Land Sea variable

[Douglas Clark]

Sajjad,

To be able to create a land/sea mask from model output, your output needs to include a variable that is only defined at land points (or takes on a distinct range of values at land points, so that you can change it to GrADS undef at land points).

In the CRU example, precipitation is only defined at land points, so that works. In your model output, you might be able to use a variable such as soil moisture since this is (presumably) not set at ocean points.

Most models can include the land/sea mask in the output.

If none of the above are available, you might have to write a new land/sea mask file (using FORTRAN, C or the likes) for use with GrADS.

Doug


>>> sajjad_met at HOTMAIL.COM 13/05/2007 15:13 >>>
Dear all

I am interested to mask the sea values from the output results of model. I
am using the LAMCON map projection in my model simulations. One of my friend
suggested to use a CRU data file to solve the problem as

Open a CRU file in grads
define pmask=const(pre,0,-a)
define pmask=maskout(pmask,pmask-1)

than open the model out put file in grads and

d var*pmask

I did this but there is a dimension error. This dimension error is because
the CRU is on regular longititude and Latitude grid where as model is using
LAMCON projection.

My question is how I can create a land sea mask variable using the model
output file or if you can help me how to solve this problem

Many thanks in advance

sajjad




>From: Douglas Clark <dbcl at CEH.AC.UK>
>Reply-To: GRADSUSR at LIST.CINECA.IT
>To: GRADSUSR at LIST.CINECA.IT
>Subject: Re: how to do averaging over land
>Date: Thu, 10 May 2007 10:42:29 +0100
>
>There is a 'direct' way to do this within GrADS:
>
>1. Create a land/sea mask variable (1 at land,0 sea)
>If you don't already have this, you could create a new GrADS file, using
>the same grid as in your existing file.
>
>2. Use aave and maskout to calculate area average at land points.
>e.g.
>set x 1
>set y 1
>set t 1 12
>define a1=aave(maskout(val1,land-0.5),lon=0,lon=90,lat=0,lat=30)
>will calculate the area average of the variable val1 at land points and
>store it in variable a1.
>maskout is used here to remove sea points (where land-0.5 is <0)
>This can be done with the time dimension varying, so you get a time series
>as the result.
>
>Doug
>
> >>> stegert at IFM.UNI-HAMBURG.DE 10/05/2007 10:03 >>>
>Dear,
>
>I don't know whether there is a 'direct' way within GrADS.
>Anyway, I calculated average values at certain ocean areas.
>
>I assume you have a certain grid of say 180x120 cells (i.e.
>1'x1' per cell for your area).
>
>1) You create a land-sea-mask within an ASCII-file
>(landsea.dat) containing a 180x120 matrix of zeros and ones
>representing land and sea, respectively.
>
>2) open your grads-file and do a loop through the map (assumably
>a 2d dataset!?)
>var = 0
>inc = 0
>j   = 120
>while (j>0)
>   'set y 'j
>   lsmask = read(landsea.dat)  * opens the dat-file
>   lsline = sublin(lsmask,2)   * read line
>   i = 1
>   while (i<181)
>     'set x 'i
>     lsval = subwrd(lsline,i)  * read value
>     if (lsval = 1)            * only when on land
>       'd parameter'           * read value from grads
>       var = var + subwrd(result,4)
>       inc = inc + 1
>     endif
>     i = i+1
>   endwhile
>   j = j-1
>endwhile
>st = close(landsea.dat)
>
>With 'var' you have the total amount over land.
>With 'var/inc' you'll get the average over land.
>Mind: In the .dat the 10°N/100°E is in the last
>line (that's why to count j from 120 to 0).
>
>You can also put this within a temporal loop to
>read different time steps as well as you can define
>different areas (defined by 0,1,2,...) and with
>elseif (lsval = 2) var2 = var2+subwrd(result,4); inc2 = inc2+1; endif
>elseif (lsval = 3) var3 = var3+subwrd(result,4); inc3 = inc3+1; endif
>etc.
>
>Hope, this helps.
>Christoph
>
>praise praise wrote:
> >
> > Dear frineds,
> >
> > My problem is to construct a time series representing an area average of
> > a parameter, say temperature, over the area say (100-130E, 10-30N).  But
> > I am only interested in the average over land.
> >
> > Thus, do anyone know how to define the variable representing an areal
> > average over land only?
> >
> > Thanks
> >
> > MC
>
>
>--
>Christoph Stegert
>Institute for Oceanography - Ecological Modelling group
>University of Hamburg, GER - (ZMK/ZMAW alliance member)
>Bundesstr.53 20146 Hamburg - +49-40/42838-7486 Room 348
>
>
>--
>This message (and any attachments) is for the recipient only. NERC
>is subject to the Freedom of Information Act 2000 and the contents
>of this email and any reply you make may be disclosed by NERC unless
>it is exempt from release under the Act. Any material supplied to
>NERC may be stored in an electronic records management system.

_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.com/

--
This message (and any attachments) is for the recipient only. NERC
is subject to the Freedom of Information Act 2000 and the contents
of this email and any reply you make may be disclosed by NERC unless
it is exempt from release under the Act. Any material supplied to
NERC may be stored in an electronic records management system.