A Machine Learning Approach to Produce a Continuous Solar‐Induced Chlorophyll Fluorescence Over the Arctic Ocean

• Published in: Journal of geophysical research. Machine learning and computation Vol. 1; no. 4
• Main Authors: Madani, Nima, Parazoo, Nicholas C., Manizza, Manfredi, Chatterjee, Abhishek, Carroll, Dustin, Menemenlis, Dimitris, Le Fouest, Vincent, Matsuoka, Atsushi, Luis, Kelly M., Serra‐Pompei, Camila, Miller, Charles E.
• Format: Journal Article
• Language: English
• Published: American Geophysical Union/Wiley 01.12.2024; Wiley
• Subjects: Ocean, Atmosphere; Sciences of the Universe
• ISSN: 2993-5210; 2993-5210
• DOI: 10.1029/2024JH000215

Phytoplankton primary production is a crucial component of Arctic Ocean (AO) biogeochemistry, playing a pivotal role in carbon cycling by supporting higher trophic levels and removing atmospheric carbon dioxide. The advent of satellite observations measuring chlorophyll a concentration (Chl_a) has provided unprecedented insights into the distribution of AO phytoplankton, enhancing our ability to assess oceanic net primary production (NPP). However, the optical properties of AO waters differ significantly from those of the lower‐latitude waters, complicating remotely sensed Chl_a retrievals. To mitigate these deficiencies, solar‐induced chlorophyll fluorescence (SIF) has emerged as a valuable tool for gaining physiological insights into the direct photosynthetic processes of the AO. However, the temporal coverage of satellite SIF data makes long‐term analysis of Chl_a photosynthetic activity challenging. In this study, we leverage satellite‐based SIF measurements from 2018 to 2021 to assess their correlation with a set of predictive factors influencing phytoplankton photosynthesis. Generally, observed SIF over the AO showed a higher correlation with normalized fluorescence line height (NFLH) compared to Chl_a. We extended the temporal coverage of the original SIF data to encompass the period from 2004 to 2020. The extended record revealed noticeable differences between SIF, and satellite‐based Chl_a, and NFLH observations. Our novel data set offers a pathway forward to monitor the physiological interactions of phytoplankton with climate changes, promising to significantly improve our understanding of Arctic waters productivity. The application of this data is expected to provide new insights into how phytoplankton respond to environmental shifts, contributing to a more nuanced understanding of their role in high‐latitude marine ecosystems. Plain Language Summary Phytoplankton communities, via means of photosynthesis, play a crucial role in the global carbon cycle by transforming carbon dioxide into organic matter. Recognizing the importance of ocean productivity is essential for effectively managing and conserving marine ecosystems, promoting sustainable fisheries, and comprehending the broader ramifications of climate change on the world's oceans. Alterations in ocean productivity, especially shifts in the abundance and composition of phytoplankton, can serve as early indicators of the health of aquatic ecosystems. While satellite observations have provided an unprecedented overview of phytoplankton distribution by estimating chlorophyll concentrations over the oceans, uncertainties persist regarding the accurate estimation of the total photosynthetic activity of organisms in the ocean. Recently, the TROPOMI satellite instrument has made solar‐induced chlorophyll fluorescence (SIF) data available, offering another metric for understanding photosynthetic activity. However, the short latency of the data record makes it challenging to assess the impact of rapid climate change in the Arctic domain. In this paper, we employ a modeling framework to extend SIF data over a more extended period, facilitating a more comprehensive assessment of ocean productivity. Key Points We extrapolated Arctic Ocean red SIF over the 2004–2020 period using a set of predictive variables that impact marine photosynthesis The reconstructed SIF data demonstrates a strong correlation with independent data records The resulting data are expected to provide new insights into assessments of Arctic Ocean productivity