[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #COD-764409]: GrADS ctl with Height (in km) as vertical scale

Hi Stefan-
 
> What you suggest is ideal. Another idea I have is similar to how you 
> delineate hPa from Pa using an if statement. What if you have another if 
> statement that says if values < 30 (or some other #), then the ZDEF has units 
> of km and to use height coordinates while opening it up? The only issue that 
> will arise with this is if people have grads files for stratospheric 
> purposes, with pressures between 50 and 10 hPa. But I am not sure how many 
> people you will encounter with this problem! Let me know if there is anything 
> I can do on my end.

Right now, if the level values are less than 10, it sets the units to be "" 
which probably comes out as null, but should come out as unitless.  I don't 
have a dataset to test this with but I'm sure I did that because I found some 
dataset that just had level numbers.  So, I'm reluctant to change that to km 
because I'm sure I did it for a reason and the proposed solution (which I coded 
and tested this afternoon) would all the user to specify whatever units they 
want for the vertical dimension.

I'll work on getting this into the IDV, but it still might take a while because 
the netCDF-Java library is in flux.

Don

> On Feb 22, 2012, at 3:22 PM, Unidata IDV Support wrote:
> 
> > Hi Stefan-
> >
> >> GrADS is able to open up the ctl/dat pair just fine using height 
> >> coordinates. ZDEF can be either in height or pressure for GrADS ctl files, 
> >> even though it typically is in pressure. I think the IDV sees the ctl file 
> >> having ZDEF as only having pressure as the only possible vertical 
> >> coordinate, while it is possible it can contain heights. So, is there 
> >> anyway to add this possibility into how the IDV opens up GrADS ctl files?
> >
> > As you note there is nothing in the Grads Data Descriptor File (e.g. ctl 
> > file) that specifies the unit of the vertical dimension.  When I wrote the 
> > GrADS IOSP, I ran into 3 types of vertical level units - hPa, Pa and none 
> > (just a level number).  The default is hPa, but if there are values > 1050, 
> > I switch to Pa.
> >
> > Since GrADS is pretty loosey goosey about what can go in a ctl file, one 
> > thing we could do is use the attribute metadata 
> > (http://www.iges.org/grads/gadoc/descriptorfile.html#ATTR) to specify the 
> > units of the vertical dimension like so:
> >
> > @ zdef String units meter
> >
> > What do you think about this?
> >
> > Could you upload you file to the IDV uploads area so I can have something 
> > to test with?  It will probably take a while before this can get into the 
> > IDV, but I'll try to work on it when I get a chance.
> >
> > Don Murray
> >
> >
> >> On Feb 22, 2012, at 11:49 AM, Unidata IDV Support wrote:
> >>
> >>> Hi Stefan.
> >>>
> >>> First off, sorry for the delayed response. We have been attending 
> >>> conferences and giving workshops.
> >>>
> >>> We believe the GrADS spec is not being adhered to. As far as we can tell, 
> >>> GrADS expects levels to be defined as pressure (not KMs). We don't think 
> >>> the IDV can help here until the ctl and data file are fixed.
> >>>
> >>> Best,
> >>>
> >>> Unidata IDV Support
> >>>
> >>>
> >>>> Full Name: Stefan Cecelski
> >>>> Email Address: address@hidden
> >>>> Organization: University of Maryland
> >>>> Package Version: 3.0u2 build date:2012-02-08 08:06 UTC
> >>>> Operating System: Mac OS X
> >>>> Hardware: Java: home: 
> >>>> /System/Library/Java/JavaVirtualMachines/1.6.0.jdk/Contents/Home 
> >>>> version: 1.6.0_29 j3d:1.5.2 fcs (build4)
> >>>> Description of problem: Hi all,
> >>>>
> >>>> I recently converted my wrf output to height levels via ARWpost to 
> >>>> create a grads ctl/dat pair. When I open the ctl file (attached) in IDV, 
> >>>> the vertical levels are automatically presumed as pressure levels, even 
> >>>> though the levels are height above ground in km. Is there anyway to 
> >>>> trick the IDV in to seeing the levels as height levels instead of 
> >>>> pressure levels?  The advantage of using height levels is evenly spaced 
> >>>> vertical cross-sections via a transect view. However, since the IDV sees 
> >>>> the levels as pressure levels, all the data is not viewable in the 
> >>>> traditional transect view display because the pressures are all between 
> >>>> 0 and 20 hPa! Thanks for any advice you might have.
> >>>>
> >>>> V/R,
> >>>>
> >>>> Stefan
> >>>>
> >>>>
> >>>
> >>>
> >>> Ticket Details
> >>> ===================
> >>> Ticket ID: COD-764409
> >>> Department: Support IDV
> >>> Priority: Normal
> >>> Status: Closed
> >>>
> >>
> >>
> >>
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: COD-764409
> > Department: Support IDV
> > Priority: Normal
> > Status: Open
> >
> 
> 
> 


Ticket Details
===================
Ticket ID: COD-764409
Department: Support IDV
Priority: Normal
Status: Open