A generalized linear mixed model analysis of a multi-vessel fishery resource survey

A generalized linear mixed model (GLMM) that treats year and spatial cell as fixed effects while treating vessel as a random effect is used to examine fishing power among chartered industry-based vessels and a research trawler, the FRV Miller Freeman, for bottom trawl surveys on the upper continental slope of U.S. West coast. A Bernoulli distribution is used to model the probability of a non-zero haul and the gamma distribution to model the non-zero catch rates of four groundfish species. The use of vessel as a random effect allows the data for the various vessels to be combined and a single continuous time-series of biomass indices to be developed for stock assessment purposes. The GLMMs fit the data reasonably well. Among the different models examined, the GLMM incorporating a random vessel × year effect had the smallest ΔAIC and was thus chosen as the best model. Also, estimated random effects coefficients associated with the industry-based vessels and the FRV Miller Freeman for each year suggests that these vessels can be assumed to be from a common random effects distribution. These results suggest that combining data from the chartered industry-based vessels and from the research trawler may be appropriate to develop indices of abundance for stock assessment purposes. Finally, an evaluation of variances associated with abundance indices from the different models indicate that analyzing these data as a fixed effect GLM may underestimate the level of variability due to ignoring the grouped nature of tows within vessels. As such, use of a mixed model approach with vessel as a random effect is a reasonable approach to developing abundance indices and their variances.     

东海北部小黄鱼异方差生长模型

为揭示同方差和异方差结构对鱼类生长模型估计精度的影响,本研究以东海小黄鱼(Larimichthys polyactis)为案例,选用2007 2009年东海北部近海常规监测的小黄鱼基础生物学测定资料,采用对数正态函数型、幂函数型和指数函数型方差结构,分别对Schnute生长模型3种基本形式的von Bertalanffy(LVB)、Gompertz和Schnute模型进行了拟合分析,并采用似然比检验和AIC准则,选择确定最优拟合模型。结果表明,在渔获群体低龄化的条件下,其生长模型的参数估计值对方差结构较为敏感;对数正态函数型、幂函数型和指数函数型方差结构函数对3种生长模型的参数估计值较为接近,参数标准误和变异系数相对较小,小黄鱼生长模型以异方差结构函数拟合的效果较同方差结构更为精确和稳健;采用幂函数型和指数函数型方差结构拟合生长模型,其AIC值较同方差条件大幅降低,生长模型拟合的各月龄组置信区间范围差距缩小,明显提高了生长模型拟合度。     

Analysis of cod catch data from Icelandic groundfish surveys using generalized linear models

Catch data from the Icelandic groundfish surveys are analyzed using generalized linear models (GLM). The main goal is to test the effects of environmental variables on the expected cod catch and to distinguish between the gamma and log-normal distributions for the error structure. Only positive catch data are included in this work, i.e. only the positive part of a delta–gamma or delta–log-normal distribution is examined. The distributions are compared via a Kolmogorov–Smirnov goodness of fit test. Polynomials are used to describe the relationship between each environmental variable and the cod catch and their effects are tested within the GLM framework (a continuous model). Finally, an attempt is made to locate temperature fronts in the ocean by estimating the temperature gradient vector at each data point. The effect of the size of this gradient vector is then tested within in the GLM framework. A stratification model with only spatial and time effects explains 80% of the variation but that comes with a high cost of degrees of freedom. Most of the tested effects are found to be significant but the continuous model only captures 45% of the total variation. The size of the temperature gradient vector is found to be statistically significant though only a small portion of the variation in the data is explained by this term.     

An application of generalized linear models in production model and sequential population analysis

In fitting production models and age-structured models to an index of the relative abundance of a fish population, errors are usually assumed to follow a log-normal or normal distribution, without any diagnostic analyses. A generalized linear model can readily deal with many types of error structures. In this paper, a generalized linear model is coupled with a production model and a sequential population model to assess the stock of the Atlantic cod (Gadus morhua) 2J3KL. This study suggests that the parameter estimates in these models can be greatly influenced by the assumption about the error structures in the estimation and that log-normal and gamma distributions are appropriate for the production model in assessing the Atlantic cod 2J3KL stock, whereas gamma distribution is appropriate for the sequential population model. We recommend that generalized linear models should be used to identify the appropriate error structure in modeling fish population dynamics.     

