Memorandum submitted by Richard S Courtney (CRU 01)
In a press release at
your Select Committee "announces an inquiry into the unauthorised publication of data, emails and documents relating to the work of the Climatic Research Unit (CRU) at the University of East Anglia (UEA)."
And that press release also says;
"The Committee has agreed to examine and invite written submissions on three questions:
-What are the implications of the disclosures for the integrity of scientific research?
-Are the terms of reference and scope of the Independent Review announced on 3 December 2009 by UEA adequate (see below)?
-How independent are the other two international data sets?"
I am writing this as a response to that invitation because - in the context of your questions - the most important email among those hacked (?) from CRU may turn out to be one that I wrote in 2003. I had forgotten it but Willis Essenbach found it among the hacked (?) emails and circulated it. My submission to you explains its meaning and significance.
The email demonstrates that 6 years ago the self-titled 'Team' knew the estimates of average global temperature (mean global temperature, MGT) were worthless and they acted to prevent publication of proof of this.
The pertinent email can be seen at
and I copy it as Appendix A for your convenience.
And, also for your convenience, I copy a draft of the paper discussed in the email as Appendix B. I do not have a completed version of it now (only a late draft that does not include the references) but I could probably get a final version from one of its 18 co-signatories. Additionally, I shall post a hard copy of this document to you.
My tabulated paragraphs of submitted evidence are as follows.
This submission concerns the importance of an email (see Appendix A) from me that was among the files hacked (?) from CRU. It demonstrates that in 2003 the self-titled 'Team' knew the estimates of average global temperature (mean global temperature, MGT) were worthless, and they acted to prevent publication of proof of this.
Climate change 'attribution studies' use computer models to assess possible causes of global climate change. Known effects that cause climate change are input to a computer model of the global climate system, and the resulting output of the model is compared to observations of the real world. Anthropogenic (i.e. man-made) global warming (AGW) is assumed to be indicated by any rise in MGT that occurred in reality but is not accounted by the known effects in the model. Clearly, any error in determinations of changes to MGT provides incorrect attribution of AGW.
I and others tried to publish a discussion paper (see Appendix B) that attempted to explain the problems with analyses of MGT. We compared the data and trends of the Jones et al., GISS and GHCN data sets. These teams each provide 95% confidence limits for their results. However, the results of the teams differ by more than double those limits in several years, and the data sets provided by the teams have different trends. Since all three data sets are compiled from the same available source data (i.e. the measurements mostly made at weather stations using thermometers), and purport to be the same metric (i.e. MGT anomaly), this is surprising. Clearly, the methods of compilation of MGT time series can generate spurious trends (where 'spurious' means different from reality), and such spurious trends must exist in all but at most one of the data sets.
So, we considered MGT according to two interpretations of what it could be; viz.
(i) MGT is a physical parameter that - at least in
principle - can be measured;
These two understandings derive from alternative considerations of the nature of MGT.
If the MGT is assumed to be the mean temperature of the volume of air near the Earth's surface over a period of time, then MGT is a physical parameter indicated by the thermometers (mostly) at weather stations that is calculated using the method of mixtures (assuming unity volume, specific heat, density etc). We determined that if MGT is considered as a physical parameter that is measured, then the data sets of MGT are functions of their construction. Attributing AGW - or anything else - to a change that is a function of the construction of MGT is inadmissable.
If the thermometers (mostly) at weather stations are each considered to indicate the air temperature at each measurement site and time, then MGT is a statistic that is computed as being an average of the total number of thermometer indications. But if MGT is considered to be a statistic then it can be computed in several ways to provide a variety of results, each of different use to climatologists. (In such a way, the MGT is similar in nature to a Retail Price Index, which is a statistic that can be computed in different ways to provide a variety of results, each of which has proved useful to economists.) If MGT is considered to be a statistic of this type, then MGT is a form of average. In which case, the word 'mean' in 'mean global temperature' is a misnomer, because although there are many types of average, a set of measurements can only have one mean. Importantly, if MGT is considered to be an indicative statistic then the differences between the values and trends of the data sets from different teams indicate that the teams are monitoring different climate effects. But if the teams are each monitoring different climate effects then each should provide a unique title for their data set that is indicative of what is being monitored. Also, each team should state explicitly what its data set of MGT purports to be monitoring.
