[Thread Prev][Thread Next][Index]

Re: masking of land data



On Wed, 2 Mar 2005, Yogesh K. Tiwari wrote:

> How to mask the land  data to see the monthly
> variation(time series) at only ocean and vice versa.

Hi Yogesh,
	What you need to do is to access one of the bathymetry files
and use it to define a land/sea mask ON THE SAME GRID as your dataset.
Try the following ... good luck,
Mick


!-------- demo of land/sea mask generation -----------

! FERRET SESSION 1 - make a fake dataset in a lat-lon region
def axis/x=100e:160e:1/units=longitude xax
def axis/y=50s:10s:1/units=latitude yax
def grid/x=xax/y=yax grd
let v=randu(x[g=grd]+y[g=grd])
save/file=mydata.nc v
quit                      ! end the ferret session

! FERRET SESSION 2
! 1) read in the datafile (with no mask of its own) ...
use mydata.nc
shade v

! 2) pick a bathymetry file (a bit finer than your data grid)
use etopo20

! 3) see how it looks ...
shade/o/pal=grey if(rose ge 0)then 1

! 4) give a name to the grid of the variable you want to examine
def grid/like=v[d=1] grd

! 5) regrid "rose" to the grid of your variable and check it out
let myrose=rose[d=2,g=grd]
shade/o/pal=black if(myrose ge 0)then 1

! 6) define the mask and use it to define vland ...
let land=if(myrose ge 0)then 1 else 0
let vland=if(land eq 1)then v[d=1]
shade vland

! 7) ... and vsea
let vsea=if(land eq 0)then v[d=1]
shade vsea





[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement