MSc thesis subject: Suitability of Global Tree Cover Map for REDD+ Performance

Forest play an important role in the new Paris climate agreement, mainly through REDD+ in tropical developing countries. REDD+ is an international mechanism for Reducing Emissions from Deforestation and forest Degradation and the enhancement of forest carbon stocks.

REDD+ was envisioned as a performance based mechanism: countries will be paid for a reduction in emissions from deforestation, forest degradation relative to an established baseline (reference level - RL). Remote sensing data can help with establishing this RL. However, national monitoring capacities are often low. Existing global datasets on forest cover can help fill the gap.

Hansen et al (2013) produced a global map in characterizing global forest This global map with annual updates on gross forest cover change can be a valuable resource for countries with limited forest monitoring capacity (Goetz et al). However like any other large scale datasets, this map has also limitations in some regions (Bellot et al 2014). A comprehensive study on its uncertainty can help on decisions to use this dataset for national scale forest resources assessments. An independent reference dataset on forested area and forest cover proportions is available for this thesis topic. The dataset is suitable for large or continental scale analysis focusing on Landsat (or Sentinel) like resolution datasets. Using this dataset, the uncertainty of global scale forest map can be investigated. A follow-up step is to take this information on uncertainty to assess the consequences for REDD+ performance and reference levels.


  • Analysis of uncertainty/accuracy global forest cover map with validation data across pan-tropics
  • Assess the impact of the uncertainty of the forest map on REDD+ reference levels and performance for selected countries



  • Knowledge of programming with R

Theme(s): Modelling & visualisation, Integrated Land Monitoring