Thus, we determined that - whichever way MGT is considered - MGT is not an appropriate metric for use in attribution studies.
However, the compilers of the MGT data sets frequently alter their published data of past MGT (sometimes they have altered the data in each of several successive months). This is despite the fact that there is no obvious and/or published reason for changing a datum of MGT for years that were decades ago: the temperature measurements were obtained in those years so the change can only be an effect of alterating the method(s) of calculating MGT from the measurments. But the MGT data sets often change. The MGT data always changed between submission of the paper and completion of the peer review process. Thus, the frequent changes to MGT data sets prevented publication of the paper.
Whatever you call this method of preventing publication of a paper, you cannot call it science.
But this method prevented publication of information that proved the estimates of MGT and AGW are wrong and the amount by which they are wrong cannot be known.
(a) I can prove that we submitted the paper for publication.
(b) I can prove that Nature rejected it for a silly reason; viz.
"We publish original data and do not publish comparisons of data sets"
(c) I can prove that whenever we submitted the paper to a journal one or more of the Jones et al., GISS and GHCN data sets changed so either
the paper was
rejected because it assessed incorrect data
But I cannot prove who or what caused this.
It should also be noted that there is no possible calibration for the estimates of MGT.
The data sets keep changing for unknown (and unpublished) reasons although there is no obvious reason to change a datum for MGT that is for decades in the past. It seems that - in the absence of any possibility of calibration - the compilers of the data sets adjust their data in attempts to agree with each other. Furthermore, they seem to adjust their recent data (i.e. since 1979) to agree with the truly global measurements of MGT obtained using measurements obtained using microwave sounding units(MSU) mounted on orbital satelites since 1979. This adjustment to agree with the MSU data may contribute to the fact that the Jones et al., GISS and GHCN data sets each show no statistically significant rise in MGT since 1995 (i.e. for the last 15 years). However, the Jones et al., GISS and GHCN data sets keep lowering their MGT values for temperatures decades ago.
Methods to correct these problems could have been considered 6 years ago if publication of my paper had not been blocked.
Additionally, I point out that the AGW attribution studies are wrong in principle for two reasons.
Firstly, they are 'argument from ignorance'.
Such an argument is not new. For example, in the Middle Ages experts said, "We don't know what causes crops to fail: it must be witches: we must eliminate them." Now, experts say, "We don't know what causes global climate change: it must be emissions from human activity: we must eliminate them." Of course, they phrase it differently saying they can't match historical climate change with known climate mechanisms unless an anthropogenic effect is included. But evidence for this "anthropogenic effect" is no more than the evidence for witches.
Secondly, they use an attribution study to 'prove' what can only be disproved by attribution.
In an attribution study the system is assumed to be behaving in response to suggested mechanism(s) that is modelled, and the behaviour of the model is compared to the empirical data. If the model cannot emulate the empirical data then there is reason to suppose that the suggested mechanism is not the cause (or at least not the sole cause) of the changes recorded in the empirical data.
It is important to note that attribution studies can only be used to reject hypothesis that a mechanism is a cause for an observed effect. Ability to attribute a suggested cause to an effect is not evidence that the suggested cause is the real cause in part or in whole. (To understand this, consider the game of Cludo. At the start of the game it is possible to attribute the 'murder' to all the suspects. As each piece of evidence is obtained then one of the suspects can be rejected because he/she can no longer be attributed with the murder).
But the CRU/IPCC attribution studies claim that the ability to attribute AGW as a cause of climate change is evidence that AGW caused the change (because they only consider one suspect for the cause although there could be many suspects both known and unknown).
Then, in addition to those two pieces of pure pseudo-science - as my paper demonstrated - the attribution studies use estimates of climate changes that are know to be wrong!
