Student information

MSc thesis topic: Quantification of biomass and nitrogen of grasslands using proximal and satellite remote sensing data

Grasslands are pivotal in the global ecosystem, serving as a vital carbon sink and offering essential ecosystem services. Precise measurement of biomass and nutrient makeup is crucial for sustaining these grasslands. Hyperspectral analysis has emerged as a promising tool for assessing grassland characteristics non-destructively. Establishing a comprehensive spectral database, spanning from close-range to (hyperspectral) satellite observations (EnMap, PRISMA), is imperative for accurately quantifying biomass and nutrients across local and global scales.

The development of robust Machine Learning (ML) models to quantify grassland traits requires sufficiently large datasets spanning a range of global geographical locations. From previous (PhD) research projects of the supervisors, a large-scale spectral database has been established with corresponding ground truth data from multiple seasons, countries, and observations collected from both proximal and satellite sensors. This database offers an opportunity to develop and evaluate robust ML models which can be generalized to a global scale.

Relevance to research/projects at GRS or other groups

The research collaboration builds on previous projects of the University of Western Australia and the University of Adelaide in Australia and the WUR in the Netherlands.

Objectives and Research questions

  • Quantify biomass and nutrients of grasslands comparing multiple hyperspectral platforms (ground-based, drone and satellite).
  • Asses the performance of Machine Learning modelling approaches across different landscapes and at different scales

Requirements

  • Previous courses on Machine Learning

Literature and information & Expected reading list before starting the thesis research

  • Gao J, Liang T, Yin J, Ge J, Feng Q, Wu C, Hou M, Liu J, Xie H (2019) Estimation of alpine grassland forage nitrogen coupled with hyperspectral characteristics during different growth periods on the Tibetan Plateau. Remote Sensing 11: 2085.
  • Pullanagari R, Dehghan-Shoar M, Yule IJ, Bhatia N (2021) Field spectroscopy of canopy nitrogen concentration in temperate grasslands using a convolutional neural network. Remote Sensing of Environment 257: 112353.
  • Wang J, Wang T, Skidmore AK, Shi T, Wu G (2015) Evaluating Different Methods for Grass Nutrient Estimation from Canopy Hyperspectral Reflectance. Remote Sensing 7: 5901-5917.

Theme: Sensing & measuring