keduba / commondatamodel.jl

CommonDataModel.jl is a module that defines types common to NetCDF and GRIB data

This package contains abstracts type definition for loading and manipulating GRIB, NetCDF, geoTiff and Zarr files. This package aims to follow the Common Data Model and the CF (climate and forecast models) Metadata Conventions.

Format	Package	read support	write support
NetCDF	NCDatasets	✔	✔
OPeNDAP	NCDatasets	✔	-
GRIB	GRIBDatasets	✔	-
geoTIFF	TIFFDatasets	✔	-
Zarr	ZarrDatasets	✔	✔

Features include:

• query and edit metadata of arrays and datasets
• virtually concatenating multiple files along a given dimension
• create a virtual subset (view) by indices or by values of coordinate variables (CommonDataModel.select, CommonDataModel.@select)
• group, map and reduce a variable (CommonDataModel.groupby, CommonDataModel.@groupby) and rolling reductions like running means CommonDataModel.rolling)

Here is minimal example for loading GRIB or NetCDF files.

import CommonDataModel as CDM
import SomeDatasets # where SomeDatasets is either GRIBDatasets or NCDatasets

ds = SomeDatasets.Dataset("file_name")

# ntime is the number of time instances
ntime = ds.dim["time"] # or CDM.dims(ds)["time"]

# create an array-like structure v corresponding to variable temperature
v = ds["temperature"]

# load a subset
subdata = v[10:30,30:5:end]

# load all data
data = v[:,:]

# load a global attribute
title = ds.attrib["title"]  # or CDM.attribs(ds)["title"]
close(ds)

Most users would typically import GRIBDatasets and NCDatasets directly and not CommonDataModel. One should import CommonDataModel only to extent the functionality of GRIBDatasets and NCDatasets.

File conversions

By implementing a common interface, GRIB files can be converted to NetCDF files using NCDatasets.write:

using NCDatasets
using GRIBDatasets
using Downloads: download

grib_file = download("https://github.com/JuliaGeo/GRIBDatasets.jl/raw/98356af026ea39a5ec0b5e64e4289105492321f8/test/sample-data/era5-levels-members.grib")
netcdf_file = "test.nc"
NCDataset(netcdf_file,"c") do ds
   NCDatasets.write(ds,GRIBDataset(grib_file))
end