Statistics

Shocking, just shocking.  Not!

It's good that Tamino did this so we have an easy reference next time somebody claims the station dropout or data adjustments are producing an artificial warming trend.

İ go through garbage on some sites about this winters weather - colder than normal so it obviously disproves global warming people claim - and it does to those who didn't believe in it anyway.

İ have seen photos of poorly sited weather stations - sure there probably were a few. Big deal - a few out of how many

You can't look at any point in time or location and point out a problem. Here in İzmir, this month the average temp has been 6 deg F above normal. The average low has been 7 deg F above normal - proving only that it has been a nice February in İzmir. İ expect that next winter will be not as nice.

@Russ - agreed. As you know, Oregon is an excellent example. Since birth I can't remember an "average" summer or winter here because our weather is always wacky. The only thing you can expect is the rain, but cooler summers happen and so do warmer winters - for years too not just recently.

Actually a large percentage, at least in the USA.  But groups like NASA GISS filter out any effects due to bad siting by doing things like comparing urban station data to nearby rural station data.  And they're very successful at it, as various analyses have shown.

All complaints about the instrumental temperature record are superficial, based on photographs instead of data.  They're qualitative instead of quantitative.

Russ wrote:
I am afraid that this logic of yours is deeply wrong. How about this scary logical construction:

For some benefit of doubt for climatology, consider that climatologists did not choose places for the weather stations. Climatology did not define the silly data sampling method either, min and max temperature, twice a day. AGW Climatology just uses whatever was/is available to gather data and make their case.

But where all these weather stations came from? They were built by meteorologists to guide people in what kind of weather is ahead, what to wear. These stations were never designed to detect trends of millidegrees per century. They were built for people, and therefore vast majority of met stations were strategically placed around places where people do live and expand. Therefore, not "few", but almost every met station is placed in wrong place (from climatology point of view), and therefore must be a subject of urban heat effect to one or another degree. And the urban bias goes mostly in one direction - up. Think about this, statistically.

Hi Al - Welcome back and İ see you are up to your normal standard of misinformation and parroted phrases you have managed to pickup somewhere!

Millidegrees per century?

 

Oh goody, AlTek is back.  And completely ignoring the content of this thread, he uses the same argument I just criticized.


It amazes me that anyone has the chutzpah to continue making this argument.  Countless studies have demonstrated that the urban heat island (UHI) effect is accounted for and filtered out.  Satellite data shows the same warming as surface stations - last I checked, there's no urban centers in orbit!  "Bad" stations show the same warming as "good" stations.  The raw data shows the same warming as adjusted data.

And yet guys like AlTek still expect us to believe the data is flawed.  And then they wonder why we call them deniers.

You guys are real funny. That's why we call you "believers".

 

Ask yourself, how it is ever possible to "account for UHI" if you cannot have a reference of what the temperatures would be without man-made alternation of surroundings of each particular location?

And how can you “filter out” anything if many stations as close as 50 miles apart show opposite centennial trends? Which one to filter out?

Take a look, for example, at temperature trends in Texas:

http://www.mitosyfraudes.org/Calen4/tex.html

 

Compare data for Albany and Haskell, they are only 63 km apart. One has a cooling of 1.1F/y, the other has warming of 1.8F/y.

 

Take another pair, Boerne and Blanco, less than 45 km apart. Again, one has a warming trend; the other has a cooling trend.

 

These stations are located so closely, but have entirely uncorrelated trends. Which one to filter? Why would one think that there is no more places where actual temperatures have one, or another trend.

 

Consider this. Area of Texas is about 700,000 km2. If two stations are 63km apart but have diametrally-opposite “warming trends”, it means that there must be at least one more station in between, to see if there is something else. So, to be statistically representative, the grid of met stations must be no bigger than 30 km apart, or one station for every 1000 km2. Thus, the state of Texas needs at least 700 stations. It has 50. One might think that station locations are randomly placed, so ten times more statistics would do the job. But how do you know this without having a properly designed grid first? You don’t. Please don’t give me these “countless studies”.

