A decision support system to predict mortality events in finfish aquaculture

Author:

Howarth LM1,Knight BR12,Symonds JE1,Waddington Z3,Davidson I1

Affiliation:

1. Cawthron Institute, Nelson 7010, New Zealand

2. Deakin University, Geelong, Victoria 3220, Australia

3. New Zealand King Salmon, Picton 7220, New Zealand

Abstract

The aquaculture industry can be impacted by mortality events triggered by marine heatwaves, pathogens, and other environmental factors. Aquaculture managers would benefit from advanced warning of mortality events so they can make decisions to maximise production and profitability. To help monitor fish health and performance, finfish farms are often equipped with an array of cameras and environmental sensors. However, analysing and interpreting all this information can be difficult. Decision support systems (DSSs) can help by simplifying multiple data sources into a single output for quick interpretation and action. Here, we present a DSS capable of providing salmon farmers with 4 wk warning of an impending mortality event. This DSS was trained on a suite of data routinely collected by New Zealand salmon farmers and provides an alert if weekly mortality is predicted to exceed 0.5%. In the final model, present mortality, water temperature, and standardised feeding rate were all found to be significantly correlated with the probability of a future mortality event. The model performed well when tested on data not included in the model-building process, suggesting that the DSS could be useful to farm managers. This study shows that even limited information can be used to construct a DSS capable of providing some advanced warning of elevated mortality risk. Given the ease with which DSSs can be adapted to ingest and predict other parameters, we see strong potential for future development and adoption of these tools by the aquaculture industry and other sectors.

Publisher

Inter-Research Science Center

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.7亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2025 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3