netcdf2dataframe

This repo contains a small utility used to transform netCDF files to geopandas.GeoDataFrames.

Why netcdf2dataframe?

⚠️ This project was initiated due to the need of colleagues. It's considered a work in progress !

Installation

pip install netcdf2dataframe

Basic usage

from netcdf2dataframe import netcdf2dataframe
path = "IMERG_total_precipitation_day_0.5x0.5_global_2021_v6.0.nc"
df = netcdf2dataframe(path, "pr")

Note that this module may end up crashing your machine, as DataFrames are not RAM-efficient objects (compared to the netCDF format). But sometimes, this is what you may need...

Full docstring:

netcdf2dataframe(path: str, target: str = 'pr', dropna: bool = True) -> geopandas.geodataframe.GeoDataFrame
    Picks a netCDF file and extracts one variable ("target") on the form of
    a GeoDataFrame.
    
    Parameters
    ----------
    path : str
        Path to netCDF file
    target : str
        Target variable. Should be any of the available variables, excepting
        the latitude, longitude or time. Please look the INFO log to now more
        on the available variables.
        Note that available variables are displayed in the log (info level).
    dropna : bool, optional
        If True, the GeoDataFrame will not contain missing values, hence
        being more frugal (less RAM consumption). The default is True.
    
    Returns
    -------
    gpd.GeoDataFrame
        GeoDataFrame of the extracted values
    
    Example
    -------
    >>> df = netcdf2dataframe(
            "IMERG_total_precipitation_day_0.5x0.5_global_2021_v6.0.nc,
            "pr",
            )
    >>> df.head()
    
             pr   time_val                      geometry
    0  0.000000 2021-01-01  POINT (-179.74998 -78.24999)
    1  0.382080 2021-01-01  POINT (-177.24998 -78.24999)
    2  1.625944 2021-01-01  POINT (-176.74998 -78.24999)
    3  1.772459 2021-01-01  POINT (-176.24998 -78.24999)
    4  1.412688 2021-01-01  POINT (-175.74998 -78.24999)

Support

In case of bugs, please open an issue on the repo.

Contribution

Any help is welcome.

This project is managed with poetry and tested with pytest. It uses black formatting with 79 characters.

Author

Thomas GRANDJEAN (DREAL Hauts-de-France, service Information, Développement Durable et Évaluation Environnementale, pôle Promotion de la Connaissance).

Licence

GPL-3.0-or-later

Project Status

Development.