Independent Summary for Policymakers
The Fraser Institute's Independent Summary for Policymakers (ISPM) was published on Feb. 5, 2007, just after the release of the Intergovernmental Panel on Climate Change (IPCC) Summary for Policy Makers, part of the Fourth Assessment Report on Climate Change (AR4). Economist and Fraser Institute Senior Fellow Ross McKitrick served as co-ordinator of the ISPM. The Fraser Institute's ISPM overview page contains an executive summary as well as a link to download the current version of the ISPM. The original version of the ISPM is also available (see section on errors and discrepancies below). For comparison, the various chapters of the IPCC WG1 AR4 report are available at the IPCC AR4 download page. The full citation of this IPCC report is: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
According to Fraser Institute news release accompanying the release of the ISPM:
"An independent review of the latest United Nations report on climate change shows that the scientific evidence about global warming remains uncertain and provides no basis for alarmism.
"In 2006, independent research organization The Fraser Institute convened a panel of 10 internationally-recognized experts to read the UN Intergovernmental Panel on Climate Change (IPCC) draft report and produce an Independent Summary for Policymakers. The result ... is a detailed and thorough overview of the state of the science." 
As to the need for an "independent" summary, the Fraser Institute claimed, in another news release, that the official IPCC Summary "is neither written by nor reviewed by the scientific community." 
Coverage and reaction
On February 6, 2007, the Financial Post (financial section of the National Post) published extensive coverage of the ISPM, including a section entitled "Inside the Science" that covered supposed gaps between the IPCC Summary for Policy Makers and the full IPCC Working Group 1 report. 
In the run up to the release of the IPCC Summary in 2007, Financial Post editor Terence Corcoran also endorsed the notion that the IPCC summary was not written by scientists and went even further: "We have, therefore, an extraordinary operating scheme in which brief sensational summary statements are produced, while the basis for the summary is kept confidential so they can get the science to correspond to the summary." 
In a Newsweek magazine commentary, ISPM co-ordinating author McKitrick echoed this idea: "This is a curious and disconcerting aspect of IPCC procedures: it needs a couple of months to revise a detailed report prepared by hundreds of scientists, to ensure it agrees with a brief summary drafted by a few dozen scientists and edited by hundreds of bureaucrats and politicians." 
An advance copy of the ISPM was leaked by Desmogblog.com on January 31, 2007, with an accompanying press release and other background information. The press release attacked the Fraser Institute's contention that the IPCC Summary was "neither written by nor reviewed by the scientific community":
"In fact, the IPCC summary was written and reviewed by some of the most senior climate scientists in the world, without political or bureaucratic input . And the Fraser Institute’s 'scientific' staff – which is led by an economist – includes a group of junior or retired scientists, most of whom have direct connections to energy industry lobby groups." 
Desmogblog continued: "Dr. Andrew Weaver, the Canada Research Chair in Climate Modelling and Analysis and a lead IPCC author, called the Independent Summary 'highly ideological.' While the Fraser Institute summary says, 'There is no compelling evidence that dangerous or unprecedented changes are underway,' Weaver counters: 'The IPCC report presents 1,600 pages of compelling evidence, that’s the whole point.'"
Realclimate.org issued a response to the ISPM, attacking its premises and detailing several errors and biases, and also made available a detailed critique in the form of an annotated version of the ISPM. 
Errors and discrepancies
This section enumerates several errors and discrepancies found in the eight "bullet" paragraphs of the summarized findings from the Fraser Intitute ISPM overview page, as well as associated sections of the ISPM itself.
- Several incorrect statements concerning tropospheric temperature trends derived from satellite data.
- Misdentification of peak temperature year in GISS and NCDC global surface temperature data sets (1998 given instead of 2005).
- Mistaken citation of projected sea level rise to 2100 of only 10-30 cm, instead of 21-48 cm given by IPCC
Several examples of "cherrypicking", inexplicable omissions and misrepresentations are also given.
References to the ISPM itself and the IPCC AR4 Working Group I report use the the abbreviations ISPM and AR4 respectively, with section numbers as appropriate. ES is used to denote Executive Summary, both for the ISPM, and for each of the AR4 chapter summaries. Links to the referenced documents are found in the opening section above.
To enable easy comparisons, emphases have been added to key quotations. These do not occur in the original quotes.
