compared to the Chev 4. Chev 4 appears more reactive to the higher frequency components. Curiously, it pulls a slight negative phase shift around 1800, perhaps in response to the large extent from 1770 through 1790. I’m curious if you used a four point transform in the FFT?

http://www.imagenerd.com/uploads/t_est_25-avCpP.gif

The moving mean plot was a real eye opener! At 1825 and more notably in the mid 1840’s it’s at a totally opposite phase from reality. It is showing a marked warming trend leading into the 1840’s – a time when unseasonably cold and rainy weather was bringing on the Ireland potato blight and the Great Hunger. What it most interesting is it’s running a 40 year spread. It’s kind of close to the 30 year spread bluegrue alludes to being just right to see the global warming. Now I get it!

I reread your reference to Upsalla-incidentally none other than Arrhenius lived there-somewhat ironic.

The grand master of interpolation is James Hansen-who was a brilliant scientist in his day but made a lot of assumptions based on a small number of poorly distributed weather stations, many with chunks of data missing.

http://pubs.giss.nasa.gov/docs/1987/1987_Hansen_Lebedeff.pdf

This paper managed to make it to the infamous Congress hearing in 1988 and was a fundation stone for the first IPCC assessment. Count the number of ‘uncertainties’

CET and Zurich Fluntern are both very old records, the latter two closely mirroring each other.

Tonyb

You clearly don’t understand the concepts in the graphs in this article,

The discussion had drifted. Here’s Motl’s second plot extended to the full Hadcrut3 data set.
http://i25.tinypic.com/wji05z.png
The pink curve is the same analysis for a simplistic model: temperature anomaly is zero until 1970, it rises linearly to +0.65°C today, add noise. Not too bad for such a simplistic model.

And per popular demand ( woodfortrees (Paul Clark) ), here is Motl’s first plot for 1998 as an end date, I consider this one to be nonsense:
http://i29.tinypic.com/wuglz7.png
Absolutely ridiculous warming rates.

P.S.: I’ll call this a day, CU.

Now, if you take the .6 out of the data (That’s -.1 +.7) There is no trend at all, and,

Necessary correction of known systematic errors, US48 only, Celsius / Fahrenheit. As I told you before. This gets boring.

as Anthony has observed many times, there is no way you can justify that adjustment.

I’d like to hear that from the source.
@ Anthony: Do you think that TOBS and SHAP are necessary corrections to the US48 temperature data or not?

As for cycles, all you are showing are variations in a single location, not global level. Secondly, would you please quantify your periods, rather than say “Hey, just look”?