Re: [ferret_users] Conservative remapping

[Russ Fiedler <russell.fiedler@xxxxxxxx>]

Hi,

Shouldn't the line

let npp_new     = npp[G=reg_grid]

read


let npp_new     = npp_partial[G=reg_grid]




Also, you might want to look at the @AVE transformation which does area
weighted averaging rather than linear interpolation. 

Russ


On Fri, 2012-05-25 at 07:06 +1000, Lev Tarasov wrote:
> Markus;
>        ferret's an awkward tool for conservative remapping. It would
> entail  just a few lines of commands with CDO
> https://code.zmaw.de/projects/cdo
> (or maybe also with NCO, choice depends on who you have around that's 
> familiar with either tool)
> 
> Lev Tarasov -   Dept of Physics and Physical Oceanography,
>  		Memorial University of Newfoundland.
>                  email: lev@xxxxxx
>                  http://www.physics.mun.ca/~lev/
>  		Tel (709)-864-2675
>  		Fax (709)-864-8739
> 
> On Tue, 22 May 2012, markus.mingel@xxxxxxxxx wrote:
> 
> > Dear All,
> >
> > I would regrid a variable defined in gC/m2 on a regular grid to a new regular
> > grid; since the variable is a flux, I should use a mass conservation. Iin order
> > to have a variable independent by the cell area I tried to multiply the
> > variable by the area before the regridding, and then I divided by the new area
> > after the regridding, however when I check the global value on the new grid
> > this is not the expected value.
> > Here a sample code:
> >
> > USE npp.nc
> >
> > DEFINE AXIS/X=-179.5:179.5:1 xaxis
> > DEFINE AXIS/Y=-89.5:89.5:1   yaxis
> > DEFINE GRID/X=xaxis/Y=yaxis  reg_grid
> >
> > let dummy = y[gy=yaxis] + x[gx=xaxis]
> >
> > ! Computing the area of the new grid
> >
> > LET r=6370*1000		! radius of earth in m
> > LET pi=atan(1.)*4.
> > LET deg2rad=pi/180.
> > LET deg2km=1000*2*pi*r/360.
> >
> > LET Final_area=xbox[g=dummy]*ybox[g=dummy]*cos(y[g=dummy]*deg2rad)
> > *deg2km*deg2km*1e-6
> > LET Original_area=xbox[g=npp]*ybox[g=npp]*cos(y[g=npp]*deg2rad)
> > *deg2km*deg2km*1e-6
> >
> > let npp_partial  = npp * Original_area
> > let npp_new     = npp[G=reg_grid]
> > let npp_final     = npp_new /  Final_area
> >
> > The global value of npp is 106 Pg, while  npp_final has a much lower value.
> > Any suggestions?
> >
> > Best
> > Markus
> >
> 
> This electronic communication is governed by the terms and conditions at
> http://www.mun.ca/cc/policies/electronic_communications_disclaimer_2012.php