Due to the current gap in continued funding from the U.S. National Science Foundation (NSF), the NSF Unidata Program Center has temporarily paused most operations. See NSF Unidata Pause in Most Operations for details.
I have a SST file in GEMPAK which I've been trying to convert to GRIBingest into the WRF model. The SST grid is initially a CED projection. gdgrib won't convert to GRIB with this projection. I have tried
converting to an LCC projection with custom grid size and spacing. I receive an error about the grid not being found in grdnav.tbl, but the conversion still works. But when I try to ingest this ino the WRF model, my SSTs get extremely distorted. If I convert my GRIB file back to GEMPAK and plot it, it plots up fine. I have now found that if I specify an existing grid by setting CPYFIL=#218, or if I specify dimensions of an existing grid using PROJ, GRDAREA, and KXKY of a grid that is found in grdnav.tbl, the conversion performs without error and the WRF model successfully ingests my file without distortion. Unfortunately no existing grid has the resolution I need over a large enough area. So my question is, when I convert to GRIB using gdgrib using custom grid dimensions, is there any critical navigation information that gets left out of the GRIB file? The way I see it, if the grid I'm interpolating to does not exist in grdnav.tbl, something gets left out of the GRIB file and WRF does not know how to successfully navigate my file, and winds up blindly interpolating my file to my model domain. Perhaps I am not setting something properly in gdgrib? Here is the settings I've been using for gdgrib: GDFILE = overlay.gem GFUNC = TMPK GDATTIM = 20021224/1800:20021225/0000 GLEVEL = 0 GVCORD = noneGBTBLS GBFILE = 2002122418_SST.grb VERCEN PDSVAL = TMPK%1 PRECSN WMOHDR CPYFIL PROJ = LCC/25;-95;25
GRDAREA = 15;-115;50;-50 KXKY = 803;433 Any help would be greatly appreciated! Thanks! Matt Graduate Student North Carolina State University Department of Marine, Earth, and Atmospheric Sciences
gembud
archives: