[Wrf-users] Converting polar projection to mercator

[Thomas Nehrkorn]

Don,

GrADS is one possible way to do this.  If you provide GrADS with data on 
a grid on some map projection that it knows about (what is called 
"preprojected data" in GrADS parlance), it will internally interpolate 
to a lat-lon grid (that you can specify), and then display the data on a 
map projection of your choice.  You may want to double-check that all 
this interpolation does not introduce unacceptable errors, particular at 
the high latitudes you are interested in.

Don Morton wrote:
> I know this isn't directly relevant to WRF, but it seems like an
> area that might be of interest to others and, besides, I could use
> some guidance! :)
> 
> I'm currently running 2 48-hour forecasts per day for the Alaska
> region ( http://weather.arsc.edu ), using polar stereographic
> projection.  We're very interested - and I've already started a
> student in this direction - of taking some of the output graphics
> and using them as time-series overlays in a GoogleEarth and/or
> NASA World Wind environment.  However, these overlays need
> to be in  a mercator-like projection (you define the E and W
> longitudes and the N and S latitudes).
> 
> I naively thought that it would be kind of trivial to take my wrfout
> netCDF files (polar projection) and generate output in a mercator
> projection via something like RIP, but now I'm thinking I was way
> too optimistic.  I'm not very well versed on projections, but
> theoretically (again, I'm being naive) it seems like all you need to
> do is apply some sort of mathematical transformation from one
> grid to another, though I can see where this might get hairy,
> because your grid points would be different and you'd have to
> interpolate, etc.,
> 
> Does anybody know of a way I might be able to do these conversions?
> I stumbled across McIDAS, but it appears to me that this is probably
> pretty costly?
> 
> Thanks,
> 
> Don Morton
> 


-- 

Thomas