None of this gives confidence that the MGT data sets provide reliable quantification of change to global temperature.
Richard S Courtney
An email from Richard S Courtney dated 23 Nov 2003 that was among those hacked or leaked from CRU
The copy of the email hacked or leaked from CRU can be seen at
and it is copied from there below. In the copy below the words of others that are quoted by Courtney are reproduced in a different font for ease of identification.
Subject: Re: Workshop: Reconciling Vertical Temperature Trends
Date: Sun, 23 Nov 2003 18:42:59 EST
Cc: firstname.lastname@example.org, email@example.com, Thomas.R.Karl@noaa.gov, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, Norman.Grody@noaa.gov, Thomas.C.Peterson@noaa.gov, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, cfk @lanl.gov, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, Heinz.Hug@t-online.de, firstname.lastname@example.org, jahlbeck@ab
The excuses seem to be becoming desperate. Unjustified assertion that I fail to understand "Myles' comments and/or work on trying the detect/attribute climate change" does not stop the attribution study being an error. The problem is that I do understand what is being done, and I am willing to say why it is GIGO.
Tim Allen said;
In a message dated 19/11/03 08:47:16 GMT Standard Time, email@example.com writes:
"I would just like to add that those of us working on climate change detection and attribution are careful to mask model simulations in the same way that the observations have been sampled, so these well-known dependencies of nominal trends on the trend-estimation technique have no bearing on formal detection and attribution results as quoted, for example, in the IPCC TAR."
I rejected this saying:
At 09:31 21/11/2003, RichardSCourtney@aol.com wrote:
"It cannot be known that the 'masking' does not generate additional spurious trends. Anyway, why assume the errors in the data sets are geographical and not "?". The masking is a 'fix' applied to the model simulations to adjust them to fit the surface data known to contain spurious trends. This is simple GIGO."
Now, Tim Osborn says of my comment;
In a message dated 21/11/03 10:04:56 GMT Standard Time, firstname.lastname@example.org writes:
"Richard's statement makes it clear, to me at least, that he misunderstands Myles' comments and/or work on trying the detect/attribute climate change.
As far as I understand it, the masking is applied to the model to remove those locations/times when there are no observations. This is quite different to removing those locations which do not match, in some way, with the observations - that would clearly be the wrong thing to do. To mask those that have no observations, however, is clearly the right thing to do - what is the point of attempting to detect a simulated signal of climate change over some part of (e.g.) the Southern Ocean if there are no observations there in which to detect the expected signal? That would clearly be pointless."
Yes it would. And I fully understand Myles' comments. Indeed, my comments clearly and unarguably relate to Myles comments. But, as my response states, Myles' comments do not alter the fact that the masked data and the unmasked data contain demonstrated false trends. And the masking may introduce other spurious trends. So, the conducted attribution study is pointless because it is GIGO. Ad hominem insults don't change that.
And nor does the use of peer review to block my publication of the facts of these matters.
A draft version of the paper that was blocked from publication and is the subject of discussion in an email from Richard S Courtney dated 23 Nov 2003 that was among those hacked or leaked from CRU
Repeated attempts to obtain peer reviewed publication of various versions of this paper were blocked in the manner explained in paragraphs 7 and 8 of the Submission to the Parliamentary Science and Technology Committee from Richard S Courtney. This draft has the same contents (except for the precise data: see pragraphs 7 and 8 of the Submission) as each of the final versions. But it is a pre-publication draft and does not include its bibliography of references.
A call for revision of Mean Global Temperature (MGT) data sets
1. Mean global temperature (MGT)
Mean global temperature (MGT) is the average temperature of the air near the surface of the Earth derived from measurements mostly made at weather stations using thermometers. This short discussion paper calls for a revision to MGT procedures and titles and for the results of that revision to be published.
One could imagine an instantaneous value for MGT but there is no method to determine it. Therefore periodic, individual measurements (mostly made at weather stations) are used to determine an average MGT for periods of time such as days, months or years. Determination of the annual MGT has particular importance because historic data is utilized to compile time-series of MGT since ~1880 and, thus, to gain an indication of the change in MGT since then. This paper comments on the several reported cumulative data sets for these annual values of MGT.
