Why Oh Why Can't We Have a Better Press Corps?

Tim Lambert of the weblog Deltoid writes:

Deltoid: Global warming totally disproved again: Steve McIntyre found an error in the GISS temperature data for the US. The GISTEMP page says:

USHCN station records up to 1999 were replaced by a version of USHCN data with further corrections after an adjustment computed by comparing the common 1990-1999 period of the two data sets. (We wish to thank Stephen McIntyre for bringing to our attention that such an adjustment is necessary to prevent creating an artificial jump in year 2000.)

How much difference did the adjustment make to the US temperature series? Well, it changed this:

to this:

Not much difference. The right hand end of the red curve has moved down a little bit, but this decade is still the warmest ever recorded in the US. The change to the global temperature series is imperceptible...

But Tobin Harshaw of the New York Times Opinionator writes:

Hottest Year Data Meltdown: by TOBIN HARSHAW: You just thought you were sweating? Among global warming Cassandras, the fact that 1998 was the “hottest year on record” has always been an article of faith. Stephen McIntyre, who runs the Climateaudit blog was always puzzled by some gaps he saw in the raw data provided by NASA that supported the claim (data compiled in part by James Hansen, the climate scientist who has long accused the Bush administration of trying to “silence” him). McIntyre says he has “reverse engineered” the data to find NASA’s algorithm, discovered that a Y2K bug played havoc with some of the numbers, and notified the space agency.

Michael Asher at DailyTech explains the fallout:

NASA has now silently released corrected figures, and the changes are truly astounding. The warmest year on record is now 1934. 1998 (long trumpeted by the media as recordbreaking) moves to second place. 1921 takes third. In fact, 5 of the 10 warmest years on record now all occur before World War II. Anthony Watts has put the new data in chart form, along with a more detailed summary of the events.

The effect of the correction on global temperatures is minor (some 1-2% less warming than originally thought), but the effect on the U.S. global warming propaganda machine could be huge.

That's all Tobin Harshaw wrote.

Really sad. I don't think Tobin Harshaw looked at the data series. I don't think he could have looked at the data series and still written those paragraphs.

I don't understand why Tobin Harshaw has his current job.

---

UPDATE: Tobin Harshaw whines:

Note: Many commenters on this post have assumed that the author intended the term “Cassandras” to be pejorative, and also that he was unaware that the predictions made by the prophetess Cassandra, daughter of King Priam of Troy, came true. Rather, the term was being used in its common modern sense: one whose dire predictions, true or false, go unheeded.

And a few comments on Harshaw:

I just talked to someone at GISS (the group Jim Hansen heads) who said they fixed it in less than a day when they did learn about it from McIntyre (who didn’t actually “reverse engineer” anything), and when they fixed it nothing much changed. They saw a 0.15 deg C change for 2000-2006 in the US numbers (since the problem was melding the US data with the global numbers) and some very minor adjustments down the line. The global mean change was zero. 2005 was still the warmest year in the corrected GISS analysis, as it is with the NCDC analysis. Amazing what a little reporting will turn up. — Posted by Ken

This post is almost entirely incorrect. NASA’s much-ballyhooed data showing that 1998 was the warmest year on record for the Earth was right. The error was in the temperature record for the contiguous 48 states (I commend Steve McIntyre’s work), not for the global temperature record. The US temperature record has, as one would expect, a much greater interannual variability than the global record. The DailyTech article fails to point this out. I hope you will make this clear. Actually, I made a mistake. NASA’s data shows that 1998 is the second-warmest year on record for the Earth. The warmest year? 2005. The top twenty hottest years on earth since 1880: * 2005 * 1998 * 2002 * 2003 * 2006 * 2004 * 2001 * 1997 * 1995 * 1990 * 1991 * 2000 * 1999 * 1988 * 1996 * 1987 * 1983 * 1981 * 1994 * 1944 — Posted by The Cunctator

The term “hottest year on record” isn’t appropriate unless you’re talking about global climate. But you’re only talking about the lower 48 states. Why would you use such misleading language if you’re trying to educate people? By the same token, why use terms like “U.S. global warming propaganda machine” and use scare quote around the word “silence.” Hansen claimed that NASA tried to silence him, not “silence” him. If you want to discuss facts, there’s plenty to discuss. If you want to use inflammatory and misleading language that is certainly your prerogative, but aren’t there more appropriate places to do that than the New York Times? NB: Cassandra was right, remember? Her curse was to accurately predict the future but not to be believed. So calling someone a Cassandra is to imply that they are correct but unpopular. — Posted by Debra

