Will Hand wrote:


"Graham P Davis" wrote in message
...
Will Hand wrote:


"Graham P Davis" wrote in message
...
Dick Lovett wrote:

On Dec 13, 10:59 pm, Dick Lovett wrote:
On Dec 13, 10:29?pm, Pete L wrote:

http://news.bbc.co.uk/2/hi/science/nature/7142694.stm

Seems rather an odd article. Looking at the figures printed it
would seem that the warming trend is slowing. Perhaps the CO2
emissions from all the flights to Bali will redress the balance
soon!

If you take the last 7 years, the linear trend shows a slight
cooling.

Disregard this. I just realise I had plotted the anomalies rather
than the actual temperatures. They do though, as Pete mentions, show
a decreasing positive trend.


One point to be born in mind with the Hadley data is that it only uses
areas where there is data. At first glance, that might seem a
reasonable thing to do but it means that the area which is warming the
fastest, the Arctic, is mostly ignored. If interpolation of
temperature anomalies is used to cover such data-sparse areas, 2005
becomes the warmest year with 2007 almost certainly relegating 1998 to
third spot.


If data is sparse how do we know the arctic is warmest fastest?


Well, I could say that it's going as forecast but that could get into a
rather circular argument. How about last summer when there was only half
as much ice as there used to be forty years ago?

Anyway, do you have a problem with the idea of interpolating anomaly data
over the Arctic? It seems a reasonable solution to the problem to me. The
Met Office used a similar system for producing an SST analysis from
sparse data.


Not per se, and I'm not disputing the Arctic has warmed a lot either, it
is the statement that it has warmed the fastest that bothers me, there are
other data sparse areas of the world too, so how do we know they haven't
warmed more if there is no data? Interpolation is an estimate, scientists
often forget that. OK I'm being a purist, but I'm an empiricist too!


We meteorologists have always had to use interpolation and the SST analysis
is (was?) no exception in that it calculated the anomaly for each of the
observations and then analysed the anomalies to produce values for all
grid-points. This field was then added to the normal to produce the SST
analysis.

As long as the analysis of the air-temperature anomalies is performed in a
similar objective fashion it should give a reasonable assessment of the
global temperature anomaly. I'm not really saying that one is better than
the other, it depends what you want, a best estimate of the global anomaly
or a perhaps more reliable anomaly for part of the globe.

As to the Arctic being affected more by global warming than anywhere else on
the planet then I'm only going on the theory for that. For example, back in
1975, Manabe and Wetherald calculated that, for a doubling of CO2,
temperatures would rise 2-3C between the Equator and 45N but more than 10C
at 80N and beyond.

My personal view is that, given the forecasts of these great temperature
changes in the Arctic, I'd prefer to have a reasonable estimate of what's
happening up there instead of a map that, for that area, says "here be
dragons".


--
Graham P Davis, Bracknell, Berks., UK. E-mail: newsman, not newsboy.
"What use is happiness? It can't buy you money." [Chic Murray, 1919-85]