Simple multi-layer greenhouse

Here are the results of my simple multi-layer greenhouse experiment, conducted in December when the weather was hot and stable, not mild and rainy as it is now. The experimental setup is shown below, with two laboratory thermometers, and a mercury one to check. One sensor was attached to a 6in black tile sitting on the EPS box, the other on the glass surface. On top were up to 5 alternating layers of EPS and picture glass, as shown below.


The temperatures are in C, and were measured by recording the maximum temperature over the period. As far as possible, I tried to obtain measurements on a clear calm day. The location is on the tropic of Capricorn in December, so the sun was virtually overhead.

Time	Layers	Tile	Glass

12:00	5	116	50
12:30	5	110	49
12:45	3	110	49
12:50	2	106.9	52.2
1:00	1	110.1	60.1
1:10	4	104.6	46.6
1:25	5	102.9	46.5

Below is a graph of the data above. Its fairly clear that the number of layers has very little effect on the temperature of the black tile. The temperature of the external glass layer does decrease however, with more layers. This I would think is due to the increased heat losses from the sides of the stack of alternating glass and EPS blocks.


Once again, conducting this with precision outside is not possible without better equipment. As I reported with an earlier post, the temperature of the tile was beginning to melt the EPS foam.

The same result, of little change in the temperature of the tile with additional layers, was reported in a post by JQ Public:

My son and I repeated the experiment as mentioned and we the same results. We then used two glass jars, one as a control and one with water vapor and got the same results. We tried the two jar experiment again, but his time we stayed indoors and used a heat lamp and got the same results. In our fourth experiment we use one jar as a control and added vinegar and baking soda to the second jar to produce CO2. After and hour into the experiment we added even more vinegar and backing soda to create even more CO2 and yet again the temperature did not increase. The mean control jar temperature was 34.87 while the experimental jar was 35.43. The mean humidity for the control was <20% (we could not measure below 20%) and the mean humidity of the experimental jar was 42.73%.

Theon eviscerates climate warming community

This is not the sort of news I usually pick up on, but I quote below the retired senior NASA atmospheric scientist Dr. John S. Theon, and former supervisor of James Hansen, both because of the relevance to modelling practise, and because he captures so exactly what has driven me out of science over the last 5 years, and onto the blogosphere.

Theon declared “climate models are useless.” “My own belief concerning anthropogenic climate change is that the models do not realistically simulate the climate system because there are many very important sub-grid scale processes that the models either replicate poorly or completely omit,” Theon explained. “Furthermore, some scientists have manipulated the observed data to justify their model results. In doing so, they neither explain what they have modified in the observations, nor explain how they did it. They have resisted making their work transparent so that it can be replicated independently by other scientists. This is clearly contrary to how science should be done. Thus there is no rational justification for using climate model forecasts to determine public policy,” he added.

If climate models are useless, what does this make the order of magnitude larger community of scientists in the climate change effects community who use these models to prophesy impending doom: dupes, enabelers, or marketers of the latest trendy idea? Sorry guys.

New tests and enhancements to WikiChecks

I have tweaked the interface of WikiChecks and added some new analysis. It will take a range of analysis before I get a good enough sample, but already there is an amazing degree of insight coming out of this technique. Below is a list of some of the new additions, and whether the last digit deviates from randomness.

PDO Monthly values, significant, excess 6’s, possibly massaged.
Fidelity Mutual Fund daily adjusted price, not significant
UBS AG hedge fund monthly returns, not significant
MSCI Barra Hedge Funds, significant, excess 0’s and 5’s, possibly rounding.
CAL Global Hedge Fund, not significant

Somebody added the PDO Monthly values, which turned up a significant excess of sixes. In other tests I have done, and in the literature, excess of 6 is a sign that human preferences are influencing the numbers.

Whoever recently added PDO Monthly values had a problem, and I have changed the interface to overcome it. Its set up so that the data sets people enter are recorded as posts, so everyone benefits from the results. So if you have a data set you want to try, be it science or finance or something else, please feel free to give it a go. I will be watching to make sure tests goes through, and improve the system with use.