Comments

[anne]

Tobin Harshaw is a deceiving fool; deceiving because using the name Cassandra is meant to discredit the person referred to, as propagandist is used repeatedly and meant to discredit Michael Moore; a fool because a simple source check would have clarified the insignificant analytical revision. Fortunately Harshaw never ever has written or assisted on an article on global warming for the New York Times.

[anne]

http://www.nytimes.com/2007/08/09/science/10cnd-arctic.html?hp

August 9, 2007

Floating Arctic Ice Shrinking at Record Rate
By ANDREW C. REVKIN

The area of floating ice in the Arctic has shrunk more than in any summer since satellite tracking began in 1979, and it has reached that record point a month before the annual ice pullback typically peaks, experts said.

The cause is probably a mix of natural fluctuations, like unusually sunny conditions in June and July, and long-term warming from heat-trapping greenhouse gases and sooty particles accumulating in the air, according to several scientists.

William L. Chapman, who monitors the region at the University of Illinois Urbana-Champaign and posted a Web report on the ice retreat today, said that only an abrupt change in conditions could prevent far more melting before the 24-hour sun of the boreal summer sets in September.

"The melting rate during June and July this year was simply incredible," Mr. Chapman said. "And then you've got this exposed black ocean soaking up sunlight and you wonder what, if anything, could cause it to reverse course."

Mark Serreze, a sea-ice expert at the National Snow and Ice Data Center in Boulder, Colo., said his center's estimates differed somewhat from those of the Illinois team, and by the ice center's reckoning the retreat had not yet surpassed the satellite-era record set in 2005.

But it was close even by their calculations he said, adding that it is almost certain that by September, there will be more open water in the Arctic than has been seen for a very long time. Ice experts at NASA and the University of Washington echoed his assessment....

[anne]

"Stephen McIntyre, who runs the Climateaudit blog was always puzzled by some gaps he saw in the raw data provided by NASA that supported the claim (data compiled in part by James Hansen, the climate scientist who has long accused the Bush administration of trying to 'silence' him)."

Further deception by the deceiving Tobin Harshaw. The New York Times has indeed documented just how the Bush admninistration sought the "silence" the courageous James Hansen. The accusations were shown by the New York Times to be entirely true.

[anne]

The attempt to silence James Hansen was simply part of a continual series of attempts by the Bush administration to silence government scientists, a series of attempts continually shown by the New York Times.

Remember this is the administration so crazed as to order government biologists, not once, but twice, not to speak about polar bears, an order that still unnerves me since I am rather fond of polar bears and wondering what they might have done to become so subversive. What is it about polar bears anyway? And should I even be mentioning them?

[ThresherK]

Who will explain to their great-grandkids, watching X Games LXXV, that there wasn't always a "Jetski to the North Pole" event?

[bigTom]

anne: good catch.

You can look at the snow/ice cover daily at:http://www.rap.ucar.edu/weather/surface/snowNESDISnh.gif

The article has pretty thoughougly discredited this politically motivated attack on the science. But you can bet the Rush Limbaughs will never get the message.

[andres]

One thing I suggest that journalists start doing is actually logging and comparing the amount of time that they spend working with Excel or with statistical software in order to review and crunch the data, in comparison to the amount of time they spend on blogs reading other people's stories as well as in their word processors typing out stories.

My guess is that there would be a significant correlation between the number crunching/total time proportion and the overall veracity of the journalist's writing. Hence the suspicion that Harshaw didn't actually spend any time adjusting and graphing the data he was writing about.

[Neil B.]

You can draw a straight line right across the 1934 peak on the revised graph, and see that it is still lower than the 200x wiggles.

Yeah, Rush was blabbering about fraudulent NASA scientists, and "Seventy years ago, it was way hotter than it is now" etc. BTW, why can't groups of people like NASA scientists sue for libel if someone defames them?

