Flooding in England due to climate change?
gerh...@aston.ac.uk
a nice colourful map) to show that even in the summer in the UK flood
risk has risen / will rise due to climate change?
I can easily understand that warm air holds a great deal more moisture
than cold air, as a chemical engineer I know my steam tables well
enough, but air in summer or near the equator is warmer than in winter
or in Canada, and that doesn't seem to correlate much with flood risk.
If there's no straightforward explanation why a global increase in
temperature should yield more flooding risk everywhere, even in places
it'll get drier, that's fine, I can accept that that might be so and
that I can't easily follow the way people have worked that out. But I
would like some clear evidence it has been worked out properly, and
that eg snow melt and frozen ground have been considered as factors
(which I gather were major factors in the floods of 1947 in the UK and
presumably would improve with milder winters).
At the moment, it just has this feel to me that people think of
tropical downpours and extrapolate that to more flooding without
really having considered the flooding issue properly.
Jim Torson
http://www.guardian.co.uk/uk_news/story/0,,2133244,00.html
Human activity linked to heavier rainfall
James Randerson, science correspondent
Tuesday July 24, 2007
<http://www.guardian.co.uk>The Guardian
Greenhouse gas emissions from human activity are causing global
shifts in rainfall patterns and contributing to wetter weather over
the UK, climate scientists say today.
Their study is the first to find a "human fingerprint" in the
rainfall changes which have been detected in a belt of the northern
hemisphere stretching from the Mediterranean to the UK to Norway.
<snip>
This concerns a study done by the Hadley Centre of the UK Met Office
and others that is published in the latest issue of Nature.
Jim
At 01:20 PM 7/24/2007, gerh...@aston.ac.uk wrote:
>Is there some quantitative measure (eg something that can be put into
>a nice colourful map) to show that even in the summer in the UK flood
>risk has risen / will rise due to climate change?
<snip>
Gareth
not all related to amount or intensity of rainfall. Drainage of
wetlands, land use change, building (esp car parks and roads), flood
defences and their maintenance etc etc, all have an effect on flood
risk. But an intensification of rainfall events is a robust prediction
for a warming world, and I suggest we're now seeing that in practice.
A meteorologist friend commented in a recent email: "The non-linearity
of the Clausius-Clapyron equation has the world by the whatsits, as
they say."
gerh...@aston.ac.uk
http://www.timesonline.co.uk/tol/news/weather/article2127599.ece
"We looked at annual rainfall trends rather than any particular
season," Dr Stott said. "In the UK wetter winters are expected which
will lead to more extreme rainfall, whereas summers are expected to
get drier. However, it is possible under climate change that there
could be an increase of extreme rainfall even under general drying."
Ok, yes it's possible, but Dr Stott doesn't seem to have looked at the
question whether it's happened or can be predicted to happen in the
UK, ie that there's more floods even in the summer when it's predicted
to get drier rather than wetter?
James Annan
Since everyone is talking about the "worst floods for a generation" it
seems pretty clear that we are seeing the typical one in 50 event that
most people can expect to see in their lifetimes. Indeed the max
rainfall was rather less than half the measured record, and rainfall
stats are much less complete than temperature.
I was wondering how the "optimal climate/optimally adapted" people would
respond to this event :-) It is well known that substantial development
has taken place in vulnerable areas with limited attention to flood
risk, rather like New Orleans. On the other hand, maybe it is "optimal"
to allow such flooding on a 50 year return basis rather than pay up
front to defend against it. People still buy the houses!
Of course the usual suspects will try their best to pin it on AGW,
because that is their get-out-of-jail card for inadequate planning and
their main "give us more money" argument. The head of the UK Environment
Agency certainly has previous form on exaggerating climate change for
whatever purpose (to the extent that the scientists who were
misrepresented went public with their criticisms, which as noted in the
resent "Alarmism ad absurdam" thread is extraordinarily rare).
James
Jim Torson
Have you read the actual paper in Nature, or are you
making assumptions based on news reports?
Obviously, it would be much better to read the paper
in Nature.
Gareth
On Jul 25, 10:32 am, James Annan <james.an...@gmail.com> wrote:
> Since everyone is talking about the "worst floods for a generation" it
> seems pretty clear that we are seeing the typical one in 50 event that
> most people can expect to see in their lifetimes
Yes, but how many times in their lifetimes?
