Monitor forest dynamics using satellite image time series
Suggested electives: Environmental Monitoring with Satellite Image Time Series
Programming experience is valuable
Monitoring forest dynamics is essential for identifying the area most at risk for forest degradation and deforestation, and to improve policies for forest conservation. Remote sensing data provides periodic temporal coverage, global datasets with various combinations of spectral bands and spatial resolution that have potential to monitor forests. However, the main two operational limitation to global forest monitoring is the existence of multiple monitoring programs based on different change detection algorithms and the lack of sufficient field measurements at the global scale. Statistical time series analysis allows to assess quantitatively temporal patterns, estimate and model temporal behaviours based on temporal relationship, as well as, detect structural changes.
The main goal of this study is to monitor forest dynamics at regional scale using Landsat/sentinel time series and to identify areas of forest degradation and select areas of priority intervention for the forest manager. Classical statistical time series analysis will be used to quantify temporal patterns and to detect changes in the time series at pixel level.
This MSc topic is very broad, and it can be focused on different ways. A more methodological approach focused on the monitoring part or a more applied approach identifying the areas of priority intervention and proposing some management practices for restoration. Within the study area different stakeholders and polices must be taken into account when developing any intervention plan. In addition, there are different data sources available for the student to consider in the analysis. In conclusion, this topic can be easily adapted to the needs of each student.