Application of dinoflagellate-specific satellite models to aid Alexandrium catenella bloom monitoring in the Bering and Chukchi seas.

Abstract

Harmful algal blooms (HABs) of the toxic dinoflagellate Alexandrium catenella are increasing in the Pacific Arctic due to ocean warming. threatening ecosystems and to coastal communities that rely on marine resources for their subsistence. This study explores the potential of the Sentinel-3 remote sensing reflectance (R_rs(λ)) to detect and quantify dinoflagellate blooms in the Bering and Chukchi seas using an A. catenella cell abundance dataset to regionally parameterize and evaluate new algorithm combinations (color indexes and principal component regression - PCR). The color indexes utilize the fluorescence (FLH), green (GLH), and blue line heights (BLH) and the spectral difference in FLH peak to identify dinoflagellate blooms. The algorithms were parameterized and validated using 45 satellite match-ups with in situ A. catenella abundances measured over summer 2022 in the North Bering and Chukchi seas. Assuming the dinoflagellate bloom is dominated by A. catenella, the dinoflagellate index DINI (GLH-based) and enhanced bloom index EBI (BLH-based) provide reliable cell abundance estimates at concentrations higher than 10,000 (R² = 0.53) and 3000 cells/L (R² = 0.67), respectively. The PCR model resolves estimates at lower cell abundances (>1000 cells/L, R² = 0.68). Despite their higher uncertainty, color index models provide early detection and tracking of potential A. catenella blooms, as demonstrated during Summers 2023 and 2024. By providing timely and accurate information on bloom dynamics, these satellite products can significantly augment HAB monitoring systems in the Pacific Arctic.