[This is not (just) a reference to extreme life extension a la Aubrey
Grey, but to the shift in probabilities of extreme events expected
with climate change]
gerh...@aston.ac.uk
> making assumptions based on news reports?
> Obviously, it would be much better to read the paper
> in Nature
Unfortunately the paper is behind a $30 subscription wall and while
Aston do subscribe to the paper copy from past experience (though I
can't remember whether it was Science or Nature I was after) I've got
that suspicion it'll be nearly a month before they'll have the issue
on the library shelves.
That doesn't prevent the paper from being extremely widely cited in
the media, one of many stories got linked by you in reply to my query.
If you have read the paper, does it actually say anything about flood
risk increasing in British summers due to climate change?
Gareth
>
> > Since everyone is talking about the "worst floods for a generation" it
> > seems pretty clear that we are seeing the typical one in 50 event that
> > most people can expect to see in their lifetimes
And I've just seen this in the Independent:
http://environment.independent.co.uk/climate_change/article2795635.ece
"According to the Environment Agency, even the previous worst case,
the extensive floods of spring 1947, which were aggravated by the vast
snow melt that followed an exceptionally hard winter, has been
surpassed. "We have not seen flooding of this magnitude before," said
the agency yesterday. "The benchmark was 1947, and this has already
exceeded it." And the 1947 floods were said to have been the worst for
200 years."
The odds are shortening, perhaps?
James Annan
Maybe, but it's important to recognise that flooding is a function not
only of rainfall, but of vulnerability:
http://news.bbc.co.uk/1/hi/uk_politics/6911119.stm
http://news.bbc.co.uk/1/hi/england/south_yorkshire/6900636.stm
The EA has at least two very large axes to grind on this subject.
James
Phil Randal
> gerha...@aston.ac.uk wrote:
> >>http://www.guardian.co.uk/uk_news/story/0,,2133244,00.html
>
> >http://www.timesonline.co.uk/tol/news/weather/article2127599.ece
>
> > "We looked at annual rainfall trends rather than any particular
> > season," Dr Stott said. "In the UK wetter winters are expected which
> > will lead to more extreme rainfall, whereas summers are expected to
> > get drier. However, it is possible under climate change that there
> > could be an increase of extreme rainfall even under general drying."
>
> > Ok, yes it's possible, but Dr Stott doesn't seem to have looked at the
> > question whether it's happened or can be predicted to happen in the
> > UK, ie that there's more floods even in the summer when it's predicted
> > to get drier rather than wetter?
>
> Since everyone is talking about the "worst floods for a generation" it
> seems pretty clear that we are seeing the typical one in 50 event that
> most people can expect to see in their lifetimes. Indeed the max
> rainfall was rather less than half the measured record, and rainfall
> stats are much less complete than temperature.
Well, speaking as one who experienced the floods first hand I'd just
like to point out that we last had one of your one in fifty year
events at the end of October / beginning of November 2000. Those
floods were the worst since the 1947 floods here. Comparing the
photos I took then with the ones I took this weekend shows comparable
flood levels, with this year's floods possibly exceeding the levels
seen in 2000 here in Worcester. My uncertainty comes from not knowing
when the river peaked on each occasion, but it was certainly the
highest I've seen it here. Because the river Avon joins the Severn
downstream from here, and was more flooded this time than in 2000,
Tewkesbury and Gloucester suffered more than we did. Hmmm, two once
in fifty year floods less than 7 years apart. OK, I know my
statistics well enough to know that that can happen.
> I was wondering how the "optimal climate/optimally adapted" people would
> respond to this event :-)
Well, I didn't comment on the "optimal adaptation" thread, because I
couldn't quite formulate what I wanted to say in a coherent manner.
My preferred comment would be "bollocks", but you folk deserve better
than that. From a natural selection perspective, selection of
survival traits tends to happen at squeeze points, not when things are
cosy and comfortable, so I'm minded to conclude that we're optimally
adapted to surviving ice ages, and not per se to our current climate.
But I'm no expert on gene pool selection through environmental
factors.
> It is well known that substantial development
> has taken place in vulnerable areas with limited attention to flood
> risk, rather like New Orleans. On the other hand, maybe it is "optimal"
> to allow such flooding on a 50 year return basis rather than pay up
> front to defend against it. People still buy the houses!
Areas which have never flooded in hundreds of years got flooded last
weekend. This really is a weak argument of yours. Tewkesbury is an
old established town, not some new careless "flood plain" development.