[Gwailo]

Harshaw's "Cassandra" whine shows that his ignorance of the classics is exceeded only by his ignorance of science.

[bigTom]

andres: At the risk of insulting writers: I suspect most of them have such poor math/statistics understanding that trying to do it themselves would have disastrous results. I think the best suggestion is to have an impartial technically smart college review the article prepublication. That should also apply to the headine writers, who I think are often distinct from the article writers.

[anne]

Thank you, Tom, and here is William Chapman's site:

http://arctic.atmos.uiuc.edu/cryosphere/ .

Notice, however, that all else here are still too cowed to mention polar bears which is no comfort to me at all.

[bigTom]

anne: that is a pretty nice site that I hadn't known about. I thought the ice looked pretty minimal, it is pretty significant that it as now at a virtual tie for the record so early in the season. The acceleration of the melt continues to exceed predictions.....

[Kaleberg]

One useful thing to remember is that if there is no trend, then the probability of setting a record decreases logarithmically with the number of data points. With a 100+ year data series, there should be long gaps between records, and records should be increasingly rare. (This is a property of that falls out of the harmonic series). The fact that we have been setting many new high temperature records in recent years indicates that warming is a trend, not an artifact.

[anne]

Karlsberg, I was thinking similarly but reading to be more sure. Also, Tom reminds me of the persisting of ocean blackness for longer periods and the resulting heat absorption. Loss of ice and snow cover can accentuate a warming pattern, as far as I understand. Chapman wonders where a reversal might come from as black ocean creates climate.

[bigTom]

There are two sorts of trends which could lead to more recent records than variation about an equilibrium. The most obvious is that the average is changing, and thus an event only a few sigma in the direction of the trend could set a record. The second is if the variance were increasing. The later would be expected to be two-sided, with new lows, and highs being made with more than the expected frequency. Of course you could blend the two trends.....

In any case, the arctic changes keep accelerating. Predictions that the summer ice would vanish by roughly 2050, which seemed pretty radical a few years ago, now seem to grossly overstate the time for that to happen.

Well more open ocean, does make more moisture available for snow, but it seems that dates of spring snow/ice melt are advancing despite that.

[Craig Nelson]

If 1934 was the warmest year in the US, we must hope that modern agricultural practices are robust enough to prevent another dust bowl since US average temperatures are predicted to be higher and (I think) seasonal variability and average wind velocity are also predicted to increase.

From the Wikipedia: "Then on May 11, 1934, a strong two-day dust storm removed massive amounts of Great Plains topsoil in one of the worst such storms of the Dust Bowl. The dust clouds blew all the way to Chicago where filth fell like snow, dumping the equivalent of four pounds of debris per person on the city."

[vtcodger]

***If 1934 was the warmest year in the US, we must hope that modern agricultural practices are robust enough to prevent another dust bowl since US average temperatures are predicted to be higher and (I think) seasonal variability and average wind velocity are also predicted to increase.***

They may be. Certainly, they are better. Yesterday NPR's All Things Considered had a long article on farming practices in Kansas. Mostly it was about water mining of the Ogallala aquifer, but it also talked about non-irrigated farming. Apparently the modern practice is to leave roots and stalks of harvested crops in the ground to reduce wind erosion and improve rainfall absorbtion.

On the other hand, the Lower Mississippi is lined on both sides with thick loess deposits composed of wind borne dust from paleo droughts -- presumably in the grasslands of the great plains. The dust bowl was far from the first time the plains have had an extended drought. And it probably won't be the last.

[vtcodger]

***Well more open ocean, does make more moisture available for snow, but it seems that dates of spring snow/ice melt are advancing despite that.***

I don't think amount of snow counts as much as the amount of time spent with ambient temperatures near or above 0C and sunlight. Snow and ice melt pretty quickly if things warm up at least here in the mid-latitudes. Moreover, I would guess that the Arctic will ice over fairly quickly and cut off evaporation as a source of moisture when the sun goes away in 5 or 6 weeks. (Anyone know how much air exchange there is from the Arctic to the warmer mid latitudes? I vaguely think not a lot).

The reduced Arctic and mountain ice is probably a good indication that global warming is real and a matter for concern.

