cc: "Klein Tank, Albert" <Albert.Klein.Tank@knmi.nl>
date: Wed, 24 Jan 2007 17:58:24 +0100
from: Geert Jan van Oldenborgh <oldenborgh@knmi.nl>
subject: Re: Precipitation trends statement IPCC 4AR SPM
to: P.Jones@uea.ac.uk

<x-flowed>
Dear Phil,

P.Jones@uea.ac.uk wrote:
>  Geert Jan,
>    The bullet points come from Fig 3.14 not
>  from the maps.  The time series show large area
>  averages. This is what the models give.

I thought the models gave gridded output (Fig SPM-6), and we are 
discussing observations here, not model output.

The boxes are not defined for someone just reading the SPM, without 
access to the full report that will be released much later.  A reader 
will interpret the statement that the Mediterranean became drier as 
meaning that all of the Mediterranean area became drier.  If you see on 
the map (not available to the reader of the SPM) that it only pertains 
to the southern half, why not make the more accurate statement that 
Northern Africa became drier?

>    To get every grid box in one of the regions
>  to all show the same sign of a trend is impossible.
>  We used the large regions to show the bigger picture
>  as I said earlier.

I am not complaining about not all grid boxes having the same sign.  I 
am complaining about the whole statement based one or two grid boxes in 
a large area, with the reader who is not yet immersed in Chapter 3 
interpreting this as a trend in the whole area, when in fact this is not 
the case.  Especially when the grid boxes have a much higher climatology 
this can easily happen.

I am also curious why you left out the big trends in southern South 
America (shown explicitly in Fig 3.14) and western Australia.  Is this 
because they are not in the model results?

>    Talk to Albert!!!!!!!!!!!!!

He told me to contact you.

	Geert Jan
</x-flowed>
