PhD Defense: Kamal Aryal
Location
Physics : 401
Date & Time
July 16, 2025, 9:00 am – 11:30 am
Description
ADVISOR: Dr. Pengwang Zhai
TITLE: Advancing Aerosol and Ocean Color Remote Sensing Over Coastal Waters Using Multiangle Polarimetry and Machine Learning
ABSTRACT: Satellite remote sensing of aerosols and ocean color is crucial for advancing our understanding of air quality, ocean ecology, biogeochemical processes and their impacts on climate change. A key challenge in ocean color remote sensing over coastal waters is atmospheric correction (AC) limited by weak aerosol information in traditional spectral radiometric measurements. The measurements from Multi-angle polarimeters (MAPs) offer a promising solution by providing rich information on both aerosol and hydrosol properties. NASA's recent PACE (Plankton, Aerosol, Clouds and ocean Ecosystem) mission carries two MAPs, HARP2 and SPEXone, for global polarimetric observations. However,processing such datasets is challenging due to the high computational cost of radiative transfer (RT) based retrieval algorithms. This limitation can be addressed by using machine learning models as functional replacements to traditional RT solvers.
The first part of this dissertation develops and validates a joint retrieval algorithm called FastMAPOL/component (Fast Multi-Angular Polarimetric Ocean coLor/component) using simulated datasets from a full physics radiative transfer software for coupled atmosphere and ocean system. The algorithm combines MAP measurements for better information content and machine learning for enhanced computational capability. It incorporates bio-optical models of coastal waters, and component based aerosol representation to better capture the coastal marine environment. It also includes modules for AC and Bidirectional reflectance Distribution Function correction to retrieve remote sensing reflectance. Validation results with HARP2-only, SPEXone-only, and combined HARP2+SPEXone simulated data confirm the algorithm’s robustness and highlight the information content of each configuration. The second part evaluates the retrieval performance of FastMAPOL/component using real measurements from the SPEXone instrument. The retrieved aerosol and ocean color products are collocated and compared against the measurements from 9 AERONET-OC stations show strong agreement. The final part of this dissertation presents neural network (NN) models for Instantaneous Photosynthetically Available Radiation (IPAR) at the ocean surface and its subsurface vertical profile. The NN models are integrated into the FastMAPOL/component algorithm to enable direct IPAR retrieval from satellite measurements. Overall this dissertation demonstrates the potential of FastMAPOL/component algorithm as a robust tool for operational aerosol and ocean color remote sensing using MAP data on a global scale.
TITLE: Advancing Aerosol and Ocean Color Remote Sensing Over Coastal Waters Using Multiangle Polarimetry and Machine Learning
ABSTRACT: Satellite remote sensing of aerosols and ocean color is crucial for advancing our understanding of air quality, ocean ecology, biogeochemical processes and their impacts on climate change. A key challenge in ocean color remote sensing over coastal waters is atmospheric correction (AC) limited by weak aerosol information in traditional spectral radiometric measurements. The measurements from Multi-angle polarimeters (MAPs) offer a promising solution by providing rich information on both aerosol and hydrosol properties. NASA's recent PACE (Plankton, Aerosol, Clouds and ocean Ecosystem) mission carries two MAPs, HARP2 and SPEXone, for global polarimetric observations. However,processing such datasets is challenging due to the high computational cost of radiative transfer (RT) based retrieval algorithms. This limitation can be addressed by using machine learning models as functional replacements to traditional RT solvers.
The first part of this dissertation develops and validates a joint retrieval algorithm called FastMAPOL/component (Fast Multi-Angular Polarimetric Ocean coLor/component) using simulated datasets from a full physics radiative transfer software for coupled atmosphere and ocean system. The algorithm combines MAP measurements for better information content and machine learning for enhanced computational capability. It incorporates bio-optical models of coastal waters, and component based aerosol representation to better capture the coastal marine environment. It also includes modules for AC and Bidirectional reflectance Distribution Function correction to retrieve remote sensing reflectance. Validation results with HARP2-only, SPEXone-only, and combined HARP2+SPEXone simulated data confirm the algorithm’s robustness and highlight the information content of each configuration. The second part evaluates the retrieval performance of FastMAPOL/component using real measurements from the SPEXone instrument. The retrieved aerosol and ocean color products are collocated and compared against the measurements from 9 AERONET-OC stations show strong agreement. The final part of this dissertation presents neural network (NN) models for Instantaneous Photosynthetically Available Radiation (IPAR) at the ocean surface and its subsurface vertical profile. The NN models are integrated into the FastMAPOL/component algorithm to enable direct IPAR retrieval from satellite measurements. Overall this dissertation demonstrates the potential of FastMAPOL/component algorithm as a robust tool for operational aerosol and ocean color remote sensing using MAP data on a global scale.
