date: Tue Dec 7 16:51:48 2004
from: Phil Jones <p.jonesatXYZxyz.ac.uk>
subject: Re: Section 3.6
Your null hypothesis (sort of only expect trends since 1970) has come
through strongly in your comments
on 3.9. If you want to go this way, then this needs to be upfront
somewhere - a reason for the trends.
This isn't the way any skeptics think, nor many reasonable
climatologists. I can see many arguments for it and a few against.
Problem is we have the long series. I know you're not saying
ignore all before 1970, but I can see it getting interpreted this way.
The UK and NZ surrounded by water will likely show this, but I would
expect that most other regions would also show this if the analyses were done.
For Holland, S and SE flow comes from the continent, so it isn't just
due to oceans all around.
Worth mentioning the 2 papers, Jim.
At 15:24 07/12/2004, Kevin Trenberth wrote:
On 3.6: Jim Hurrell has been working on impacts for NAO stuff. Should be
Phil some reactions below.
On Tue, 7 Dec 2004, Phil Jones wrote:
> Jim and Kevin,
> At least some of the sentiments of the paragraph will still be there.
> The main point I
> was trying to make is that there is variability in the
> relationships/influences over time. I'll
> omit the para from 3.9.
> Showing the various indices over as long periods as possible will be
> very useful as
> will correlations over different periods. If you just put a marker down
> for this in the ZOD,
> I can get onto it next year.
I agree but I also think we need to think clearly about the null hypothesis.
Viz, natural variablity dominates till about 1970 and only since then is
there a reason for trends. We can help clarify matters if we express things
5 indices should all go on one A4 page and
> still be clear - could
> be just on one column of the 2 column printed page and still be visible.
I think this will have to be FOD nor ZOD. Agree we should try to tidy up but
for now lets get place holders.
> We will need to
> iterate a bit on this to make sure we have all the indices we want and as
> many of the
> versions as possible (within reason, a cutoff could be only those updated
> in near real time).
> I'm not disputing that the SOI and NAO (for example) explain a lot of
> variability and do
> themselves vary a lot, but over the longest of periods (from the mid-19th
> century) neither
> has much long-term trend. If either do, it is only since the 1970s and
> maybe this is the
> point we should be making.
Yes, see above.
If you accept that Nino3.4 can be produced
> back to the
> mid-19th century, it will have a larger long-term trend than SOI.
> Circulation does change,
> but as the ocean (and air) warms advection brings warmer air.
N3.4 is not the perfect ENSO index precisely because of climate change. We
are experience consequences right now where NOAA has decalared an El Nino
based on N3.4 but no change in convection to move east of dateline and no
ENSO: it is not yet hanging together. In part I think this relates to
warming and competition from Indian ocean etc. Mind you i suspect it will
happen still. So point here is no index is perfect. SOI is prob better for
Interannual variability but may not do trend part well.
> Over the very long-term (over the UK and also Holland from the
> late-18th century), the
> warming that has occurred is not explained by more preferential advection
> from some
> directions. Air from all 8 points of the compass (with the exception of
> NE) shows significant
> increases in temperature on days now than similar days 200 years ago.
> Even the NE shows an
> increase, but it isn't significant because it doesn't occur too often, so
> samples are small. The
> long-term warming is spread around the compass roughly equally. Mild
> winters still require +ve NAO
> values, but they are warmer now than they were in the 1910s/1920s when
> the NAO had almost
> equally positive values.
Doesn't this partly reflect the strong maritime influence: UK temps, like NZ
ones, are similar to SSTs in surrounding waters and thus all directions flow
over those waters. Is that too simple an explanation?
> At 01:29 07/12/2004, Jim Renwick wrote:
> >Hi Phil:
> >Thanks for your paragraph - I've incorporated a modified version of it in
> >But, I wonder about
> > > This can only be true, up to a point: as the mass of the atmosphere cannot
> > > change, only be redistributed; the long-term trend of all atmospheric-based
> > > indices must tend to zero (as boundary conditions have been similar
> > over the
> > > last few millennia). On century timescales, therefore, circulation indices
> > > can hardly explain any of the long-term trends in temperature and
> > > precipitation, as they should have little trend.
> >I'm sure you're right on a globally-integrated basis, but regionally, I'm
> >not so
> >sure. Changed teleconnections can alter the regional energy distributions, so
> >surely could contribute to long term trends? If GHG increase acts to
> >modify the
> >meridional temperature gradient on long time scales (centuries) then we
> >might see
> >multi-century trends in the zonal winds and hence the annular modes. Would
> >not be related to century-scale trends in surface climate? Also, should
> >change imply changed tropical diabatic heating (more El Nino, say), then
> >we would
> >get different (more frequent or stronger) extratropical Rossby wave
> >forcing, so a
> >change in the nature of the such things as the PNA (say), leading to
> >changes in storm tracks and regional precip etc.
> > Jim Renwick, Climate Variability
> > National Institute of Water & Atmospheric Research
> > Private Bag 14901, Wellington, NEW ZEALAND
> > J.RenwickatXYZxyza.co.nz Ph: +64 4 3860343 hm: 9732523
> > http://www.niwa.co.nz Fax: +64 4 3862153 or 3860574
> Prof. Phil Jones
> Climatic Research Unit Telephone +44 (0) 1603 592090
> School of Environmental Sciences Fax +44 (0) 1603 507784
> University of East Anglia
> Norwich Email p.jonesatXYZxyz.ac.uk
> NR4 7TJ
Kevin E. Trenberth e-mail: trenbertatXYZxyzr.edu
Climate Analysis Section, NCAR www.cgd.ucar.edu/cas/
P. O. Box 3000, (303) 497 1318
Boulder, CO 80307 (303) 497 1333 (fax)
Street address: 1850 Table Mesa Drive, Boulder, CO 80303
Prof. Phil Jones
Climatic Research Unit Telephone +44 (0) 1603 592090
School of Environmental Sciences Fax +44 (0) 1603 507784
University of East Anglia
Norwich Email p.jonesatXYZxyz.ac.uk