Share Email Print
cover

Proceedings Paper

Comparison of leaf-on and leaf-off ALS data for mapping riparian tree species
Author(s): Marianne Laslier; Antoine Ba; Laurence Hubert-Moy; Simon Dufour
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Forest species composition is a fundamental indicator of forest study and management. However, describing forest species composition at large scales and of highly diverse populations remains an issue for which remote sensing can provide significant contribution, in particular, Airborne Laser Scanning (ALS) data. Riparian corridors are good examples of highly valuable ecosystems, with high species richness and large surface areas that can be time consuming and expensive to monitor with in situ measurements. Remote sensing could be useful to study them, but few studies have focused on monitoring riparian tree species using ALS data. This study aimed to determine which metrics derived from ALS data are best suited to identify and map riparian tree species. We acquired very high density leaf-on and leaf-off ALS data along the Sélune River (France). In addition, we inventoried eight main riparian deciduous tree species along the study site. After manual segmentation of the inventoried trees, we extracted 68 morphological and structural metrics from both leaf-on and leaf-off ALS point clouds. Some of these metrics were then selected using Sequential Forward Selection (SFS) algorithm. Support Vector Machine (SVM) classification results showed good accuracy with 7 metrics (0.77). Both leaf-on and leafoff metrics were kept as important metrics for distinguishing tree species. Results demonstrate the ability of 3D information derived from high density ALS data to identify riparian tree species using external and internal structural metrics. They also highlight the complementarity of leaf-on and leaf-off Lidar data for distinguishing riparian tree species.

Paper Details

Date Published: 2 November 2017
PDF: 13 pages
Proc. SPIE 10421, Remote Sensing for Agriculture, Ecosystems, and Hydrology XIX, 104210X (2 November 2017); doi: 10.1117/12.2278424
Show Author Affiliations
Marianne Laslier, LETG-RENNES COSTEL, CNRS (France)
Univ. Rennes 2 (France)
Antoine Ba, LETG-RENNES COSTEL, CNRS (France)
Univ. Rennes 2 (France)
Laurence Hubert-Moy, LETG-RENNES COSTEL, CNRS (France)
Univ. Rennes 2 (France)
Simon Dufour, LETG-RENNES COSTEL, CNRS (France)
Univ. Rennes 2 (France)


Published in SPIE Proceedings Vol. 10421:
Remote Sensing for Agriculture, Ecosystems, and Hydrology XIX
Christopher M. U. Neale; Antonino Maltese, Editor(s)

© SPIE. Terms of Use
Back to Top