In a major paper published in the British science journal "Nature" (Vol.382, 4 July 1996, p.39-46) the top players in the Greenhouse Industry (Benjamin Santer of the IPCC, Tom Wigley of NCAR, Philip Jones of CRU, John Mitchell of the U.K. Hadley Centre, A. Oort and R. Stouffer of GFDL among others) all lent their names to a paper titled "A Search For Human Influences On The Thermal Structure Of The Atmosphere". This paper was trumpeted by the Greenhouse Industry as the final `proof' that Greenhouse was already here, proved not just by models, but also by actual observed data.  And it was little surprise that the `observed data' agreed with the models.

 They claim to have found the imprint of human influence in observations of upper troposphere temperatures as recorded by sonde balloons, matched these observations with what their model would predict under similar conditions and found the very match they were `searching' for.

 This result then inspired the much quoted claim that there was "... a discernible human influence on global climate", a remark made in the notorious Chapter 8 of the 1995 IPCC Report, a remark added later to the report after the meeting of drafting scientists in Madrid.

 Here's how they found their "discernible human influence ..."

 Santer et al choose the dates in the upper graph as a basis on which to compare observed conditions against those that the models would predict. Since the models predict upper troposphere warming under enhanced Greenhouse conditions, it was necessary to show that observed data agreed with the models, thus validating those models and proving that the Greenhouse human fingerprint was already evident.

 When the full available time period of radio sonde data is shown (Nature, vol.384, 12 Dec 96, p522) we see that the warming indicated in Santer's version is just a product of the dates chosen. The full time period shows little change at all to the data over a longer 38-year time period extending both before Santer et al''s start year, and extending after their end year.

 And which version should we trust? The simple rule in all cases like this is -

The longer the time span of a data series, the more reliable is the underlying trend

 It was 5 months before `Nature' published two rebuttals from other climate scientists, exposing the faulty science employed by Santer et al. (Vol.384, 12 Dec 1996). The first was from Prof Patrick Michaels and Dr Paul Knappenberger, both of the University of Virginia, who said in part -

"When we examine the period of record used by Santer et al. (1st graph) in the context of the longer period available from ref.5 (2nd graph), we find that in the region with the most significant warming (30-60⁰ S. 850-300 hPa), the increase is largely an artefact of the time period chosen"

 The second rebuttal was from a German scientist, Gerd R. Weber, who drew attention to the fact that even the period of warming chosen by Santer et al could itself be explained thus -

"Regarding the role of natural factors, the early years of the period 1963-87 were substantially influenced by tropospheric cooling (and stratospheric warming) following the eruption of Mount Agung, whereas the end of that period was influenced by several strong El Nino events, which have led to some tropospheric warming and stratospheric cooling, particularly in the southern subtropics of the lower latitudes. Therefore the general tropospheric warming and stratospheric cooling trend between 1963 and 1987 has been accentuated by widely known natural factors and could at least partially be explained by them."

 In other words, even the warm trend selected out by Santer et al was itself largely explainable by known natural events and not induced through any man-made cause.

 So, did Santer et al  really discover a "discernible human influence on global climate" ?  Hardly. The obvious intent inherent in the paper's title, mounting external pressures for some unambiguous sign of human climatic impact, and the choice of a time period which just happened to show a warming phase in an otherwise neutral longer-term record, indicates only that there is today "a discernible human influence on global climate change science".