A Monte Carlo study of the seagrass-induced depth bias in bathymetric lidar.

A bathymetric lidar survey is the most cost efficient method of producing bathymetric maps in near shore areas where the ocean bottom is both highly variable and of greatest importance for shipping and recreation. So far, not much attention has been paid to the influence of bottom materials on the lidar signals. This study addresses this issue using a Monte Carlo modeling technique. The Monte Carlo simulation includes a plane parallel water body and a flat bottom with or without seagrass. The seagrass canopy structure is adopted from Zimmerman (2003). Both the surface of the seagrass leaves and the bottom are assumed to be Lambertian. Convolution with the lidar pulse function followed by the median operator is used to reduce the variance of the resultant lidar waveform. Two seagrass orientation arrangements are modeled: seagrass in still water with random leaf orientation and seagrass with a uniform orientation as would be expected when under the influence of a water current. For each case, two maximum canopy heights, 0.5 m and 1 m, three shoot densities, 100, 500, and 1000, and three bending angles, 5, 25, and 45 degrees, are considered. The seagrass is found to induce a depth bias that is proportional to an effective leaf area index (eLAI) and the contrast in reflectance between the seagrass and the bottom material.