Following the previous post here about alarmist stories of Arctic ice melt, Seymour Laxon, the scientist behind the recent spate of ‘Arctic melting quicker than we thought’ stories, replied in the comments.
“Paul Matthews let me know by twitter that I was wrong to say the measurements were based just on Cryosat2. In the interview, he explains that the data were produced by using Cryosat2 and NASA’s Icestat satellite. Either way, however, the data he refers are measurements still only taken since 2010, which I still believe is far too short a time series to say anything about trends, let alone safely projecting them.”
As I stated on Today the results come from combining data from CryoSat-2 with earlier measurements by NASA’s ICEsat satellite (2003-2008). The statement in Andrew Orlowski’s article that my results are based only two years of data is therefore untrue (I have e-mailed him to point this out).
In the interview, Laxon referred to measurements taken ‘this decade’, which I presumed to mean since 2010. It doesn’t seem unreasonable to me to take from this understanding that Laxon’s predictions about the decline of Arctic sea ice were therefore based on a data series which is too short to be meaningful. (I am still not confident that even the 10 year sample is sufficient, but that is by-the-by as far as this post is concerned).
Most of my criticism of environmental journalists has been their apparent laziness in establishing the facts of the stories they aim to report on — of taking scientists’ claims at face value, and of failing to subject any back story, or coincident claims to criticism and scrutiny. Most of the coverage of developments in environmental science are lifted straight out of the press release, and seen through the pre-existing alarmist narrative, as I discussed in my article for Spiked this week. So it would seem that I had been pretty sloppy myself, for not having established precisely what Laxon’s method was.
Except there was nothing to find. I looked. And I looked. And I looked. I could not find a press release from Laxon’s places of work, or organisations associated with them. I could not find any new research published anywhere. No mention of it was made on the Cryosat2 website. I even asked on Twitter — where is the science?
I replied to Laxon by email, and in the comments:
Thanks for your comment, Seymour.
Could you explain where the research is published, so that we can see how you have produced these results from the data?
Laxon replied by email:
As I have made clear in the media the results are preliminary.
They results are now being finalised for publication (hopefully very soon). The paper will
report on what the data tells us has happened to Arctic ice volume over the period 2003-12.
But ‘preliminary results’ are not results, as I pointed out in the previous post. Why would anyone be interested in a premature announcement of results, and how come they generated so many headlines? I asked Laxon:
So how did the media get hold of these preliminary results? Did you
send them to newsdesks? Or did it go out on a PR newswire? And why
didn’t you wait until you had the final results before you commented
I was approached by the media (it is well know that I am working with CryoSat) and agreed to talk as long as it was made clear that the results were preliminary.
In any case the first estimates of volume from CryoSat were published on the BBC back in April, and presented at the Royal Society in front of the press, at the request of the European Space Agency.
For your information the data have been processed in a more or less identical manner to that described in Kwok et al JGR, 2009 and Giles et al GRL, 2008.
I am doing my best to get the paper and data out there as soon as possible.
At this point, I am not sure who this reflects on the worst: journalists who seem hungry for the story, or the scientist who is prepared to give it to them. Surely an experienced scientist like Laxon ought to know that research like this would provoke a great deal of interest and debate, which would be better served with the benefit of the full and final analysis, not simply headline figures about ‘preliminary results’. I asked Laxon:
I can see by the raft of results produced by google that many journalists picked up the story.
Which media organisation/journalist approached you first in this most recent case of your ‘preliminary results’? I presume that the rest followed after just one newspaper(?) reported it? I’m trying to understand how they found out that you had ‘preliminary results’.
I’m also a bit confused about why you didn’t explain that you couldn’t comment until the preliminary results had been made more concrete.
They were talking to one of my colleagues who was aware of my results. Scientists often share results prior to publication, that’s how science progresses.
It’s common in any case for journalists to come and talk to scientists after conference presentations which may show unpublished results and write articles about them afterwards.
If I had doubts that the final numbers might be substantially different then I might have been more cautious. However the data have been processed using well established and documented (i.e. published in peer review journals) procedures and validated using various ground data. Some of the numbers (such as the agreement of ice volume with PIOMAS) were presented to the press back in April.
Nice talking to you but I really do need to get on and finish the paper which I hope to submit in the next week or so.
That seems to be the end of the correspondence from Laxon. I replied:
Thanks for your email, but it didn’t quite answer my question. I wondered who you had spoken to first — which newspaper/journalist?
No doubt scientists and journalists talk to each other. But most scientific developments that I am aware of, are announced to the press via press release, embargoed until the publication of the
peer-reviewed work. This process has its own problems, of course. But it surely is preferable to the obvious problems that would be created by unpublished, un-peer-reviewed research being reported in the media. Your integrity and honesty notwithstanding, we nonetheless have only your word for the soundness of your method, and of the data itself, and are left none the wiser about how either developed. I look forward to your research being published (when?) but in the meantime, the headlines have been generated, without the scrutiny that stories which [have] such far-reaching implications surely deserve? in spite of your comment that this is how science progresses, I don’t see how science can progress if it is done like this.
