gistim 0.1

Installation requirements

In terms of development requirements, the packages are independent as well. The plugin requires:

• A modern (>=3.0) QGIS installation, within the context of OSGeo4W. A basic QGIS installation (as can be gotten here suffices to run the plugin, but will not suffice for full development capabilities.

The QGIS installation comes with its own Python installation and interpreter. This installation does not provide a package manager such as conda. This complicates the distribution of complex binary dependencies. Hence, the gistim package (and TimML) should run in a different interpreter. This requires:

• A modern Python version (>=3.6)

• GeoPandas

• Xarray, which in turn requires numpy and pandas

• Rasterio

• rioxarray

• netCDF4

Rasterio, rioxarray, and netCDF4 are optional dependencies but output options are severely limited without them. Both geopandas and rasterio requires GDAL, which is a heavy and complex binary dependency. It is highly recommended to install these packages via conda in a separate environment – the specification is included in the environment.yml of this repository.

See the installation instruction below.

Installing the plugin in QGIS

There are a few ways to install this plugin:

• Copy the directory ./plugin/qgistim to your local QGIS plugins directory. On Windows, this should be located at:

  c:\Users\{username}\AppData\Roaming\QGIS\QGIS3\profiles\default\python\plugins

  The location can be found in the QGIS GUI via:

  QGIS menu > Settings > User Profiles > Open Active Profile Folder

• Alternatively, zip the ./plugin/qgistim directory. Then, in the Plugins menu, under “Manage and Install Plugins…”, find “Install from ZIP”, and enter the path to the zipfile – this unzips the files and copies them to the directory mentioned above.

The plugin is good to go now. Find the Qgis-Tim entry in the Plugins menu, and click it to open a docked menu on the right side of your QGIS screen.

Installing the Python package

The recommended way of installing is using conda, in a separate conda environment. This environment (called tim) can be setup by downloading the environment.yml and running the following command in the anaconda prompt:

conda env create -f environment.yml

After creation, the conda environment can be activated by running the following command in the anaconda prompt:

conda activate tim