Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Quantifying the spatial pattern of landscapes has become a common task of many studies in landscape ecology. Most of the existing software to compute landscape metrics is not well suited to be used in interactive environments such as Jupyter notebooks nor to be included as part of automated computational workflows. This article presents PyLandStats, an open-source Pythonic library to compute landscape metrics within the scientific Python stack. The PyLandStats package provides a set of methods to quantify landscape patterns, such as the analysis of the spatiotemporal patterns of land use/land cover change or zonal analysis. The implementation is based on the prevailing Python libraries for geospatial data analysis in a way that they can be forthwith integrated into complex computational workflows. Notably, the provided methods offer a large variety of options so that users can employ PyLandStats in the way that best supports their needs. The source code is publicly available, and is organized in a modular object-oriented structure that enhances its maintainability and extensibility.
Karen Scrivener, Fabien Jacques Michel Georget, William Wilson
Ludger Weber, Alberto Ortona, Manoj Kondibhau Naikade
Devis Tuia, Alex Hubertus Levering