I strongly suspect that they are justifying the “good” stations and the “bad” not by adjusting the bad stations down, but by adjusting the good stations up. That is the only way I can explain the bottom line. (They are also burying it under offset rather than absolute values.)

They also use USHCN2 comparisons without noting the change from the USHCN1 procedure. They come up with a much higher upward adjustment than last time and they do not show how much of the adjustment comes from each step in the procedure. That makes me very, very suspicious.

I am continuing the analysis. But so far, I don’t trust the historical network. And if it weren’t for RSS and UAH, I wouldn’t trust them with the current tracking either.

Imagine you were asked (in the days before planes and satellites) to produce a relief map of the United States – how many readings would you need to take? You would not need to measure the height of absolutely every square inch, but just make a sample. But, on the other hand, a sample of just 100 measures of height above sea-level would not give a decent map. I would guess that you would need a sample containing many millions of measurements before a detailed map could be produced .

Perhaps you were asked to work out the average height above sea level of the USA. In how many places would you need to take data in order to come up with a reasonable estimate? To my simple mind that problem seems very similar to trying to find the average temperature – except of course that, unlike height, temperature is constantly changing.

Christ. Half again worse than their previous outrage. I picked over those adjustments step by step and they were an scandal to the jaybirds. looks as if I am going to have to do the same for this USHCN2 adjustment outrage.

I have been to the NOAA page. They have seen the error of their ways. It was easy to pick apart USHCN1. It was ridiculous on the face of it. But they are much cleverer this time. They graphed each item of the overall adjustment for anyone to see. It was one of the most-cited graphs by the skeptics. And for good reason.

But there ain’t no graphs on THIS page. Just a phony-baloney bromide job gussied up to sound reasonable.

I found no obvious link to any powerpoint presentation. If it’s there, they are hiding it very well. I can’t find it.

Rev, would you PLEASE post the link to the powerpoint.

I need to find the same adjustment data for USHCN2 that they provided for version . I need to pin down the adjustments piece by piece the same way they had it in USHCN1. Even those maps only have the overall, not the piece-by-piece. You can’t pin them on it alone–except in tandem with USHCN1, which, at the least, I intend to do.

“Trust, but verify”.

Not possible.

The more I think about this the madder I get.

This is not going to go unsung. I swear it.

REPLY: The link to the powerpoint is there in the post as “watts-visit”

but here is the URL:

http://wattsupwiththat.files.wordpress.com/2008/05/watts-visit.ppt

The basic flaw in the unadjusted maps is that they show TOO MUCH WARMING.

Why do you ask? #B^U

Obviously they are making a hash out of SHAP (and probably FILENET, as well).

It is an outrage! I shall tell everybody!

Are these stations that form the bulk of the data private? I don’t think the Joseph station is on the grid.

“Perhaps Philip B can explain why opinion pollsters like to get a sample of over 1000 if it has no more real signficance than one of 100.”

I will take a crack at it. The first reason is to enable analysts to look at subgroups within the general population. If we take a sample of 1000, we still want to get over 100 samples (for example) from each race, and if we want to look at how white women look at an issue versus black men, then we need a valid sample by race and gender. You can see how we quickly get over 1000 in required total sample size if the analysis for each subgroup is to be valid. The BLS surveys 30,000 households to get valid statistics for each state, for each race, for each gender, for each metropolitan area.
The second reason is that a 96% confidence level is not good enough to make a projection on national T.V. election night. You need to well over 99%.

- start with 50 volunteers, each of whom is assigned a couple of dozen stations. This should cover all the stations.

- each volunteer will plot the graph for each of their stations, and note which, if any, of them seems to have a step change

- any stations that are flagged at this stage will be looked at by a second group to confirm that a step change happened

- the second group will issue a list of suspect stations

I’m not sure exactly what to do after that. I suggest a 2 step process. Submit the list and documentation to NCDC and/or GISS, via Anthony. If NCDC/GISS don’t act on it, they’ll hand us a major PR coup. We can go public with the list and show people that “the top ten all time hottest years in the USA” are the result of garbage data.

Hadley CRU http://www.cru.uea.ac.uk/cru/data/temperature/hadcrut3gl.txt

RSS ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tlt_anomalies_land_and_ocean_v03_1.txt

and UAH http://vortex.nsstc.uah.edu/public/msu/t2lt/tltglhmam_5.2

Off topic, but important.

THE HEARTLAND INSTITUTE

19 South LaSalle Street #903
Chicago, IL 60603
phone 312/377-4000 · fax 312/377-5000
http://www.heartland.org

MEDIA ADVISORY: Polar Bear Decision Defies Scientific Evidence
Author: Dan Miller
Published by: The Heartland Institute
Published in: News Releases
Publication date: May 2008

(Chicago, IL — May 14, 2008) The U.S. Department of the Interior decided today to list polar bears as a “threatened” species under the Endangered Species Act. The decision was based on predictions that future global warming will negatively affect polar bear populations.

