Thompson and Gabrielli
Lonnie Thompson, senior research scientist at Ohio State University’s Byrd Polar Research Center, and his colleague Paolo Gabrielli, have just been awarded a three-year $588,000 grant from the NSF’s Division of Atmospheric and Geophysical Sciences “to assess the human impact on the chemical characteristics of the glaciers in the Himalaya and the Tibetan Plateau from the pre-industrial era to the present time”:
Gabrielli and Thompson will use an existing set of unique ice cores retrieved from Guliya (Western Tibetan plateau), Naimona’nyi and Dasuopu (Central Himalaya), Puruogangri and Dunde (Central and Northern Tibetean plateau, respectively) to analyze for a large suite of trace elements. These data will allow discrimination of the natural background components (e.g. crustal, volcanic constituents) from the anthropogenic components (e.g. fossil fuel combustion and non-ferrous metal production) of aerosol deposited to these glaciers over time.
The spatial and temporal characterization of atmospheric pollution at high elevations in the Himalaya and the Tibetan Plateau is very much needed because recent studies suggest that atmospheric “brown clouds” deposition to the Himalayan glaciers may affect their energy balance, resulting in an acceleration of ablation. Knowledge of the initial quality of the meltwater, resulting from the ongoing shrinking of the glaciers in the Himalaya, is also important for planning the availability of water resources for millions of people who live downstream from these glaciers.
Ultimately, this study will serve as a source of fundamental information for policy makers trying to mitigate the impact of trace metals in the environment.
See the press release for further details.
Under previous NSF grants, Thompson has collected a truly amazing amount of valuable data on ice cores. However, he has also been notoriously lax about providing definitive archived versions of his measurements. See, for example, IPCC and the Dunde Variations, Juckes, Yang, Thompson and PNAS: Guliya, Gleanings on Bona Churchill, and Mann on Irreproducible Results in Thompson (PNAS 2006).
Fortunately, since Jan. 2011, all NSF proposals must now include a Data Management Plan detailing how any data collected will be archived for public access, so that we can expect any findings under the new grant to be promptly archived. However, according to an NSF representative who recently spoke at OSU, this requirement merely formalizes a long-standing policy that the results of NSF research, including any “metadata” standing behind the bottom line results, must be made public so that others can use it and/or replicate the final results. Thompson is therefore still obligated to archive the results of his past NSF studies.
Perhaps just by coincidence, in the last 10 months Thompson has archived long-overdue data for Guliya on 2/08/12, for Dasuopu Core 3 on 6/16/11, and two cores from Puruogangri on 8/24/11, on NOAA’s paleoclimatic data website at ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/trop/.
(Thompson also reloaded data for Quelccaya Core 1 on 1/13/12. However, it’s not clear whether this a revision of the original file dating back to 1997, or if it is just a new upload of the original file.)
Update: Steve has observed that the new Quelccaya file states right up front,
Note: This file was reformatted 13 January 2012 to provide column
separation between columns 4 and 5. No data values were changed.
The previous version of this file lacked separation between columns
4 and 5 (at AD 871 and at AD 615 through AD 617), potentially
causing errors reading the data.
Update 4/27
Per the request below by Kenneth Fritsch, Here are graphs of the data on 6 cores back to 1000AD that was used in Thompson’s 2003 Climatic Change article. The newly archived Puruogangri data was used in the PNAS 2006 7-core index that goes back to 0AD, but not in the CC index.
(Click on image for larger view.)
These graphs are from my paper “Posterior Confidence Intervals in Linear Calibration Problems: Calibrating the Thompson Ice Core Index,” which was discussed in my earlier CA post, “Calibrating ‘Dr. Thompson’s Thermometer’”.
Update 4/29:
Here are Ken’s plots of the new Puruogangri Core 1 and 2 data:
While Core 2 is rather flat, Core 1 shows a Current Warm Period, but also suggests a Medieval Cool Period, preceded by a Dark Ages Warm Period. It would have been useful if Thompson had provided a concordance of inferred age versus depth so that the dating assumptions could be reviewed.
Update 5/2
Ken has also plotted for comparison the data from Quelccaya Core 1 and Summit Core, as shown below:
The d18O readings from the two Quelccaya cores clearly tell a more coherent story than the two cores from Puruogangri shown above. Their average (as used in MBH99) would presumably have less noise than either series by itself (as in Thompson’s CC03 article).
Both cores also show the attenuation of noise before 1000 AD that characterizes Core 1. This leads me to suspect that the H2O molecules may be able to migrate slowly through the ice. In the later layers this doesn’t make much difference, but in the earlier layers, which are both thinner and have been around longer, it may be causing differences in d18O to average out, creating the appearance of flatter temperatures than really occurred.
Also, if H2O molecules can migrate slowly through ice, it would be interesting to know whether CO2 can also be absorbed from air bubbles into ice, given enough time. This would greatly distort estimates of atmospheric CO2 from ice core records if true.