 

Now, think about what you wrote, "The raw data shows the same warming as adjusted data." Don’t you see the utter nonsense of this statement? Asphalt covers are facts. Aerodrome concrete is a fact. UHI is a fact. “Correcting” UHI _must_ show different, reduced warming.

 

Then you mention “satellite data”. Do you have any clue how many odd operations is required to get “global estimate” from swath of satellite “brightness”, and how many assumptions is involved? For starters, try this:

http://www.drroyspencer.com/2010/01/how-the-uah-global-temperatures-are-produced/

I will let you know my opinion on all this.

 

Cheers,

 - AlTek

 Hi Al - İ treat you the same as a screwball on another site - different topic and all but he is hung up on certain things. İ just scroll on past - don't even bother to read his or your comments.

Have a great day! 

Anyone who seriously argues that the temperature record is faulty is beyond hope, demonstrating no interest whatsoever in the scientific data or serious discussion.  You might as well argue that the evolutionary theory is wrong because the fossil record is faulty.

There is simply no reasoning with people who make unreasonable arguments.

What did you find in my arguments "unreasonable"? Could you be a bit more specific?

Are you familiar with Nyquist-Shannon sampling theorem, which specifies necessary requirements for discretization and sampling of continuously changing signals?

Do you know the extent of spatial spectrum of surface temperature distribution? No, you don't. My example above demonstrates that 50 stations across Texas are vastly inadequate for scientific evaluation of the field of surface temperatures, at least by a factor of 10. And I am not even talking about validity of crippled time-sampling methods emplyed in the net of weather stations.

Anyone who uses statistics to evaluate trends in variable spatio-temporal fields without adhering to scientifically-established requirements is ignorant. If data are gathered without regard for these known requirements, the data are not "scientific", period. Consequently, all results from this faulty analysis must be questioned. Misrepresentation of data by a factor of 10 (or more, nobody knows the real extent) is a very good reason to question the issue.

I do have a genuine interest in serious discussion of scientific data, but have no respect whatsoever for unscientifically-acquired garbage and for people who have no clues about established sciences and peddle conclusions that are not scientifically justifiable.

I'm only going to explain it again once, against my better judgment, since I know from experience you're just going to continue ignoring facts you don't like.

1) Satellites show the same warming trend as surface stations.  If you really think both data sets happen to be wrong by the same amount, you're delusional.



2) Numerous peer-reviewed studies have demonstrated that the UHI effect is accounted for and filtered out.  Google Scholar is your friend (well, not your friend, but a useful resource, unless you want to maintain your AGW denial, of course).

3) Studies have also shown that supposed 'bad' stations show the same warming trend as supposed 'good' stations, proving #2.  In fact they have shown there is a slight bias in the surface temperature record - a cooling bias.

You have to be certifiably insane, ignorant, or in complete denial to continue to question the reliability of the temperature record at this point.  It has been repeatedly demonstrated to be accurate.

Dana wrote:
I have already stated my scientific grounds on why the "global surface temperature" is prone to uncertainties and is an easy subject of selective treatment. As I explained, the original data are already unscientific, and no effort to massage them can improve the situation.

Regarding "repeatedly demonstrated" accuracy of the derived globals, you might want to familiarize yourself with this small effort of data "re-analysis" of HadCRU Temperatures . The analysis shows a shift of about 0.1C for earlier data, for allegedly the same data set but released at different times, creating an image of stronger, accelerating  "warming". So much for "reliability" of records.

On a less-scientific level of remarks, recent developments at CRU (Climategate) do not trigger any higher confidence in their data. Therefore, the fact of four data sets showing "same amount" of "warming trend" does not speak well for the whole issue, especially knowing that satellite readings are re-calibrated from nearly the same data sets (from even more undersampled ballon soundings).

