Saturday, November 22, 2008

Bias In Global Warming Data?




Late one evening a policeman came upon a man who was crawling on hands and knees under a lamp post. "Whatch doing?" asked the cop. "Looking for a quarter I lost." "Where'd you drop it?" "Oh, in the alley back there." "Why you look'in here?" "THE LIGHT IS BETTER!"

As I've stated in previous postings (Carbon Tax, Definitive Guide to Global Warming, Nuclear Power in France, Correlation and Causation) I know Global Warming is a real phenomena, a substantial fraction of which is due to human release of sequestered carbon (oil, coal, natural gas). However, I believe the problem has been hyped by those who have their own agendas and are therefore "looking where the light is" to find evidence to back up their theory that we are near some kind of "tipping point" to global disaster.


I never imagined that the basic data record for the surface temperature of the Earth over the past few decades might be biased. However, that seems to be the case, at least for some measurements.

The top photo from http://wattsupwiththat.com/ (with annotations by me) shows the official NOAA Climate-Monitoring site in Clarinda, Iowa. Note that it is adjacent to a wastewater treatment plant, where warm water vapor could distort the readings. It is also quite close to what appears to be a methane gas burnoff torch associated with the wastewater plant.


The second photo is of the official NOAA Climate-Monitoring site in Fairbury, Nebraska. Note it is adjacent to a barbeque. It is also close to a house and a roadway where car exhaust could disort temperature reading.

Furthermore, a major data-reporting error occured early this month. According to http://wattsupwiththat.com/ as well as http://www.realclimate.org/ the official NASA GISS (Goddard Institute of Space Studies) worldwide surface temperature readings for October 2008 contained 90 sites (all or mostly in Russia) where September data was reported instead of October data, ressulting in errors of several degrees.

This data error affected a small percentage of worldwide sites. Due to this error, the average worldwide surface temperature for October was interpreted to be the warmest October on record. The worldwide surface data set goes back to 1880. GISS has corrected the error, see http://data.giss.nasa.gov/gistemp/.

There are a number of responsible and reasonable internet sites that report on climate change, some of which generally support the consensus that human activity is responsible for most global warming (particularly the http://www.realclimate.org/) and others that are skeptical (such as http://wattsupwiththat.com/). It is the skeptical sites that first noticed the error in the October data from Russia as well as the inappropriate locations of some official climate-monitoring stations.


It seems NASA simply took the reported October data without sufficient examination. How could they overlook such a drastic increase in temperatures from Russia when other sites in Europe and Asia reported little change? How could they miss 90 sites that had the exact, digit-for-digit data as in the previous month? Perhaps they were so pleased to have their theory of global warming confirmed that they failed to investigate further?


What if the faulty dataset had been of the previous January rather than September? I'll bet that if the dataset showed significant cooling rather than warming someone at NASA would have noticed the error!

It seems to me there should be standards for placement of official NOAA Climate Monitoring stations. For example, they should be at least 100 meters from a direct source of heat. Government employees regularly visit these stations for maintenance and they should notice if someone has placed a barbeque right underneath one! Of course, sometimes civilization may encroach on an existing site with a new house or a new factory or power plant or road. They should notice and move the site! According to http://wattsupwiththat.com/2008/11/19/how-not-to-measure-temperature-part-74/#more-4209 there are dozens of official sites that are located in highly questionable places, many put there within the past decades.

By the way, the site in the second photo was placed in 1998, obviously after that old house was already there. The site in the top photo was placed in 1985, most likely also after the wastewater treatment plant, with its methane burnoff torch, was already in place!

Ira Glickstein


11 comments:

Howard Pattee said...

Ira,
A few selected anecdotal cases are not meaningful. For example, start with
Wikipedia on global warming and read some of the links.

Howard

Ira Glickstein said...

I agree, Howard, that anecdotal evidence is not proof that the Earth surface temperature readings that indicate significant Global Warming are faulty.

However, if the main source of that data, NASA/NOAA GISS, has published faulty data without noticing it, and if a bunch of unpaid volunteers have discovered the recent erroneous data and have also taken photos of some 74 official NOAA Climate-Monitoring sites, many installed in the past couple of decades, that appear to be located such that they accentuate the close-by heat from human barbeques and wastewater treatment plants, it is reasonable to question the reliability of the entire enterprise.

