Elastic and inelastic LiDAR pulse return phenomenology in coastal underwater biological substrates

Matthieu Huot^1,2,3*Fraser Dalgleish⁴Michel Piché⁵Philippe Archambault^1,2,3

• ¹Laval University, Quebec, Canada
• ²Takuvik Joint International Laboratory, Department of Biology, Faculty of Science and Engineering, Laval University, Quebec City, Quebec, Canada
• ³Québec Océan, Laval University, Québec, Quebec, Canada
• ⁴BeamSea Associates, Loxahatchee, United States
• ⁵Centre d'optique, photonique, et lasers, Laval University, Quebec, Quebec, Canada

In the context of current and future climate-related environmental changes, the development of innovative underwater substrate detection, classification and imaging methods at large spatial scales is key in monitoring and understanding changes from stresses occurring in coastal ocean areas. This development will help understand the spatial distribution and abundance patterns of marine primary producers and ecosystem service providers such as macroalgae, eelgrass and other important ecosystem components such as coral, and can provide insights into future ecosystem response and better management practices. The objective of the current work is to describe an analysis of data acquired by full waveform underwater fluorescence LiDAR, designed for detecting, imaging, and generating 3D point clouds of inert and biological substrates capable of fluorescence. Since the instrument is designed as a small form-factor AUV payload operating at standoff distances of 5-10 m, we chose to implement full-waveform (2.5 Gs/s), pulsed 532 nm laser, capable of generating 1 ns pulses of up to 2.5 uJ at a 200 kHz repetition rate to generate elastic (532 nm) and inelastic (685 nm) 3D point clouds for underwater benthic mapping. Analysis of these acquired waveforms has shown opportunities for improving the point cloud density, by identifying multiple returns within the same waveform, when present. Pulse return processing methods such as Gaussian decomposition and Richardson-Lucy deconvolution are evaluated on data acquired during LiDAR sea-trials over various bottom substrates. As the present LiDAR beam footprint is relatively small to maximize energy density for longer range detection and potential fluorescence response, the number of detected returns per pulse ranges from one in the case of a bare benthic substrate and up to 2 or 3, in areas where for example macroalgae, kelp, corals and/or other substrates characterized by a vertical structure are present.