[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Model vs. Observed SST
- Subject: Model vs. Observed SST
- From: firstname.lastname@example.org (James Orr)
- Date: Wed, 13 May 1998 22:22:28 +0200 (CEST)
Just to clarify matters, all OCMIP-1 participants did indeed use the
same K (i.e., Kw*alpha), for which both alpha and Sc were determined
from the observed SST climatology (Levitus, 1982). On the other hand, all
participants (except perhaps Hadley) used the model SST to compute pw.
I agree with Niki Gruber that the same temperature field should be used for
determining both the solubility (alpha) and the Schmidt number (Sc).
I do not feel strongly about using observed or modeled temperature fields
for CO2 runs. This is important of course only if and where there are
substantial deviations between the two fields. It is less important for
CO2 than for CFC's as you mention. I would be happy to go with your
suggestion to use modeled temperatures throughout. Although that approach
would mean that each model would have its own unique 2-D field of K's, I
would hope that differences there would not be so important. For CO2, a
difference of 2O degrees C results in a Kw*alpha which is only about 7%
different. For CFC's, temperature sensitivity is greater, but still if
models differ by say 1 degree from the observations, I'd estimate the
difference in K (calculated from observed vs. model SST) to be of the
order of 5%. I'd hope that modeled vs. observed SST would not generally
surpass 1 degree C, considering that models are restored to observed SST.
To be sure to work out related differences that might come out later on,
we will need to ask all participating groups to supply their monthly
fields of temperature and K. The same fields were supplied for OCMIP-1.
In conclusion, if we are to use model temperatures thoughout, we will
simply provide each group with one 2-D field (a*(U^2 + V), and subroutines
to calculate the Sc and alpha for each tracer (CO2, CFC's, O2, etc.).
Jean-Claude has already produced that 2-D field as well as subroutines to
calculate Sc and alpha for the CFC's. After interpolation to the standard
1 degree grid, we will make them available by ftp, probably before the
week is out.
Best regards, Jim
James Orr ; Lab. des Sciences du Climat et de l'Environnement
mailto: email@example.com ; Unite Mixte de Recherche CEA-CNRS
Tel: (33) (0)1 69 08 77 23; Address: LSCE, CEA Saclay
Fax: (33) (0)1 69 08 77 16; Bat. 709 - Orme
Home: (33) (0)1 30 41 31 72; F-91191 Gif-sur-Yvette Cedex - FRANCE