Modular microcomputer software to estimate fish population parameters, production rates and associated variance

Abstract– Microcomputer software to estimate fish population parameters, production rates, and associated variances was developed using established procedures. Pop/Pro Modular Statistical Software is a series of portable programs written in English that run on Macintosh^(R) or IBM^(R) (and compatible) computers. Population parameters can be estimated using either a single-census mark-recapture method or a removal method. Production and component parameters can be estimated with either the instantaneous growth rate method or the increment summation method. Output may be stratified by fish length, age or sampling interval and includes variance estimates for all parameters. The software was developed for use with fish data but may be applicable to other taxa.     

Integrating the standardization of catch-per-unit-of-effort into stock assessment models: testing a population dynamics model and using multiple data types

Standardization of catch-per-unit-of-effort (CPUE) data can be integrated into stock assessment methods. We apply this method to the stock of trevally (Pseudocaranx dentex) off the west coast of New Zealand to address: (1) whether the stock assessment model explains all of the annual variation in the CPUE data, and (2) the impact on the assessment results of how the catch-at-age data are weighted. If not all of annual variation in CPUE is explained by the stock assessment model, the assessment may be statistically inadequate. The inadequacy may be in the representation of the population dynamics, in the relationship between CPUE and abundance, or due to additional variation in CPUE left unexplained by the independent variables. Catch-at-age data often have too much influence on the estimated abundance trajectory, so the sample size used in the catch-at-age likelihood function is often reduced when applying age-structured stock assessment methods. The integrated approach automatically places more weight on the CPUE data compared to the catch-at-age data, and may therefore provide an alternative to arbitrarily downweighting catch-at-age data.     

Inherent bias in using aggregate CPUE to characterize abundance of fish species assemblages

We have analyzed the practice of assessing an assemblage of fish species in a multispecies fishery on the basis of aggregate catch per unit effort (CPUE), which is the summed catch of all species per unit of effort. We show that at the onset of fishing or of a large positive or negative change in fishing effort, aggregate CPUE will be hyper-responsive, that is, relative change of aggregate CPUE will be greater than that of aggregate abundance. We also show that as the fishery reaches equilibrium, the aggregate CPUE in most circumstances will continue to be hyper-responsive, with a greater relative change from its value at the start than the aggregate abundance. However, there are less likely circumstances in which the aggregate CPUE will be hyper-stable compared to aggregate abundance. The circumstances leading to hyper-responsiveness or hyper-stability depend on the distribution of productivity and fishery vulnerability parameters among the species in the aggregation.     

A computer vision system to analyse the swimming behaviour of farmed fish in commercial aquaculture facilities: A case study using cage-held Atlantic salmon

Knowledge of fish behaviour plays an important role in aquaculture farm management. Video systems are the most common and cost-effective way of observing behaviours in commercial aquaculture operations. However long term observation is not feasible due to a limited ability to analyse footage manually. This paper describes preliminary findings obtained via computer vision software that was developed to automatically analyse fish movement and behaviours in aquaculture sea cages. Results show that the system is capable of detecting fish shapes in video recordings and from these shapes quantifying changes in swimming speed and direction continuously throughout the day. Also variations between days were detected and these may have been associated with the daily shift in the tidal cycle. The system has the potential to act as an alarm to farm operators, informing them about unusual fish behaviours on a continuous, real-time basis. It also has potential to assist in the evaluation of fish welfare.     

Are the long‐term fluctuations in Atlantic bluefin tuna (Thunnus thynnus) population related to environmental changes?

We tested whether synchronous, long‐term fluctuations in Atlantic bluefin tuna (Thunnus thynnus) trap catches, collected from the ancestral Mediterranean and Atlantic trap fishery, might be related to large‐scale environmental change. Nine time series of trap catches of more than 80 yr long were compared with long time series of three preselected environmental variables, i.e. the North Atlantic Oscillation (NAO), the Length of the Day Index (LOD, a proxy of the atmospheric circulation index) and the temperature. Spectral analyses of trap catches, LOD and temperature displayed similar spectra with peaks at low frequencies, whereas those of the NAO exhibited a broad band spectrum. Regression analyses and tests of correlation did not reveal any clear relationship between trap catches on the one hand and NAO and LOD on the other hand. In contrast, long‐term fluctuations in trap catches appear to be closely and negatively related to long‐term trends in temperature. Underlying processes that could explain such a relationship are discussed, with special focus on changes in migration patterns of the Atlantic bluefin tuna.     

