[Thread Prev][Thread Next][Index]

Re: Sigma to pressure level



Hi Jagadish,
Welcome to Ferret!
Your commands for ZAXREPLACE  look correct.  I have just a few
comments:

In the DEFINE AXIS command we normally use /Z=lo:hi:delta
The /DEPTH qualifier is the element of the command that tells Ferret
that the axis is positive downwards.  Ferret will take the arguments
/Z=1000:0:25 and use them as if you had said /Z=0:1000:25.

The expression tk*0.0 is a variable with the same grid as tk but
with value 0.  So adding the term tk*0.0 in the definition of variable
pk forces pk to have the same shape and size as tk.  This topic is
discussed in the Users Guide in Chapter 3, section 2.2 "Multi-dimensional
Expressions"  which you can find by looking up "conformability" in the
index.

I don't know quite what you mean in your question about the averaging.
Do you want to take a collection of x-y locations and find the average of a
variable at those points? (If so then the term would be points rather than
grids.)  You can use the SAMPLEXY function to find the value of a variable
at a set of points:

!   First define the x and y locations as lists of values (constant arrays)
!   Variable xpts is all of the x coordinates, ypts is the y coordinates

yes? let xpts = {37, 37, 38, 38}
yes? let ypts = {36, 37, 35, 36}

!  SAMPLEXY the variable "var" at the locations defined by xpts and ypts
!  The output of SAMPLEXY is a list on an abstract X axis, one for
!  each element of xpts and ypts.  You can average this result in X to
!  get the average over the list of points.

yes? let  varpts = SAMPLEXY(var, xpts, ypts)
yes? let var_avg = varpts[x=@ave]

var_avg will have the same grid in Z and T as the input variable var.

Ansley Manke

jagadish karmacharya wrote:

Dear ferret users,

I have started using ferret recently. I am facing couple of problem in using the software.

1) I have data is sigma coordiante with the variables in listed below. I want to present my result (varaibles) in pressure level. for that purpose i have used following command

yes? use ATMpentd_91AVG.CDF

yes? define axis/Z=1000:0:25/unit=millibar/depth plev
> yes? let ptop=100 ! as my model top is 100mb
> yes? let pk=tk*0.0+z*(ps-ptop)+ptop ! i don't know why one has to put tk*0.0 in the equation because it's value is zero anyway; there is blank display if i omit tk*0.0 in the equation
> yes? let h_on_p =ZAXREPLACE(hgt,pk,z[gz=plev])
> yes? shade/l=1 h_on_p[z=500] ! gives the hgt( geo. pot. ht.) for 500 mb level
> yes? let rh_on_p =ZAXREPLACE(rh,pk,z[gz=plev]) ! similarly for other variable say rh
> yes? list/l=1 rh_on_p[z=850] ! list the rh( rel. humidity) for 500 mb level

The output with this set of commands seems to be ok but i am not certain. can somebod verify if this is the correct set of command to get output in pressure levels with original file in sigma coordiante.

2 Next I want to get arial average of a number of neighbouring grids (however they don't form a rectangle) for example say grids having (I,J) values (37,36),(37,37),(38,35),(38,36). what will be the best method?

thanking in advance for your help

jagadish

_____________________________________________________

yes? show data

currently SET data sets:

1> ./ATMpentd_91AVG.CDF (default)

name       title                         I              J       K        L

TK   Temperature              1:63        1:66   1:23     1:30

RH   Relative Humidity     1:63        1:66   1:23     1:30

HGT Geopotentail Height 1:63       1:66    1:23    1:30

.......etc

 


Do you Yahoo!?
Read only the mail you want - Yahoo! Mail SpamGuard.

[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement