Enhancing downwelling diffuse attenuation coefficient retrieval at 490 nm through a Neural network and water classification system

The downwelling diffuse attenuation coefficient (Kd) is a key optical parameter for ocean color science, yet accurately estimating Kd across diverse water types remains a challenge due to variations in water constituents. To address this, the study developed a neural network-based method (Kd_NN) to...

Full description

Saved in:
Bibliographic Details
Main Authors: Tianle Yao, Tianjie Zhan, Jilin Men, Lan Zhang, Liqiao Tian
Format: Article
Language:English
Published: Taylor & Francis Group 2025-07-01
Series:Geo-spatial Information Science
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/10095020.2025.2514177
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The downwelling diffuse attenuation coefficient (Kd) is a key optical parameter for ocean color science, yet accurately estimating Kd across diverse water types remains a challenge due to variations in water constituents. To address this, the study developed a neural network-based method (Kd_NN) to directly estimate the diffuse attenuation coefficient at 490 nm (Kd(490)) from remote sensing reflectance (Rrs). Kd_NN was trained with globally collected Kd(490) measurements, representing a wide range of bio-optical conditions. Notably, all measurements were classified into 30 water types and quality-controlled using OC_3S to ensure the quality of the training dataset. The results show that Kd_NN delivers robust performance across various water types, reducing the Mean Absolute Percentage Error (MAPE) by approximately 10% for most water types. Spatio-temporal analysis further reveals that Kd_NN effectively addresses the underestimation issue observed in NASA’s standard algorithm in coastal waters and captures real oceanic variations regardless of changes in water types. In low- and mid-latitude regions, the Relative Percentage Difference (RPD) between our algorithm and NASA’s remains within ± 5%, whereas in coastal and polar regions the RPD rises to 20% or more, suggesting that the results in these regions need to be revisited. Moreover, an analysis of 20 years of MODIS data demonstrate that Kd_NN offers substantial improvements over NASA’s method, which underestimates Kd(490) by about 33% in China’s coastal waters. Consequently, Kd_NN has the potential to provide high-quality global Kd(490) inversion products, supporting marine ecosystem monitoring and management.
ISSN:1009-5020
1993-5153