[Global Change: 3786] New multiple regression analysis of climate change

[Robert I Ellison]

This is a link to Judith Lean's recent paper on multiple regression
analyisis of major modes of climate variability. She uses ENSO,
anthropogenic influences, solar variability and volcanic aerosols.
Judith Lean's methodology is given in a 2007 paper (reference 22) also
freely available.

http://www3.interscience.wiley.com/cgi-bin/fulltext/123222295/PDFSTART

I quite like Judith Lean's paper - I think it is a model of clarity.
I think the methodology is hugely promising. Simple, robust,
refreshingly empirical and captures a major aspect of nonlinear
climate variability through ENSO. And yes I do note that she
concludes that anthropogenic influences are predominant.

I don't think she is wrong as such - although the forward projection
is dodgy enough to be amusing. It does show convincingly the
problematic nature of climate projections. We should anticipate the
cooling influence of ENSO on this decade's temperatures to continue
for a while yet. Our abilities to predict volcanos decades in advace
is obviously not huge at this stage. Solar variability should be
projected as probabilities of change rather than as a copy of cycle
23.

I do think there is another obvious influence on climate that should
be included in the multiple regression analysis. See the discussion
on clouds at Professor Ole Humlum's site -

http://www.climate4you.com/ClimateAndClouds.htm

Cloud cover decreased by about 5% from the mid 1980's to 2000 and has
increased about 3% since on ISCCP-FD data. That is about 0.3 degrees
net warming to 2000 and 0.18 degrees net cooling since. A major
influence by any estimation. This is more likely to be largely an
ENSO feedback rather than a 'global warming feedback' - but
nonetheless should be included as a factor in the regression
analysis.

Cheers
Robert


--
You received this message because you are subscribed to the Google Groups Global Change ("globalchange") newsgroup. Global Change is a public, moderated venue for discussion of science, technology, economics and policy dimensions of global environmental change.

Posts will be admitted to the list if and only if any moderator finds the submission to be constructive and/or interesting, on topic, and not gratuitously rude.

To post to this group, send email to global...@googlegroups.com

To unsubscribe from this group, send email to globalchange...@googlegroups.com

For more options, visit this group at http://groups.google.com/group/globalchange

[David B. Benson]

On Apr 25, 4:09 pm, Robert I Ellison <rob...@robertellison.com.au>
wrote:

> This is a link to Judith Lean's recent paper on multiple regression
> analyisis of major modes of climate variability. She uses ENSO,
> anthropogenic influences, solar variability and volcanic aerosols.
> Judith Lean's methodology is given in a 2007 paper (reference 22) also
> freely available.
>
> http://www3.interscience.wiley.com/cgi-bin/fulltext/123222295/PDFSTART

She doesn't reference the work by Professor Tung
and co-workers on temperature changes over the
average solar cycle. The papers may be obtained
from Professor Tung's web site.

Otherwise, certainly a fine contribution.