> Of course the usual suspects will try their best to pin it on AGW,
> because that is their get-out-of-jail card for inadequate planning and
> their main "give us more money" argument.
Ad-hominem attacks against straw men, you should know better. Shame
on you.
The problem's like the cigarette and cancer conundrum. You can't
prove which cigarette (if any) caused the cancer, but there's still a
demonstrable link. Similarly whilst climate change could produce
intensified storms and changes in the jetstream, it is impossible to
"prove" any single incident. Absence of evidence, however, is not
evidence of absence.
> The head of the UK Environment
> Agency certainly has previous form on exaggerating climate change for
> whatever purpose (to the extent that the scientists who were
> misrepresented went public with their criticisms, which as noted in the
> resent "Alarmism ad absurdam" thread is extraordinarily rare).
He was wrong once so he's always wrong? Logical fallacy there.
Phil
gerh...@aston.ac.uk
What metric do they use to assess how bad a flood is? Area inundated?
Number of houses flooded?
http://hadobs.metoffice.com/hadukp/charts/hadukp_daily_plots.html
http://hadobs.metoffice.com/hadukp/charts/hadukp_ts_plots.html
July doesn't look all that bad precipitation wise, and in fact, there
seems to be a downward trend in the historical data with the 10
wettest July's all before 1900.
I've been looking at the historical data, and with a bit of digging I
managed to undig some all time records:
1. For Enland and Wales as a whole, maximum daily total for May (25
mm), beating the previous record by 5 mm; however, in a number of
other months higher maximum daily totals can be found, eg for August,
where the record was established in 1986 (43 mm), and it seems this
particular series is less than a hundred years old.
2. Then we've got to dig into regional totals and for the North East
there's some substantial records for June, both for monthly total and
maximum daily. However, when looking at August instead, the records
for June look less impressive.
Based on the precip data, we shouldn't be having 50 year floods this
July, let alone 200+ year floods.
Gareth
On Jul 25, 1:00 pm, James Annan <james.an...@gmail.com> wrote:
> Maybe, but it's important to recognise that flooding is a function not
> only of rainfall, but of vulnerability:
As I pointed out in my first post...
Perhaps you'd care to expand on the "maybe", being our local
probability expert?
James Annan
>> It is well known that substantial development
>> has taken place in vulnerable areas with limited attention to flood
>> risk, rather like New Orleans. On the other hand, maybe it is "optimal"
>> to allow such flooding on a 50 year return basis rather than pay up
>> front to defend against it. People still buy the houses!
>
> Areas which have never flooded in hundreds of years got flooded last
> weekend. This really is a weak argument of yours. Tewkesbury is an
> old established town, not some new careless "flood plain" development.
The issue is (partly) the increased speed of upstream run-off due to
increased "hardening" of the surface. I don't know how significant this
effect is in your particular case, but I don't think it is reasonable to
dismiss it as lightly as you seem to be doing above.
>> Of course the usual suspects will try their best to pin it on AGW,
>> because that is their get-out-of-jail card for inadequate planning and
>> their main "give us more money" argument.
>
> Ad-hominem attacks against straw men, you should know better. Shame
> on you.
I think it is merely a statement of the obvious. If people agree that it
was all "unprecedented" then that absolves them of any responsibility of
planning for rainfall that, as I mentioned, did not approach the
observed daily record.
> The problem's like the cigarette and cancer conundrum. You can't
> prove which cigarette (if any) caused the cancer, but there's still a
> demonstrable link. Similarly whilst climate change could produce
> intensified storms and changes in the jetstream, it is impossible to
> "prove" any single incident. Absence of evidence, however, is not
> evidence of absence.
I think we all agree that increasing CO2 will change rainfall patterns
in some way - it would be bizarre and implausible to claim that it will
not have any effect whatsoever. Indeed I rather wonder what all the fuss
is about the recent Nature paper - surely they all know that the
difference between statistically significant, and insignificant, is not
in itself significant? Maybe not, it seems. But anyway.
The issue is, I would say, whether such rainfall could reasonably have
been anticipated and planned for (answer: rather better than it was) and
to what extent it will change in the future (answer: probably an
increase in total rainfall and in extremes, although the summers are
supposed to be drier, and the long-term changes will depend on
cumulative emissions). Obviously improved estimation of the latter will
have some influence on the former, but probably not a great deal over
the next 3 decades (say).
