Re: [ferret_users] Define axis/from_data

[vivek shilimkar <vivek.shilimkar@xxxxxxxxx>]

Thank you all for your help.

As pointed out by Martin Schmidt, the newly defined variable P does have four dimensions. and it is not suitable to define an axis.

And I also understand about the auto-conversion of units is not possible.

I am looking for way to convert the Z axis of CMIP5 model data, which is on model grid, to standard pressure level.

I found the ZAXREPLACE(V,ZVALS,ZAX) function in ferret  may have solution to my problem.

If you have any better and easier suggestion please let me know.

Regards,

Vivek

On Tue, Feb 26, 2019 at 3:16 AM Ansley C. Manke <ansley.b.manke@xxxxxxxx> wrote:

Also, regarding units.  Some measures of units are automatically converted when working with coordinates; for instance a variable with an on axis that has units of meters can be regridded to an axis with units of meters and Ferret/PyFerret will handle the conversion. And in the horizontal directions, longitude/latitude are handled with areas computed on the globe, and calendar time is converted whether the units are seconds, days, etc.  Other units are allowed but would not be converted.  So that's all that this message about units is telling us:

 *** NOTE: (fyi) units are not subject to auto-conversion: PA

Ansley

On 2/22/2019 9:52 AM, Martin Schmidt wrote:

Dear all,

I think b*ps inherits all dimensions:

 B        vertical coordinate formul          ...           ...       1:95      ...

 PS       Surface Air Pressure              1:192     1:96      ...       1:120

So, p has dimension 192 x 96 x 95 x 120 and is not suitable to define an axis.

It should be something like:

let zval = z[gz=B]
define axis/z/depth/unit="Pa" pax=zval
(depth means posive down, hopefully :-) not sure)

Best,
Martin

Am 22.02.2019 um 18:05 schrieb Satyesh Ghetiya:

Hi,

Following may give you some hints :

(1) Is pressure unit allowed ?  (I think 'K' means only depth is allowed in distance unit )

(2)  In _expression_ "let P = ap+(b*ps)" .  b*ps is 'missing data' everywhere. Because there are no common valid points in 3 D which can give answer. (missing data * good data = "" data)

^ This may not solve problem , but will give you path to solve.

Regards,

Satyesh

On Fri, Feb 22, 2019 at 9:32 AM vivek shilimkar <vivek.shilimkar@xxxxxxxxx> wrote:

Dear all,

I have been trying to define new vertical axis for my data.

However, I am getting following error in ferret.

currently SET data sets:

    1> ./cl.nc  (default)

 name     title                                          I            J         K         L

 AP       vertical coordinate formul         ...          ...       1:95      ... 

 B        vertical coordinate formul          ...           ...       1:95      ...

 PS       Surface Air Pressure              1:192     1:96      ...       1:120

 AP_BNDS  vertical coordinate formul  1:2       ...         1:95      ...

 B_BNDS   vertical coordinate formul    1:2       ...         1:95      ...

 CL       Cloud Area Fraction                 1:192     1:96    1:95      1:120

let P = ap+(b*ps)

define axis/from_data/z/unit="Pa" pressure_axis=P[k=1:95]

           *** NOTE: (fyi) units are not subject to auto-conversion: PA

**ERROR: improper grid or axis definition: data for DEFINE AXIS/FROM_VARIABLE is ambiguous

define axis/from_data/z/unit="Pa" pressure_axis=P[k=1:95]

Command file, command group, or REPEAT execution aborted

Am I doing it right way? Or is it not possible to define axis this way?

Regards,

Vivek

--

G Satyesh