Decreasing uncertainty in catch rate analyses using Delta-AdaBoost: An alternative approach in catch and bycatch analyses with high percentage of zeros

The gillnet data of walleye (Sander vitreus), yellow perch (Perca flavescens), and white perch (Morone americana), collected by a fishery-independent survey (Lake Eire Partnership Index Fishing Survey, PIS) from 1989 to 2008, contained 75–83% of zero observations. AdaBoost algorithm was applied to the model analyses with such fishery data for each species. The 3- and 5-fold cross-validations were conducted to evaluate the performance of each candidate model. The performance of the delta model consisting of one generalized additive model and one AdaBoost model (Delta-AdaBoost) was compared with five candidate models. The five candidate models included: the delta model comprising two generalized linear models (Delta-GLM), the delta model comprising two generalized linear models with polynomial terms up to degree 3 (Delta-GLM-Poly), the delta model comprising two generalized additive models (Delta-GAM), the generalized linear model with Tweedie distribution (GLM-Tweedie), and the generalized additive model with Tweedie distribution (GAM-Tweedie). To predict the presence/absence of fish species, the performance of AdaBoost model was compared in terms of error rate with conventional generalized linear and additive models assuming a binomial distribution. Results from 3- and 5-fold cross-validation indicated that Delta-AdaBoost model yielded the smallest training error (0.431–0.433 for walleye, 0.528–0.519 for yellow perch and 0.251 for white perch) and test error (0.435–0.436 for walleye, 0.524 for yellow perch and 0.254–0.255 for white perch) on average, followed by Delta-GLM-Poly model for yellow perch and white perch, and Delta-GAM model for walleye. In the prediction of the presence/absence of fish species, AdaBoost model had the lowest error rate, compared with generalized linear and additive models. We suggested AdaBoost algorithm to be an alternative to deal with the high percentage of zero observations in the catch and bycatch analyses in fisheries studies.     

Proposal for stock-recruitment relationship for Japanese sardine <Emphasis Type="Italic">Sardinops melanostictus</Emphasis> in North-Western Pacific

ABSTRACT:   This study assessed the stock-recruitment relationship (SRR) for the Japanese sardine Sardinops melanostictus in the North-western Pacific. Of the 20 SRR models investigated, the Akaike information criterion (AIC) was the minimum (AIC = 551.2) when the data were separated into two groups (A and B) and the log-normal distribution was applied as the error term. Group A was constructed with SRR data from 1976–1987 and 1992–2004. Group B consisted of data from 1988–1991. The AIC minimum model was R  = 22.8 S  ×  e ^ε for Group A, where R , S , and ε denote the recruitment of sardine (individual number of 0-year old fish), spawning stock biomass (SSB), and error term, respectively. This model indicated that recruitment was proportional to the SSB and that no density-dependent effect operated over the range of SSB investigated (51 000–11.3 million t). Recruitment was markedly higher (lower) when the sea surface temperature (SST) of the Kuroshio Extension area in February was low (high). The essential SRR can simply be expressed as R  = 22.8 S  ×  e ^ε with the level of recruitment deviating from the model to a greater or lesser degree depending on the environmental conditions.     

