[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: minor corrections


I would agree that sea level atmospheric pressure need not be included in
the CFC protocols, but for additional reasons.

Your arguments seem appropriate for the vertical perspective for "natural" 
tracers.  Supplemental reasons can be given when considering horizontal
variability. Horizontal temperature variation has a much larger effect on
the gas transfer coefficient for CFC's than for CO2.  For CO2, the
temperature variation in the gas transfer coefficient's solubility and
Schmidt number terms almost cancel one another out; such is not so for
CFC's.  Variability in CFC concentrations is much larger due to changes in
temperature T than due to changes in atmospheric pressure P at sea level. 
For CFC's, modelers need to take into account temperature driven
variability but can live with a 3% error due to changes in P. 

Furthermore, one must consider anthropogenic tracers.  In a 1-D vertical
sense, the dynamic range for CFC's is much larger than for anthropogenic
CO2 and post-nuclear C-14 simply because measurements can be made with
much greater precision, not because of a relatively greater recent
increase in the atmosphere.  So vertical reasoning won't work here; 
however, horizontal arguments given above still seem appropriate.

- -- Jim

   James Orr                       Laboratoire des Sciences du Climat 
   LSCE, CEA Saclay                             et de l'Environnement    
   Bat. 709 - Orme                 Unite Mixte de Recherche CEA-CNRS    
   F-91191 Gif-sur-Yvette Cedex    Phone:   (33) (0)1 69 08 77 23      
   FRANCE                          Fax:     (33) (0)1 69 08 77 16
                                             |    |     
   mailto:  orr@cea.fr            (country code)  (when in France)      

On Tue, 23 Jun 1998, Ray Najjar wrote:

> Oceanographers can now routinely measure pCO2 in surface waters to within a few
> microatmospheres, DIC to a few umol/kg and oxygen with even better precision.
> I have been using the guideline that if a process affects something by more
> than typical measurement precision, we should include it in OCMIP simulations.
> The reason is that we want to be able to *evaluate* our models with
> measurements, not just intercompare the models.
> Sea level pressure varies from about 1020 mb at 30S to about 990 mb at 60S,
> a 3% difference.  This translates into a pCO2 difference of about 10 uatm and
> a corresponding DIC difference of about 6 umol/kg  (using a buffer factor of
> 10 and a mean surface DIC of 2000 umol/kg.  The O2 differnce would be
> about 10 umol/kg.  These adjustments are just large enough, I think, that
> pressure effects should be included in O2 and CO2 gas transfer. 
> For CFCs the pressure effect is effectively much smaller because of the large
> dynamic range in CFC concentrations due to the rapid increase from zero in the
> past few decades.
> With regard to the carbonate chemistry, it is important to include phosphate
> and silicate equilibria because it will impact pCO2 by more than 5 uatm, as
> Chris and I noted in our protocol description and as Bob reiterated in a
> recent email.
> These modifications are not going to make or break OCMIP, but they are not
> trivial, they are systematic corrections, and are easily included in
> calculations, so let's do them.
> One could argue that cool-skin corrections and bubble-induced supersaturation
> should be included as well, but the problem is that these effects are not well
> characterized and easy to model.  So I suggest we not include these effects
> until good parameterizations are developed.
> Ray