Simple Greenhouse Proofs

A reported increase in the longwave downward radiation in the Swiss Alps, proves the ‘‘theory’’ of greenhouse warming with direct radiation observations according to this paper, “Radiative forcing – measured at Earth’s surface – corroborate the increasing greenhouse effect”, by Rolf Philipona, Bruno Durr, Christoph Marty, Atsumu Ohmura and Martin Wild.

Supposed direct observational proofs of the enhanced greenhouse effect have been reviewed here in the past.

  1. Rahmstorf, who claimed climate responding faster than expected on the basis of a dubious graph with no statistical test;
  2. Harries who claimed to detect the greenhouse effect from CO2 spectral brightening but whose later (underreported) publications were much more equivocal;
  3. Soden, whose claims to have detected increase in specific water vapor from
    spectral brightening were reported as proof in the IPCC AR4, despite conflicting evidence.

Any comments on this radiative proof from the radiation experts here?

Hansen's Regression to Zero

While reading Hansen’s latest mailout I came upon an intriguing reference that I followed up. I suspect this paper is as important as Douglass et al. in describing an important way the models do not agree with the observations. It may be more important, in redefining the role of the Sun in recent warming.

His mailing contains a massive revision of his estimate of the rate of warming down from 0.2C per decade to 0.15C per decade. Near the end of his mailing he notes:

Solar irradiance has a non-negligible effect on global temperature [see, e.g., Reference 7, which empirically estimates a somewhat larger solar cycle effect than that estimated by others who have teased a solar effect out of data with different methods].

Reference 7, a paper by Tung, K.K., J. Zhou, C.D. Camp, Constraining model transient climate response using independent observations of solar-cycle forcing and response, recently published in Geophys. Res. Lett. in 2008 is freely available. It uses the geographical distribution of global temperature in response to the 11 yr solar cycle to isolate the transient climate response (TCR). Noting previously that the 19 IPCC models have a very large range in TCR, it compares the empirically derived TCR with the TCR of all the major IPCC climate models. The result is damning:

[14] The TCRs of 19 coupled atmosphere-ocean GCMs in IPCC AR4 listed in Table 1 fall within the rather low range of 1.2–2.2 K with the exception of one, and thus fail the lower constraint of 2.5 K determined by ERA-40, GISS and HadCRUT3. The only exception is the Japanese MIROC (hi-res), with a TCR of 2.6 K. All models fail the higher constraint of 3.6 K determined by the NCEP data.

The emphasis was added by the authors. As I understand it, they are saying that GCM’s have grossly underestimated solar forcing. Like the tropospheric ‘hotspot’ due to GHG’s in Douglass et al., they are not even in the right ballpark. The problem, it seems, is with the rate of transfer of heat into the ocean.

[16] It is seen that most of the current generation of general circulation models assessed by IPCC AR4 have too low a transient climate response as compared with the observed range. This is consistent with the independent finding by Forest et al. [2006] that these models simulate too large an ocean heat uptake as compared to observations of ocean temperature changes during the period 1961–2003. See Raper et al. [2002] and Meehl et al. [2004] for different views on how ocean heat uptake affects TCR. This excessive heat into the oceans tends to reduce the transient climate response for the atmosphere, but does not affect the modeled equilibrium climate sensitivity, which was calculated with a slab ocean in thermal equilibrium with the atmosphere.

The last sentence notwithstanding, there is an argument that underestimating TCR must lead to overestimating GHG forcing in the recent past. This would be a confirmation of the AGW skeptical view that recent response to solar forcing has been grossly underestimated, and GHG forcing exaggerated.

For the paper to be acknowledged by James Hansen himself is intriguing. Perhaps backpedaling is his way of avoiding jousting with jesters (climate skeptics).

Global Temperature Graphs

The next step in the statistical forensics process is to breakdown the data in ways that reveal where the anomolous divergences are coming from. Here I am indulging in classical scientific reduction methodology by examining overall phenomena in terms of the sum of its parts.

