Re: WRF NMM geography

From: David Brown <dbrown_at_nyahnyahspammersnyahnyah>
Date: Sun Apr 24 2011 - 10:53:29 MDT

Hi Ahmed,
I am attaching a revised version of your script that plots the data twice: first using the coordinate arrays to project the data into a regular lat-lon projection, and second in its native projection, which happens to be a rotated lat-lon projection. I modified some of the contour settings to highlight the alignment of the data to the map. Here are some things to note:
1) You will get incorrect results if you set tfDoNDCOverlay True if you have not matched the map projection settings to the actual projection of the data.
2) tfDoNDCOverlay should be set for the contour plot and not for the map plot.
3) You should not use the min/max latitude and longitude values when using the "corners" map limit mode unless you know the data is in a regular lat/lon projection. As your dataset illustrates, the corner points are not always the minimum and maximum coordinate values.
4) Your data file does contain metadata with information about the projection, but it is not necessarily obvious. Even though you have acquired it as NetCDF data, some of the information it contains is based on the same tables used for GRIB. In particular the global attribute MAP_PROJ is set to 203. This number indicates that the data uses the
"Arakawa semi-staggered E-grid on rotated latitude/longitude grid-point array (2-dimensional)", as indicated in Table 6 of the GRIB 1 specification at Since it is a rotated lat/lon grid you need to set the NCL map projection resources mpCenterLonF and mpCenterLatF correctly to match the projection of the data. These values also appear to be contained in the global metadata as CEN_LAT and CEN_LON. The second frame of the revised script uses this information to match the map projection fairly reasonably to the pre-projected data.

On Apr 24, 2011, at 5:58 AM, ahmed lasheen wrote:

> Hello
> I am trying to draw the output of the geogrid.exe , but a portion of the plot doesnot appear in the east of the map.
> i have attached the NCL script and the nc file .
> thanks in advance
> --
> ===============
> Ahmed Lasheen
> Junior researcher at Cairo Numerical Weather Prediction Center (CNWP)
> ===============
> <plot.ncl><>_______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:

ncl-talk mailing list
List instructions, subscriber options, unsubscribe:

Received on Sun Apr 24 10:53:51 2011

This archive was generated by hypermail 2.1.8 : Tue May 03 2011 - 14:47:35 MDT