Since 1979, satellite data has been used to analyze atmospheric trends. Most of the cited studies have been produced by two teams: Mears and Wentz at Remote Sensing Systems (RSS) and Christy and Spencer at the University of Alabama (UAH). [AR4 3.4.1, ISPM 2.1a]
Overall tropospheric warming: ISPM (first version)
The very first finding in the original news release for the ISPM (and the original version of ISPM overview) contains the following statement: "Data collected by weather satellites since 1979 continue to exhibit little evidence of atmospheric warming, with estimated trends ranging from nearly zero to the low end of past IPCC forecasts."  This was virtually identical to the statement found in the ISPM [ISPM v.1 2.1b] and echoed the summary statement: "Globally-averaged measurements of atmospheric temperatures from satellite data since 1979 show an increase of 0.04°C to 0.20°C per decade over this period" [ISPM v.1 ES].
However, the IPCC gives a range of "0.12°C to 0.19°C per decade for MSU estimates of tropospheric temperatures." [AR4 3 ES] The error in the ISPM was apparently due to the citation of of raw T2 channel satellite trends, which are uncorrected for the effect of stratospheric cooling [AR4 184.108.40.206]. Moreover, the two ISPM statements above refer to the vague term "atmospheric warming" without even distinguishing between the troposphere and stratosphere. The error was pointed out to the Fraser Institute by an anonymous Sourcewatch contributor.
In an amendment dated March 2, 2007, the ISPM overview now acknowledges "a problem in the wording of 2.1b" and goes on to state: "The trend coefficients of specific interest reflect processing to remove an estimated stratospheric component and fall in the narrower interval 0.12-0.19 C/decade. The wording of Sct 2.1b has been expanded and revised to remove any misunderstanding on this point."
The key finding in the overview, original news release and ISPM Executive Summary have all been corrected to remove the above error. The overview (and amended news release) now read: "Data collected by weather satellites since 1979 continue to exhibit some evidence of lower atmospheric warming, with estimated trends ranging near the low end of past IPCC forecasts." The news release is still dated February 5, 2007 and no indication is given that it has been amended. No acknowledgment has been made by the Fraser Instiute concerning this correction in a key finding of the ISPM.
The ISPM overview claims: "There is no significant warming in the tropical troposphere (the lowest portion of the Earth’s atmosphere), which accounts for half the world’s atmosphere, despite model predictions that warming should be amplified there." [see also ISPM 2.1c].
However, one of the two satellite-based tropical tropospheric estimates (RSS) shows a trend of 0.18°C per decade. [AR4 Fig. 3.18] Moreover, as acknowledged in the ISPM, models predict amplified warming in the tropical upper troposphere, not the tropical troposphere as a whole. [ISPM 2.1c, AR4 10.3.4] There are no estimates for temperature trends in the upper troposphere in the satellite-based data sets found in AR4.
Moreover, models predict that amplified warming in the upper tropical troposphere would accompany long-term warming, no matter whether the forcing is anthropogenic (greenhouse gases) or natural (solar).  As noted in the section below, it is the combination of tropospheric warming and stratospheric cooling that most clearly distinguishes greenhouse gas induced warming from that resulting from natural external forcings.
The ISPM goes on: "One of the available satellite data sets shows trends consistent with increased warming at higher altitude in the tropics ... while others do not." [ISPM 2.1c]
However the IPCC states: "After 1987, when MSU channel 3 became available, Fu and Johanson (2005), using RSS data, found a systematic trend of increasing temperature with altitude throughout the tropics." [AR4 220.127.116.11.2] This is the only satellite data based study cited in AR4 that addresses the issue at all.
The ISPM also states: "Adjusting T2 data to remove an estimated contribution from the stratosphere yields tropospheric trend coefficients ranging from about 0.12 oC to 0.19 oC per decade, depending on the method." [ISPM 2.1c] But the difference between UAH and RSS results appears to be mainly due to differing tropospheric data filtering and calibrations independent of adjustments to remove spurious stratospheric cooling; a comparison of the T2 and TMT (mid-troposphere) trends also show similar differences in trends between the UAH and RSS. [AR4 Fig. 3.18]
The IPCC gives nearly equal weight to observed trends of tropospheric warming and stratospheric cooling (AR4 3 ES). However, as noted above, the ISPM did not even mention stratospheric trends at first, and only peripherally in the second version, but still without explicit mention of the observed cooling trend [ISPM 2.1c]. In general, the observed trends are consistent with IPCC model future projections of "relatively uniform warming of the troposphere and cooling of the stratosphere". [AR4 10.3.2.1, AR4 Fig. 10.7]
Surface temperature trends
The ISPM overview states: "Temperature data collected at the surface exhibits an upward trend from 1900 to 1940, and again from 1979 to the present. Trends in the Southern Hemisphere are small compared to those in the Northern Hemisphere."
