Title :
Characterising Reedbeds Using LiDAR Data: Potential and Limitations
Author :
Onojeghuo, A.O. ; Blackburn, G.A.
Author_Institution :
Lancaster Environ. Centre, Lancaster Univ., Lancaster, UK
Abstract :
Reedbeds are dominated by a small number of plant species, but are extremely valuable habitats for faunal biodiversity. However, reedbeds often exist in small patches distributed across landscapes and for most regions there is a lack of information about their location and condition. This paper investigates the potential of using LiDAR-derived elevation and intensity data to characterise reedbeds. A Leica ALS50 was used to acquire data for reedbeds during the leaf-off phenological period and the study site encompassed a wide range of canopy development. For reedbeds there was a lack of multiple LiDAR returns and ground returns, which limited the ability to acquire information on canopy structure or terrain elevation. Nevertheless, the first return LiDAR data was able to generate an accurate digital surface model and subsequent canopy height model, as validated using field measurements (RMSE 0.47 m; average difference 0.09 m (5% of average height)). LiDAR intensity data displayed specular reflection effects within reedbed areas, but off-nadir imagery was successfully used for mapping reedbeds, non-reedbed vegetation and water bodies.
Keywords :
lakes; optical radar; remote sensing by laser beam; vegetation mapping; Leica ALS50; canopy height model; digital surface model; faunal biodiversity; first return lidar data; leaf off phenological period; lidar derived elevation data; lidar derived intensity data; lidar intensity data; plant species; reedbed characterisation; Accuracy; Data models; Global Positioning System; Laser radar; Lasers; Remote sensing; Vegetation mapping; Canopy height model; LiDAR; Reedbed; digital surface model; intensity; land cover map;
Journal_Title :
Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of
DOI :
10.1109/JSTARS.2012.2212235