The previous post in the series identified significant divergence in the distribution of the last digits of two global temperatrue data sets, from GISS (Pr<0.05) and CRU (Pr<0.01). Two other data sets based on satellite data were cleared of non-randomness, from RSS and UAH.

LuboÅ¡ Motl confirmed my results on GISS and CRU. Steve McIntyre initally disagreed, but then found an order of magnitude mistake in his calculations which he reported in a comment here. So there can be no doubt that the anomaly in distribution of digits in these datasets is real. This can be caused by many factors, only one of which is ‘manipulation’ of the data. How do we find the cause?

The graphs below show changes in chi-sq values (red) over the time scale of the GISS and CRU temperature series (blue) from 1880 to the present. I show them now to indicate where I am going. Sorry they are very basic but I am developing the code in php from scratch, so it can be used on the WikiChecks website. I used a 100 data point window, and plotted the significance of the dirvergence from a uniform over time (red).

The regions where the distribution of digits diverges is shown clearly, and will be the basis for more detailed examination.

GISS temperature and digit divergence.


CRU temperature and digit divergence.


UBS returns show significant management

Cross posted at WikiChecks.

I pasted in monthly data from the Swiss bank UBS and found significant management. The file used was this. The digit frequency shows an excess of zeros and ones and a deficiency of 7s and 8s. One possible explanation is that figures slightly below a whole number have been boosted to slightly above a whole number (eg. 3.9% to 4.1%).

For comparison, I looked at overall returns from a number of funds here. These fund returns showed no signs of management.

Interestingly, a google search on ‘UBS fraud’ gives plenty of hits. UBS was charged with fraud in July 2008 by the SEC. The linked article states of the charges:

Of course, the Commission’s complaint are only allegations, thus far unproven, and UBS has not yet responded to the compliant. However, if true, the allegations are serious, and provide significant insight into a corporate mindset at UBS which put its profits ahead of the well being of its customers, and its own employees.

Announce: New fraud detection website

Detecting ‘massaging’ of data by human hands is an area of statistical analysis I have been working on for some time, and devoted one chapter of my book, Niche Modeling, to its application to environmental data sets.

The WikiChecks web site now incorporates a script for doing a Benford’s analysis of digit frequency, sometimes used in numerical analysis of tax and other financial data.

I have posted some initial tests on the site: random numbers and the like. I also ran each of the major monthly global temperature indices through the site: GISS, RSS, UAH and CRU. The results, listed from lowest deviation to highest are listed below.

Continue reading Announce: New fraud detection website

Dr Roy Spencer's new blog

Dr Roy Spencer has a new blog. His latest post describes a study demonstrating another possible negative feedback produced by clouds. Of more interest to me, he exposes the bias in the academic publication system, due to no explicit mention of the possible relevance of this negative feedback to moderating warming in climate models. Simply, he thinks it would not have got published if it did.

Probably if a positive feedback was described, it would have been published in Science, not the Journal of Climate. Its great to hear insiders describing the details of the mechanism whereby the AGW view is propagated.

Another thing I like is that the blog is self-titled, Roy W. Spencer, Ph.D. I have put his blog on my Google reader subscription list.

Markopolos, Dan diBartolomeo, and Koppel vs. Madoff

Another way to predict — lie about your success rate.
Mathematical analysis in 1999 showed Madoff’s returns were impossible and repeated warnings went unheeded. Competitor Markopolos complained to the SEC’s Boston office in May 1999, saying it was impossible for the kind of profit Madoff was reporting to have been gained legally. Markopolos reached his conclusion with the help of mathematicians like Dan diBartolomeo, whose analysis of the Madoff’s methods in 1999 helped fuel Markopolos’ suspicions.

“As the market goes up and down, this strategy should have done a little better or a little worse, just like everybody else,” he said. “Instead, it appeared to be indifferent as to whether the market went up or down. They made money all the time.”

