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

20010827: radar data projections



Wil,

You can determine the latitude and longitude of each grid point
using GFUNC=mul(latr,rtd) and GFUNC=mul(lonr,rtd) respectively.
Other values such as MSFX and MSFY (map scale factors in X and Y)
are available.

The latr and lonr values are in radians as shownb in the GPARM 
help, and the RTD constant multiplier can be used in the mul()
function to convert the values to degrees.

The location of the grid points are constant once the grid is
created in GDRADR. The GAREA and PROJ in GDLIST determine the mapping 
bounds as you would see displayed with GDMAP.

If you are looking for a rectilinear map projection, then you may
want to use CED rather than MER- I have no idea as to your other
application data- but, CED would give you constant spacing in the
X and Y rows, whereas the MER will be stretched poleward.
The default for ang1;ang2;ang3 will be 0;0;0 in proj unless otherwise
specified when you create the grid.

Steve Chiswell
Unidata User Support


>From: Wil Wollheim <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200108271715.f7RHFG121983

>
>Hi Steve,
>
>We've been able to work with GDRADR and GDLIST to start looking at
>the gridded radar data - thanks for helping us get started on that.  But
>now we have an additional question regarding how to project it correctly
>to match other datalayers that we have.
>
>We've used GDRADR to convert the NEXRADIII data into a grid and
>then exported the grid with GDLIST to an ascii table.  We specified the
>mercator projection in GDRADR and did not specify any additional
>projection in GDLIST.  Is the exported ascii table then
>also in mercator or is it in geographic?   
>
>We assumed the data was in geographic (because only latitude and
>longitudes are specified in GDINFO), imported it into Arc, and compared 
>this map to the same nexrad data displayed in nmap.  The two maps are
>close, but seem to be off by a couple of grid cells.
>
>If the exported table is in mercator, how do we find out the other
>projection parameters (e.g. central meridan, true scale latitude)
>
>We thought that one way to get the areas of the grid cells in Mercator
>would be to create a geographic grid with the same number of cells,
>use ARC to project that grid into Mercator, then export the grid to
>ASCII and use the parameter values from the ASCII grid in our dumped
>GDLIST grid. Would this work for determining the correct projection
>parameters?
>
>Thanks again for your help!
>
>Wil Wollheim
>
>
>Water Systems Analysis Group
>Complex Systems Research Center
>Morse Hall
>University of New Hampshire
>Durham NH 03824
>Voice: (603) 862-0812
>FAX: (603) 862-0587
>E-mail: address@hidden
>
>
>
>