date: Tue, 10 Nov 2009 08:47:42 -0700
from: Tom Wigley <wigley@ucar.edu>
subject: Re: CC Essay
to: Phil Jones <p.jones@uea.ac.uk>

<x-flowed>
Ah ha --- thanks.

Tom.

++++++++++++

Phil Jones wrote:
> 
>  Tom,
>    This was related to whether the CC Essay should say anything about 
> global T estimated from Reanalyses.
>  I did send you a paper while ago about ERA-INTERIM from 1989-2008.
>    The extended reanalysis (by Gil Compo) is from 1891-2005. It uses 
> only SST and surface pressure as input.
>    Bottom line is this extended reanalysis (20CRv2) gets about the same 
> trend as HadCRUT3 for the 105 year period.  The model gets the SSTs, but 
> also changes in forcing due to CO2, volcanoes and the sun. It uses the 
> SST that NCEP got. So it isn't using the newer SST dataset that Jim 
> Hurrell et al talked about. It is, though, the extended reanalysis that 
> Jim talked about at the end of his paper in J. Climate in 2008 - 
> attached this.
> 
>  Cheers
>  Phil
> 
> 
> At 14:29 10/11/2009, you wrote:
>> Phil,
>>
>> I can't remember what this is in response to.
>> Can you remind me?
>>
>> Tom.
>>
>> ++++++++++++++++++++++
>>
>> Phil Jones wrote:
>>>
>>>>  Tom,
>>>      I've got some info back from Gil Compo. I can't pass on though, 
>>> but I can summarize.
>>>  I've given him loads of suggestions to track down some of the 
>>> differences. As large-zonal-band time series none are that large, but 
>>> he needs to look at maps.
>>>    On trends though for 50S to 70N (for land) his extended reanalyses 
>>> are very similar to HadCRUT3 and to GISS. For some reason he's yet to 
>>> use the NCDC series.
>>>     Correlations between his series (20CRv2) and HadCRUT3 for this 
>>> large domain is 0.85 and for GISS it is 0.81.   GISS/HadCRUT3 
>>> correlate (all over 1891-2005) to 0.98.   The correlation reduce to 
>>> 0.69, 0.60 and 0.95 if you detrend them.  They'd also drop a little 
>>> further (by about 0.02) if you looked at 20-70N.  Correlations would 
>>> be improved if you could drop the 1890s.
>>>    Despite always being pressure only they give better agreement from 
>>> the 1960s onwards. I suspect this is just down to more input surface 
>>> pressure data then.
>>>    The odd periods are the 1920s and the period from 1940-60. For the 
>>> latter if the SSTs were adjusted they would look much better. The 
>>> 1900s, 1910s for some reason look amazingly good.
>>>   It looks as though there won't be a paper to refer before we have 
>>> to submit the CC piece. I've asked him to plot land minus ocean when 
>>> he has some time.
>>>  Cheers
>>>  Phil
>>>
>>> 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.jones@uea.ac.uk
>>> NR4 7TJ
>>> UK 
>>> ---------------------------------------------------------------------------- 
>>>
>>
> 
> 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.jones@uea.ac.uk
> NR4 7TJ
> UK 
> ----------------------------------------------------------------------------                                                                                 
> 


</x-flowed>