Use of random forests and support vector machines to improve annual egg production estimation

The delta-generalized additive model (Delta_GAM) is commonly used for analyzing zero-inflated continuous data, and has been widely applied in egg production methods (EPMs). It consists of two GAMs: one with a binomial distribution to estimate the probability of non-zero values, and the other with a log-normal distribution (Delta_LN model) or a gamma distribution (Delta_LG model) to model the continuous non-zero values. However, the rather restrictive distribution assumptions are not fulfilled for egg production data. In this study, we modified the Delta_GAMs using two machine learning techniques: random forest (Delta_RF) and support vector machines (Delta_SVM). We applied the tenfold cross-validation procedure to compare the performance of these four models using root mean square error (RMSE) and the EPM survey data of small yellow croaker Larimichthys polyactis, mullet Liza haematocheilus and gizzard shad Konosirus punctatus from Haizhou Bay, China. Both the Delta_RF and Delta_SVM models showed superior performance to that of the Delta_LN and Delta_LG models. Predicted spatial and temporal distributions varied among the models, although predictive performance varied little. The annual egg production was predicted and estimated with large uncertainty. We propose that machine learning techniques such as RFs and SVMs be used to model zero-inflated continuous data from EPM surveys, which tend to provide a more reliable estimation of annual egg production (AEP).     

The effect of agricultural fertilizer on growth rate of benthic diatoms

The aim of this work was to consider a non-conventional medium for benthic diatoms culture, formulated with liquid agricultural fertilizers and with the ‘f/2’ medium used as control. Three strains of benthic marine diatoms (Nitzschia laevis, Nitzschia thermalis var. minor and Navicula incerta) were grown with these two media, in batch cultures in 18-l plastic containers. There were no significant differences in mean cell concentration and growth rate between three diatoms raised with agricultural fertilizers compared to the standard ‘f/2’ medium. The cost for non-conventional culture media is less than 1/8 the cost of the ‘f/2’ medium.     

Linear mixed-effects models for weight–length relationships

Length and weight data are often analyzed in fisheries science to derive a parametric weight–length relationship for estimating biomass and to develop indices of condition for comparing the ‘wellness’ of different populations of fish. However, analysis of such data often ignores the inherent spatial and temporal grouping of the observations, and hence, the data hierarchy. This paper proposes the use of linear mixed-effects models as an effective means of analyzing and comparing weight–length relationships and indices of condition when there are many groups. The use of simple linear regression (where grouping is ignored), ANCOVA (where group effects are incorporated as fixed-effects), and linear mixed-effects models (where group effects are random-effects) are compared using data for Atlantic sea scallops (Placopecten magellanicus). The group means of residuals is proposed as a measure of relative weight or index of population condition. Linear mixed-effects models should be used to analyze grouped data because the variability among groups is ignored in simple linear regression and ANCOVA. Also, it is important that explanatory variables be incorporated in analyses of grouped data because their influence may mask the true differences among groups.     

A generalised linear model for catch allocation: an example from Australia''s Northern Prawn Fishery

Two biologically distinct species, Penaeus semisulcatus and P. esculentus, are caught in the tiger prawn component of Australia's Northern Prawn Fishery. These species are recorded and marketed together as ‘tiger prawns’. A generalised linear model is used to separate the logbook catches of ‘tiger prawns’ into catches by individual species, even though only fragmentary historical survey data are available. We also consider generalised additive model and generalised linear mixed model approaches, and the extra benefits they might allow.     

Exemple de système de réchauffement de l''eau de mer utilisable dans les dispositifs aquacoles

This paper deals with the possibility of heating cold seawater entering a seaside aquacultural nursery during winter. The proposed model contains a ‘warm’ water source and two heat exchangers in series (heating and pre-heating). The model evolves for the system a set of values for various pairs of exchanger surface areas. The values of the working temperature are calculated for different flowrate ratios and for several surface areas given a set value of 15°C for the ‘warm’ water source. The paper further develops the optimising of the ratio S₂/S₁ for minimum capital investment and takes the further step of investigating the possibility of failure to achieve a set T_i for a specific pair of heat exchangers during the cold season at a specific seawater site.     