Three different research teams provide values of MGT that are widely used (e.g. Jones et al., GISS, GHCN). They present their results as 'anomalies' from a set value (usually the average MGT of a specified period of years e.g. 1961-90). The 'anomaly' is obtained by subtracting this average temperature value from the determined MGT. Use of anomalies permits direct comparison of the results between teams, because temperature subtractions can be used to adjust the start points of the data sets for comparison.
One important use of data sets of MGT anomalies is in 'attribution' studies of climate change. Attribution studies model the effects that can alter climate, e.g. changes to solar radiance, atmospheric injection of volcanic aerosols, etc.). Differences between the model results and the observed changes to MGT are usually attributed to anthropogenic climate change (AGW). Any errors in the MGT data sets will clearly affect the results of attribution studies which use those data sets.
There are significant variations between the results of MGT calculated by the different teams that compile them. The teams each provide 95% confidence limits for their results. However, the results of the teams differ by more than double those limits in several years, and the data sets provided by the teams have different trends. Since all three data sets are compiled from the same available source data (i.e. the measurements mostly made at weather stations using thermometers), and purport to be the same metric (i.e. MGT anomaly), this is surprising. Clearly, the methods of compilation of MGT time series can generate spurious trends (where 'spurious' means different from reality), and such spurious trends must exist in all but at most one of the data sets.
The three MGT time series are shown in Figure 1.
Figure 1. Mean global temperature anomalies and trends normalized to a common start value as indicated by three teams (after Jones et al., GISS and GHCN).
In this figure, the trends (in °C/decade) and the 2SD trend error are
GHCN: 0.076 ± 0.010
Jones: 0.064 ± 0.007
GISS: 0.048 ± 0.006
The Jones trend is significantly different from the GISS trend (p<0.05), and the GHCN trend is very significantly different from the GISS trend (p<0.01).
The data sets in Figure 1 are derived from the same available source data and, therefore, the differences between the data sets in Figure 1 demonstrate either:
(a) that they are monitoring different
Each team claims to provide a true MGT, but their results differ. Each uses a unique method to derive an indication of some changes to the lower atmosphere, and these methods clearly each provide a different indication of the changes. Therefore, the minimum required amendment to MGT usage is for each team, and others who refer to their data, to use a unique title for the metric that they provide (e.g. 'GISS Surface Temperature Index' and 'GHCN World Warmth Index').
In addition, and noting the importance of MGT time series for bodies such as the IPCC, other changes to the determination of these time series are warranted also, against two different understandings of MGT. Either:
(i) MGT is a physical parameter that - at least in principle - can
These two understandings derive from alternative considerations of the nature of MGT:
1. If the MGT is assumed to be the mean temperature of the volume of air near the Earth's surface over a period of time, then MGT is a physical parameter indicated by the thermometers (mostly) at weather stations that is calculated using the method of mixtures (assuming unity volume, specific heat, density etc).
2. If the thermometers (mostly) at weather stations are each considered to indicate the air temperature at each measurement site and time, then MGT is a statistic that is computed as being an average of the total number of thermometer indications.
The following discussions consider MGT according to each of these alternative understandings.
2. Consideration of MGT as a physical parameter with a unique value for each year
The MGT data sets provided by the various teams are often presented on the same graph (e.g. by IPCC) under the same heading, and there has been no public objection to this by any of these teams. This suggests that the teams agree MGT is a physical parameter that indicates a unique value for the average temperature of the air near the surface of the Earth for each year. But, the data sets provide significantly different trends, and in each of several pairs of years the annual change to MGT differs between the data sets by more than double the calculated 95% confidence limits of each data set. This paradox can be explained by application of measurement theory.
When the measurement sites are considered as being the measurement equipment, then the non-uniform distribution of these sites is an imperfection in the measurement equipment. Some measurement sites show warming trends and others cooling trends and, therefore, the non-uniform distribution of measurement sites may provide a preponderance of measurement sites in warming (or cooling) regions. Also, large areas of the Earth's surface contain no measurement sites, and temperatures for these areas require interpolation.
