Date of Award
Campus Access Dissertation
Doctor of Philosophy (PhD)
Environmental Sciences/Environmental, Earth & Ocean Sciences
Robert F. Chen
This study is dedicated to address several issues and the challenges in ocean color remote sensing. The objects of this study include: (1) investigation of the ship perturbation for in situ optical measurements; (2) investigation of the shading error of SBA system and the correction of its shading error; and (3) an innovative progressively image process scheme for retrieving bathymetry from high spatial resolution ocean color satellite products.
The water color is an indicator of water properties and it could be observed remotely even from the space. From the satellites in the space, the water color information at large spatial scale could be acquired swiftly. Because of the dynamic of water, the investigation of global water requires large spatial coverage and short revisit period. Ocean color remote sensing is the only method that could satisfy such requirements. This study covers the issues of ocean color remote sensing specifically at in situ radiometric measurements and shallow water remote sensing.
In situ radiometric measurements is an essential part in ocean color remote sensing and provides data for calibration and algorithm developments. Ship perturbation and self-shading error are the two issues that bring uncertainties to in situ measurements. Thus we exam the ship perturbation and the self-shading error (with skylight blocked approach (SBA) as demonstration) by Monte-Carlo simulations based on radiative transfer model. After that, from the simulations, a practical guidance is given about how to limit ship perturbation. Besides, built on simulated results, a scheme is developed to correct the self-shading error of SBA.
Shallow water remote sensing requires data with high spatial resolution to capture spatial variances. Besides that, because of the complex optical environment, to achieve a reliable retrieval of water and bottom properties from a single pixel, sufficient (i.e., more than 7 bands) spectral information is required. However, none of the ocean color satellite satisfies these two simultaneously. By incorporating spatial correlation at the image, an progressively image processing (PIP) scheme is proposed. For high spatial resolution data, a significant improvement in bathymetry retrieval is found compared with the traditional method that ignoring spatial correlation.
Shang, Zhehai, "Challenges and Solutions of Ocean Color Remote Sensing: In Situ Radiometric Measurements and Shallow Water Inversion" (2019). Graduate Doctoral Dissertations. 808.