These statements are essentially correct, but the corresponding ISPM passages contain misleading details and implications, as outlined below.
Peak year and overall temperature trend
The ISPM states: "The global temperature statistic produced by the Goddard Institute for Space Studies (GISS) and the National Climate Data Center (NCDC) was slightly higher in 2005 than at any time since 1998, while that produced by the Hadley Center peaked in 1998 and has been slightly lower ever since." [ISPM 2.1d] This seems to suggest that 1998 was the peak year in all three data sets, whereas in fact 2005 was the warmest year in the instrumental record for both GISS and NCDC. There would also appear to be an unstated implication that temperature may have reached a plateau or even decreased since 1998. However, the ISPM fails to mention that the smoothed temperature statistic for the combined data sets continued to show an upward trend through 2005. [AR4 3.2.2]
Other key indicators of continued warming trend are mentioned in AR4, but omitted in the ISPM:
- "2002 to 2004 are the 3rd, 4th and 5th warmest years in the series since 1850"
- "Eleven of the last 12 years (1995 to 2006) ... rank among the 12 warmest years on record since 1850."
- "Surface temperatures in 1998 were enhanced by the major 1997–1998 El Niño but no such strong anomaly was present in 2005." [AR4 3 ES]
Long-term persistence and trend significance
The ISPM states: "The significance of trends in temperature and precipitation data is likely to have been overstated in previous analyses", due to possible long-term persistence of accumulated climate changes. [ISPM 2.1g]
This is a misrepresentation of the corresponding statement in the AR4:
"Hence, the statistical significances of REML AR1-based linear trends could be overestimated (Zheng and Basher, 1999; Cohn and Lins, 2005). Nevertheless, the results depend on the statistical model used, and more complex models are not as transparent and often lack physical realism. Indeed, long-term persistence models (Cohn and Lins, 2005) have not been shown to provide a better fit to the data than simpler models." [AR4 Appendix 3.A]
Thus, the ISPM statement is at complete odds with AR4 and should have been relegated to the following sidebar on long-term persistence.
Although the ISPM correctly notes that trends in the southern hemisphere are lower than those found in the northern hemisphere, it fails to note that this difference is in accord with IPCC model projections, due to smaller extent of continental land mass in the south. [ISPM 2.1c; AR4 10.3.2.1]
Extreme weather events
Both the ISPM overview and the ISPM itself state: "Perceptions of increased extreme weather events are potentially due to increased reporting. There is too little data to reliably confirm these perceptions." [ISPM 2.3a] In support of these assertions, the ISPM draws on several potential problems cited in the background section of the extreme weather events section of the AR4 report. These include possible increased anecdotal reporting of extreme events through increased use of technology like camcorders, or scarcity of data for certain land regions. [ISPM 2.3a, AR4 3.8.1].
However, the assertion that there is "too little data" to support any "perceptions" or, more to the point, actual trends of extreme weather events, is not borne out by the rest of section AR4 3.8 as detailed below.
Extremes of temperature and precipitation
According to AR4 tables 3.7 and 3.8, extreme weather trends documented with "very likely" confidence include:
- Decrease in low-temperature days/nights and frost days over 70% of global land area (1951–2003) [AR4 18.104.22.168]
- Increase in high-temperature days/nights over 70% of global land area (1951–2003) [AR4 22.214.171.124]
Other trends cited with "likely" confidence include (but are not limited to):
- Increase in warm spells (heat waves) globally (1951-2003) [AR4 FAQ 3.3]
- Increase in heavy precipitation events in mid-latitudes, disproportionate with changes in mean (1951-2003) [AR4 126.96.36.199]
- Increase in total area affected by drought since the 1970s [AR4 3.3.4 and FAQ 3.3]
None of the above trends are discussed or even mentioned in the ISPM.
Cyclones (hurricanes and extreme storms)
The ISPM states: "Since 1970, there is some evidence of increased tropical cyclone intensity in both hemispheres, but a decrease in total tropical storm numbers, and no clear global pattern." [ISPM 2.3b] The increase in tropical cyclone intensity is indeed confirmed in AR4, but there is no significant trend in frequency since the 1970s [AR4 3.8.3, FAQ 3.3, Table 3.8].
