Coast-Algae : Deep learning algal bloom mapping using 30-m Harmonized Landsat-Sentinel at global nearshore waters

Explore coastal algal blooms everywhere!

Coastal problem

Algal blooms, defined as a rapid, large-scale accumulation of micro- or macroalgae in the upper water column, can occur across diverse coastal environments, with more than 5,000 species documented globally

Bloom events can emerge abruptly and are projected to increase with climate change, emphasizing the urgent need for a scalable coastal algal bloom monitoring (Anderson et al., 2009, 2012, 2019; Barton et al., 2016; Gobler, 2020; Hallegraeff et al., 2021).

Dai et al. (2023) demonstrated a global-scale bloom mapping using MODIS and unsupervised anomaly detection, which is one of the first studies providing daily bloom-affected area masks

How does it works?

Harmonized Landsat-Sentinel

Integrating both data sensors into a single virtual constellation presents opportunities to enhance the revisit rate and capture the temporal variability of optically active water constituents

Algal Bloom Training dataset

A globally distributed algal bloom patch dataset (24,265 training and 12,255 validation patches, each 30m 256×256 pixels) was generated in coastal algal bloom-prone hotspots worldwide

Vision Transformers

Vision Transformers (ViTs) have gained attention in remote sensing for their ability to model global relationships via self-attention mechanisms, surpassing the locality limitations of CNNs

User-friendly data visualization

This web interface will allow access to all coastal tiles with an easy request of per-pixel time-series bloom count and descriptive statistics, such as monthly and annual occurrence, in gridded information

Outcomes

Publications

Lima, T. M., Martins, V. S., Paulino, R. S., Caballero, C. B., Maciel, D. A., & Giardino, C. (2025). A General Spectral Bandpass Adjustment Function (SBAF) for Harmonizing Landsat-Sentinel over Inland and Coastal Waters. Science of Remote Sensing, 100225.

Lima, T. M., Martins, V. S., Paulino, R. S., Caballero, C. B., Barbosa, C. C., & Ashapure, A. (2025). AQUAVis: Landsat-Sentinel Virtual Constellation of Remote Sensing Reflectance (Rrs) Product for Coastal and Inland Waters. Science of Remote Sensing, 100294.

Portal sample