In 2005, he submitted a report to the SEC saying it was “highly likely” that “Madoff Securities is the world’s largest Ponzi scheme.” The report highlights 29 “red flags” about Madoff’s business, among them the returns of a third-party hedge fund managed by Madoff’s firm which had negative returns in just seven on the 174 months Markopolos analyzed. His warnings were heard too late, and he’s becoming a symbol of a botched oversight of fraudulent dealings by governmental authorities.

Rich Koppel, founder and managing director of technology provider said Fund-of-hedge fund managers (FOHFs) that lost big in the alleged Madoff $50 billion securities fraud could have avoided the debacle if they had deployed technology to gather information and monitor the trading strategies and results from hedge funds they were invested in.

“Lack of technology is a factor, or failure to use technology effectively is a factor,” said Rich Koppel, founder and managing director of technology provider youDevise Limited in an interview yesterday. Koppel maintains that using technology would have enabled FOHFs to avoid the huge losses that resulted from investments in Madoff funds.

Another parallel to make of this form of prediction with global warming, nobody listens to whistle-blowers, until its too late. Numerous well-credentialed physicists have shown using mathematics that sensitivity of climate to CO2 doubling can be no greater than 1C, and that the high rates of warming shown by the IPCC projections are only possible with unrealistic models. Even though the empirical data confirm low sensitivity, it falls on deaf ears.

For an example of this strategy, see the Met Office getting it wrong but claiming the opposite!

Help, its raining swans!

Nassim Taleb, author of “The Black Swan”; gives us another example of how to predict. His strategy is to predict eventualities that are possible only remotely, yet are highly consequential. This is also called the Chicken Little Strategy – ‘the sky is falling’. Like global warming.

Models and agents argues this approach (like anthropogenic global warming), is largely disengenuous. Like climate change, the current economic crisis is not a black swan. The world’s economic history has suffered numerous credit and banking crises. Not only are credit crises predictable; they can be a no-brainer if they involve extending huge loans to people with no income, no jobs and no assets.

Taleb, like Prof. Garnaut, also recommends that we buy insurance against black swans—that is, investments with a tremendous (though still highly remote) upside but limited downside. For example, you could buy insurance against global warming by converting coal-fired electricity plants to solar, and look good if it happens. Only if global warming is catastrophic, everyone suffers the same fate.

I think models and agents gets it right. Ultimately, the problem lies in our failure to use our brain. Only “using our brain” does not mean taking out dubious insurance. It means identifying strategies that are diversified, productive and viable, improve our quality of life and create jobs. It also means testing our models, and overriding them when they fail. The big problems you find in academia are:

1) People often grossly exaggerate confidence in models, usually because they take their assumptions literally. they act as though the model is exactly reality.

2) Academia is so specialized, one may understand his narrow area well, but little else.

3) Intuition and high level thinking is grossly under-rewarded.

NASA Hathaway slides the goalposts

Prediction is dangerous to your reputation.

If you don’t make a clear prediction (a climate cycle, a solar cycle, a financial trend…) then you are just doing your best. What comes does not damage your reputation.

One way to predict is to reproject on a regular basis, called ‘moving the goalposts’. David Hathaway of NASA illustrates this strategy, as show in a recent post at WUWT.

Here is Hathaway’s most familiar graphic, which has an active sun in the background. Perhaps it is time to update that background to something more reflective of the times…..oh wait, read on.

Here in this graphic, from we can see how much has changed since Hathaway’s last prediction update in October 2008:

Click for a larger image

Note that Hathaway did indeed change background graphics from October to January. Its just not quite the smooth and nearly featureless ball we see today.

Hathaway’s predicted Cycle 24 maximum in March 2006: 145
Hathaway’s predicted Cycle 24 maximum in October 2008: 137

Hathaway’s predicted Cycle 24 maximum in January 2009: 104

I’d say that represents a sea change in thinking, but the question now is: How low will he go?

I was looking for a substantial quote from Hathaway in his prediction page, but it appears he is being quite conservative in his language, focusing mostly on methodology, not the prediction itself. I don’t blame him, he’s in a tough spot right now.