The ISPM also states: "Data are too sparse, and trends inconsistent, to identify a pattern in extratropical cyclones." [ISPM 2.3c] However, the AR4 Table 3.8 ascribes "likely" confidence to a finding of "net increase in frequency/intensity and poleward shift in track" for extreme extratropical storms over land in the northern hemisphere since 1950. [AR4 3.8.4, 3.5, FAQ 3.3]
Precipitation, snow and Arctic ice
The ISPM overview states: "There is no globally-consistent pattern in long-term precipitation trends, snow-covered area, or snow depth. Arctic sea ice thickness showed an abrupt loss prior to the 1990s, and the loss stopped shortly thereafter."
Each of these assertions is discussed at greater length below.
The ISPM states: "There is no globally-consistent pattern in long-term precipitation trends." [ISPM 2.2a] This is followed by a few examples of inconsistent regional trends, as well as a statement concerning a lack of trend in overall total precipitation. The significance of a presence or absence of a "globally-consistent pattern", or indeed the definition of that term, remain unexplained.
However the trends in precipitation noted in AR4 include:
- Increased precipitation over land in north of 30°N since 1900 [AR4 3 ES, 3.3.2, FAQ 3.2]
- Decreasing precipitation in the tropics since the 1970s [AR4 3 ES, 3.3.2, FAQ 3.2]
- Substantial increases in heavy precipitation events [AR4 3 ES, 3.3.2, 3.8, FAQ 3.2]
- Increased drought, especially in the tropics and sub-tropics, since the 1970s [AR4 3 ES, 3.3.4, FAQ 3.3]
None of these trends are discussed or even mentioned in the ISPM.
Snow and snow cover
The ISPM summary states: "There is no globally-consistent pattern in snow-covered area or snow depth." [ISPM 2.2b] This represents a marked contrast to the AR4 chapter summary which states: "Snow cover has decreased in most regions, especially in spring and summer." [AR4 4 ES]
In discussing overall northern hemisphere (NH) snow cover trends the ISPM pays special attention to the month of October: "Over the 1966 to 2004 interval, mean Northern Hemisphere snow cover in October showed a statistically insignificant decline. But over the entire span of available data (1922 to 2004) the mean Northern Hemisphere snow cover in October shows a statistically significant increase." [ISPM 2.2b]
The other 11 months are summarized more succintly: "Over the 1966 to 2004 interval, mean Northern Hemisphere snow cover trended downward in spring and summer, but not substantially in winter." [ISPM 2.2b]. 10 of those months showed decrease in NH snow cover and five were statistically significant (March, April, June, July and August). [AR4 Table 4.2]
Moreover, the trends for the two available months other than October in the interval 1922 to 2004 (March and April) are not mentioned at all in the ISPM. They show a significant decrease. [AR4 Table 3.2]
The ISPM states: "Arctic sea ice thickness showed an abrupt loss prior to the 1990s, and the loss stopped shortly thereafter." [ISPM ES] Sea ice thickness, along with various uncertainties associated with its measurement and attribution, are discussed at great length in the ISPM (as well as in AR4). [ISPM 2.5d, 2.5f, 2.5g; AR4 4.4.3]
However, the AR4 gives even more prominence to arctic ice extent, and summarizes: "Satellite data indicate a continuation of the 2.7 ± 0.6% per decade decline in annual mean arctic sea ice extent since 1978. The decline for summer extent is larger than for winter, with the summer minimum declining at a rate of 7.4 ± 2.4% per decade since 1979. Other data indicate that the summer decline began around 1970." [AR4 4 ES; see also 4.4.2] The ISPM does not mention arctic sea ice extent at all.
Sea level rise
The ISPM overview states: "Models project an increase of roughly 20 centimeters over the next 100 years, if accompanied by a warming of 2.0 to 4.5 degrees Celsius."
This corresponds to a projected range of only 10-30 cm sea level rise, as elaborated in the ISPM: "Models project that a doubling of CO2 levels in the atmosphere (A1B scenario), if accompanied by a warming of 2-4.5°C, will cause a sea level increase of about 20 centimeters, plus or minus 10 cm over the next 100 years." [ISPM 5.3d]
However, the actual projected range for sea level rise for the A1B scenario is in fact 21-48 cm. [AR4 10 ES, 10.6.5, Figs. 10.32 & 10.33]. The lower ISPM range would appear to be correspond to the projected rise from thermal expansion alone without any contribution from net land ice melt, given the ISPM citation of AR4 Fig. 10.6.1 (now renumbered AR4 Fig. 10.31).