>
>> The head of the UK Environment
>> Agency certainly has previous form on exaggerating climate change for
>> whatever purpose (to the extent that the scientists who were
>> misrepresented went public with their criticisms, which as noted in the
>> resent "Alarmism ad absurdam" thread is extraordinarily rare).
>
> He was wrong once so he's always wrong? Logical fallacy there.
She, actually.
Of course I am not making the absurd claim that anyone who ever made a
mistake is always going to be wrong in that future - that would rule out
any chance of any of us ever getting anything right. But past form is
certainly something I take into account when listening to people's
views, and I think it is entirely natural and reasonable to do so, every
bit as much with the alarmists as with the septics.
James
James Annan
Are you asking about the probability of precipitation extremes, or of
flooding? Given the changes in the built environment, there is not a 1-1
correspondence over time.
The predictions are for drier summers and wetter winters, with an
associated increase in extremes. I'm not sure if the increase in
extremes is supposed to be confined to the winter, but I would not be
surprised if this is the case. In order to see increased summer
extremes, the variability would have to increase rapidly enough to
overcome the trend (rather than adding to it as in winter).
The observed increase in annual precipitation is about 6cm per century
in this latitude band. I don't think there is any reason to expect
substantially more extreme ppt in summer in the current climate,
compared to (say) 50 years ago, but I would happily defer to an expert
analysis. In contrast, I'm sure we can expect hotter summers than were
common over the last century, with a significantly enhanced probability
of extreme heat (and associated drought).
Obviously with more vulnerable housing, there is going to be more flood
damage even with the same precipitation.
James
Gareth
> it seems pretty clear that we are seeing the typical
> one in 50 event that most people can expect to see in their lifetimes
And on Jul 25, 1:55 pm, James Annan <james.an...@gmail.com> wrote:
> Are you asking about the probability of precipitation extremes, or of
> flooding? Given the changes in the built environment, there is not a 1-1
> correspondence over time.
I'm asking abut both, really. Your 50 year flood event turns out to be
worse than 1947, which was a 200 year event. Now, obviously, a 200
year event followed 60 years later by a ?300 year event doesn't mean
the probabilities have necessarily changed, but it is suggestive given
the predicted changes to the hydrological cycle.
As far as precipitation extremes are concerned, we have (as I said) a
robust prediction of intensification, and it looks as though that is
being seen. Although the actual amounts involved in the UK floods may
not themselves be records, the rate of precipitation over a wide area
may well be. It will be interesting to see the post-flood Met Office
analysis. (Some figures on the NZ environment blog:
http://www.newscientist.com/blog/environment/2007/07/flooding-in-england-whats-to-be-done.html).
As to the summer extremes - wouldn't that be exactly when you'd expect
rainfall records to be set? Warmer air, more moisture, thunderstorms,
flooding. Perhaps we should be looking at changes in the occurrence of
"thundery rain"
AdamW
towards 2080 is predicted to be drier summers generally, but more
extreme rainfall events when they do occur within that.
It is worth noting that it's not just the rainfall on the day that
counts. For example, Sheffield had its highest rainfall month on
record in June (the month it flooded) with the previous record easily
beaten around about the time of the floods. This meant that the rivers
were already running high, and the ground saturated when the (second)
rainfall event occurred. The area also experienced heavy floods (not
as bad I think) around 2000ish but I don't know much about those.
As regards to planning, well claiming that they are unprecedented or
caused by global warming (whatever the truth of the matter), is not
get-out for the EA or whoever, as the global warming predictions are
for more and heavier rainfall in winter, which would effectively cause
the same outcomes, so they need to be planned and prepared for anyway.
The main difference with a summer event (as someone pointed out
elsewhere) is that the caravan parks on the flood plains tend to have
higher occupancy.
Cheers,
Adam
gerh...@aston.ac.uk
> rainfall records to be set? Warmer air, more moisture, thunderstorms,
> flooding. Perhaps we should be looking at changes in the occurrence of
> "thundery rain"
The record for the daily total happens to fall in August in England,
it's not hard to see why that shouldn't be beaten inspite of overall
drier summers.
Interestingly, two month combined totals aren't listed on the Met
Office historical records page
http://hadobs.metoffice.com/hadukp/charts/hadukp_ts_plots.html
They've got 11 regions and 12 months though, with each month having a
minimum rainfall record, a maximum rainfall record, and a daily
maximum rainfall record. That makes for 11*12*3 or 396 possible
historical records.