Anthony, you are too kind. The update from Hathaway makes no mention of the previous failed prediction.

By moving the goalpost, you never have to acknowledge you missed it. If you move goal, you don’t need nearly as many excuses.

The thing about goals is that moving them maintains your reputation, in the short run.

It seems to me, though, that the people who really care about the science, who lead, who grow and discover things… those people use prediction to show what they don’t know, where they went wrong, reject bad theories and work to fix it. Hathaway is a consensus scientist, whereby the models of other researchers are weighted according to their reliability at any time. Nobody is wrong in this fluid consensus model. No theory is ever falsified. Everyones get funded next year.

The following quote from Hathaway’s update illustrates how consensus science makes a mockery of scientific method.

We have employed these methods along with several others to determine the size of the next sunspot cycle using a technique that weights the different predictions by their reliability. [See Hathaway, Wilson, and Reichmann J. Geophys. Res. 104, 22,375 (1999)]

December 2008 Global Temperature Falls

Latest results from RSS for global temperature in the lower atmosphere show a decline in December 2008 to 0.174 from 0.216 in the previous month. Two early leaders in the ‘Guess the monthly global temperatures’ competition have emerged: CoRev and Jan Pompe. Below are the questions so-far, and all the punters with at least one correct prediction of the direction of monthly global temperature.

What is the point of this competition? Well, I think there is a big difference between talking about prediction, and actually doing it. Real tests of skill provide an opportunity to observe your motivations, biases and reactions. I notice that the total number of votes dropped from November to December. Could this be due to the realization that expectation for monthly global temperature direction is based more on intuitive certitude (bias) than science?

In the unlikely case the somebody does display real skill at prediction, how would they do it? My guess is that skill would be achieved by some version of ‘front-running’, using information not widely available to other participants. In any event, its a prediction experiment.

Voting is open for January temperatures. You can place your vote below.

AIMS GBR coral findings merely opinion

Jennifer Marohasy concurs that the AIMS GBR study presents level 5 evidence (merely expert opinion) that measured decline in coral growth is due to anthropogenic global warming.

Indeed no data is presented to suggest the PH (a measure of acidity) of GBR waters has changed.

Confronted with a lack of evidence in support of this hypothesis – that ocean acidification has caused the drop in growth rates – the researchers suggest in the paper “synergistic effects of several forms of environmental stress” and implicate higher temperatures. But no data is presented in the paper to contradict the well established relationship between increasing temperature and increasing growth rates – though various confusing statements are made and it is suggested that global warming has increased the incidence of heat stress in turn reducing growth rates – while at the same time the researchers acknowledge higher growth rates in northern, warmer, GBR waters.

In a further swipe at global warming zealotry, squarely at the exaggeration of low quality of evidence, she quotes:

Marine Biologist Walter Starck has perhaps aptly described the research as part of “the proliferation of subprime research presenting low value findings as policy grade evidence” and has suggested this has “science headed in the same direction as Wall Street.”

Interestingly, Queensland Premier, Anna Bligh, has decided the “massive decline in the reef’s growth” will require new laws.

CSIRO and AIMS race each other to the bottom of the science quality ladder.

What can be done? I have frequently advocated the adoption of a quality of evidence rating modeled on the Oxford Evidence Scale, applied to all research with policy implications, particularly global warming. This could easily be implemented on an institution-wide (CSIRO, AIMS, etc.) basis.

The main benefits would be a quick indication of how much weight to place on a particular piece of research, and a quantitative measure of improvements in research quality over time.

In the comments section of Jennifer’s post, Walter Starck indicates the downturn may be an artifact of the statistical smoothing methodology (sound familiar?).

The purported slowing of coral growth rates on the GBR appears to be a remake of “The Hockeystick” with Splines playing the role of Principal Component Analysis. De’ath et al. offer no actual growth data in the paper or online supporting evidence. What they present are splines derived from the data. These depict a dramatic reduction in growth after 1990. During this period the GBR suffered two severe bleaching events in 1998 and 2002. Both were associated with El Niño induced extended summer calms and resultant SST spikes.