Accordingly, the measurement procedure to obtain the MGT for a year requires compensation for the imperfections in the measurement equipment. A model of the imperfections is needed to enable the compensation, and the teams who provide values of MGT each use a different model for the imperfections (i.e. they make different selections of which points to use, they provide different weightings for e.g. effects over ocean and land, and so on). So, though each team provides a compensation to correct for the imperfection in the measurement equipment, each also uses a different and unique compensation model.
The large differences between the results generated by each team demonstrates that the compensation models used - by all except at most one of the teams - must contain large errors that generate:
(A) spurious trends to MGT with time, and
But the fact that all the teams calculate their errors demonstrates that each of the teams thinks that its particular model is correct.
Use of an analogy may assist understanding of the problem posed by use of an imperfect compensation model. Figure 2 shows that an area to be measured by an optical system will have a distorted image if viewed at the wrong angle. But - if the viewing angle is known - then the true shape of the image can be calculated so it can be measured. However, a distorted image will still be measured if the compensation applies a wrong correction for the angle. In terms of MGT, the non-uniformity in results between the teams is analogous to different "wrong correction angles". Perhaps most important, the magnitude of the errors to MGT resulting from imperfect compensation models cannot be known, because there is no independent calibration for MGT.
Figure 2. Use of a compensation model in image analysis.
MGT time series are often used to address the question,
"Is the average temperature of the Earth's surface increasing or decreasing, and at what rate?"
If MGT is considered to be a physical parameter that is measured then these data sets cannot give a valid answer to this question, because they contain errors of unknown magnitude that are generated by the imperfect compensation models.
We know that the different compensation models create significantly different results for MGT. Importantly, we also know that the imperfect compensation models provide spurious trends of unknown magnitude and sign in the MGT data sets. And finally we know that the MSU data sets indicate strong warming trends that do not occur in the available MSU satellite (e.g. Christy - full reference needed) and weather balloon radiosonde (e.g. Angell - full reference needed) data sets for lower atmosphere temperature.
A result that is a function of its construction is a serious error. If MGT is considered as a physical parameter that is measured, then the data sets of MGT are functions of their construction. Attributing AGW - or anything else - to a change that is a function of the construction of MGT is inadmissable.
3. Consideration of MGT as a statistic with a variety of possible useful values
The issues raised in Section 2 (above) might be resolved by considering MGT as a statistic (as described in Section 1 above) which does not have a unique value. According to this consideration MGT is not measured - it is calculated from measurements - and, therefore, it is not correct to use measurement theory when considering MGT. Thereby, the arguments advanced in Section 2 (above) become invalid because they are based on measurement theory.
However, if MGT is considered to be a statistic then it can be computed in several ways to provide a variety of results, each of different use to climatologists. In such a way, the MGT is similar in nature to a Retail Price Index, which is a statistic that can be computed in different ways to provide a variety of results, each of which has proved useful to economists. If MGT is considered to be a statistic of this type, then MGT is a form of average. In which case, the word 'mean' in 'mean global temperature' is a misnomer, because although there are many types of average, a set of measurements can only have one mean.
Importantly, if MGT is considered to be an indicative statistic then the differences between the values and trends of the data sets from different teams indicate that the teams are monitoring different climate effects. In this case, there is no reason why the data sets should agree with each other, and the 95% confidence limits applied to the MGT data sets by their compilers may be correct for each data set. Similarly, the different trends indicated by the MGT data sets and the MSU and radiosonde data sets could indicate that they are also monitoring different climate effects.
To treat the MGT as an indicative statistic has serious implications. The different teams each provide a data set termed mean global temperature, MGT. But if the teams are each monitoring different climate effects then each should provide a unique title for their data set that is indicative of what is being monitored. Also, each team should state explicitly what its data set of MGT purports to be monitoring. The data sets of MGT cannot address the question "Is the average temperature of the Earth's surface increasing or decreasing, and at what rate?" until the climate effects they are monitoring are explicitly stated and understood. Finally, the application of any of these data sets in attribution studies needs to be revised in the light of knowledge of what each data set is monitoring.
