Suspicious of those computer models. If the global financial crisis achieves nothing else it should make all of us non-mathematical geniuses at least a little suspicious every time we read of powerful computer models predicting this or that. Much of the mayhem that struck the financial markets was the result of the brilliant modelling that created investments designed to eliminate risk — some of that wizardry rewarded with Nobel prizes if my memory is correct — turning out to be wrong.
The occasional doubts I have about the pace and the likely impact of global warming come from that experience. I find more convincing the rather more mundane methods like those employed by Professor Peter Wadhams, from the University of Cambridge. Professor Wadhams has been studying the Arctic ice since the 1960s and who reported this week on the findings of the Catlin Arctic Survey.
This survey was not the result of some computer simulation but good old fashioned measurements by an expedition led by explorer Pen Hadow which drilled holes through the ice during the course of a 435km trek over the ice earlier this year. The team’s measurements found that the ice-floes were on average 1.8m thick — typical of so-called “first year” ice formed during the past winter and most vulnerable to melting. The survey route — to the north of Canada — had been expected to cross areas of older “multi-year” ice which is thicker and more resilient.
The BBC reports Professor Wadhams saying:
The Catlin Arctic Survey data supports the new consensus view – based on seasonal variation of ice extent and thickness, changes in temperatures, winds and especially ice composition – that the Arctic will be ice-free in summer within about 20 years, and that much of the decrease will be happening within 10 years.
That means you’ll be able to treat the Arctic as if it were essentially an open sea in the summer and have transport across the Arctic Ocean.
Not exactly cheery news for polar bears but at least slightly better than the last report I came across based not on drilling but on sophisticated computer modelling. Back in December 2007 the BBC was reporting the end of summer Arctic ice by 2013 — just four years from now — with Professor Wieslaw Maslowski of the Naval Postgraduate School, Monterey, California, claiming previous projections had underestimated the processes now driving ice loss.
The number crunching by the Maslowski team, which included co-workers at NASA and the Institute of Oceanology, Polish Academy of Sciences (PAS), incorporated “more realistic” representations of the way warm water is moving into the Arctic basin from the Pacific and Atlantic oceans than other models that variously produced dates for an open summer ocean that, broadly speaking, went out from about 2040 to 2100.
The National Snow and Ice Data Center (NSIDC), which is part of the Cooperative Institute for Research in Environmental Sciences at the University of Colorado at Boulder, reported this month that at the end of the Arctic summer, more ice cover remained this year than during the previous record-setting low years of 2007 and 2008. However, sea ice has not recovered to previous levels. September sea ice extent was the third lowest since the start of satellite records in 1979, and the past five years have seen the five lowest ice extents in the satellite record.
Don’t hit him again please. South Australian Liberals who thought they just might benefit from the Premier getting into a spot of bother while on a fund raising night out were sadly disillusioned this morning. Rather than the assault on Mike Rann and stories about an association with the assailant’s wife hurting his political standing, Labor has gone further ahead.
The Advertiser poll published this morning gives the governing party a primary vote of 39% — up one percentage point on the September figure and four points higher than a year ago. As to his personal following, the recent publicity seems to have done no harm at all.
Our Crikey election indicator has Labor well and truly favoured to be returned at the election due in March.
Federally the probability of a Labor win keeps improving with the Crikey Indicator now assessing it as a 78% chance.
A quarter or a half. The Reserve Bank has certainly got the message across that interest rates are on the way up with the only question seeming to be whether the rise will be o.25% or 0.5%. The Crikey Interest Rate Indicator gives the following probabilities:
Don’t want to be pedantic, but the derivatives are not designed to eliminate risk. They are a risk transfer function. The risk still exists but it is given to someone else for economic benefit. One of the functions of the financial system. Also, while you can reduce risk using these exotic securities you can never effectively get rid of all risk.
You are right about computer modelling though. It is always based on certain assumptions that often bear no similarity to the real world. It’s funny that journalists always report on the conclusions, but never the assumptions underlying these models. Take em all with a grain of salt. (but that wouldn’t be news-worthy)
Good to see some healthy scepticism about computer modelling. Models reflect the quality of the data and the assumptions you build in to them (GIGO), and so should only be used to demonstrate possible scenarious that might result from projecting out those assumptions. They should not be used to generate supposedly authoritative predictions of the future, as the future is always uncertain.
The Arctic ice study is at least based on actual data, but the implications are again far from clear. Take a lok at these celebrated photos of nuclear subs in open sea at the North Pole taken in 1958, 1959, 1962 & 1987: http://wattsupwiththat.com/2009/04/26/ice-at-the-north-pole-in-1958-not-so-thick/
Like many supposedly unprecendented climate events, it has all happened before.
It is in no way reasonable to judge climate models on the performance of financial/economic models. Economic modelling uses baseless assumptions and conditions of the kind that the natural sciences would never allow.