Global Forest Fragmentation

Code and data repository for the analysis of global forest fragmentation

Download all files View on GitHub


This repository contains the source code and data products of the global assessment of forest fragmentation using ALOS PALSAR data. A manuscript accompanying this repository is currently in peer review.

The files in this repository are modified maps based on ALOS PALSAR Forest-NonForest (FNF) tiles, which can be requested for academic purposes at the following website.
Please have a look at the code if you are interested in reproducing the results or want to generate similar products at smaller resolutions. The landscape metrics were calculated using adapted code from Jung et al. (2016). You can find the original code at the following repository developed for usage within the QGIS suite.

Jung, M. (2016). LecoS — A python plugin for automated landscape ecology analysis. Ecological Informatics, 31, 18–21.

Available metrics

See below for links for download and description of metrics at 0.1 degree resolution. Every netCDF file contains five subdatasets representing the years 2007,2008,2009,2010 and 2015. Global maps within those datasets are projected in WGS 84 longitude-latitude projection. Every file should be self-describing, but have a look at the source code or description if in doubt.
(Note that the Mean Patch area and Percentage of Like adjacencies metric were not included in the analysis of the accompanied manuscript.)

Mean Patch Area Download Description
Patch density Download Description
Edge density Download Description
Largest patch index Download Description
Landscape division index Download Description
Fractal dimension index Download Description
Average inner edge distance Download Description
Percentage of Like adjacencies Download Description

How to visualize netCDF files?

The author recommends the excellent software panoply (freely available here). Alternatively netCDF files can also be opened and visualized in GIS suites such as ESRI ArcGIS or QGIS or alternatively in scripting languages such as R or python.

Support or contact:

This work was mostly done by @Martin-Jung), a PhD researcher at the University of Sussex (UK) investigating land dynamics and their influence on biodiversity. If you want to contact the author personally please visit the following adress:
If you want to help developing LecoS of found a bug in the calculation of the landscape metrics, please report it on the LecoS repository.