# Re: Convergence Plotting issues

From: Mary Haley <haley_at_nyahnyahspammersnyahnyah>
Date: Wed Aug 28 2013 - 14:55:50 MDT

Hi Brian,

I think the problem may be that you are trying to treat 2D XLAT/XLONG as 1D coordinate arrays by doing this:

XLAT = a->XLAT_M(n,:,:)
XLONG = a->XLONG_M(n,:,:)

lat = XLAT(:,0)
lon = XLONG(0,:)
lat@units = "degrees_north"
lon@units = "degrees_east"
lat!0 = "lat"
lon!0 = "lon"
lat&lat = lat
lon&lon = lon

;Assign U and V lat/lon arrays
u_plane!1 = "lat"
u_plane!2 = "lon"
u_plane&lat = lat
u_plane&lon = lon

v_plane!1 = "lat"
v_plane!2 = "lon"
v_plane&lat = lat
v_plane&lon = lon

If XLAT and XLONG are truly repeating their values along one dimension, then that's fine. But, if they are not, then you should do this instead:

u_plane@lat2d = XLAT
u_plane@lon2d = XLONG

v_plane@lat2d = XLAT
v_plane@lon2d = XLONG

I don't know about your contour data, however, because I can't tell if they have 1D coordinate arrays or 2D lat/lon coordinates.

If the above suggestion doesn't help, then it would really help if you could also provide the data files. You can use our ftp:

To get dash patterns for negative contours for the RUC plots, try:

res_at_gsnContourNegLineDashPattern = 1 ; some value from 1 to 16, see http://www.ncl.ucar.edu/Document/Graphics/Images/dashpatterns.png

--Mary

On Aug 27, 2013, at 9:25 PM, brianjs @iastate.edu wrote:

> Good Afternoon,
>
> I attempted to plot divergence/convergence for winds at four specific levels above the surface layer and have had success with WRF data, but not with RUC. As can been by the two WRF images (Upper_Air_Convergence plots), I obtained the product I wanted, but with the other two images (RUC_Wind_Conv plots), the divergence/convergence contours do not seem to portray anything realistic compared to their WRF counterparts. Despite the RUC data being at 13 km and the WRF at 4 km, I can't imagine the difference in resolution would have this much of an impact on the convergence contours. Most likely, I am probably missing a key function somewhere that was automatically used in a WRF overlay function with the WRF code. Also, WRF-NCL seems to have an automatic contouring scheme that separates positive values with solid contours and negative values with dashed contours. What would be the NCL contour function that does this manually? I would need to implement it in the ruc script. I sear
ched the NCL functions only to find changes in color schemes and not line types.
>
> Attached are the two codes, the first for the successful WRF plots (convergence.ncl) and the other for the RUC plots in question (ruc_conv.ncl). I appreciate any input that can be provided! Please also copy my email into the reply (brianjs@iastate.edu) in case I do not receive the notification email (which has happened to me on occasion).
>
> Brian Squitieri