New techniques for the assessment and optimal management of growth and standing crop variation in the cultured freshwater prawn, Macrobrachium rosenbergii

The current ‘traditional’ prawn pond management system really does not ‘manage’, but rather accommodates prawn production characteristics and was instituted with a minimum knowledge of prawn biology and husbandry. As such, the traditional system is not optimal because it does not rely on a strong empirical knowledge base. Prawn culture has succeeded up until now because Macrobrachium rosenbergii is easily cultured and will give 500–1000 kg ha⁻¹ year⁻¹ without much effort. This is sufficient in many areas of the world but not in others where economic conditions have break-even production requirements of nearly 1 tonne greater than this. The traditional system relying on selective harvesting with large seine nets is seriously inefficient which not only leads to lost revenue but under-manages pond growth since unculled large animals suppress the growth of unculled smaller ones. The degree to which this occurs was not known to the designers of the traditional system who had no way of knowing (as we do now from our research results) the extremely large compensatory growth capability of small prawns in the absence of large ones. The traditional system also does not manage sexual dimorphic growth because no technology exists which can be used to create monosex broods or manipulate the sex ratio in ponds. Accurate production models are not available because traditional ponds are rarely sampled and/or rarely drained. This, along with inefficient harvests, results in a co-mingling of cohort stocking classes. It is impossible to get accurate survival and growth data because of this and because it is so time consuming to sample commercial ponds and process the data on a regular basis. This paper describes four new engineering and ‘bioengineering’ techniques which are under development and can overcome all the drawbacks of the traditional system: (1) surgical sex reversal to create monosex broods; (2) genetic tagging of stocking and resident cohorts to assess survival and growth; (3) size grading and 100% efficient harvesting using pond draining and a machine grader-harvester; and (4) a semi-automated computer assisted prawn sample data management system which uses sonic digitization of prawn sample data.     

环境因子对东南太平洋智利竹䇲鱼渔场时空分布异质性影响

基于2012—2018 年4—8 月我国东南太平洋智利竹?鱼 (Trachurus murphyi) 渔捞日志数据,应用地理权重回归模型 (GWR) 探究智利竹?鱼渔场资源分布与环境因子的空间异质性关系.结果表明,环境因子海面温度基于GWR 模型回归的拟合优度为0.54,校正的拟合优度为0.34,赤池信息准则 (Aka...     

基于SELECT模型的南海金线鱼刺网选择性研究

为研究流刺网网目尺寸对金线鱼(Nemipterus virgatus)的选择性, 2008年5月于南海北部渔场用网目尺寸分别为40 mm、45 mm、50 mm、55 mm和60 mm的刺网进行了共同捕捞试验。在SELECT(Share Each LEngthclasss Catch Total)模型的架构下, 用Normal、Gama 和Bi-normal概率模型拟合刺网对金线鱼的选择性, 根据赤池信息指数(Akaike's Information Criterion, AIC)值评价并选出最佳拟合模型。结果表明, 试验共捕获金线鱼185尾, 金线鱼众数体长随网目尺寸的增加而增大; Gama的估计分隔模型AIC值在所有模型中最小(466.55), 将其作为最佳拟合模型; 刺网对金线鱼的相对作业强度(Pi)大小顺序依次为40 mm60 mm55 mm45 mm50mm; 40 mm、45 mm、50 mm、55 mm和60 mm刺网对体长150 mm金线鱼的选择率分别为98.39% 、80.94% 、39.30% 、13.70% 和3.87%。该研究认为将50 mm作为金线鱼刺网的最小网目尺寸不足以达到保护渔业资源的目的, 建议将南海区金线鱼刺网最小网目尺寸增加到55 mm。     

Alternative error distribution models for standardization of catch rates of non-target species from a pelagic longline fishery: billfish species in the Venezuelan tuna longline fishery

