Abstract:The Galician estuaries are Europe's foremost supplier of mussels, generating millions of euros annually and offering substantial employment opportunities for its population. One of the most critical threats to shellfish production is the occurrence of harmful algal blooms (HABs), contaminating bivalves with phytotoxins. To successfully tackle this problem, there needs to be a collaborative effort between the scientific community and decision-makers to establish a dynamic and effective monitoring system. This could enable early warnings and preventive actions to avert the loss of millions of tons of shellfish. Remote sensing, despite its limitations, requires commitment and effort by experts to devise effective methods for detecting target optical constituents mixed with other undesired target but that exhibit strong signals. Therefore, the essential necessity arises to identify approaches for mitigating the shortcomings of the undertaken efforts. The objective of this study is to assess the main environmental drivers of potential harmful genera ( Pseudo-nitzschia , Dinophysis , Alexandrium , and Gymnodinium ) in the Rías Baixas from 2015 to 2022, developing regression-based models and customizing Generalized Additive Models (GAMs) to investigate their spatial-temporal dynamics. Risk-susceptible bloom zones were identified in the river mouth of the Ría Pontevedra and from the center to offshore of the Ría Vigo. Early upwelling events triggered peaks in chlorophyll-a (Chl-a), driving Dinophysis grazing on phytoplankton communities mainly dominated by diatoms, flagellates, and ciliates. Subsequently, the upwelling intensity favoring Pseudo-nitzschia or Alexandrium minutum growth as larger diatoms declined. A. minutum exhibited elevated incidence over the past five years. Gymnodinium catenatum presented a rare occurrence across the three studied estuaries, which limited the assessment of its spatial dynamics in the region. This study emphasizes the need to integrate remote sensing evaluation of high-risk bloom areas (July to September), in-situ cell count collection, and enhanced efforts for forecasting future critical occurrences of HABs.

Forecasting harmful algae blooms: Application to Dinophysis acuminata in northern Norway

A demographic model to forecast Dinophysis acuminata harmful algal blooms

Advances in the Early Warning of Shellfish Toxification by Dinophysis acuminata

Seasonal Early Warning of Impacts of Harmful Algal Blooms on Farmed Shellfish in Coastal Waters of Scotland

Algal bloom forecasting with time-frequency analysis: A hybrid deep learning approach.

Deep-Learning-Based Approach for Prediction of Algal Blooms

Machine Learning in Extreme Value Analysis, an Approach to Detecting Harmful Algal Blooms with Long-Term Multisource Satellite Data.

Harmful algal bloom forecasting. A comparison between stream and batch learning

Potentially toxigenic phytoplankton patterns in the northwestern Iberian Peninsula

Predicting coastal algal blooms with environmental factors by machine learning methods

Vertical Development of a Prorocentrum Donghaiense Bloom in the Coastal Waters of the East China Sea: Coupled Biophysical Numerical Modeling

Environmental drivers and prediction of Karenia mikimotoi proliferation in coastal area, Southeast China

Probabilistic Prediction of Algal Blooms from Basic Water Quality Parameters by Bayesian Scale-Mixture of Skew-Normal Model

A Novel Single-Parameter Approach for Forecasting Algal Blooms.

Operational modelling to assess advective harmful algal bloom development and its potential to impact aquaculture

Prediction of harmful algal blooms in large water bodies using the combined EFDC and LSTM models

Employing hybrid deep learning for near-real-time forecasts of sensor-based algal parameters in a Microcystis bloom-dominated lake

A Fully Unsupervised Machine Learning Framework for Algal Bloom Forecasting in Inland Waters Using MODIS Time Series and Climatic Products

Prediction on daily spatial distribution of chlorophyll-a in coastal seas using a synthetic method of remote sensing, machine learning and numerical modeling

Molecular forecasting of domoic acid during a pervasive toxic diatom bloom

Harmful Algal Blooms Prediction with Machine Learning Models in Tolo Harbour