This project simulates Landsat data and evaluates the performance of recovery indicators with respect to data and disturbance characteristics.

363 commits | Last update: July 13, 2021

Cite this software

Choose a version:
[[ releases.length > 0 ? releases[selectedIndex].doi : conceptDOI ]]
Copy to clipboard
Choose a reference manager file format:
Download file

What BenchmarkRecovery can do for you

  • Generates synthetic, but realistic time series simulating forest recovery after a fire
  • Tests several resilience indicators over the time series

The context of this project is the study of the recovery of tropical forests after an abrupt disturbance (typically a forest fire) using satellite images as a data source.

The speed of recovery after a disturbance is known to be correlated with the concept of resilience. This is true not only for forests, but for many dynamical systems. To put it simply: forests that recover fast are more resilient. Forests that recover slowly may be in danger of permanent disappearance.

The specialized literature proposes different metrics for measuring the recovery speed. The performance of these metrics depends on many factors. Some of them are natural, such as the intensity of the perturbation or the seasonality. Others are technical, such as the sampling frequency or the spatial resolution.

Read more
No tags available
Programming Language
  • R
  • Apache-2.0
Source code


  • Pablo Rodríguez-Sánchez
    Netherlands eScience Center
  • Wanda de Keersmaecker
    Wageningen University and Research

Information for page maintainers

OAI-PMH metadata:
citation metadata: