date: Tue, 29 Apr 1997 09:51:23 +0100
from: Nick Brooks <n.brooks@uea.ac.uk>
subject: Re: rainfall - temp analysis
to: Mike Hulme <m.hulme@uea.ac.uk>

Mike

Thanks for your reply. Comments on rainfall anomalies noted. Padding at the
extremities of the curves is something I'm addressing: the IDL routines
automatically "fill in" at the ends - I'll look at the maths of it and
think about which year to use as the start of each curve as we discussed in
the last meeting. 

As for the requirements for the meeting/report, by "draft chapter outline"
I presume you mean a draft structure of the whole thesis, as opposed to a
single draft chapter?

I've mailed Mick and Steve, and I'll let you know when I hear from them -
both your dates are OK for me, but 16 May would have to be after 12 noon.

Cheers

Nick

Climatic Research Unit 
School of Environmental Sciences
University of East Anglia
Norwich NR4 7TJ
U.K.

Direct line  +44 1603 592702
Office        +44 1603 592722
Fax           +44 1603 507784
E-mail       n.brooks@uea.ac.uk

what's possible?  For this meeting we should have from you (a day or
> two ahead) a fairly firmed up work plan for the next 12 months, a draft
> chapter outline for the PhD, your latest thoughts about the modelling,
and
> any other progress you wish to discuss.  Before the meeting I would like
to
> see a redraft of the GFDL paper.
> 
> Cheers,
> 
> Mike
> 
> 
>  16:56 28/04/97 +0100, you wrote:
> >Mike
> >
> >Another question to do with the pertinence of my GFDL analyses. Now that
I
> >have all my data in standard anomaly form, I am considering the Sahel
> >rainfall / global temperature correlations. As the correlations will be
> >unaffected by the units, I don't need to redo the analysis to take
account
> >of the fact that the rainfall is now in standardised anomaly form.
However,
> >as it stands the correlatations are with annual, rather than seasonal
mean
> >temperatures. My instinct is to keep it like this, bearing in mind the
fact
> >that (i) we don't know much about the rainfall - temperature
relationships
> >apart from on a broad spatial scale, and know nothing at all about the
> >intra-annual variations in these relationships in the model, and (ii)
the
> >seasonality of the rainfall in the model is not realistic - so
identifying
> >a season in which temperature anomalies are important for a (poorly
> >modelled) monsoon phenomenon would be difficult. Obviously we'll know
more
> >if a seasonally based analysis is done, but then identifying a season of
> >particular importance would involve quite a bit of analysis itself -
maybe
> >this will be worth doing later but I think not now for an initial
> >publication, when a first analysis should suffice.
> >
> >What do you think?
> >
> >Nick
> >
> >Climatic Research Unit 
> >School of Environmental Sciences
> >University of East Anglia
> >Norwich NR4 7TJ
> >U.K.
> >
> >Direct line  +44 1603 592702
> >Office        +44 1603 592722
> >Fax           +44 1603 507784
> >E-mail       n.brooks@uea.ac.uk
> >
