Thesis subject

MSc thesis topic: Integrating Hyperspectral and LiDAR Data for Tree Species Mapping and Canopy Trait Validation in Forest Ecosystems

Forest ecosystems are critical for biodiversity and carbon storage, and understanding the structural and functional characteristics of trees is essential for sustainable management and conservation. The combination of hyperspectral remote sensing data with LiDAR technology, more specifically Airborne Laser Scanning (ALS), offers a promising approach for tree species mapping and canopy trait validation.

Hyperspectral data provides detailed spectral information that allows for the extraction of key physiological traits, while ALS offers precise structural data on canopy height, density, and biomass. This study focuses on integrating both datasets to improve the accuracy of tree species maps and to validate canopy traits such as leaf area index (LAI), biomass, and chlorophyll content, inferred using Radiative Transfer Models (RTMs).

Background

The integration of hyperspectral data with ALS has the potential to revolutionize forest monitoring and tree species mapping. Hyperspectral sensors are able to capture fine spectral details across a broad range of wavelengths, which are essential for identifying specific tree species and extracting canopy traits. However, the accuracy of these extracted traits, such as biomass and LAI, can be limited by the complexity of canopy structures and the resolution of remote sensing data. ALS, with its ability to precisely measure canopy structure, offers an effective way to validate and refine the canopy traits derived from hyperspectral imagery. This study will explore how integrating these two data sources can enhance tree species classification, improve the retrieval of functional traits, and provide more reliable forest ecosystem maps.

Relevance to research/projects at GRS or other groups

This research aligns with ongoing projects at the Geo-Information Science and Remote Sensing (GRS) group, focusing on the use of remote sensing data for forest monitoring. The combination of hyperspectral data and ALS will support GRS’s goal of advancing forest management techniques, biodiversity monitoring, and carbon flux analysis. Additionally, the Joint Research Center (JRC) is collaborating in this research, leveraging their expertise to enhance forest monitoring through the integration of different remote sensing technologies. This collaboration strengthens the research's potential to contribute to more effective biodiversity monitoring and sustainable forest management practices.

Objectives and Research questions

The objective of this research is to investigate how the integration of hyperspectral and ALS data can improve canopy traits estimation in forest ecosystems and tree species mapping. The study aims to use RTMs to extract tree species-specific physiological traits from hyperspectral data and validate them with structural data from ALS.

How can hyperspectral data and ALS be combined to improve the accuracy of tree species classification in forest ecosystems, incorporating functional plant traits to better understand the biochemical composition?
What are the key canopy plant traits (e.g., LAI, chlorophyll) that can be extracted from hyperspectral data, and how can ALS data be used to validate these traits?
How can RTMs be optimized to better integrate hyperspectral and ALS data for accurate retrieval of main functional traits in trees?

ALS enables accurate individual tree segmentation, can RTMs benefit from an individual tree-based rather than pixel-based approach?

Requirements

Required: GeoScripting, Remote Sensing, Advance Earth Observation
Optional: Spatial Modelling and Statistics, Deep Learning

Literature and information

Savinelli, B., Tagliabue, G., Vignali, L., Garzonio, R., Gentili, R., Panigada, C., & Rossini, M. (2024). Integrating Drone-Based LiDAR and Multispectral Data for Tree Monitoring. Drones, 8(12).
Gregory P. Asner, Susan L. Ustin, Philip Townsend, Roberta. (2024). Forest Biophysical and Biochemical Properties from Hyperspectral and LiDAR Remote Sensing. In P.S. Thenkabail (Ed.), Remote Sensing Handbook, Volume IV: Vol. IV (pp. 96–124). CRC Press.
Dalponte, M., Kallio, A. J. I., Ørka, H. O., Næsset, E., & Gobakken, T. (2022). Wood Decay Detection in Norway Spruce Forests Based on Airborne Hyperspectral and ALS Data. Remote Sensing, 14(8).

Expected reading list before starting the thesis research

Shi, Y., Skidmore, A. K., Wang, T., Holzwarth, S., Heiden, U., Pinnel, N., Zhu, X., & Heurich, M. (2018). Tree species classification using plant functional traits from LiDAR and hyperspectral data. International Journal of Applied Earth Observation and Geoinformation, 73, 207–219.

Xu, L., Shi, S., Gong, W., Shi, Z., Qu, F., Tang, X., Chen, B., & Sun, J. (2022). Improving leaf chlorophyll content estimation through constrained PROSAIL model from airborne hyperspectral and LiDAR data. International Journal of Applied Earth Observation and Geoinformation, 115.

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