Imaging Science Thesis Defense: Harnessing airborne and spaceborne LiDAR for large-area assessment of canopy structural complexity
Imaging Science Ph.D. Thesis Defense
Harnessing airborne and spaceborne LiDAR for large-area assessment of canopy structural complexity
Zoom Link here
Tahrir Ibraq Siddiqui
Imaging Science PhD Candidate
Rochester Institute of Technology
Abstract:
Forests play a critical role in carbon cycling, biodiversity conservation, and ecosystem resilience, yet large-area monitoring of ecologically meaningful forest structure remains challenging. This dissertation investigates how airborne and spaceborne light detection and ranging (LiDAR) systems can be used to characterize and map canopy structural complexity (CSC) metrics linked to productivity and other ecosystem functions.
First, simulated airborne laser scanning (ALS) acquisitions over Harvard Forest were used to assess how collection parameters affect CSC retrieval. ALS-derived estimates of outer-canopy complexity were strongly correlated with reference values (R = 0.78–0.86, whereas whole-canopy complexity remained difficult to recover due to limited subcanopy sampling in nadir-oriented scans. Second, high-density ALS data from seven NEON forest sites were used to model plot-level net primary production (NPP). Three ALS-derived CSC metrics explained 77% of NPP variance in deciduous forests and 76% in evergreen forests, demonstrating that airborne LiDAR can provide strong biome-wide predictors of forest production when forest types are modeled separately.
Finally, this dissertation develops deep-learning methods for extending fine-scale CSC retrieval to high-altitude and future spaceborne waveform LiDAR observations. An encoder–decoder Transformer was developed to reconstruct high-resolution canopy profiles from NASA LVIS-F waveforms using co-located ALS data as reference, improving correlations with ALS-derived CSC metrics from R = 0.62 to 0.84 and from R = 0.76 to 0.90. Building on this, a generative diffusion framework integrates LVIS waveforms with optical imagery to reconstruct high-resolution, voxelized 3D forest plots using ALS point-cloud volumes as reference. Together, these contributions advance LiDAR-based approaches for large-area monitoring of forest structure, productivity, and ecosystem function.
Intended Audience: Beginners, undergraduates, graduates. Those with interest in the topic.
To request an interpreter, please visit myaccess.rit.edu
Event Snapshot
When and Where
Who
This is an RIT Only Event
Interpreter Requested?
No