Characterization of Ecosystem Structure in Tropical Forests Using Point Clouds Derived from LIDAR and Drone to Support Interpretation of Radar Imaging Data Sets
Abstract
Tropical forests contain diverse habitats with a high degree of endemism, but these structurally complex systems remain difficult to characterize by common biometry and remote sensing methods. Persistent cloud cover over the Neotropics, including the Chocó of South America and the Amazon basin, limits the utility of spaceborne optical remote sensing to obtain timely information on land use change. Rapid changes in land use and climate are fundamentally altering water and carbon cycles and biodiversity in these ecoregions. As a result, the lowland Chocό rainforest remains one of the most poorly characterized tropical biodiversity hotspots in terms species distribution, ecosystem structure, and ecosystem function. Across the Andes mountains, White sands forests in the Amazon basin are loosely delineated unique biogeophysical systems, but support distinct vegetative regimes and ecological communities. Synthetic Aperture Radar (SAR) is an active microwave sensor that provides measures of vegetation structure independent of cloud cover. Deriving and validating quantitative structural measures from such remote sensing data requires statistically-meaningful ground reference data that accurately describes structure across the landscape. We aim to characterize tropical forest ecosystem structure using a combined point cloud approach derived from scanning lidar and aerial drone photogrammetry to support interpretation of radar imaging datasets. Terrestrial Laser Scanning (TLS) backpack mounted scanning lidar provides simultaneous localization and mapping (SLAM) processing to create 3D point clouds from line-scan recordings. Small unmanned aerial vehicles (sUAV) acquire high resolution surface reflectance imagery of forest canopy to produce point clouds from multi-view photogrammetry. Combined forest plot collections from TLS and sUAV represents a novel method to derive traditional biometry measurements for comparative analysis and characterization of Ecosystem Structure, an Essential Biodiversity Variable.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.B11E2382T
- Keywords:
-
- 0439 Ecosystems;
- structure and dynamics;
- BIOGEOSCIENCES;
- 0480 Remote sensing;
- BIOGEOSCIENCES;
- 1294 Instruments and techniques;
- GEODESY AND GRAVITY