Variation in lat/long distances

[Kelly Koehn]

I'm working with some NEXRAD radar data and reading the binary data and drawing it on a map.  I'm using Mapserver on the backend to produce the map.  What i'm basically doing is calculating 124 nautical miles (142 statute miles) from a center point (the radar) and then drawing my radar data on top of that.  I found this nifty little formula on the net that supposedly converts nautical miles to statute miles but it doesn't seem to really be working as expected.

The formula is as follows:
(((sm * 0.0254) * 63360) / 1852) / 60 = nm
where sm = statute miles and nm = nautical miles

At that point I take the value that I receive above and do the appropriate + or - calclations from the radar's location (lat/long) and get the extent for the map which is pushed through the URL.  What I've noticed is that the latitude seems to be correct but the longitude always falls short 20 or 30 miles from what it should be (almost cutting off a county on either side).  Has anyone else experienced this?  I've checked the x and y distances and they are both equal.  I'm producing a 460 x 460 GIF image that uses the nationalatlas.gov SHP files.  I'm using the county and states sets.

I can post my .map files and anything else that might be helpful (including sample images).

Thanks
Kelly Koehn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/mapserver-users/attachments/20060521/23614f55/attachment.html