Student information

MSc thesis topic: Forest biomass mapping in the Netherlands with satellite radar

Accurate mapping and monitoring of forest biomass is needed in many applications, including climate research where more accurate estimates of carbon fluxes are needed, mainly carbon emissions due to deforestation and carbon sequestration in forests [1]. While biomass can be estimated with spaceborne laser scanning sensors (e.g., NASA GEDI [2]), these are sampling systems with a low revisit frequency and without wall-to-wall mapping capabilities. Synthetic aperture radar (SAR) satellites do not have that limitation, and can provide cloud-free, global wall-to-wall data with a high revisit frequency.

While P- and L-band SARs (70 cm and 30 cm wavelengths, respectively) are traditionally considered most suitable for this task due to their better canopy penetration capabilities, the data availability is currently very limited for these systems. Meanwhile, dense time series of global C-band SAR (wavelength: 6 cm) data are available for free thanks to the European Sentinel-1 (S1) system and a recent study has shown that repeat-pass S1 coherence has a good potential for forest biomass mapping in some areas [3].

The combination of S1 and spaceborne laser scanning data from GEDI is especially promising, with S1 providing wall-to-wall monitoring capabilities and GEDI providing possible reference biomass measurements for training.

In this thesis, you will compare 6- and 12-day S1 coherence data from 2014-2022 over the Netherlands with forest biomass estimates from field inventories and the NASA GEDI mission. You will consult meteorological data and S1 metadata to determine if and under which circumstances forest biomass can be estimated from S1 coherence. Then, you will define, train, and validate a simple biomass estimation approach that can be used with future S1 acquisitions and existing ancillary data to map and monitor biomass of the Dutch forests, also outside of the GEDI coverage.

Software: ESA Snap, R/Python/Matlab, ...

Objectives

  • Study time series of S1 coherence against reference biomass measurements from GEDI and field inventories
  • Determine which meteorological conditions and spatial baselines are most suitable for biomass estimation
  • Develop and test a simple biomass estimation algorithm.

Literature

  1. Le Quéré, Corinne, et al. "Global carbon budget 2018." Earth System Science Data 10.4 (2018): 2141-2194.
  2. Patterson, Paul L., et al. "Statistical properties of hybrid estimators proposed for GEDI—NASA’s global ecosystem dynamics investigation." Environmental Research Letters 14.6 (2019): 065007.
  3. Cartus, Oliver, et al. "Sentinel-1 coherence for mapping above-ground biomass in semiarid forest areas." IEEE Geoscience and Remote Sensing Letters 19 (2021): 1-5.

Requirements

  • Advanced Earth Observation course
  • Geo-scripting course (Good knowledge in scripting is an asset; e.g. Google Earth Engine, R, Python, JavaScript)

Theme(s): Sensing & measuring; Modelling & visualisation