Candid Admissions On Shortcomings In The Land Surface Temperature Data [GHCN and USHCN] At The September Exeter Meeting

At the meeting in Exeter, UK September 7-9, 2010 ,

Surface temperature datasets for the 21st Century

there were several  candid admissions with respect to the robustness of the global and USA surface temperature record that are being used for multidecadal surface temperature trend assessments (such as for the 2007 IPCC report).

These admissions were made despite the failure of the organizers to actually do what they claimed when they organized the meeting. In their announcement prior to the meeting [and this information has been removed in their update after the meeting] they wrote

“To be effective the meeting will have to be relatively small but, as stated above, stringent efforts will be made to entrain input from non-attendees in advance.”

In asking colleagues (such as my co-authors on our 2007 JGR paper)

Pielke Sr., R.A., C. Davey, D. Niyogi, S. Fall, J. Steinweg-Woods, K. Hubbard, X. Lin, M. Cai, Y.-K. Lim, H. Li, J. Nielsen-Gammon, K. Gallo, R. Hale, R. Mahmood, S. Foster, R.T. McNider, and P. Blanken, 2007: Unresolved issues with the assessment of multi-decadal global land surface temperature trends. J. Geophys. Res., 112, D24S08, doi:10.1029/2006JD008229

which has raised serious issues with the USHCN and GHCN analyses, none of us were “entrained” to provide input.

Nonetheless, despite the small number of individuals who were invited to be involved, there still are quite important admissions of shortcomings.

These include those from Tom Peterson

who stated in slide 8

“We need to respond to a wide variety of concerns – Though not necessarily all of them”

[from Introductory remarks - Tom Peterson];

Matt Menne, Claude Williams and Jay Lawrimore who reported that

“[GHCN Monthly]Version 2 released in 1997….but without station histories for stations outside the USA)”

“Undocumented changes [in the USHCN] can be as prevalent as documented changes even when extensive (digitized) metadata are available”

“Collectively station changes [in the USHCN] often have nearly random impacts, but even slight deviations from random matter greatly”

“Outside of the USA ~60% of the GHCN Version 3 average temperature trends are larger following homogenization”

“There is a need to identify gradual as well as abrupt changes in bias (but it is may (sic) be problematic to adjust for abrupt changes only)”

“Automation is the only realistic approach to deal with large datasets”

“More work is required to assess and quantify uncertainties in bias adjustments”

“Critiques of surface temperature data and processing methods are increasingly coming from non traditional scientific sources (non peer reviewed) and the issue raised may be too numerous and too frequent for a small group of traditional scientists to address”

“There is a growing interest in the nature of surface temperature data (reaching up to the highest levels of government)”

from Lessons learnt from US Historical Climate Network and Global Historical Climate Network most recent homogenisation cycle – Matt Menne;

and Peter Thorne from Agreed outcomes – Peter Thorne who wrote

“Usage restrictions

Realistically we are not suddenly going to have open unrestricted access to all withheld data. In some areas this is the majority of the data.”

There are very important admissions in these presentations.  First, outside of the USA,  there is inadequate (or no) publicly available information on station histories, yet these data are still used to create a “homogenized” global average surface temperature trend which reaches up to the “highest level of government”.  Even in the USA, there are undocumented issues.

While the organizers of the Exeter meeting are seeking to retain its leadership role in national and international assessments of the observed magnitude of global warming, it is clear that serious problems exist in using this data for this purpose. We will  post information on several new papers when ready to introduce readers of this weblog to quantification of additional systematic biases in the use of this data for long-term surface land temperature trend assessments.

There is a need, however, to accept that the primary metric for assessing global warming and cooling should be upper ocean heat content, since from 2004 onward the spatial coverage is clearly adequate for this purpose (e.g. see).  While there, of course, is a need for regional land surface temperature measurements including anomalies and long-term trends, for obtaining a global average climate system heat content change the oceans are clearly the more appropriate source of this information.

Comments Off

Filed under Climate Change Metrics

Comments are closed.