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

Re: Bug in LambertAzimuthalEqualArea?



On Mar 17, 2011, at 08:59, John Caron wrote:

> On 3/16/2011 7:18 PM, Robert B. Schmunk wrote:
>> John,
>> 
>> Between NJ 4.0.49 and 4.2.21, a bug seems to have crept into the
>> class for projecting data on a Lambert Azimuthal Equal Area grid.
>> I am wondering if this resulted from the "earth_radius attribute
>> on grid_mapping variable must be in meters" item in the change
>> history under version 4.1.15.
>> 
>> The problem that I'm seeing is that in an old copy of Panoply
>> which uses NJ 4.0.49, data on an LAEA grid are plotted as one
>> expects. But with the same data in a newer Panoply which uses
>> NJ 4.2.21, I find that the result returned by the latLonToProj
>> method needs to be scaled by a factor of 1000 if the data are
>> to be plotted correctly. (Presumably the projToLatLon method is
>> also off by a factor of 1000, in the other direction, although
>> I haven't tested that yet.) The LambertAzimuthalEqualArea object
>> is being obtained by calling getProjection() on the variable's
>> coordinate system and is not constructed from scratch.
>> 
>> The dataset I am working with, and which seems to be tagged
>> correctly, is located at
>> http://www.giss.nasa.gov/staff/rschmunk/open/azim_equal_area.nc
>> and includes some weather radar data for the United States.
>> 
>> rbs
>> 
>> 
>> 
>> --
>> Robert B. Schmunk, address@hidden
>> NASA Goddard Institute for Space Studies, 2880 Broadway, New York, NY 10025
>> 
> hi robert:
> 
> im not sure whats going on. the latest version of toolsui is displaying that 
> file correctly.
> 
> are you opening the file as a GridDataset or some other way?


John,

I'm using the following for all datasets

NetcdfFile ncf = NetcdfDataset.acquireFile (url.toString ( ), null);
NetcdfDataset ncd = new NetcdfDataset (ncf);


I checked some other datasets with projected grids later last night
and found there was a similar problem with some others, on various
grid schemes. But some still plotted okay. I haven't quite figured
out why the difference.

rbs




--
Robert B. Schmunk, address@hidden
NASA Goddard Institute for Space Studies, 2880 Broadway, New York, NY 10025