So maybe Andrew Orlowski and I were wrong to suggest that the alarmist headlines were based on a ‘half-baked data set‘. But maybe we weren’t. The alarmist narrative has been served, with no opportunity given to interrogate the data, the analysis of the data, and the reporting of the analysis in the media.
And there is no doubting the effect of this story. A Google search for web pages published in the last week for the terms ‘seymour’, ‘laxon’, ‘arctic’ returns 996 results:
BBC interview – Arctic ice melting faster than ever?
Arctic ocean losing 50% more summer ice than predicted
Arctic sea ice could disappear within 10 years as global warming increases speed of melting
Arctic ice could vanish in 10 years, scientists warn
Arctic ice could vanish within 10 years: Scientists
An ice-free Arctic ocean in the summer within 10 years? It’s possible
Arctic losing ice much faster than expected
Arctic Sea Ice Could Vanish In 10 Years: Study
Rate of Arctic summer sea ice loss is 50 per cent higher than predicted
Arctic ice thinning faster than thought
‘Arctic sea ice could disappear within 10 years’
And so on. Hundreds of headlines, read by millions of people. Because Seymour Laxon had a cup of tea with a journalist pal, and revealed the results of his unpublished, un-peer-reviewed work.
This is the anatomy of climate alarmism. For years it has been the claim of environmentalists that their arguments were based on peer-reviewed literature published in credible, scientific journals, and that their critics didn’t enjoy the authority that institutional science gave them. But now we see that all it takes for a story to snowball, are, in such conditions, the premature words of a single scientist, about his ‘preliminary results’. The truth is that the avalanche was ready for the first tiny impulse that would send it cascading throughout the media, across the world. And that is how unpublished, untested results from one study, told by just one man, get turned into stories about ‘science’ detecting ‘greater ice loss than we expected’. If Laxon didn’t know it, he was naive.
Laxon published the following comment on the story at the Register.
Get the facts right Andrew
The statement in this article that these new results rely on just two years of data is, quite simply, false. If you wish to know why then listen to my Today interview (http://news.bbc.co.uk/today/hi/today/newsid_9744000/9744378.stm) where I state that the trends are derived by combining CryoSat-2 volume estimates with earlier (2003-2008) volume estimates from NASA’s ICESat mission [Kwok, JGR, 2009].
I also state that one must be cautious in extrapolating these trends forwards.
Andrew Orlowski responds:
A cautious scientist would be expected to go through the peer review process. You, by contrast, haven’t even published this work yet. It is not available for scrutiny. Nevertheless, you are willing to appear on the national media making dramatic long-term claims, based on *new* data of less than two years observations.
You have been anything but cautious.
Your science may be well turn out be sound, but until it has been independently scrutinized, we just don’t know. Your argument boils own to: “Trust me, I’m a scientist.”
William Connolley — who can’t even keep a job at Wikipedia, let alone Real Climate, such an angry rodent is he — writes at his blog,
Weasel, Rat, Stoat, that Orlowski and I are ‘stupid’, ‘liars’ and ‘idiots’. He then deletes the word ‘stupid’ from my comment there, for its ‘incivility’. Environmentalists have never been very good at sticking to the standards they set for others. but that’s a trivial instance of it. The more important issue is that the emphasis on peer review and publication in credible journals is what made climate science better than anything the sceptics could throw at it.
Seymour Laxon joins in, amongst the commenters there, who seem much more interested in hurling abuse — and hackneyed abuse at that — than engaging in debate. Says Laxon:
Ben >>[in which I ask him for “the science” — something he wasn’t able to produce]
Ben has not actually asked me any questions about the science . If you look at the e-mail exchange you will see he has only asked “Could you explain where the research is published, so that we can see how you have produced these results from the data?”
Anyway for Ben and anyone else who’s actually interested in the science the ICESat paper on trends in ice volume (K09) is available to download here:
In paragraph 39 the 2003-8 trends from ICESat are provided: “The trend in ice volume is -1237/-862 km3/a (fall/ winter).”
What I have done it to combine this ICESat time-series with 2 winters of CryoSat data processed in the same way as described in Giles et al., GRL, 2008, and validated in a similar manner the to comparisons shown in K09, figure 4.
So go and read those two papers and if you have any questions about the “science’ they describe (you’ll need to understand those papers to understand mine as the methodologies are more or less the same) then let us know.
In addition why not go here: file: http://psc.apl.washington.edu/wordpress/research/projects/arctic-sea-ice-volume-anomaly/data/ to download a simple ASCII text data. Once downloaded select out the September data (more or less day 244-274) since 2003, average the data for each year, and then use Excel to tell you what the slope in the data is. Then you’ll have your own trend in Arctic volume to report back.
PS. There are a dozen scientists and engineers on the paper which describes the CS-2/ICESat results.
So Laxon’s answer is: do your own science.
It’s a bit late for that now, though, isn’t it. His hundreds of headlines have been created over the last week.