POSTER SESSION 4
10:30–12:30, Thursday, September 17
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
ABSTRACT 1251 | POSTER 100
HYPERSPECTRAL GAINS IN OCEAN-COLOUR REMOTE SENSING: KD RETRIEVAL WITH THE PACE OCEAN COLOR INSTRUMENT
Different satellite missions have employed a wide range of passive ocean-colour sensors, with varying numbers, positions, and widths of spectral bands. These differences directly impact the performance of bio-optical algorithms, including those used to retrieve the diffuse attenuation coefficient for downwelling plane irradiance (Kd). In this study, we evaluate a new algorithm for estimating Kd averaged over the first attenuation depth (<Kd>1), developed for the NASA Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission. The algorithm builds on the neural network framework of Jorge et al. (Remote Sens. Environ., 263, 2021), and takes advantage of the increased spectral resolution of the PACE Ocean Color Instrument (OCI), using 11 input bands compared to 5 input bands in heritage multispectral configurations. We compare the PACE implementation with versions adapted for OLCI and MODIS sensors to assess the impact of increased spectral information on Kd retrieval. Particular attention is given to spectral regions where limited band coverage in multispectral sensors led to reduced representation of Kd variability. Performance is evaluated using both newly available PACE observations and hyperspectral in situ measurements. This analysis provides insight into the benefits of hyperspectral observations for improving Kd retrieval and advancing ocean-colour applications.
Daniel Schaffer Ferreira Jorge, LOG – Laboratoire d’Océanologie et de Géosciences, [email protected], http://orcid.org/0000-0002-4812-9239
Cedric Jamet, LOG – Laboratoire d’Océanologie et de Géosciences, [email protected]
Hubert Loisel, LOG – Laboratoire d’Océanologie et de Géosciences, [email protected]
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