Experts contacted by The Heartland Institute note global temperatures have not risen in the past 10 years, and scientists with the United Nations’ Intergovernmental Panel on Climate Change predict temperatures will cool for at least the next 10 years. Moreover, polar bear populations have been increasing during recent decades.

The expert statements below can be quoted directly, or the experts can be contacted for additional information at the telephone numbers and email addresses provided below.

“The U.S. Fish and Wildlife Service has just taken its place alongside Miss Cleo and the Psychic Friends Network in terms of a complete divorce from scientific reality. FWS apparently believes it has the clairvoyance to forecast sharp declines in polar bear populations even though temperatures for most of the past 10,000 years have been warmer than today and polar bears have flourished. Moreover, global polar bear populations have been rising for decades, even as temperatures have recovered from the end of the Little Ice Age 100 years ago.

“The only plausible basis for ruling polar bears as threatened is blind faith in alarmist computer models that have been no more accurate than Chicken Little’s claim that the sky is falling. Compare the alarmist computer models to the real world. Global temperatures have not risen one bit during the past decade. Before that, for 30 of the preceding 50 years, global temperatures fell. And now even IPCC scientists are predicting global temperatures will cool for at least the next decade.

“Only by completely ignoring real-world scientific evidence and jumping head-first into the world of special-interest group propaganda can one justify listing polar bears as a threatened species.”

James M. Taylor
Senior Fellow for Environment Policy
The Heartland Institute
taylor@heartland.org
941/776-5690

“This decision represents a conflict between politics and science. Polar bear populations have been increasing in recent decades, so there is no current problem. The concern is based on forecasts. However, the government forecasts used to support the decision violate basic scientific principles, and thus provide no scientific support for the listing.

“There are no scientific forecasts that would suggest a reduction in polar bear populations. It would be improper, then, to designate polar bears as endangered. Application of proper forecasting methods suggests a small short-term rise in polar bear populations followed by a leveling off. We provide full disclosure to support these statements at publicpolicyforecasting.com and at theclimatebet.com. In the long term, science will prevail.”

Scott Armstrong
Professor
Wharton School of the University of Pennsylvania
armstrong@wharton.upenn.edu
215/898-5087

“Canadians, who manage two-thirds of all polar bear populations, just reviewed their listing status and decided not to up-list the bear to a more serious status. Activists are attempting to politically interfere and change that reasonable and informed decision so today’s U.S. listing would not look extreme, unwarranted, and political, which it is.

“The listing is lunacy because carbon dioxide emissions–the real target of activists–are surging worldwide, and unless all other countries cut their carbon emissions, atmospheric concentrations will continue to rise even if the entire West shuts down its emissions. If the United States were to go 100 percent CO2 emissions-free, just the projected growth in China’s and India’s emissions would replace U.S. ’savings’ in about a decade.

“The self-inflicted economic wound of making the use of carbon fuels more expensive in the United States than in China will merely transfer carbon emissions and jobs to that regime, which already has one of the worst environmental records in the world, and will deploy the profits toward the continued expansion of its own network of uniquely dirty, coal-fired power stations, to the detriment of the environment, without any benefit to the climate or polar bears whatsoever.”

Robert Ferguson
President
Science and Public Policy Institute
bferguson@sppinstitute.org
703/753-7846

Today we have electronic devices that can measure and transmit data either by wire or wireless communications. Those data can now be automatically recorded and stored by remote electronic devices and reviewed at the leisure of the observer. There is no longer any requirement that the data be accessible every day. Access would be required only in case of malfunction and having a pair of measuring devices in each housing would allow a visit to be scheduled.

So basically, being able to walk to the thermometer is no longer a requirement today and taking the measurement 100 feet into the troposphere might give a better idea of “climate” vs. local micro-climate.

“Any statistician will tell you 10 samples is a huge improvement over 1 sample. 100 samples is a modest improvement over 10 samples. 1000 samples has no real significance over 100 samples.
. . . we don’t need 1200 stations . . .”

I believe your point would be valid if we were talking about samples from the same population. However, the station in Lampass may be different from a different population than the one at Detroit Lakes. Yes, they are both on the same planet, but climate trends in one place may not match climate trends in another place. I have not seen a good statistical discussion of how many “samples” are needed to determine global climate direction. Of course, that raises the question whether it is legitimate to talk about THE global climate when climate changes might be regional.

If data is for some reason incorrect or contaminated in some manner, wouldn’t the best approach be to throw it out and not use it? Why not determine which rural stations are empirically sound and restrict the data from any site which is not sound on its own merits.

This really seems very ridiculous to me and just moves from one problem situation to another.

Rex, your URL’s don’t seem to work.

Jack Koenig, Editor
The Mysterious Climate Project
http://www.climateclinic.com