Parent Marianne I, Stryhn Henrik, Hammell K Larry, Fast Mark D, Vanderstichel Raphaël
Department of Health Management and Centre for Veterinary Epidemiological Research, Atlantic Veterinary College, University of Prince Edward Island, Charlottetown, Prince Edward Island, Canada.
Department of Pathology and Microbiology, Atlantic Veterinary College, University of Prince Edward Island, Charlottetown, Prince Edward Island, Canada.
J Aquat Anim Health. 2024 Dec;36(4):355-373. doi: 10.1002/aah.10235.
The primary objective was to construct a time series model for the abundance of the adult female (AF) sea lice Lepeophtheirus salmonis in Atlantic Salmon Salmo salar farms in the Bay of Fundy, New Brunswick, Canada, for the period 2016-2021 and to illustrate its short-term predictive capabilities.
Sea lice are routinely counted for monitoring purposes, and these data are recorded in the Fish-iTrends database. A multivariable autoregressive linear mixed-effects model (second-order autoregressive structure) was generated with the outcome of the abundance of AF sea lice and included treatments, infestation pressures (a measure that represents the dose of exposure of sea louse parasitic stages to potential fish hosts) within sites (internal) and among sites (external), and other predictors. The treatments were categorized by duration and type.
The effect of mechanical treatments decreased with increasing sea surface temperature. In-sample predictions had good accuracy. A one-standard-deviation increase in the external infestation pressures (EIP) produced a significant relative increase in the abundance of AF sea lice by 5% when other model predictors were kept constant. Sites separated by short seaway distances had stronger EIP than sites with more considerable distances.
This model may be helpful for managers and farmers in implementing sea lice mitigation strategies on salmon farms in the Bay of Fundy.
主要目标是构建一个时间序列模型，用于预测2016 - 2021年加拿大新不伦瑞克省芬迪湾大西洋鲑鱼养殖场成年雌性（AF）海虱鲑鳟鱼虱的数量，并展示其短期预测能力。
为监测目的，定期对海虱进行计数，这些数据记录在Fish-iTrends数据库中。使用AF海虱数量作为结果生成了一个多变量自回归线性混合效应模型（二阶自回归结构），该模型包括处理方式、养殖场内部（场内）和养殖场之间（场外）的感染压力（一种表示海虱寄生阶段对潜在鱼类宿主暴露剂量的指标）以及其他预测因子。处理方式按持续时间和类型进行分类。
随着海面温度升高，机械处理的效果降低。样本内预测具有良好的准确性。当其他模型预测因子保持不变时，场外感染压力（EIP）增加一个标准差会使AF海虱数量显著相对增加5%。海路距离短的养殖场比距离较远的养殖场具有更强的EIP。
该模型可能有助于管理人员和养殖户在芬迪湾的鲑鱼养殖场实施海虱缓解策略。
