1 Executive Summary

FastMAPOL (Fast Multi-Angle Polarimetric Ocean and Land) is a nonlinear least-squares retrieval algorithm developed for the NASA PACE mission. It retrieves aerosol and ocean optical properties from multi-angle polarimetric observations acquired by the PACE polarimeters, including HARP2 and SPEXone.

The algorithm uses neural networks to emulate polarized radiative transfer, significantly improving computational efficiency and enabling operational-scale processing. FastMAPOL produces a comprehensive suite of environmental data records along with associated uncertainty estimates.

This document describes the theoretical framework, forward modeling, inversion methodology, and data products generated by FastMAPOL.

2 Introduction

The PACE mission includes three instruments:

• Ocean Color Instrument (OCI)
  
• Hyper-Angular Rainbow Polarimeter-2 (HARP2)
  
• SPEXone polarimeter

Multi-angle polarimetric observations provide enhanced sensitivity to aerosol microphysical properties and ocean optical parameters. FastMAPOL is designed to exploit this information content through a physics-based inversion framework accelerated by machine learning.

3 Scope of Document

This Algorithm Theoretical Basis Document (ATBD) describes the methodology used in FastMAPOL to retrieve aerosol and ocean properties from PACE polarimeter measurements.

The document includes the following components:

• Algorithm architecture and components
  
• Aerosol optical model
  
• Ocean bio-optical model
  
• Land surface model
  
• Neural network forward model
  
• Retrieval methodology and uncertainty quantification
  
• Level-2 data product definition
  
• Validation using HARP2 and SPEXone observations

4 Document Organization

The remainder of this document is organized as follows:

• Section 2 describes the forward radiative transfer model
  
• Section 3 presents the aerosol model
  
• Section 4 describes the ocean bio-optical model
  
• Section 5 introduces the land surface model
  
• Section 6 details the neural network forward model
  
• Section 7 describes the inversion methodology
  
• Section 8 defines the Level-2 data products
  
• Section 9 presents validation results
  
• Appendices provide additional technical details

5 Document Information