Alternative error distributions were evaluated for calculating indices of relative abundance for non-target species using catch and effort data from commercial fisheries. A general procedure is presented for testing the underlying assumptions of different error distributions. Catch rates, from an observer program, of billfish caught mainly as bycatch in a pelagic tuna longline fishery in the Western Central Atlantic were standardized. Although catches of billfishes are not common in pelagic tuna longline fisheries, these fisheries are one of the main sources of fishing mortality for these stocks in the central Atlantic due to the magnitude and spatial extent of longline fishing effort. Billfish CPUE data are highly skewed with a large proportion of zero observations. Delta distribution models can accommodate this type of data, and involve modeling the probability of a non-zero observation and the catch rate given that the catch is non-zero separately. Three different Delta models were compared against other error distributions, including the lognormal, log-gamma, and Poisson. Diagnostic checks and deviance table analyses were performed to identify the best error distribution and the set of factors and interactions that most adequately explained the observed variability. The results indicated that the Delta-lognormal model (a binomial error distribution for the probability of a non-zero catch and lognormal error for the positive catch rates) complied best with the underlying characteristics of the data set. Analyses of catch rates for blue marlin, white marlin and sailfish confirmed the spatio-temporal nature of their distribution in the central Atlantic and Caribbean Sea. Also, the analyses indicated that catch rates of billfish differed among fishing vessel types; larger vessels had a higher probability of catching blue marlin, the more oceanic-oriented species, and lower probabilities of catching the more coastal-oriented species white marlin and sailfish. Standardized catch rates indicated in general a lower relative abundance for blue and white marlin in the most recent years, although estimated confidence intervals overlap through the years especially for white marlin.     

海州湾双斑蟳栖息分布特征与环境因子的关系

为了解双斑蟳栖息分布规律,实验根据2011—2016年多个季度航次在海州湾进行的渔业资源和环境调查数据,采用广义线性模型(GLM)、广义可加模型(GAM)以及随机森林3种物种分布模型(SDMs)方法,结合AIC(akaike information criterion)准则、累积偏差解释率和交叉检验等评判指标筛选和构建了双斑蟳栖息分布模型,并分析了环境因子对双斑蟳分布的影响。结果显示,3种模型在解释因子与响应变量间的关系上基本一致;其中GAM在模型拟合上具有优势,而随机森林的预测性能明显高于传统的GLM和GAM。双斑蟳相对渔获量在年份和月份间的变异性最为显著,两个因子的解释率分别在18%和3.8%以上。水深和表层盐度对双斑蟳资源分布的影响较大,均与双斑蟳相对丰度呈正相关关系;双斑蟳分布总体呈现冬季相对较高,夏季东北部海域高、西南部低的特点,与海州湾水深分布特点基本一致。本研究还根据FVCOM(finite-volume coasta ocean model)模拟环境数据,利用随机森林分布模型估计了双斑蟳在海州湾海域2011年各个季节的空间分布,为渔业资源的开发和保护提供依据。     

Effects of pond preparation and feeding rate on production of Penaeus monodon Fabricius, water quality, bacteria and benthos in model farming ponds

Fibreglass pools with sediment were used as model farming ponds to investigate the interactive effects of pond preparation and feeding rate on prawn production, water quality, bacterial dynamics, abundance of benthos and prawn feeding behaviour. Pools were either fertilised 1 month (prepared) or 2 days (unprepared) prior to stocking and either ‘high’ or ‘low’ feeding rates were used. The ‘high’ rate was 5.0% (range 4–8%) wet prawn biomass/day and was similar to that recommended for commercial farms. The ‘low’ rate was 2.5% (range 2–4%) wet prawn biomass/day. Juvenile Penaeus monodon (2.0–7.5 g) were stocked at 15 prawns/m² and were cultured for 71 days. With the exception of one prepared, high feeding-rate pool where mass mortality (> 80%) of prawns occurred following an interruption to aeration, prawn survival was high (> 86%) and was unaffected by preparation, feeding rate or their interaction. Pond preparation improved growth and biomass gain by about 20%. Growth was 4% higher with the higher feeding rate but biomass gain was not affected and, as food conversion ratio was much worse, use of the lower feeding rate offers considerable scope to reduce production costs, especially during cooler periods. There was no interaction in relation to growth between pond preparation and feeding rate. Meiofauna were more abundant, and prawns grew faster, in prepared pools than unprepared pools at the start of the experiment. However, changes in bacterial dynamics or meiofauna abundance over time did not explain reductions in prawn growth over time. In general, water quality was reduced in pools receiving the high feeding rate compared with low feeding rate pools. Other interactive effects of pond preparation and feeding rate on water quality, bacteria, benthos and prawn feeding behaviour are discussed.     

