[ExI] scaling global warming

[Anders Sandberg]

On 24/01/2013 05:26, spike wrote:
>
> A thought occurred to me today as I was reading about a deep lingering 
> cold snap in upstate New York. We discuss the notion of global 
> warming, which is based on an average temperature over the year.  But 
> the thermometers are not evenly distributed.  They are generally more 
> concentrated near population centers.  But it is entirely likely that 
> the concentration factor is insufficient: the distribution of 
> thermometers should be proportioned according to the local human 
> population.
>

The unequal distribution of temperature stations in time and space is a 
big headache for climate modelling. You need to be very careful in 
interpreting measurements because of it. I had great fun playing around 
with the Met office climate data a few years back:
http://www.aleph.se/andart/archives/2009/12/digging_into_data.html
http://www.aleph.se/andart/archives/2009/12/significant_anomalies.html
The concentration near population centres also means the urban heat 
island effect must be controlled for, both in time and space - the extra 
thermometers do not necessarily contribute quality data. It is hard and 
tricky work.

>  It just doesn't seem fair somehow.  More people should equal more 
> data in the global average temperature.
>
>

No, it should mean more weight in evaluating how good or bad it was. 
Multiply the local change with population density and you get a better 
measure.

Suppose all global warming happened in Topeka, Kansas. About 127,000 
people would be affected, but the rest of the world would be fine (and 
the topekans could move from their boiling city): it would essentially 
be no difference. The big temperature increases in the far north right 
now are not so bad for humans since nearly nobody lives above the polar 
circle - if their effects remained local they would only be a concern in 
regards to local ecosystems and economic exploitability. However, the 
feedback effects are worth watching for - they are not well understood 
yet, but potentially important for weather and climate patterns across 
the industrialized - and densely populated - north. And anything that 
messes up the Monsoons is Bad News.

Generally, scaling importance with population affected seems to be a 
good heuristic for setting priorities.

-- 
Anders Sandberg,
Future of Humanity Institute
Oxford Martin School
Faculty of Philosophy
Oxford University

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20130124/0bf2eac4/attachment.html>