[ncl-talk] Regrid PRISM data to a low resolution rectilinear grid

[Dave Allured - NOAA Affiliate]

Brandon, attached is a PRISM to Netcdf translator which includes a high
accuracy generator for the 1-D source grid coordinates.  Please see how
this is done in the section "Write lat and lon coordinates".

The latitudes are reversed to South to North for my final product, so you
would need to either reverse the latitudes or the data grids to align them
correctly.  Lower down in this example, you can see where I am flipping the
data grids to South to North.

If you follow this example, you will end up with grids with standard 1-D
lat and lon coordinate variables, which should be accepted by ESMF.


On Fri, May 8, 2020 at 3:00 PM Brandon Fisel via ncl-talk <ncl-talk at ucar.edu>
wrote:

> Hello,
>
> I am following the PRISM example (
> https://www.ncl.ucar.edu/Applications/Scripts/prism_1.ncl) to read .bil
> format PRISM data into NCL. There are no latitude or longitude variables
> available to read from the .bil files.  I would like to regrid the PRISM
> data to another rectilinear destination grid (~12km) using conservative
> remapping.
>
> Many of the examples show a source grid that is read in before the
> regridding step.  I am not familiar with regridding from a grid that
> supplies no lat/lon to another grid that does.  Below is an example of the
> data written to netCDF with some additions.
>
> dimensions:
>         time = UNLIMITED ; // (1 currently)
>         lat = 621 ;
>         lon = 1405 ;
> variables:
>         double time(time) ;
>                 time:calendar = "standard" ;
>                 time:units = "days since 1955-01-01 00:00:00" ;
>         float p(time, lat, lon) ;
>                 p:units = "mm day-1" ;
>                 p:missing_value = -9999.f ;
>                 p:_FillValue = -9999.f ;
>
> Recent ESMF regrid attempts have resulted in: (0)     get_src_grid_info:
> can't determine what type of source grid you have.
>
> Is there a way I can specify the areal dimensions of the source grid, or
> lat/lon corners as the source grid?
>
> -------------------------------------------------------------
> Brandon
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20200508/231aed7c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bil-to-nc.prism.daily.ncl
Type: application/octet-stream
Size: 26765 bytes
Desc: not available
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20200508/231aed7c/attachment.obj>