DEVELOPMENT OF A MULTIWAVELENGTH AIRBORNE POLARIMETRIC LIDAR FOR VEGETATION REMOTE SENSING
Date of this Version
A Multiwavelength Airborne Polarimetric Lidar (MAPL) system has been developed for vegetation remote sensing purposes to support the Airborne Remote Sensing Program at the University of Nebraska. The MAPL design and instrumentation are described in detail. Characteristics of the MAPL system include dual-wavelength detection, lidar waveform capture, and polarimetric measurement, which provide enhanced opportunities for vegetation remote sensing, compared to current sensors. Vegetation canopy models, for a single tree and for the three dimensional forests, are developed for airborne lidar applications; and efforts are made towards modeling the vegetation lidar waveform. This study also develops the theoretical analysis for the vegetation canopy lidar, by combining the lidar equation with the canopy models. The noise sources that may affect the performance of the MAPL are discussed and the ground signal to noise ratio (SNR) under the vegetation cover conditions is also investigated. The optical alignment of the MAPL system is briefly introduced. Polarimetric calibration of the system by laboratory measurements of the Stokes parameters of various materials is discussed. Range detection ability was field tested and calibrated. To validate the system's ability for vegetation canopy detection, extended ground tests were performed on the separate trees and the forest detection. Efforts were also made toward a tree species discrimination algorithm applying the MAPL data. Backscattered polarimetric returns, the depolarization ratios, polarimetric normalized difference vegetation indices, and polarimetric band ratios are presented in this dissertation; many of them arc obtained for the first time.
Advisor: Ram M. Narayanan