Historical temperature reconstructions
Variability and uncertainty
The ISPM overview states: "Natural climatic variability is now believed to be substantially larger than previously estimated, as is the uncertainty associated with historical temperature reconstructions." The ISPM itself echoes this satement, while specifying that the "previous estimations" are those of the Third Assesment Report (TAR). [ISPM 3.2a]
Regarding variability, the ISPM fails to mention that the IPCC found that the larger "natural climatic variability" is almost all in the direction of cooler temperatures, relative to "previous estimations", for the past millenium: "The additional variability shown in some new studies [since the Third Assesment Report] implies mainly cooler temperatures (predominantly in the 12th to 14th, 17th and 19th centuries), and only one new reconstruction suggests slightly warmer conditions (in the 11th century, but well within the uncertainty range indicated in the TAR)." [AR4 6 ES].
There is no specific justification for the claim that "uncertainty associated with historical temperature reconstructions" is larger in TAR than in AR4, as no specific citation is given in the ISPM. However, the essential conclusion of exceptional warming in the last fifty years is considered more, not less, certain in AR4 than in TAR.
"The TAR pointed to the ‘exceptional warmth of the late 20th century, relative to the past 1,000 years’. Subsequent evidence has strengthened this conclusion. It is very likely that average Northern Hemisphere temperatures during the second half of the 20th century were higher than for any other 50-year period in the last 500 years. It is also likely that this 50-year period was the warmest Northern Hemisphere period in the last 1.3 kyr, and that this warmth was more widespread than during any other 50-year period in the last 1.3 kyr." [AR4 6 ES]
In 2006, following a request from the U.S. Congress, the National Research Council issued a report entitled "Surface Temperature Reconstructions for the Last 2,000 Years" (NRC). In a "Supplementary Information" section, the ISPM contends: "The National Research Council recommended that proxies sensitive to precipitation be avoided in temperature reconstructions and, in particular, that strip-bark bristlecones and foxtails be avoided." [ISPM 3.2a].
However, this would appear to be an imprecise summary of the NRC recommendations. The NRC stated that "[U]sing proxies sensitive to hydrologic variables ... should be done only if the proxy–temperature relationship has climatologic justification." (NRC, p. 116-7) It should be clarified as well that the NRC did not recommend against all use of bristlecones and foxtails as proxies, but rather stated that "strip-bark samples" of these proxies should be avoided [emphasis added].
The ISPM goes on: "However, none of the IPCC reconstructions for the past millennium observe the National Research Council recommendations." There is no citation or detail given for this assertion, but it is important to note that this finding is not contained in the NRC report itself (which did review all of the reconstructions cited in the IPCC). Indeed, in a live online discussion of the "hockey-stick" controversy, Gerald North, chairman of the NRC report panel stated: "I feel certain that the most recent studies by Cook, d'Arrigo and others do take this [strip-bark problem] into account." 
According to the IPCC, attribution of observed climate change to anthropogenic forcings such as greenhouse gases can be established through the use of computer models. [IPCC 9, ISPM 6]
The ISPM summary states: "These attribution studies do not take into account ... potentially important influences like aerosols, solar activity, and land use changes." This would appear to be an error, as even the ISPM itself states: "Studies have concentrated on what are believed to be the most important forcings: greenhouse gases, direct solar effects, some aerosols and volcanism."
The ISPM generalizes as follows: "Attribution studies to date do not take into account all known sources of possible influence on the climate." [ISPM 6.3c] This is true of most, but certainly not all, of the cited attribution studies, as even the ISPM admits that some studies do take into account forcings such as black carbon and land use. Moreover these studies "continue to detect a significant anthropogenic influence on 20th century temperature changes." [ISPM 6.3c; see also AR4 188.8.131.52 and Table S9.1]
- Terence Corcoran, "Politics first, science second", National Post (Canada), January 27, 2007
- Richard Littlemore, "DeSmog Leaks Advance Copy of Think Tank's IPCC Attack", Desmogblog, January 31,2007
- Terence Corcoran, "Inside the Science", National Post (Canada), February 6, 2007