Plot on map projections (with coastlines and political boundaries) using matplotlib.

Basic requirements are the following:

• Python 2.6 (or higher)
• matplotlib
• numpy
• pyproj
• pyshp

Optional requirements include:

• OWSLib. It is needed for the method Basemap.wmsimage.

• Pillow. It is needed for the methods Basemap.bluemarble, Basemap.etopo, Basemap.shadedrelief and Basemap.warpimage.

The basemap-data and basemap-data-hires packages are available in PyPI and can be installed with pip:

python -m pip install basemap-data
python -m pip install basemap-data-hires

Precompiled basemap binary wheels for Windows and GNU/Linux (architectures x86 and x64, Python 2.7 and 3.5+) as well as for MacOS (architectures x64 and arm64, Python 3.9+) are also available in PyPI:

python -m pip install basemap

Otherwise, you will need to install basemap from its source hosted on GitHub as indicated in the following steps:

1. Install pre-requisite Python modules:

   • cython
   • numpy

2. Download the basemap source code and move to the packages/basemap folder:

   git clone --depth 1 https://github.com/matplotlib/basemap.git
   cd basemap/packages/basemap

3. Build the GEOS library. You may use the helper provided in utils, (please note that you need CMake and a working C compiler in advance):

   export GEOS_DIR=<your desired location>
   python -c "import utils; utils.GeosLibrary('3.6.5').build(installdir='${GEOS_DIR}')"

   or you can link directly to the system library if it is already installed. GEOS_DIR must point to the GEOS installation prefix; e.g. if libgeos_c.so is located in /usr/lib and geos_c.h is located in /usr/include, then you must set GEOS_DIR to /usr.

4. Build and install the basemap binary wheel:

   python -m pip install .

   On Linux, if your Python was installed through a package management system, make sure that you have the Python header Python.h required to build Cython extensions (e.g. on Debian-like systems, you should have the package python-dev installed).

5. Check that the package was installed correctly by executing:

   python -c "from mpl_toolkits.basemap import Basemap"

The source code and data assets are under the following licenses:

• basemap: MIT.
  • GEOS bundled dynamic library is under the LGPL-2.1-only license.
• basemap-data: LGPL-3.0-or-later.
  • The EPSG file and the JPG images are also under the MIT license.
• basemap-data-hires: LGPL-3.0-or-later.

For a full description, please visit the README and LICENSE files of each package.

See https://matplotlib.org/basemap/

See scripts in examples directory for example usage.

Read the FAQ and/or email the matplotlib-users mailing list if you have problems or questions.

Ben Root [email protected]

Víctor Molina García (@molinav)

Special thanks to John Hunter, Andrew Straw, Eric Firing, Rob Hetland, Scott Sinclair, Ivan Lima, Erik Andersen, Michael Hearne, Jesper Larsen, Ryan May, David Huard, Mauro Cavalcanti, Jonas Bluethgen, Chris Murphy, Pierre Gerard-Marchant, Christoph Gohlke, Eric Bruning, Stephane Raynaud, Tom Loredo, Patrick Marsh, Phil Elson, and Henry Hammond for valuable contributions.

The Basemap.fillcontinents method doesn't always do the right thing. Matplotlib always tries to fill the inside of a polygon. Under certain situations, what is the inside of a coastline polygon can be ambiguous, and the outside may be filled instead of the inside. A workaround is to change the map projection region slightly or mask the land areas with the Basemap.drawlsmask method instead of filling the coastline polygons (this is illustrated in the ortho_demo.py example).