Much better than that temperature data. Am I the only one that sees the data as quite noisy with a probable 10-11 year cycle? It's clear that the 1930s and post 1990 are warm. Yes, there is probably an upward trend. But it doesn't look anything like the CO2 injection trend See http://en.wikipedia.org/wiki/Image:Co2-temperature-plot.svg

My guess it that greenhouse gase induced global warming is barely showing up in the temperature data (so far) and that you'd have to be a Hogwart's trained magician to isolate the affect much less quantify it. So what's the point in arguing about it?

BTW, what is that green dot in the lower right corner of the chart? It looks like a data point (1997 or 1998?), and easily the coldest year on the chart. But the various discussions indicate that 1995, 1996, 1997, 1998 and 1999 are all amongst the warmest years. My guess is that some people are using single year numbers whereas others are using five year averages.

I'm not wild about Tobin Harshaw's "journalism", but maybe it's not all that egregious if you are talking -- as he appears to be -- single year values. IMO, anyone -- pro or anti global warming -- ought to have the good sense to look at those charts, resolve to come back to them in a couple of decades and move on to data that is less ambiguous.

[stewart]

The green dot is 1993 (or 92, I can't recall which) after Mount Pinatubo injected large amounts of ash and sulfate aerosols into the atmosphere, leading to a brief cooling trend, until it came out of suspension. Mean temperature trends track much better to CO2 plus sulfates than to anything else (rising sulfates from dirty coal use in the 50's to'75 led to that temporary decline in temperatures). Claims that solar activity, sunspots, the number of pirates, and other third variables are linked to the current temperature changes have been disposed of by now. The argument that 'climate always changes' or 'we're coming out of the little ice age still' rejects all the evidence for relative climate stability over the last 4,000 years. It is interesting to contemplate that 6,000 years ago, the Sahara looked rather like the Great Plains today, and that the end of the Old Kingdom in Egypt was probably linked to climatic events.

[Charles]

Doesn't anyone use linear least squares anymore?

VT Codger and Augustus. The little green point in the corner is neither 1998 nor 1992 nor even 1993.

It is part of the legend of the graph, associated with the phrase "annual mean."

[anne]

http://www.nytimes.com/2007/05/01/us/01climate.html?ex=1335672000&en=96d5885b0b356799&ei=5090&partner=rssuserland&emc=rss

May 1, 2007

Arctic Sea Ice Melting Faster, a Study Finds
By ANDREW C. REVKIN

Climate scientists may have significantly underestimated the power of global warming from human-generated heat-trapping gases to shrink the cap of sea ice floating on the Arctic Ocean, according to a new study of polar trends.

The study, published online today in Geophysical Research Letters, concluded that an open-water Arctic in summers could be more likely in this century than had been estimated in the latest international review of climate research released in February by the United Nations Intergovernmental Panel on Climate Change.

"There are huge changes going on," said Julienne Stroeve, a lead author of the new study and a researcher at the National Snow and Ice Data Center in Boulder, Colo. "Just with warm waters entering the Arctic, combined with warming air temperatures, this is wreaking havoc on the sea ice, really." ...

[Andrew Revkin has been tracking research on climate change in the arctic and antarctic for years. The research finding are continually shaping a story for us.]

[anne]

http://www.nytimes.com/2005/11/25/science/earth/25core.html?ex=1290574800&en=d5078e33050b2b0c&ei=5090&partner=rssuserland&emc=rss

November 25, 2005

Rise in Gases Unmatched by a History in Ancient Ice
By ANDREW C. REVKIN

Shafts of ancient ice pulled from Antarctica's frozen depths show that for at least 650,000 years three important heat-trapping greenhouse gases never reached recent atmospheric levels caused by human activities, scientists are reporting today.

The measured gases were carbon dioxide, methane and nitrous oxide. Concentrations have risen over the last several centuries at a pace far beyond that seen before humans began intensively clearing forests and burning coal, oil and other fossil fuels.

The sampling and analysis were done by the European Program for Ice Coring in Antarctica, and the results are being published today in the journal Science.

The evidence was found in air bubbles trapped in successively older ice samples extracted from a nearly two-mile-deep hole drilled in a remote spot in East Antarctica called Dome C.

Experts familiar with the findings who were not involved with the research said the samples provided a vital long-term view of variations in the atmosphere and Antarctic climate. They say the data will help test and improve computer models used to forecast how accumulating greenhouse emissions will affect the climate....

