Considerations for Achieving Cross-Platform Point Cloud Data Fusion across Different Dryland Ecosystem Structural...

Area of Science:

• Ecological monitoring
• Remote sensing
• Geospatial analysis

Background:

• Dryland ecosystems require high-resolution, multi-temporal remote sensing for vegetation analysis.
• Existing 3D remote sensing techniques (lidar, SfM photogrammetry) have limitations in vegetation volume and extent detection.
• Combining different platforms and techniques may overcome individual method weaknesses.

Purpose of the Study:

• To compare detection biases of lidar and structure from motion (SfM) photogrammetry.
• To evaluate these techniques across ground-based, small unmanned aerial systems (sUAS), and manned aircraft platforms.
• To assess the potential of combining platforms for improved vegetation measurement.

Main Methods:

• Comparison of two 3D remote sensing techniques: lidar and SfM photogrammetry.
• Deployment across three platforms: ground-based, sUAS, and manned aircraft.
• Analysis of data for bare earth characterization and vegetation (grass, shrubs) detection and height measurement.

Main Results:

• Aerial lidar excelled at bare earth mapping in dense herbaceous cover but missed fine vegetation.
• Terrestrial lidar and high-resolution near-distance SfM (ground, sUAS) accurately detected fine vegetation.
• Combined data from multiple platforms yielded more accurate vegetation measurements (height, canopy cover) than single techniques.

Conclusions:

• No single 3D remote sensing technique or platform is optimal for all dryland vegetation monitoring tasks.
• Combining lidar and SfM data from various platforms enhances accuracy for both bare earth and vegetation characterization.
• SfM photogrammetry, especially from sUAS, offers cost-effective, repeatable monitoring, though post-processing can be a bottleneck for large areas.