Add the seasons and the annual average for the 11 regions and the 11
possible consecutive two month combined totals, for minimal and
maximal rainfall and we have another 11*16*2 or 352 possible
historical records.
Ok, they are't entirely independent, but still. So, if June isn't a
record, well, let's see whether June + May are, or at least are for
England and Wales, or for Scotland, and if that still doesn't yield a
record, let's add April or the first half of July.
It isn't entirely surprising that this should yield records, even if
the underlying probability distribution has actually shifted slightly
to make a particular record (May/June combined total) less likely.
A cursory glance at the historical data seems to suggest a much
clearer (downward) signal for July than for May or June, though, so
there doesn't have to be a contradiction between drier summers and
wetter combined May/June totals.
AdamW
> Ok, they are't entirely independent, but still. So, if June isn't a
> record, well, let's see whether June + May are, or at least are for
> England and Wales, or for Scotland, and if that still doesn't yield a
> record, let's add April or the first half of July.
I would say that *generally* they will be quite closely linked. If
June isn't a record, but you get one by adding May, either May is
likely a record in its own right, or June is also very high, if not a
record. You won't get a record two month period without two high
months. Now, add in the fact that the values are calculated from spot
measurements, and the second highest may well have been the highest in
reality. Thus rather than just talking about record months, it is
often advisable to see the differences and rainfall pattern as well.
Also, the whole thing is complicated by the fact that there are two
types of rainfall (convective and dynamic), and spot values are
affected by speed of movement of the system (as the S Yorks floods
were).
IMHO it partly (as well as variations from the average over time)
comes down to if we continue to get 50, 100, 200 etc. year events, in
the same spots, within the short spans of time (currently within the
same decade). I might start to ponder the reasons. ;)
Tom Adams
Well, one would think that the insurance industry would be more on top
of
this than anyone. But they are struggling with it, so of course
governments
will be struggling too.
Perhaps it was always an illusion that the climate did not change from
decade to decade, but now few in responsible positions believe it.
The whole point is that anticipation is harder than it use to be.
gerh...@aston.ac.uk
> months.
There's another method to generate more records, reduce the length of
the data series:
http://news.bbc.co.uk/1/hi/uk/6270346.stm
"Last month was the wettest June the UK had seen since detailed
records began in 1914."
That's only a 93 year data series, rather than the 241 years available
for England and Wales (where the June record of 1860 still stands).
In Scotland the data series is much shorter (in the case of the Hadley
Centre data the series only starts in 1931) and funnily enough the
June record was set in 1938 and still stands (this year's total of 115
mm for Scotland is well short of the 167 mm in June 1938).
> Now, add in the fact that the values are calculated from spot
> measurements, and the second highest may well have been the highest in
> reality.
Yes, past records may have been overlooked due to poor spatial
coverage missing the highest rainfall areas.
> IMHO it partly (as well as variations from the average over time)
> comes down to if we continue to get 50, 100, 200 etc. year events, in
> the same spots, within the short spans of time (currently within the
> same decade). I might start to ponder the reasons. ;)
If 200 year events now regularly happen within 10 years, we've got a
very strong signal in the data making detection and attribution an
easier task. If we have thousands of data series, it's not that
surprising to find a few where two 200 year events happen within 7
years of each other.
I am happy enough with a well grounded theoretical prediction absent
such a strong signal.
I am also happy enough to accept that the average summer might get
drier, but the number of wet outliers simultaneously might go up
sufficiently to compensate. But is that actually predicted?
And I fail to see it in the historical data.
Maybe there's a strong signal for some measure of flooding. But I
haven't seen anybody present such data, and a breathless statement in
the Independent that the 1947 floods have now been exceeded in
magnitude, and those were said to have been 200 year floods, so that
we are now in unprecedented territory, doesn't quite cut it with me
(exceeded by what measure? That one of hundreds of possible river
flooding sites exceeded its previous high? That more homes got
flooded? More area? That it just feels worse???)
AdamW
> "Last month was the wettest June the UK had seen since detailed
> records began in 1914."
>
> That's only a 93 year data series, rather than the 241 years available
> for England and Wales (where the June record of 1860 still stands).
>
Some background here: http://www.climate-uk.com/page5.html
> > Now, add in the fact that the values are calculated from spot
> > measurements, and the second highest may well have been the highest in
> > reality.
>
> Yes, past records may have been overlooked due to poor spatial
> coverage missing the highest rainfall areas.