[anne]

Remember, this is all hush hush....

http://www.nytimes.com/2007/03/08/washington/08polar.html?ex=1331010000&en=861771605d9c648b&ei=5090&partner=rssuserland&emc=rss

March 8, 2007

Memos Tell Officials How to Discuss Climate
By ANDREW C. REVKIN

Over the past week, biologists and wildlife officials received a cover note and two sample memorandums to be used as a guide in preparing travel requests. Under the heading

"Foreign Travel — New Requirement — Please Review and Comply, Importance: High,"

the cover note said:

"Please be advised that all foreign travel requests (SF 1175 requests) and any future travel requests involving or potentially involving climate change, sea ice and/or polar bears will also require a memorandum from the regional director to the director indicating who'll be the official spokesman on the trip and the one responding to questions on these issues, particularly polar bears."

The sample memorandums, described as to be used in writing travel requests, indicate that the employee seeking permission to travel "understands the administration's position on climate change, polar bears, and sea ice and will not be speaking on or responding to these issues." ...

[anne]

Particularly, yes, we have no polar bears, and if we had polar bears I would be the last to tell and I haven't told yet and won't tell, so don't accuse me of telling about polar bears particularly because I haven't.

[anne]

So that we understand that attempt to cover up the firece polar bear gay sex controversy (can she say that?), was a typical sadness in an Interior Department that has just witnessed the resignation of Deputy Assistant Secretary Julie A. MacDonald, not for the polar bear gay sex cover up but for what the inspector general's agency found was "browbeating" government biologists who thought they knew more about species protection than the appropriate industry lobbyists to whom Ms. MacDonald was giving reseach findings. Huh?

[anne]

http://www.nytimes.com/2007/07/21/washington/21interior.html?ref=washington

July 21, 2007

U.S. Agency May Reverse 8 Decisions on Wildlife
By JOHN M. BRODER

WASHINGTON — The Interior Department said Friday that it would review and probably overturn eight decisions on wildlife and land-use issues made by a senior political appointee who has been found to have improperly favored industry and landowners over agency scientists....

[No polar bears here, particularly no sexually active polar bears however oriented.]

[andres]

bigTom: you may be right, but innocent naif that I am, I'm still shocked. Don't undergraduate and graduate-level journalism curriculums require at least one course in statistics and/or quantitative analysis? Or do they just go through the motion of teaching this part of journalism, so that they turn out innumerate journalists like Harshaw?

Charles: "Doesn't anyone use linear least squares anymore?" Only if they want to look shockingly old-fashioned to their economist/statistician colleagues. But in addition to fashion, I have found that real world data includes any number of problems that often make the use of least squares problematic, including (1) outliers which often force you to either use observation-specific dummy variables that have no theoretical value or to use a non-least squares loss function in order to reduce their influence on the regression, and (2) serious autocorrelation problems in time series data which are better corrected with ARIMA models than by modifying least squares.

[andres]

anne, your obsession with polar bear sex (either orientation, and you forgot to mention the issue of positions) is starting to worry me. Suggestion: if you have overweight relatives, think of them having sex and you'll stop worrying about the hairy quadrupeds in no time.

Of course, birds are a refreshing alternative, although I've yet to figure out if the damn things can have sex while staying airborne :-o

[anne]

Actually, though there are many species I know of no bird that can have intercourse in the air. There are migratory birds that spend astonishing amounts of time in the air, and there are displays in the air, but I would guess not. The genetic advantage would seem minimal. Possibly the more interesting question is why birds do not get pregnant; which they do not get.

[andres]

What?! Not even hummingbirds? I am dismayed :-(

Then again, what do I know? If polar bears were anything like humans, their apparatus would shrivel up in the arctic cold and they would never reproduce. Which goes to show that I should stick with political economy and leave biology and reproduction alone.

[Guy in Jersey]

"At the risk of insulting writers: I suspect most of them have such poor math/statistics understanding ... the best suggestion is to have an impartial technically smart college review the article prepublication"

Big Tom: Assuming that your quant skills are strong, you are showing us that you have the reverse problem of the writers? Or did you actually mean that a "technically smart college" should "review the article"?