A Review of Recent Machine Learning Advances for Forecasting Harmful Algal Blooms and Shellfish Contamination

Susana Vinga,Marta B. Lopes,Pedro Reis Costa,Rafaela C. Cruz,Ludwig Krippahl

doi:10.3390/jmse9030283

Abstract

Harmful algal blooms (HABs) are among the most severe ecological marine problems worldwide. Under favorable climate and oceanographic conditions, toxin-producing microalgae species may proliferate, reach increasingly high cell concentrations in seawater, accumulate in shellfish, and threaten the health of seafood consumers. There is an urgent need for the development of effective tools to help shellfish farmers to cope and anticipate HAB events and shellfish contamination, which frequently leads to significant negative economic impacts. Statistical and machine learning forecasting tools have been developed in an attempt to better inform the shellfish industry to limit damages, improve mitigation measures and reduce production losses. This study presents a synoptic review covering the trends in machine learning methods for predicting HABs and shellfish biotoxin contamination, with a particular focus on autoregressive models, support vector machines, random forest, probabilistic graphical models, and artificial neural networks (ANN). Most efforts have been attempted to forecast HABs based on models of increased complexity over the years, coupled with increased multi-source data availability, with ANN architectures in the forefront to model these events. The purpose of this review is to help defining machine learning-based strategies to support shellfish industry to manage their harvesting/production, and decision making by governmental agencies with environmental responsibilities.

Highlights

Shellfish farming and shellfish harvesting from natural seed banks have been growing in recent years as a response to the increasing worldwide demand for seafood products [1]
The results showed that the proposed autoregressive integrated moving average (ARIMA)-deep belief network (DBN) model could reliably forecast the future trend of algal biomass, outperforming several other models, such as DBN, feed-forward neural network (FFNN; method described below), and ARIMA combined with an FFNN
Harmful algal blooms (HABs) are a natural phenomenon known from ancient times, their modeling and prediction remain a challenging task

Summary

Introduction

Shellfish farming and shellfish harvesting from natural seed banks have been growing in recent years as a response to the increasing worldwide demand for seafood products [1]. The frequency and intensity of HAB events are increasing in response to global warming and other climate change conditions, with consequences to environment and shellfish safety. During these events, shellfish may accumulate high concentrations of natural toxins and remain contaminated and unsafe for human consumption, jeopardizing shellfish farming [4]. In the particular context of HAB and shellfish contamination forecasting, the goal is to make use of available time-series data via statistical and machine learning methods to make predictions on future events and guide decision on harvesting permissions (Figure 1). The following review covers past and current directions in time-series forecasting of algal bloom events and shellfish contamination by marine biotoxin. These models have the ability to handle big data and are flexible in selecting/excluding features

Objectives

Methods

Findings

Conclusion