Aquaculture Europe 2026

September 28 - October 1, 2026

Ljubljana, Slovenia

Add To Calendar 01/10/2026 09:45:0001/10/2026 10:00:00Europe/ViennaAquaculture Europe 2026A DECADE OF HABREPORTS: EARLY WARNING OF HARMFUL ALGAL BLOOMS AND BIOTOXINS FOR AQUACULTUREGallery 1The European Aquaculture Societywebmaster@aquaeas.orgfalseDD/MM/YYYYaaVZHLXMfzTRLzDrHmAi181982

A DECADE OF HABREPORTS: EARLY WARNING OF HARMFUL ALGAL BLOOMS AND BIOTOXINS FOR AQUACULTURE

Keith Davidson1*, Dmitry Aleknik1, Gregg Arthur2, Steve Gontarek1, Gary Groves1, Alan MacDonald1, Tim Szewczyk1, Callum Whyte1

1Scottish Association for Marine Science (SAMS), Oban, PA37 1QA, Scotland, UK

2University of the Highlands and Islands Shetland, Port Arthur, Scalloway, Shetland, ZE1 0PX, Scotland, UK

Email: keith.davidson@sams.ac.uk

 



Introduction

Harmful algal blooms (HABs) are increases in the density of certain phytoplankton species that are harmful to humans or our use of the marine environment. In many locations, these blooms are of particular concern to the finfish and shellfish aquaculture industries. The phytoplankton genera that are detrimental to these two related sectors are, typically, distinct. Some HAB species cause "shellfish poisoning" that results from the human consumption of shellfish that have ingested toxic cells and then bio-accumulated the toxin within their flesh. Farmed fish are impacted, in general, by different HAB species through physical interference, deoxygenation or ichthyotoxicity.

The negative impacts of a human poisonng event on the shellfish industry can be far reaching. The economic consequences of HABs for the finfish industry can be very high, for example a a massive fish kill in Chile in 2016 following a bloom of the dictyochophyte Pseudochattonella caused the mortality of 39 million salmon with a cost of $ (USD) 800 million.

Early warning of the timing, location and magnitude of HABs and their associated biotoxins is therefore of great importance to the aquaculture industry to allow mitigation to be put in place.

Methods, results & discussion

Since 2015 SAMS has produced early warning and risk assessments of HABs and associated biotoxins in Scottish waters for the shellfish and finfish aquaculture industries. Since 2017 these have been provided on the www.HABreports.org web site (Davidson et al. 2021, Figure 1), with a parallel Malaysian (My-HABs) alerts system being developed in 2021/2.

This presentation will discuss the development and content of HABreports including its traffic light alerts, and development of a mobile phone-based App for those users who do not routinely sit at a desktop.

An important component of our alert system is mathematical models. Lagrangian based unstructured grid (FVCOM) based models are used to simulate the spatial and temporal development of existing blooms to evaluate if and when aquaculture sites may be impacted. In addition, statistical ensemble models using Bayesian and other maching leading techiques are being used within HABreports to predict the onset of blooms at aquaculture prior to their occurrence (Szewczyk et al. 2025).

High resolution and taxonomically detailed monitoring of HAB taxa is crucial to early warning and risk assessment. Recent developments of our alert system have therefore incorporated automated phytoplankton imaging using two Imaging FlowCytobots sited in the aquaculture intensive Shetland Islands. AI based analysis of these image data using a neural network based on the Multi-Axis Vision Transformer allows real time reporting of HAB taxa at a temporal frequency of less than an hour (Groves et al. 2026).

The HABreports system therefore provides rapid, spatially and temporally resolved, monitoring and model based information that informs husbandary practices and business planning within the aquaculture industry protecting both human and fish health.

Figure 1: screen grab of www.HABreports.org web site

Acknowledgment

We thank our funders and partners including UKRI, SAIC, Food Standards Scotland, Cefas, The Marine Directorate of the Scottish Government, The Data Lab, Scottish Sea Farms, MOWI, Bakkafrost and the EU Horizon Europe project OCCAM.

References

Groves GJJ, Arthur G, Bresnan E, Whyte C, Arce P, Davidson K (2026), Automatic enumeration of chains of marine diatoms using "You Only Look Once"—a machine learning approach, Journal of Plankton Research, 48, https://doi.org/10.1093/plankt/fbaf064

Szewczyk TM, Aleynik D, Davidson K (2025) Ensemble models improve near-term forecasts of harmful algal bloom and biotoxin risk, Harmful Algae, 142, 102781, https://doi.org/10.1016/j.hal.2024.102781.

Davidson K, Whyte C, Aleynik D, Kurekin A, Gontarek S, McNeill S, Miller PI, Porter M, Saxon R, Swan S (2021) HABreports: Online early warning of harmful algal and biotoxin risk for the Scottish shellfish and finfish aquaculture industries. Frontiers in Marine Science https://doi.org/10.3389/fmars.2021.631732