I fully accept that we are in a Global Warming cycle that started some fifty to sixty years ago and is likely to continue for some time into the future. I also accept that human activities, particularly the large-scale burning of sequestered carbon (coal, oil, natural gas) have contributed significantly to the CO2 in the atmosphere and thus greenhouse heating of the surface.

I looked carefully at the Wikipedia link you provided where they claim "that most of the increase since the mid-twentieth century is 'very likely' due to the increase in anthropogenic greenhouse gas concentrations."[Emphasis added]

My only issue with the above statement is the word "most". I do not believe that it has been proved that anthropogenic (human-caused) greenhouse gas has generated over 50% of the current Global Warming.

That web page has a link to Wikipedia Solar Variation. If you look carefully at that web page, and go down seven graphs, you will see that we are at or near maximums for Precession, Obliquity, and Eccentricity.

That Wikipedia webpage does not make much of these longer-term cycles but some Global Warming skeptics point to them as the only possible explanation for the undoubted Global Warming periods that occurred well before the advent of humans on Earth.

The well-known ice core data prove that the Earth has experienced Global Warming periods (followed, not preceeded by high atmospheric CO2), far more severe that we have experienced so far, about every 100,000 years or so over the past 400,000 to 600,000 years covered by available ice core data. Humans have been around only about 100,000 years and the Industrial Revolution only a few hundred years ago.

Precession, Obliquity, and Eccentricity do not directly increase the amount of Solar energy that falls on the Earth, but mainly where it falls. Global Cooling occurs when the Earth's axis is more perpendicular to the Sun and the opposite occurs when more Solar energy falls on one of the poles.

If you look at the second figure on the web page you linked to, you will see that current Global Warming is concentrated in the upper latitudes of the Northern Hemisphere (indicated by brown and orange) at the very time Global Cooling has occured in Equatorial regions and, especially, at the South Pole (indicated by light and dark blue).

While by no means proof, that indicates that the current Global Warming cycle is due more to the orientation of the Earth's axis than to human-caused CO2.

That said, I am still in favor of a punitive Carbon Tax as the most effective way to reduce human burning of sequestered coal, oil and natural gas. (Any takers?)

I am also strongly in favor of conservation which is why, for nearly five years, we have had only one car, a hybrid Prius, that gets actual 45 to 54 MPG. Our next car, in a few years, will be a plug-in Prius hybird that will use gasoline only if we have to take long trips. We also use our electric golf cart for local travel and I do 40-50 miles on my bike each week.

Ira

joel said...

Howard said to Ira,
A few selected anecdotal cases are not meaningful. For example, start with
Wikipedia on global warming and read some of the links.

Ira already said he was convinced of global warming and major human contribution, so I don't think he needs anymore convincing. I, on the other hand, am not sure whether or not the global temperature in rising or falling on the time scale of decades. Looking at the history of planetary temperature one thing clear is that it never stays stays the same. It's always going up or down, so it wouldn't be surprising if it were going up at the moment. The contribution of industrialization is quite another story. It is difficult to determine the load and most of all, how much is due to an anti-industrialization political bias.

Ira has presented evidence that measurements are sometimes careless and that inaccuracy is ignored if it favors the global warm if hypothesis. They are examples, not anecdotal evidence. Ira did not say that a friend of his knows a guy who purposely placed a thermometer near a methane burn-off vent. He showed us photographic evidence of bad experimental practice. That's data not anecdote. When one starts dismissing data we don't like by calling it anecdotal, we are not acting as scientists. Every bit of data requires an explanation and must be supported, otherwise cause and effect is a useless concept.

Here's wikipedia's official word on the subject:
The expression anecdotal evidence has two quite distinct meanings.
.......................
"(1) Evidence in the form of an anecdote or hearsay is called anecdotal if there is doubt about its veracity: the evidence itself is considered untrustworthy or untrue.

(2) Evidence, which may itself be true and verifiable, used to deduce a conclusion which does not follow from it, usually by generalizing from an insufficient amount of evidence. For example "my grandfather smoked like a chimney and died healthy in a car crash at the age of 99" does not disprove the proposition that "smoking markedly increases the probability of cancer and heart disease at a relatively early age". In this case, the evidence may itself be true, but does not warrant the conclusion.

In both cases the conclusion is unreliable; it might happen not to be untrue, but it doesn't follow from the "evidence"."
.........................
I think that the appropriate response to a charge that measurements of surface temperature may be unreliable, is to go back and examine every site and determine whether or not thermometer are properly placed. A willingness to accept shoddy data is a symptom of a researcher being too emotionally involved with the outcome of the research. It seems to me that the pictures that Ira presents cast doubt on all the data. It's the researchers' burden to demonstrate that those cases were atypical or all the data must be scrapped. With respect -Joel
P.S. I found this link interesting It emphasizes agility. http://smarteconomy.typepad.
com/smart_economy/2008/11/
global-cooling---a-growing-
business-case.html

joel said...