>
TBH I have no idea. I know that they very nearly lost the entire
rainfall series after a hardware change in the '80s/90s, and I did
some work on the move a newer, more robust, database in the early
'90s, but have had no connection with it since. Which is a shame
otherwise I'd have a direct access to the data and be able to talk to
the people responsible for the series.
> If 200 year events now regularly happen within 10 years, we've got a
> very strong signal in the data making detection and attribution an
> easier task. If we have thousands of data series, it's not that
> surprising to find a few where two 200 year events happen within 7
> years of each other.
True, but we are looking at regions here, rather than cities so the
numbers of series involved would not necessarily be that high. That
said I do take the point (in fact I think I made it) that it's still
too early to draw a conclusion.
>
> I am also happy enough to accept that the average summer might get
> drier, but the number of wet outliers simultaneously might go up
> sufficiently to compensate. But is that actually predicted?
Here's where I read it:
www.metoffice.gov.uk/research/hadleycentre/pubs/brochures/2005/clim_green/slide44.pdf
It's a little ambiguous and there are caveats there. I admit that I
should get round to tracking down the original paper...but then it'd
just get added to the "to be read" pile.
I think the UK impacts study of 2004 (forget it's proper title) also
discusses it.
gerh...@aston.ac.uk
Thanks a google search with the reference gave me this:
http://prudence.dmi.dk/public/publications/FreiEtAl_subm.pdf
And it's got what I was after, namely predictions for 1 day
precipitation totals in summer (it would be nice to have the 1 month
and 2 month totals as well, for comparison with this year and all, but
that's maybe a little too much to ask).
The colourful figures are on page 47, and they seem to indicate that
there is some disagreement between different models, but there'll
either be little change or some decline in 1 day precipitation events
with a present 50 year return period (your link also gave a nice map,
the only thing that was lacking was an explanation of what was meant
by "heavy rainfall events").
James Annan
> The whole point is that anticipation is harder than it use to be.
The future isn't what it used to be, that's for sure :-)
James
Timothy Chase
> Greenhouse gas emissions from human activity are causing global
> shifts in rainfall patterns and contributing to wetter weather over
> the UK, climate scientists say today.
>
> Their study is the first to find a "human fingerprint" in the
> rainfall changes which have been detected in a belt of the northern
> hemisphere stretching from the Mediterranean to the UK to Norway.
>
> <snip>
>
> This concerns a study done by the Hadley Centre of the UK Met Office
> and others that is published in the latest issue of Nature.
Beautiful...
I have been able to find newspaper articles, but I am looking for more
tech material.
Nature is better than Discover, even better than Scientific American!
All kidding aside, I trust the work coming out of Hadley.
gerh...@aston.ac.uk
> > shifts in rainfall patterns and contributing to wetter weather over
> > the UK, climate scientists say today.
It is rather frustrating to have access to stories about a paper, but
not the paper itself.
Now the Hadley Centre have got this graph on their site:
It seems to indicate that in fact annual rainfall over the UK will be
little changed (which I gather is because wetter winters get
compensated by drier summers).
In so far as flooding in the UK correlates with 5 day winter
precipitation totals with a present return period of 50 years, the
link I found earlier thanks to Adam would seem to give a helpful
quantification, namely and roughly reading off their figure, said
return period should fall from 50 years to between 30 and 45 years
depending on the exact location in the UK.
http://prudence.dmi.dk/public/publications/FreiEtAl_subm.pdf
I've got this niggly suspicion that the Nature paper is about
attribution and that attribution can be made primarily due to changed
rainfall patterns outside the UK, and it's got very little to say
directly about flooding in the UK.
And while the above paper does have something relevant to say, being 2
years old is maybe too old as far as the media is concerned and what
it does have to say about flooding is far too "unsensationalist" to
hang a story on.
James Annan
Authors are usually happy to send copies of the paper if asked politely
(and most journals explicitly endorse this sort of behaviour). People
like me might also be persuaded by a sly email :-)
FWIW the url works for me from home, which seems like a mistake. Have
you actually tried?
http://www.nature.com/nature/journal/v448/n7152/full/nature06025.html
But briefly, the "detection" is only in latitude bands, and only annual
average (because finer sampling is too noisy). It just talks about a
small but detectable increase in annual average in this latitude - I
think I already mentioned a figure of 60mm per century. It has no direct
relationship to any prediction of changes in summer extremes in the UK,
although arguably it marginally raises confidence that the models are
doing plausible things.
James