The dubious sea-level study by Church et al.

Global warming advocate produces data to support global warming

I am sure there will be more expert comments than mine forthcoming but I just want to point out what seem to me to be some of the more obvious implausibilities in the latest Australian research (GRE 2006) that reports a study of sea-levels from 1870 on. I reproduce the Abstract hereunder:

Multi-century sea-level records and climate models indicate an acceleration of sea-level rise, but no 20th century acceleration has previously been detected. A reconstruction of global sea level using tide-gauge data from 1950 to 2000 indicates a larger rate of rise after 1993 and other periods of rapid sea-level rise but no significant acceleration over this period. Here, we extend the reconstruction of global mean sea level back to 1870 and find a sea-level rise from January 1870 to December 2004 of 195 mm, a 20th century rate of sea-level rise of 1.7 ¤ 0.3 mm yr?1 and a significant acceleration of sea-level rise of 0.013 ¤ 0.006 mm yr?2. This acceleration is an important confirmation of climate change simulations which show an acceleration not previously observed. If this acceleration remained constant then the 1990 to 2100 rise would range from 280 to 340 mm, consistent with projections in the IPCC TAR.

There is a popular summary of the research here. A few excerpts:

What we found is that sea levels are rising and increasing with time," the CSIRO study's co-author John Church said. "It means there will be increased flooding of low-lying areas when there are storm surges. "It means increased coastal erosion on sandy beaches. We're going to see increased flooding on island nations." ... Greenhouse gas emissions should be reduced by 50 per cent by 2050, Mr Church said. "If not, climate change will continue and increase in magnitude," he said.

By examining tidal data, Mr Church said sea levels rose by 19.5cm between 1870 and 2004. The increases accelerated with time, averaging 1.7mm a year in the 20th century and 1.8mm in the past 50 years. Mr Church said sea increases were previously based on climate change models. He said his team's research was the first to document rises based on extensive historical tidal data, allowing predictions on sea-level increases to be made with greater precision.

Many island nations are already feeling the impact of rising seas. In Vanuatu and Papua New Guinea, increased sea levels have forced hundreds of islanders to abandon vulnerable coastal homes for higher ground, according to the United Nations and news reports.

So they report that sea levels have risen nearly eight inches between 1870 and now! That for a start seems to me to be a nonsense. Such a large rise in relatively recent times would surely have led to worldwide comments about what was once land now being swamped and I know of no such widespead comments or examples of flooding. Land does rise and fall for various reasons (e.g. in coastal California and Eastern England and perhaps the Maldives) but flooding due to sea-level rise has just not happened as far as I can see. And while sea-levels in some Pacific islands may have risen (though the Vanuatu claim is a fraud), in others the levels have fallen! (See here and here).

And how do Church et al. reconcile their "reconstruction" of sea-levels with the actual evidence provided by John Daly's `Isle of the Dead' (Tasmania), tide gauge from 1841 -- which shows a sea-level that is HIGHER than today? No doubt the actual 1841 observation was "wrong" and the modern reconstruction is "right"

In view of Church's obvious enthusiasm for global warming theory, we should also perhaps keep in mind this report:

When a trial of 908 volunteers found that using anti-inflammatory drugs could reduce the risk of mouth cancer, it caused considerable excitement among cancer researchers. The Harvard School of Dental Medicine described the study as impressive, claiming it might lead to earlier identification of pre-cancerous cells. Conducted by Dr Jon Sudbo, a previously-published researcher and cancer expert from the well-respected Radium Hospital in Oslo, Norway, the study was published in The Lancet, one of the world's most respected medical journals. So it came as a shock when revealed earlier this month that Sudbo's study was fiction, based on 908 patients who did not exist.

To make matters worse, the fraud was not discovered by The Lancet or his colleagues, but by Camilla Stoltenberg, a director of epidemiology at the Norwegian Institute of Public Health in Oslo. Sudbo said the study was based on information collated from a public health database. Stoltenberg, responsible for the database, knew it did not contain the sort of information Sudbo cited. Confronted, Sudbo admitted he falsified the data. He also admitted that other studies on oral cancer, in the New England Journal of Medicine in 2004 and The Journal of Clinical Oncology in March last year, were also fake.