4. Additional considerations
Clearly, the issues in Sections 2 and 3 (above) need to be properly evaluated. However, all the above issues assume the source data (mostly obtained from weather stations) used to obtain MGT are correct. Though there are in fact serious reasons to doubt the quality of much of this source data, their consideration is outside the scope of this paper.
Another important consideration is that it is not self-evident that MGT is the most useful or even a valid indicator of global climate change. Doubts exist, because it can be argued that a temperature reading is not sufficient to define the quantities of heat stored in a volume of air. What needs to be resolved is the difference between:
(a) the amount of energy the Earth has received
from the Sun; and
(b) the amount of heat retained in the atmosphere by the greenhouse gases.
We conclude that the use of time-series of temperature change alone has the potential to be a strongly misleading indicator of global climate change.
5. Summary and Conclusions
We have discussed the validity of the ways in which MGT is reconstructed and interpreted. However, we are also concerned about the uses to which MGT time series are put in the public discussion about possible human-caused global warming.
Whatever the interpretation placed on the reconstruction of
MGT over the last ~100 years, the following seminal points remain true:
most-cited data sets (CRU, GISS, GHCN) differ significantly in the temperature
trends that they portray;
These rates of temperature change are significantly higher
than the rates for the lower atmosphere measured by satellite MSUs (insert
rate) and weather balloon radiosondes (insert rate).
The pattern of all three MGT estimates between 1900 and 2005
signally fails to correlate with the pattern of human production of CO2.
In contrast, more than 60% of the variance in the MGT temperature signal
correlates with solar variability.
In the face of these facts, the degree to which the debate on global warming is being influenced by the publicizing of alarmist temperature scenarios - based on unverified, deterministic computer models - and by the encouragement of public hysteria about atmospheric CO2, is of great concern to us. That concern is deepened by the fact that senior governmental science advisors, once-influential science journals and distinguished science academies all currently continue to fuel such public alarmism.
Courtney, R S Independent Consultant on Energy and Environment ; an IPCC Peer Reviewer.
Aksberg, A H, Author of Fearless Climate
Baltutis, J S CDR, USN (RET) BA, Mathematics; MS, Operations Research (RET)
Ball, T Environmental Consultant; former Professor
Boehmer-Christiansen, S Reader Dept. of Geography,
Böttiger, H Independent publisher
Bijkerk, A Independent
Quaternary Palaeo-Climate Researcher; Environmental Consultant; Lieut.-Colonel
of the Royal
Carter, R M Professor of Geology, Marine Geophysical Laboratory,
Ellsaesser, H W US
Air Force (RET 1963);
Ferreyra, E President of the Argentinean Foundation for a Scientific Ecology; Independent researcher in climatology
Hissink, L Consulting Geologist; Editor of Australian Institute of Geosciences News
Hughes, W S Geologist; since 1991has studied global temperature trend compilations; author of web pages http://www.warwickhughes.com/
Jelbring, H PhD thesis "Wind
Controlled Climate"; President Inventex
McLean, J Data analyst, independent climate researcher
Moura, R G Electrical engineer and meteorologist
Thoenes, D Professor (em) Chemical and Process Engineering, Eindhoven University of Technology (1979-1995)
Rorsch, A Professor (em) Molecular Genetics Leiden University (1967-1997); member of the board of the Netherlands Organization of Applied Science TNO (1980-1995)
van der Lingen, G J Geologist/paleoclimatologist, Climate Change Consultant Geoscience Research and Investigations New Zealand (GRAINZ
B Senior Marine Researcher
(RET) Geological Survey of
Most of the signatories are members of Sceptical Climate Science - Climatesceptics - the global scientific discussion group for climate scientists and other participants interested in discussing pro and con 'mainstream', neutral, alternative, critical and sceptical views in climatology, but the signatories represent only themselves.
Richard S Courtney