SALPA: Spaceborne LiDAR Point Adjustment for Enhanced GEDI Footprint Geolocation

2025-11-17

AI Summary

Large geolocation errors (5–15 m) in spaceborne LiDAR footprints—e.g., NASA’s GEDI—significantly limit the accuracy of forest structure and carbon stock estimation. To address this, we propose SALPA, a multi-paradigm optimization framework for robust georegistration. SALPA integrates gradient-based optimization (L-BFGS-B), evolutionary computation (genetic algorithms), and swarm intelligence (particle swarm optimization), coupled with five distance metrics—including a novel area-weighted metric—leveraging only globally available digital elevation models and geoid data. It operates in a continuous solution space without requiring high-resolution auxiliary datasets. On complex and flat terrain, SALPA improves GEDI geolocation accuracy by 15–16% over baseline, outperforming the state-of-the-art GeoGEDI by an additional 0.5–2%. The framework is platform-agnostic and establishes a reusable, open-data-driven geolocation correction paradigm for spaceborne LiDAR remote sensing.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services

Abstract

Spaceborne Light Detection and Ranging (LiDAR) systems, such as NASA's Global Ecosystem Dynamics Investigation (GEDI), provide forest structure for global carbon assessments. However, geolocation uncertainties (typically 5-15 m) propagate systematically through derived products, undermining forest profile estimates, including carbon stock assessments. Existing correction methods face critical limitations: waveform simulation approaches achieve meter-level accuracy but require high-resolution LiDAR data unavailable in most regions, while terrain-based methods employ deterministic grid searches that may overlook optimal solutions in continuous solution spaces. We present SALPA (Spaceborne LiDAR Point Adjustment), a multi-algorithm optimization framework integrating three optimization paradigms with five distance metrics. Operating exclusively with globally available digital elevation models and geoid data, SALPA explores continuous solution spaces through gradient-based, evolutionary, and swarm intelligence approaches. Validation across contrasting sites: topographically complex Nikko, Japan, and flat Landes, France, demonstrates 15-16% improvements over original GEDI positions and 0.5-2% improvements over the state-of-the-art GeoGEDI algorithm. L-BFGS-B with Area-based metrics achieves optimal accuracy-efficiency trade-offs, while population-based algorithms (genetic algorithms, particle swarm optimization) excel in complex terrain. The platform-agnostic framework facilitates straightforward adaptation to emerging spaceborne LiDAR missions, providing a generalizable foundation for universal geolocation correction essential for reliable global forest monitoring and climate policy decisions.

Problem

• Corrects geolocation uncertainties in spaceborne LiDAR forest data
• Overcomes limitations of existing waveform and terrain correction methods
• Enhances global forest monitoring accuracy for climate policy decisions

Innovation

• Multi-algorithm optimization framework for geolocation correction
• Uses gradient-based evolutionary and swarm intelligence approaches
• Operates with globally available elevation and geoid data