大西洋鲑循环水养殖系统弧菌总数快速预测模型  

Dynamics and predictive modeling of Vibrio spp. in a recirculating aquaculture system for Atlantic salmon(Salmo salar L.)

作　　者：傅松哲[1] 涂俊凌[1] 夏斌[1] 李贤[2] 刘鹰[2] 

机构地区：[1]南昌市疾病预防控制中心,江西南昌330038 [2]中国科学院海洋研究所,山东青岛266071

出　　处：《中国水产科学》2015年第2期269-277,共9页Journal of Fishery Sciences of China

基　　金：国家科技支撑计划课题(2011BAD13B04);公益性行业(农业)科研专项经费项目(201003024);江西省青年科学基金(20132BAB215027)

摘　　要：为探明弧菌总数与水质参数之间的定量关系,对大西洋鲑循环水养殖系统中弧菌总数和水质参数的动态变化进行了为期12个月的系统监测。采用多元线性回归法对弧菌总数与水质参数进行了相关性分析,并通过蒙特卡罗模拟法模拟了关键水质参数随机性变化对弧菌总数的影响。结果表明,温度,盐度和化学需氧量(COD)对弧菌总数有显著影响(P<0.01)。采用多元线性回归法建立了温度,COD,盐度和弧菌总数之间关系的数学表达式:弧菌总数对数值=–7.24+0.355×COD+0.323×温度+0.129×盐度(R2=0.694,P<0.01)。通过蒙特卡洛模拟验证了温度,COD,盐度和弧菌总数动态变化的相关性,结果表明,COD与弧菌总数的动态变化最为相关(R2=0.756),COD是大西洋鲑循环水养殖系统中促进弧菌生长的最大风险因素。为了验证预测模型的有效性,对大西洋鲑循环水养殖系统中弧菌总数和水质参数进行了12个月的测定,并将S–T模型的预测值与本研究的预测值进行了比较。结果表明,本研究模型与观测数据基本吻合(P<0.01)。多元线性回归模型和蒙特卡洛模拟法可以作为大西洋鲑循环水养殖系统快速评估弧菌数量的重要工具。Aquaculture is one of the world＇s fastest growing food production sectors. The outbreak of infectious bacterial diseases in aquaculture is a major concern in the industry. Among the pathogens causing these diseases, Vibrio species are responsible for several diseases in cultured animals. Models describing the growth and survival of pathogenic Vibrio spp. populations represent a promising tool for improving predictions of a Vibrio outbreak. Traditionally, the control of vibriosis involved the indiscriminant use of antibiotics at high concentrations, resulting in negative effects on aquaculture ecosystems and development of antibiotic-resistant pathogens. Thus, there is an urgent need to develop alternative techniques to prevent a Vibrio outbreaks in aquaculture. To describe the dynamics of Vibrio spp. populations and conduct a risk assessment, the concentration of Vibrio spp. and a range of microbial, physical, and chemical indices were monitored every week for 12 months in a recirculating aquaculture system（RAS） containing Atlantic salmon（Salmo salar L.）. A Monte Carlo simulation was used to model the effects of environmental factors on the concentration of Vibrio spp. This allowed characterization of the likelihood of harm resulting from the pathogen. In this simulation, uncertainty about the concentration of Vibrio spp. was modelled with a PERT distribution. Tornado sensitivity analysis, used to compare the relative importance of variables, was conducted in Model Risk 4.0. Multiple regression analysis revealed a positive relationship to temperature, chemical oxygen demand（COD）, and salinity, and yielded a good fit to the observed data（R2=0.694）. A stepwise multiple regression yielded the following formula： lg（Vibrio number）= –7.24＋0.355× COD＋0.323×temperature＋0.129×salinity. The model basically reflected the dynamic trend of Vibrio spp. abundance at the sampling site. The predicted number of Vibrio spp. agreed well with the observed data. The predicted time of peak abundance w

关 键 词：弧菌总数(TNV) 预测模型 多元线性回归 蒙特卡罗模拟 

分 类 号：S94[农业科学—水产养殖]