To return the favor, you have to be certifiably lacking congnitive abilities not to see these incoherences. But you feel free to believe in all this, especially when all this fuzz about global warming has little-to-no realtionship to increase in atmospheric CO2.

Hi Al - Should İ post the Andy Capp Comic strip as a reply to your? İt does sound a bit like the little drunk English guy.  

No dude, it's not happening.  I'm not arguing an indisputable fact with you.  You might as well be arguing that 2+2=3, scientists used to think 2+2=4, but they're 'hiding the decline'.  As I predicted, you're ignoring every point I made because you're convinced you're right even though you're completely demonstrably 100% wrong.

Your position is "you haven't proven why my argument is wrong".  Your argument is wrong because I've proven your conclusion is wrong.  I've shown the sum of the numbers is 4, and you're still trying to argue that if you manipulate the twos, you can make them add up to 3.

I strongly suggest you reconsider your position, because anybody who continues to dispute the accuracy of the instrumental temperature record at this point loses all credibility in this 'debate'.  You might as well argue that the evolutionary theory is wrong because god told you so.  The argument you're making is completely wrong and indefensible.

I do agree that HadCRU is flawed.  By excluding the Arctic, it underestimates the recent warming slightly.  That's why I prefer NASA GISS, which is the gold standard of global temperature records IMO.

Damn, you are right!  The uncorrected data shows absolutely no warming from 1900 to 2000!  Those duplicitous bastards fiddled the data to suggest the planet got warmer!

Great call AlTek!  You (and Stevie Wonder) certainly have a keen eye for when a minor change in data analysis makes a huge difference in the conclusions!  Clearly, without that reanalysis there is no indication of warming, therefore absolutely no reason to think that CO2 might cause a surface radiative forcing and warm the planet (aside from all that physics and what-not (clearly wrong though, because you say it is)). 

Go buy yourself a beer.  You've earned it.  Well, Stevie Wonder did really, but get yourself one anyway for pointing out this now 3-year old trenchant piece of analysis that changed the face of climate science. 

Quote:
This would be a meaningful objection, except that by comparing the data from the website you cite above with the actual data from the US GHCN network you can show the data on the skeptic website are incorrect.  So all you've done is provide a correct analysis based on faulty input data.  The real data for Texas don't show nearly as much regional variability and the spatial scales of the sampling network is just fine. 

The question you ought to ask yourself, if you're so smart, is why you were so easily taken in by someone peddling bad data.  Perhaps because they agreed with what you want to believe and you're not really as objective as you think you are?  Ya think?  

I don't think we can really trust anything you say with respect to climate.  Whatever analysis you provide will be slanted with bad data or a fatal flaw somewhere, that you can't/won't acknowledge for emotional reasons. 

(just decided to re-visit your nice website. Wow, what a progress!)

gcnp58, why don't you take a deep breath and calm down? Even Dana disagrees with your assessment of data I use. Dana wrote:

Guess what? The cite you denigrated uses the GISS data:

 

Haskell

Albany:

Details of their statistical approach might be discussed, but the fact remains - Albany cooled down, while Haskell warmed up. All 66km apart. Even after having 3 beers.

If you are not lazy or dumb, you are welcome to study the trends close, yourself - GISS - TEXAS.

The fact remains: surface data are undersampled to derive any confidence about trends. This is all noise so far.

Cheers,
 - AT

Al - We are sure you use data - we are just not sure what you use it for.

Quote:
And they *all* show warming from about 1975 onward.  I'm not sure what you want in terms of spatial or temporal synchronization.  How long would that warming trend have to go on before you would begin to question whether what you think you know to be true is in fact an emotional bias on your part?  Probably that would never happen, because you are not skeptical, you are in denial.  

Anyway I notice you no longer point me back to the skeptic site.  I guess even that data was a little rugged for you to swallow, but you did try didn't ya?  Thank you for giving up trying to convince people that parts of Texas cooled over the last 25 years when in fact it warmed everywhere.  You're showing brief glimmers of intellectual honesty, which I didn't expect. 

