map projections and WRF regridded to NARR plotting issues

From: jonathan meyer <tallj03_at_nyahnyahspammersnyahnyah>
Date: Wed Nov 02 2011 - 13:33:50 MDT

Hi All,
I am battling plotting resources for a script that takes 4 various WRF simulation, each using a different map projection, and regrids them all into NARR's domain to make difference plots. I have two main questions regarding this script.
First Question: I am having plotting issues:
My script takes each regridded difference field and plots it using the original domains default projection (mercator domain plots over where the wrf mercator domain would have been). The problem I've encountered is with the Lambert and Rotated Lat-Lon projections and how they are plotted. Both the mercator and regular Lat-Lon projection plot correctly but the other two appear to be crunched within a smaller area that does not match with the original domain. My script plots correctly if I compare the individual simulations against each other (ie, regridding all 4 to mercator works fine as does regridding to lambert or any of the other projections) but for some reason, the regridding into NARR's domain causes the exact same resources to plot incorrectly. When looking at the image attached, you can see that the incorrect panels contain the same size of information as the other (all contain roughly the same amount of missing data due to difference in size of original domain and NARR's domain) but the incorrect panels are compressed in space.
Below is the list of map resources for each projection type as well as the plotting function used

Second Question:
This is regarding the actual projection resources I need to use for the Rotated Lat-Lon projection. When comparing wind components, the other three simulations typically have comparable fields for the entire domain but the rotated lat-lon projection, when using the resources below (#4), the upper corners exhibit stronger magnitude winds with directions that disagree with the other three simulations types. This leads me to believe the plotting resources are not correct and create a twisting effect in those corners, but I have been unable to find any examples of what resources/values should be used for a rotated lat-lon projection. I have copied my domain's projection settings from my namelist.wps into the bottom portion of this email so someone can take a look at advise on the appropriate plotting resources.

Thanks in advance!!!

------------------------------------problem script plotting resources for 4 specific projection types---------------------------------------------
1) Lambert res1 = res res1@mpProjection = "LambertConformal" res1@mpLambertParallel1F = 39.5 res1@mpLambertParallel2F = 39.5 res1@mpLambertMeridianF = -93.3 res1@tfDoNDCOverlay = True res1@mpLeftCornerLatF = Lam_min_lat res1@mpLeftCornerLonF = Lam_min_lon res1@mpRightCornerLatF = Lam_max_lat res1@mpRightCornerLonF = Lam_max_lon 2) Lat-Lon res2 = res res2@mpProjection = "Mercator" res2@mpLambertParallel1F = 39.6 res2@mpLambertParallel2F = 39.6 res2@mpLambertMeridianF = -93.3 res2@tfDoNDCOverlay = False res2@mpLeftCornerLatF = Lat_min_lat res2@mpLeftCornerLonF = Lat_min_lon res2@mpRightCornerLatF = Lat_max_lat res2@mpRightCornerLonF = Lat_max_lon 3) Mercator res3 = res res3@mpProjection = "Mercator" res3@mpLambertParallel1F = 42.5 res3@mpLambertParallel2F = 42.5 res3@mpLambertMeridianF = -93.3 res3@tfDoNDCOverlay = False res3@mpLeftCornerLatF = Mer_min_lat res3@mpLeftCornerLonF = Mer_min_lon res3@mpRightCornerLatF = Mer_max_lat res3@mpRightCornerLonF = Mer_max_lon
4) Rotated Lat-Lon res4 = res res4@mpProjection = "LambertConformal" res4@mpLambertParallel1F = 39.5 res4@mpLambertParallel2F = 39.5 res4@mpLambertMeridianF = -93.3 res4@tfDoNDCOverlay = True res4@mpLeftCornerLatF = Rot_min_lat res4@mpLeftCornerLonF = Rot_min_lon res4@mpRightCornerLatF = Rot_max_lat res4@mpRightCornerLonF = Rot_max_lon
        plot(0) = gsn_csm_vector_map(wks, diffX1_YR(:,:), diffY1_YR(:,:), res1) plot(1) = gsn_csm_vector_map(wks, diffX2_YR(:,:), diffY2_YR(:,:), res2) plot(2) = gsn_csm_vector_map(wks, diffX3_YR(:,:), diffY3_YR(:,:), res3) plot(3) = gsn_csm_vector_map(wks, diffX4_YR(:,:), diffY4_YR(:,:), res4) ;############################################################################################################## resP@gsnMaximize = True resP@gsnPanelLabelBar = True gsn_panel(wks, plot, (/2,2/), resP)

--------------------------------------------------------------------------------------------------
-------------Rotated Lat-Lon Domain settings-----------------
&geogrid parent_id = 1, parent_grid_ratio = 1, i_parent_start = 1, j_parent_start = 1, e_we = 260, e_sn = 138, geog_data_res = '10m', dx = 0.2878281, dy = 0.2878281, map_proj = 'lat-lon', ref_lat = 39.5, ref_lon = -93.3, pole_lat = 50.5, pole_lon = -180.0, stand_lon = 93.3,

                                               

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk

Received on Wed Nov 2 13:34:03 2011

This archive was generated by hypermail 2.1.8 : Fri Nov 04 2011 - 08:43:41 MDT