|
In general, hyper-spectral sensor, LiDAR and high spatial resolution satellite imagery for underwater monitoring are
dependent on water clarity or water transparency that can be measured using a Secchi disk or satellite ocean color data.
Optical properties in the sea waters of South Korea are influenced mainly by a strong tide and oceanic currents, diurnal,
daily and seasonal variations of water transparency. The satellite-based Secchi depth (ZSD) analysis showed the
applicability of hyper-spectral sensor, LiDAR and optical satellite, determined by the location connected with the local
distribution of Case 1 and 2 waters. The southeast coastal areas of Jeju Island are selected as test sites for a combined
underwater experiment, because those areas represent Case 1 water. Study area is a small port (<15m) in the southeast
area of the island and linear underwater target used by sewage pipe is located in this area. Our experiments are as
follows: 1. atmospheric and sun-glint correction methods to improve the underwater monitoring ability; 2. intercomparison
of water depths obtained from three different sensors. Three sensors used here are the CASI-1500
(Wide‐Array Airborne Hyperspectral VNIR Imager (0.38-1.05 microns), the Coastal Zone Mapping and Imaging Lidar
(CZMIL) and Korean Multi-purpose Satellite-3 (KOMPSAT-3) with 2.8 meter multi-spectral resolution. The
experimental results were affected by water clarity and surface condition, and the bathymetric results of three sensors
show some differences caused by sensor-itself, bathymetric algorithm and tide level. It is shown that CASI-1500 was
applicable for bathymetry and underwater target detection in this area, but KOMPSAT-3 should be improved for Case 1
water. Although this experiment was designed to compare underwater monitoring ability of LIDAR, CASI-1500,
KOMPSAT-3 data, this paper was based on initial results and suggested only results about the bathymetry and
underwater target detection.
|
