Post by socold on Sept 13, 2010 0:01:52 GMT
You aren't getting it Socold!
1. Every single link you posted is not an observation record but a reconstruction record.
Phil Jones takes temperature data from different greenland stations and performs an analysis to produce a single graph of temperature from those individual records. What part of the process makes it a "reconstruction"?
Weighting is employed to produce a single graph of temperature elevating poorly sampled region errors over well sampled region errors. A course in upper division statistics would explain what that means to you.
It is an effort to arrive at an absolute value and its a constructed value, not a measured value.[/QUOTE]
On one hand you disown these "reconstructions", but on the otherhand you depend on them for your arguments. Eg earlier you claimed Greenland hadn't warmed. How do you know that if not for "reconstructions" of Greenland temperature? And in your next argument you make a claim about US temperature that also depends on "reconstructions" of US temperature.
If you want to explain global warming from its fundamental basis of increasing global heat content you need to understand why the best sampled area of the world has not warmed over the past 70 years.
While you might think it's a good idea to dismiss temperature records you are actually digging yourself into a corner. Whole swathes of arguments are going to become inaccessible to you and even worse it's going to make it harder for you to dismiss manmade global warming. The less you know the less certain you can be of the stuff you are claiming.
Until that is answered that fact alone implies huge uncertainties in the global climate record.
That's your view. Remember I accept GISTEMP and HadCRUT as roughly accurate portrayals of global temperature, so I don't accept such huge uncertainties in these records. I don't think you truely realize the ramifications of your actions. If you don't accept "reconstructions" then you can't know what temperature has done in regions such as Greenland and the US in the past 100 years. That also of course applies to the globe too. Without "reconstructions" you don't know what global temperature has done in the past 100 years.
With no reconstructions for all you know temperature might have correlated with co2 perfectly in the past 100 years. Uncertainty is a double edged sword. It's easy to weild and use dismiss scientific results you don't like, but you'll find it also hits back at you by making certain arguments against manmade global warming impossible. If you run back far enough and disown enough science as useless and uncertain you'll reach a terrifying position of darkness where all you know is that co2 is rising in the atmosphere, but you can't see what's going on with temperature, sea level, etc. In many ways ignorance is worse than knowing stuff about the climate, as little as we do know.
A thorough audit of Jones and Hansen would reveal the biases in the temperature record to the extent they exist, which undoubtedly they do considering the attitudes of these people and give you a far better perspective on the remainder of the issues than we have right now.
"Undoubtably they do"? Since when are scientific conclusion drawn from vague senses of "attitudes of these people"? Here's some facts instead: GISTEMP and HadCRUT source code and input data is available. If you want to audit it knock yourself out.
The CRU raw data issue has sort of fallen from view but it is a real issue. I suspect where ever you start your audit it is going to come back to some of that missing data.
It's fallen from view because it's not a real issue. The downloadable HadCRUT source code operates on GHCN raw data. That shows the same result as the reported CRU one for years. Therefore the result does not depend on any CRU raw data issue.
You are focused on coming up with a "greenland" temperature. I wouldn't even attempt that without far more complete GIS data than we currently have. I also don't think there is any need to come up with it.
Yet up a bit you claimed Greenland had not warmed.
What I would do is simply average the stations after reviewing the individual stations and classifying them in a way similar to the WUWT station classification system so you could do some analysis for station accuracy. Now I am aware that simply averaging stations creates the potential for regional biases but I would classify regional biases as being "weather" from the perspective of a global warming trend.
Averaging implictly weights using equal weighting. What you've done is weight all stations with the same weight. What about spacial correlation? What if you have 3 stations within a mile of each other and another station 100 miles away? Are you going to average them anyway therefore giving undue weight to the area with the 3 stations, or are you going to spacially weight them? In any case what does your average represent? The region? If so then you've infilled/smoothed data, because those stations are only samples of the region.
The data smoothing exercises like Hansen's and Jones' merely serve as a measure of the degree of regional weather bias in the record and provide a away of feeding data into a GCMs but I don't see any basis for declaring a smoothed dataset as being a more reliable predictor of a climate trend than an averaged dataset and the averaged dataset will be clear of anthropogenic bias provided bias does not enter into the selection of the dataset
Your method smoothes too. If you are claiming you can just average 4 stations at each corner of Greenland and call the result "Greenland temperature" then you have implicitly smoothed the data for those 4 stations all over greenland. Any graph of "greenland tempeature" will have to smooth data. There are not thermometers covering every nanometer of Greenland.
Akasofu presents a reasonable approximation of climate and weather cycles in his reconstruction. It is far from complete, primarily because it does rely greatly upon the climate reconstruction work of Jones and Hansen, but he does outline where some of the problems are.
This is another argument you cannot use. Akasofu's work doesn't just "rely greatly" upon "reconstructions", it depends on them. If you have no graph of global temperature then you can't do what Akasofu did. If you don't know what global temperature did in the past 100 years, then you can't know it cycled. It might even have shot up in an exponential curve.
Hansen has argued that absolute temperatures are not as important as measuring the trend. Yet the Jones and Hansen methods of gridding and smoothing data is trying to get to an "absolute" anomaly in the presence of regional weather pollution of longer periods than the existence of an adequate global instrument record. (which in fact may still be lacking). So the extension of Hansen's argument applies to his own work in comparison to an unweighted averaging of stations to determine a trend. He did nothing to fix the inherent basic problem with uncontrolled variables. In fact such an averaging would make a good test of the gridded dataset trend. Since AGW theory produces different warming by latitude the fact that the best sampled region is in the mid latitudes any latitudinal error from averaging stations should be reduced. At any rate any error from averaging may well produce far less error than using questionable data from a few remote outposts to estimate warming for the unsampled areas of the region.
Averaging the absolute recorded temperatures of all stations in the GHCN dataset has been done:
rhinohide.wordpress.com/2010/07/10/trb-0-01-ghcn-simple-mean-average/
There are severe problems with this method though, the solution to which is the purpose of gridding and converting station records into anomalies before averaging. Your audit will find this out.
Of course nobody is going to do that because they don't want people to know there are problems with the dataset.
They aren't stopping anyone from doing it.
To go ahead and grid the temperatures like Hansen does pretty clearly results in biases favored by Hansen. So he accepts it and defends it. But only inside his own egotistical mind has he proven anything. The glaring fact is he has done nothing whatsoever to explain the weather biases in the record.
Then don't do it like Hansen does. Be a hero and show that a different valid method produces wildly different results. Show that hansen's result depends on his method.
Tell you what, go onto WUWT and Climate Audit right this instance and demand they produce their own global temperature record. That's what's needed. Nothing from me, nothing from the IPCC. Nothing we say will ever change your mind. Any audit will be met with calls of "whitewash". So do yourself a favor and demand that the only people who can satisfy you do an audit. The only way you are ever going to admit the global temperature records are not biased by Phil Jones and Hansen is if someone like steve mcintyre makes a temperature record and then you'll see it looks remarkably like hadcrut and gistemp and you'll have no excuses left. That's the only way, so stop insisting an audit happens here and go and demand one where it counts.