Non‐native marine species in north‐west Europe: Developing an approach to assess future spread using regional downscaled climate projections

1. Climate change can affect the survival, colonization and establishment of non‐native species. Many non‐native species common in Europe are spreading northwards as seawater temperatures increase. The similarity of climatic conditions between source and recipient areas is assumed to influence the establishment of such species, however, in a changing climate those conditions are difficult to predict. 2. A risk assessment methodology has been applied to identify non‐native species with proven invasive qualities that have not yet arrived in north‐west Europe, but which could become problematic in the future. Those species with the highest potential to become established or be problematic have been taken forward, as well as some that may be economically beneficial, for species distribution modelling to determine future potential habitat distributions under projected climate change. 3. In the past, species distribution models have usually made use of low resolution global environmental datasets. Here, to increase the local resolution of the distribution models, downscaled shelf seas climate change model outputs for north‐west Europe were nested within global outputs. In this way the distribution model could be trained using the global species presence data including the species' native locations, and then projected using more comprehensive shelf seas data to understand habitat suitability in a potential recipient area. 4. Distribution modelling found that habitat suitability will generally increase further north for those species with the highest potential to become established or problematic. Most of these are known to be species with potentially serious consequences for conservation. With caution, a small number of species may present an opportunity for the fishing industry or aquaculture. The ability to provide potential future distributions could be valuable in prioritizing species for monitoring or eradication programmes, increasing the chances of identifying problem species early. This is particularly important for vulnerable infrastructure or protected or threatened ecosystems.     

A partial-reuse system for coldwater aquaculture

A model partial-reuse system is described that provides an alternative to salmonid production in serial-reuse raceway systems and has potential application in other fish-culture situations. The partial-reuse system contained three 10 m³ circular ‘Cornell-type’ dual-drain culture tanks. The side-wall discharge from the culture tanks was treated across a microscreen drum filter, then the water was pumped to the head of the system where dissolved carbon dioxide (CO₂) stripping and pure oxygen (O₂) supplementation took place before the water returned to the culture tanks. Dilution with make-up water controlled accumulations of total ammonia nitrogen (TAN). An automatic pH control system that modulated the stripping column fan ‘on’ and ‘off’ was used to limit the fractions of CO₂ and unionized ammonia nitrogen (NH₃---N). The partial-reuse system was evaluated during the culture of eight separate cohorts of advanced fingerlings, i.e., Arctic char, rainbow trout, and an all female brook trout × Arctic char hybrid. The fish performed well, even under intensive conditions, which were indicated by dissolved O₂ consumption across the culture tank that went as high as 13 mg/L and fish-culture densities that were often between 100 and 148 kg/m³. Over all cohorts, feed conversion rates ranged from 1.0 to 1.3, specific growth rates (SGR) ranged from 1.32 to 2.45% body weight per day, and thermal growth coefficients ranged from 0.00132 to 0.00218. The partial-reuse system maintained safe water quality in all cases except for the first cohort—when the stripping column fan failed. The ‘Cornell-type’ dual-drain tank was found to rapidly (within only 1–2 min) and gently concentrate and flush approximately 68–88% (79% overall average) of the TSS produced daily within only 12–18% of the tank’s total water flow. Mean TSS concentrations discharged through the three culture tanks’ bottom-center drains (average of 17.1 mg/L) was 8.7 times greater than the TSS concentration discharged through the three culture tanks’ side-wall drains (average of 2.2 mg/L). Overall, approximately 82% of the TSS produced in the partial-reuse system was captured in an off-line settling tank, which is better TSS removal than others have estimated for serial-reuse systems (approximately 25–50%). For the two cohorts of rainbow trout, the partial-reuse system sustained a production level of 35–45 kg per year of fish for every 1 L/min of make-up water, which is approximately six to seven times greater than the typical 6 kg per year of trout produced for every 1 L/min of water in Idaho serial-reuse raceway systems.