Let's try that again

http://smarteconomy.typepad.com/smart_economy/2008/11/global-cooling---a-growing-business-case.html

Ira Glickstein said...

Thanks Joel for your supportive comments regarding the proper reaction to even a few examples of what appear to be unreliable data collection and analysis methods.

As you point out, I do not claim any purposeful malpractice by NASA/NOAA GISS scientists. At worst, the examples call into question some minor carelessness that could bias their data by a fraction of a degree. My main point is that some global warming skeptics are providing a public service by checking published data for anomalies and are sincere in trying to help NASA/NOAA GISS improve their methods.

As for Howard's use of the word "anecdotal", he and I have used that word in a non-perjorative sense. For example, in my PhD dissertation (which Howard supervised) I present numerical evidence for my optimal span hypothesis in a dozen different domains and sub-domains.

Some of the data, when analyzed using the "Student-t test" were shown to have a high statistical confidence level of 90% to 99%, depending upon the quantity of data points and their distribution.

Other data was based on a smaller number of observations and, if it had less than 90% confidence, it was termed anecdotal in my dissertation.

In my Google Knol on Optimal Span I have a table near the end called: Statistical Comparison of Optimal Span Hypothesis with Empirical Data which presents numerical results. Under "Human Cognition" I report that One-Dimensional sensory stimuli span averages 6.5 and, according to the "Student-t" test, there is a 95% statistical confidence it is within the range 5.24 to 7.76. I was able to achieve 95% confidence because the data I used from a famous paper by George Miller included 20 samples.

In the same table I provide Two- and Three-Dimensional results of 12.5 and 17.2 but note that these are anecdotal.

Ira Glickstein

PS: I will have a look at your link to http://smarteconomy.typepad.com/smart_economy/2008/11/global-cooling---a-growing-business-case.html and will comment on that later.

Howard Pattee said...

My point is that in controversies of this type where the complexity is too great and the evidence insufficient to allow definite conclusions, the best bet should be on the “preponderance of evidence” of experts, not on relatively small subsets of the data or lay opinions found on the internet.

Ira has by his energy-saving behavior apparently placed his bets on such scientific consensus even though he has found some questionable evidence that should be corrected, but that was not significant enought to make him change his behavior.

Steve Ruberg said...

Well since I work for NOAA I should probably respond to this! However, I can't really help much in regard to the Climate Change discussion - my lab works on Great Lakes water chemistry, physics, and biology - but I'll give my 2 cents.

In regard to placing Climate Monitoring sites in such ridiculous locations, many times when things look this silly it turns out to be some kind of hoax. But supposedly smart people do incredibly questionable things - or sometimes technicians take the easiest path when setting up data collection sites. NOAA scientists have also been accused of being too conservative by those who want the world to believe that we should panic about climate warming. The global temperature forecast models that NOAA puts out are pretty conservative (note the GFDL line (NOAA) in the forecast from Wiki) compared to others. Sometimes you just can't please anybody.

Just to clarify these sites are only measuring temperature and rainfall - not CO2 (http://www.ncdc.noaa.gov/oa/climate/research/ushcn/ushcn.html). My understanding is that global temperature is estimated from an average of thousands of observations from around the world - not just these 74 - so if a few were located badly they'd likely not contribute significantly to the error. Anytime that I have seen discussions of CO2 - the measurements have come from Mauna Loa which is supposed to represent a global average - which it should since it's located in the Pacific Ocean far away from continents where human generated CO2 comes from.

When you look at the ice core data it appears that global CO2 levels are higher now than at any time in the last 800,000 years. We know that we are dumping large amounts of carbon into the atmosphere (electric generation, cars, etc) so humans apparently contributed to these high levels. When we are the only little blue dot in the near vicinity in this corner of the universe that can support life - we should be concerned.

So reducing our dependence on carbon-based energy might improve our planet (there are also indications that increased CO2 levels might be making the oceans more acidic). That's a good thing. And the bonus here - the strategic importance of reducing our dependence on foreign energy sources could be great.

Ira Glickstein said...