Look, I know you're too smart to even consider something you think is wrong, but a lot of us who are more able to consider things objectively, well, we're snickering a little at you. 

gcnp58 wrote:
Really? All? I was curious and looked a bit deeper. For example, into stations (1),  (2),  (3),  (4),  (5), (6),  (7),  (8),  (9),  (10)...

The following question immediately arises: why would all these stations that show clear decline in temperatures over several decades would suddenly stop reporting their data into GISS database? Huh? Homogenization anyone? Or because they would "add to decline"?

Or these ones that do not show any warming for more than a half of century: (11),  (12) ???

Quote:
LOL! Initially you had an excuse that the skeptical side has allegedly bad data. Now when I pointed out that the data are from GISS database (2000 variant) and started to use the original source, you are complaining again? Your intellectual honesty is really over the top.

So, objectively looking, all this is a random walk across random selection of technically impaired data.

And the emotional outburst is all yours, and is clearly misaddressed. Instead of snickering, next time try harder, and study data more carefully and objectively.

All I can do when reading an AlTek post is shake my head in disbelief.  I don't know how you have the patience to argue with this guy, gcnp.  Maybe you just like arguing more than I do   But it's clear at this point you would make more progress arguing with a brick wall.

Quote:
Do you mean that some other ignorant AGW activist has an opinion on data processing?

Dana wrote, quote:
Could not agree more. When your arguments have a sophistication of a brick wall, there will be no progress, it is for sure.
But judging by your silence about my compilation from GISS data, you seem to agree that "gcnp58" was grasping at pathetic emotional straw.

Cheers,
 - AT

He's baiting you Captain (and me, and gcnp) - I suggest we all just ignore him.  No sense in feeding the trolls.  Ad hominems don't warrant a response anyway.  AlTek is Dunning-Kruger personified.

AlTek, if you want to keep believing that 2+2=3, go right ahead.

Quote: No, I am not "baiting". I am simply trying to demonstrate to audience how narrow and ridiculously incomplete is your point of view on global warming issues, that the conclusions you are peddling are completely unwarranted from scientific standpoint. That's it.

(could it be that I am using a wrong smiley?)
Not sure about your arithmetical example, but I could be perfectly fine with 2*2=0.

Cheers,
 - AlTek

I'd be interested to see a refutation of the points made in the Tamino article that Captain linked to. Dismissing him outright as "some other ignorant AGW activist" doesn't really count. 

Dawei,
I think the author of observation that temperatures are more like a random walk, "VP", has pretty much refuted the G. Foster's pontifications. IMO, Mr. Foster has no points, only a nit-picking. He clearly does not know physics and methods of it, and confuses abstract mathematical constructions with physical approximations of reality that is usually limited in time and space. In part, he argues silly things like since the random walk is theoretically unbounded while historical observations show that the temperatures are well bounded, then VP therefore must be wrong. 

In case you do not remember, MIT professor Carl Wunsch also came to similar conclusion that all climatological time series are better approximated by some random process and are not any responses to Milankowitch forcing.

I cannot take seriously a person who claims that four cycles in a time series is "safe ground" to conclude that the motion is periodic. I think he is a better musician than statistician.

I do understand your frustration with data. You want to see a clear climate response to alleged CO2 forcing. Clearly, the proposed physics of greenhouse forcing dictates that each and every spot on Earth must have this warming trend on time average, since every ground spot faces a well-mixed and fluid atmosphere. Unfortunately, Mother Nature does not want to cooperate with your goals to control carbon, all you measures appear as a noise, where nearly any trend is possible depending on start and stop points. Why the forcing has so little response is not clear. I tried to suggests few possible physical factors why observations do not meet expectations, but was met with hostility and ridicule from your cohorts, with solid dogmatism of a "brick-wall" . Not that I care too much, but it is a bit frustrating.