Simulation and Optimization-based Method of Environmental Parameter Distribution from Airborne Bathymetric LiDAR Data

A bathymetric LiDAR system's emitted laser pulse is affected by many system specific and environmental variables, including sea water scattering and absorption, seafloor reflectance and rugosity, in-air attenuation, electronic bandwidth and noise, among others. These factors influence the captured, digitized waveform which can be used to estimate seafloor depth and sea water properties. Understanding how these parameters influence bathymetric LiDAR waveforms and extracting estimates of these parameters is important in developing more accurate analysis techniques as well as creating data products that are associated with georectified coordinates. However, estimating more than ten parameters from a collection of waveforms that may or may not have correlated parameters is challenging. This paper details a post-processing technique used to estimate bathymetric environmental parameters through robust simulation of airborne LiDAR data that models laser beam propagation in a specific bathymetric environment. To create an initial waveform, the simulator is seeded with known parameters and with reasonable estimates for unknown parameters. Then, using optimization algorithms, the unknown parameters are iteratively adjusted, creating a waveform that minimizes error between it and an associated truth waveform. By estimating these unknown values for many waveforms within a geographic area, the distributions of the environmental factors can be characterized for future analysis.