Project

General

Profile

Forums » Support »

Combining shapefiles and netCDF files

Added by Miles Sowden over 2 years ago

The post https://code.mpimet.mpg.de/boards/2/topics/9139 raises an interesting observation. Their problem is related to one shapefile and cropping the netcdf accordingly.

If the shapefile contained countries, with a unique country ID and the netCDF file contained global weather data what would the easiest way be to extract the weather data by country?
The obvious answer would be a big for loop of code generating a file per country.

But is it possible to take the shapefile and use the histogram function to define levels (one per country)
Then multiply the multi-level country mask with the 2D weather file
then summarize by level.

My problem for next week :-)


Replies (5)

RE: Combining shapefiles and netCDF files - Added by Karin Meier-Fleischer over 2 years ago

Hi Miles,

you can use NCL to convert the shapefile to netCDF or you can extract the weather data using the shapefile polygons with Python.

RE: Combining shapefiles and netCDF files - Added by Miles Sowden over 2 years ago

Hi Karen,

Sorry if I wasn't clear. If I have a shapefile of countries (for instance) and I converted that to a netCDF where each country is a different ID what is the easiest way to mask by each country and extract and summarise the weather data by country.

RE: Combining shapefiles and netCDF files - Added by Karin Meier-Fleischer over 2 years ago

Ok, let's say your netCDF shapefile shp_file.nc has a variable NAME_ID containing integers 0 to 195 (one unique id for each country).

Create the mask file for country NAME_ID == 2 (set ones for selected country area and missing value for the other areas:

cdo -expr,'NAME_ID = ((NAME_ID == 2)) ? 1.0 : (NAME_ID/0.0)' shp_file.nc mask_id_2.nc

Then use the mask file to mask your weather data:

cdo -mul mask_id_2.nc weather_data.nc masked_weather_data.nc

See also the article https://code.mpimet.mpg.de/boards/53/topics/10933

RE: Combining shapefiles and netCDF files - Added by Miles Sowden over 2 years ago

Karin,

This will get me the mask for country 2 but I want a mask for each of the 195 countries. I then want for each country to get some weather statistics.
So is the easiest a huge big for loop?

    (1-5/5)