Thanks Steve for joining in. I knew you moved to NOAA after you left the government agency where you were my customer and I was hoping you would comment. Sorry you are not in the part of NOAA that supervises these temperature-measuring stations or they would be better done.

The site I linked to Watts Up seems serious and claims millions of hits and shows hundreds of mostly sane comments.

It links to an interesting slide show that gets into some detail on what may have artificially raised temperature readings over the past decade.

For example, sensors are sheltered in wood boxes that may be bare wood or whitewashed or painted with latex. Their experiments show that sunny day temperature readings are up to ten degrees F higher in latex-painted boxes than in the open air, and seven degrees higher than whitewash. So, when a government agency decided to upgrade from whitewash to latex that affected experimental conditions and inadvertently raised sunny day temperatures by several degrees C.

Photos of the insides of these boxes show various electronics and radios right there with the temperature sensor. If, in the past, temperature readings were reported via phone lines, the switch to radios obviously increased the heat in the box and therefore the average readings.

They show photos of stations that are well-sited and contrast that with ones that are obviously picking up heat from nearby buildings and traffic and so on. Some sites are on gravel and other unnatural surfaces or near waste-treatment plants, air-conditioner exhausts, etc. I do not believe anyone has claimed these are trick photos.

It is important to understand that Global Warming science deals in fractions of a degree per decade. If new measuring stations have been added in the past couple decades and if conditions have changed for many of the existing stations, and if these additions and changes ave boosted temperatures by even a few degrees per station, the overall effect may be as large as the claimed Global Warming!

You say "sometimes technicians take the easiest path when setting up data collection sites." Right. Perhaps some sites were OK thirty years ago before an air-conditioner was installed in the adjacent building or a new road or parking lot was added. I'm sure the guy who decided latex paint would last longer than whitewash and save money had no idea he was changing the experimental conditions.

Any chance you could use your position as a NOAA employee to contact some fellow NOAA scientists and get their personal reaction to some of the more responsible skeptic sites? I know there are plenty of irresponsible websites out there making ridiculous claims on both sides and I would like to get some assurance that the few I follow are responsible.

Ira Glickstein

Steve Ruberg said...

Ira said, "Any chance you could use your position as a NOAA employee to contact some fellow NOAA scientists and get their personal reaction to some of the more responsible skeptic sites?"

Yes I'll see if I can contact someone about this.

Ira Glickstein said...

Thanks Steve for agreeing to try to contact some NOAA experts regarding responsible websites on both sides of the Anthropogenetic Global Warming (AGW) issue. ("Anthropogenetic" means human-caused).

Right now I am monitoring a few websites on an almost daily basis. I believe they are all very responsible and span both sides of the AGW issue.

The most responsible site I have found that tends to support AGW arguments is:

Real Climate, "Climate science from climate scientists." I quoted from Real Climate in most of my Topics about Global Warming because when I agree with them I believe you can "take it to the bank" (and I find I DO agree quite often.)

Global Warming (Accuweather) appears to be a bit more hyped up about the problem, but still reasonable. They are "a company with a tradition of innovation and accuracy ... more than just weather information—we give you insights useful to you in planning your life ... more than 100 forecast meteorologists and proprietary forecast models ... "

The following websites are a bit skeptical of AGW as the primary cause of Global Warming while, IMHO, being reasonable and scientific:

Watts Up run by "a former television meteorologist who spent 25 years on the air and who also operates a weather technology and content business..." This site contains frequent scientific commentary related to "puzzling things in life, nature, science, weather, climate change, technology, and recent news."

Surface Stations project by the same guy who does Watts Up (Anthony Watts :^) This site contains scientific surveys of climate-monitoring sites.

Climate Audit run by a guy who has "been in business nearly all my working life, most recently in financing and promoting mineral exploration projects. ... In terms of occupation, right now, this is what I'm doing. No one's paying me to do this and there is a substantial opportunity cost for me personally in doing this, but I enjoy it and can afford to do it for a while."

Ira Glickstein

Ira Glickstein said...

Here's a unique way to reduce Global Warming worries - Shut down any climate monitoring station closer than 10 meters from an anthropogenetic source or move it to 100 meters distance.

Here's one in Napa, CA only three meters from an air conditioner exhaust. It seems a few more questionably located sensors are discovered and posted each week.

The most charitable explanation is that these temperature monitors were placed decades ago and buildings grew up around them. A less charitable explanation is that some AGW proponents have purposely ignored the 100 meter standard when placing new stations in order to inflate the measured warming.

Ira Glickstein