Splines are only a curve fitting tool and various different splines can be constructed to fit the same data. In this instance it is apparent that the knots chosen for construction of these particular splines are ones which result in a sharp downturn after 1990 due to the hiatus in growth from the bleaching events. This is obvious from the fact that neither the bleaching events nor the subsequent recovery appear as distinct changes in the curves but have both been smoothed into a sharp decline continuing down to the end in 2005. That the data set ends in 2005 also helps in exaggerating the decline by truncating the ongoing recovery in growth after the 2002 bleaching.

Interestingly this study comes only a few days after the report that recovery of corals from the Boxing Day tsunami has been found to be occurring much faster than expected.

It would be productive to get the raw data plotted and reanalyzed. Any takers?

GBR Coral Growth Study from AIMS

This newly released study from the Australian Institute of Marine Science in Townsville is getting a lot of press. An interview with the author Glen De’ath by the ABC claims a tipping point for coral growth has already been reached in 1990. claims the growth of coral in Australia’s Great Barrier Reef has slowed its lowest rate in at least 400 years as a result of warming waters and ocean acidification.

The claims are made on the basis of data apparently showing the rate of calcification from 328 colonies of massive Porites corals from 69 reefs of the Great Barrier Reef (GBR) has declined by 14.2% since 1990. These data would seem to be worthwhile checking by independent sources. Abstract from Science below, a News of the Week article by Elizabeth Pennisi. As the claims are based essentially on an upside-down hockey stick, there seems to be a contradiction in the abstract.

Declining Coral Calcification on the Great Barrier Reef
Glenn De’ath,* Janice M. Lough, Katharina E. Fabricius

Reef-building corals are under increasing physiological stress from a changing climate and ocean absorption of increasing atmospheric carbon dioxide. We investigated 328 colonies of massive Porites corals from 69 reefs of the Great Barrier Reef (GBR) in Australia. Their skeletal records show that throughout the GBR, calcification has declined by 14.2% since 1990, predominantly because extension (linear growth) has declined by 13.3%. The data suggest that such a severe and sudden decline in calcification is unprecedented in at least the past 400 years. Calcification increases linearly with increasing large-scale sea surface temperature but responds nonlinearly to annual temperature anomalies. The causes of the decline remain unknown; however, this study suggests that increasing temperature stress and a declining saturation state of seawater aragonite may be diminishing the ability of GBR corals to deposit calcium carbonate.

Australian Institute of Marine Science, Townsville, Queensland 4810, Australia.

How can calcification simultaneously increase linearly with increasing sea surface temperature, and at the same time, have remained relatively constant over the last 400 years? Either sea surface temperatures were constant, or calcification rate did not respond to temperature.

Update: jae pointed to some contradictory studies.
Crabbe, M.J.C., Wilson, M.E.J. and Smith, D.J. 2006. :

Crabbe et al. report, first of all, that the Quaternary corals they studied appear to have grown “in a comparable environment to modern reefs at Kaledupa and Hoga,” except, of course, for the air’s CO2 concentration, which is currently higher than it has been at any other time throughout the entire Quaternary, i.e., the past 1.8 million years. Second, the results of their measurements indicate that the radial growth rates of the modern corals are 31% greater than those of their more ancient Quaternary cousins, in the case of Porites species, and 34% greater in the case of Favites species.

Review of long-term coral data sets:

So what did Bessat and Buigues find? First of all, they found that a 1°C increase in water temperature increased coral calcification rate at the site they studied by fully 4.5%. Then they found that “instead of a 6-14% decline in calcification over the past 100 years [as] computed by the Kleypas group, the calcification has increased, in accordance with [what] Australian scientists Lough and Barnes [found].” They also observed patterns of “jumps or stages” in the record, which were characterized by an increase in the annual rate of calcification, particularly at the beginning of the past century “and in a more marked way around 1940, 1960 and 1976,” stating once again that their results “do not confirm those predicted by the Kleypas et al. (1999) model.”

There seems a variation from other results in the field.