The scandal comes less than a month after the Science retracted two papers by leading stem cell researcher Hwang Woo-suk after it was revealed he faked most of his ground-breaking work on cloning. The scams are by no means the only examples of fabricated research (see next page). However, the breathtaking nature of Sudbo's actions has raised questions about the effectiveness of peer review and journal editors' ability to identify misleading research.....

Professor Judy Black, chair of the National Health and Medical Research Council's research committee, agrees peer review has limitations. "When you get an article to review, you go on the data in front of you. You can look at the researcher's methods and see if you can detect differences between your methodology and theirs, but if two people do the same experiments and get different results, it doesn't mean one is fraudulent."

Black believes fraudulent research may also go unnoticed because peer reviewers and colleagues are reluctant to "dob in" fellow researchers. "People don't necessarily speak up about it. There is research that people know is fabricated, and they haven't dobbed the person in because everyone knows what happens to whistleblowers." However, Black says reviewers have no choice but to assume researchers' work is legitimate. "The onus is on the researcher to be honest and not falsify research."

Initial results from what might be the largest study of the practice of peer review ever conducted shows this faith may be misplaced. Three medical journals, The Lancet, the British Medical Journal and the Annals of Internal Medicine, have allowed a team of researchers from the University of California, San Francisco, to attend editorial meetings, look at reviewer comments and follow the progress of more than 1000 articles from submission to rejection or publication. The team hasn't released its final report, but initial findings indicate authors frequently fail to disclose funding sources and potential conflicts of interest in submitted manuscripts, until asked to do so by journal editors.

Black says the pressure to have work published in a high profile journal may tempt some researchers to take shortcuts with their research. "There's no doubt that being published in a high profile journal has a big impact. More and more people are going to be judged on their productivity and the number of studies they have published, and that determines funding. "If you get a research grant and if you have a lot of publications in high impact journals out of that grant, then your likelihood of getting another grant is increased." As a result, the competition to be published is intense....

Last year Richard Smith, editor of the BMJ for 25 years, wrote an editorial saying he suspected fraud "is probably happening on quite a large scale, and we just have inadequate mechanisms for sorting this out." Last year the US Office of Research Integrity, a federal agency responsible for investigating scientific misconduct, received 265 allegations of falsified research.

Michael Callaham, vice-president of the World Association of Medical Editors, says there is little anyone can do to eliminate fraud: "Journals could ask for all sorts of corroborating materials, but reviewers, who are mostly unpaid, and editors, who are mostly underpaid, would not be able to confirm their authenticity, and would not have the time to review them." ....

Jefferson says there is very little evidence that peer review is effective - a fact editors are reluctant to consider.... Jefferson says the medical publishing industry requires "radical change." "The first thing that needs to happen is that editors who maintain that peer review is infallible need to understand that, at best, it's untested. There are also far too many journals and some of them are publishing irrelevant or misleading research.

And despite the angst, research is still coming out in some of the world's leading scientific journals that should never have passed even the limited barriers of peer review. Note the summary below of the latest gem from The Lancet:

"Eating your greens will do more than please your mother: new evidence shows five servings of fruit and vegetables a day can slash your risk of having a stroke by 26 per cent. A review of previous studies, conducted by British and Australian experts, found that even eating between three to five 80g servings a day cut strokes by 11per cent, compared with people who ate fewer than three servings a day. The authors said that while a reduction in stroke from fruit and vegetable consumption was already known, this was the first time researchers had been able to quantify the benefit. The findings suggested that heeding recommendations on fruit and vegetable intake could save lives and prevent thousands of strokes a year.... Their review, published yesterday in The Lancet, looked at the results of eight previous studies that together involved more than 250,000 people who were followed up for an average of 13 years....

The study authors conceded their results might be affected by observational bias. People who ate a lot of fruit and vegetables were probably likely to share other characteristics known to reduce stroke risk - being less likely to smoke or be overweight, and more likely to exercise and to have lower intakes of salt and saturated fat.

The second paragraph as excerpted above shows, of course, that the study proves precisely nothing.

And I suppose that it is just too curmudgeonly of me altogether to point out that one third of what is published in even the most prestigious journals subsequently turns out to be wrong.

Comments? Email John Ray

No comments:

Post a Comment

All comments containing Chinese characters will not be published as I do not understand them