An xcube plugin for generating data cubes from Sentinel Hub.
First install xcube
,
then the xcube_sh
plugin.
While xcube
is not yet available from conda-forge, install from sources:
$ git clone https://github.com/dcs4cop/xcube.git
$ cd xcube
$ conda env create
$ conda activate xcube
$ python setup.py develop
Once xcube
will be available from conda-forge:
$ conda create --name xcube xcube>=0.3
$ conda activate xcube
You cannot use the xcube_sh
plugin without specifying your Sentinel Hub
credentials:
$ export SH_CLIENT_ID=<your client ID>
$ export SH_CLIENT_SECRET=<your client secret>
While xcube_sh
is not yet available from conda-forge,
install it from sources. We'll need to update the xcube
environment first,
then install xcube_sh
:
$ conda activate xcube
$ conda install -c conda-forge oauthlib pip
$ pip install requests_oauthlib
$ git clone https://github.com/dcs4cop/xcube-sh.git
$ cd xcube-sh
$ python setup.py develop
Once xcube_sh
will be available from conda-forge:
$ conda activate xcube
$ conda install -c conda-forge xcube-sh
$ pytest
Check available xcube CLI extensions added by xcube_sh
plugin:
$ xcube sh --help
$ xcube sh gen --help
$ xcube sh info --help
Generate an xcube-compatible ZARR data cube from Sentinel-2 L1C data:
$ xcube sh info
$ xcube sh info S2L1C
$ xcube sh gen S2L1C -o cube.zarr -b B08 -g 10.237174,53.506205,10.271174,53.540205 -r 6.640625e-05 -t 2017-08-01,2017-08-31 -p 1D
$ python
>>> import xarray as xr
>>> cube = xr.open_zarr('cube.zarr')
>>> cube
>>> cube.B08
>>> cube.B08.isel(time=22).plot.imshow()