Testing a model to track fish migrations in polar regions using pop‐up satellite archival tags

The use of pop‐up archival satellite tags (PSATs) to geolocate marine fishes in polar regions is challenging due to the brevity of periods during which there is a defined sunrise and sunset. Models using other environmental parameters are thus required to supplement geolocation data in the estimation of marine migratory routes. The objective of this work was to create a simple method that would estimate the migratory pathways of Atlantic salmon (Salmo salar) in polar regions using temperature and depth recordings. Validated geolocations from PSATs were used to test and constrict the model. The model’s predicted migratory routes were within 100 km of the light‐based geolocations calculated by the tags. By constraining the trajectories through the geolocations, bias was reduced. Sensitivity analyses demonstrated that slight alterations of the location and timing of the start and end points did not affect the mean migratory route estimates. This method is a management tool that can determine the primary habitat areas for any surface‐ or bottom‐dwelling marine species – especially in polar regions, where other methods may be impossible.     

Confidence interval estimation of CPUE year trend in delta-type two-step model

Abstract:   A procedure is suggested for estimation of the approximate confidence intervals of the extracted catch per unit effort (CPUE) year trend in the delta-type two-step model used for CPUE standardization with a lot of zero-catch data. This method is a simple way to combine the Taylor expansion and delta method and is suitable for practical use. This model was applied to the catch and effort data with more than 80% zero catch for silky shark in the North Pacific Ocean caught by Japanese training vessels. As a result, realistic values of the 95% confidence interval of CPUE year trend are obtained. A method for left–right unsymmetrical interval estimation based on the asymptotic normality of the natural logarithm of CPUE is also suggested. In the example of silky shark, both CPUE year trends obtained from these two methods are similar.     

Using coupled fish behavior–hydrodynamic model to investigate spawning migration of Japanese anchovy,Engraulis japonicus,from the East China Sea to Taiwan

Adult Japanese anchovies (Engraulis japonicus) migrate from the East China Sea to the coastal region of Taiwan to spawn around late winter and early spring and, later, their larvae constitute important fisheries in Taiwan. However, their migration route and its mechanism remain unclear. To investigate their spawning migration, we used a coupled fish behavior–hydrodynamic modeling approach. The physical field is simulated by the Pacific Ocean adaptation of the TaIwan Multi‐scale Community Ocean Model (TIMCOM) and the fish migration by Lagrangian tracer tracking with the aid of approximation of fish swimming behavior. We investigated three fish behavioral scenarios: (i) passive tracking of the current, (ii) swimming along with the current, and (iii) swimming along with the current and then changing to swimming toward the optimal spawning temperature. The comparison with and without Changjiang discharge is used to investigate the impacts of discharge reduction due to the Three Gorges Dam. Our results suggest that spawning migration of Japanese anchovy from the East China Sea to Taiwan may be aided by the China Coastal Current and that adult anchovies cannot reach the spawning site by passive advection alone. Thus, the swimming behavior of anchovies is crucial during the spawning migration, as it provides extra velocity and the orientation to the favorable spawning grounds. In addition, the adult anchovy is unlikely to reach the coastal area of Taiwan without Changjiang discharge. Our findings indicate that a coupled fish behavior–hydrodynamic model can help understand the influences of physical environment on the migration of Japanese anchovies.     

Initial design for a fish bioenergetics model of Pacific saury coupled to a lower trophic ecosystem model

A fish bioenergetics model coupled with an ecosystem model was developed to reproduce the growth of Pacific saury. The model spatially covers three different oceanographic spatial domains corresponding to the Kuroshio, Oyashio, and interfrontal (mixed water) regions. In this coupled model, three (small, large, and predatory) zooplankton densities which were derived from the lower trophic level ecosystem model were input to the bioenergetics model of saury as the prey densities. Although certain model parameters were imposed from other species’ bioenergetics, several model parameters were estimated from observational data specific to Pacific saury. The integrated model results reproduced appropriate growth rates of Pacific saury. Model sensitivities to water temperature and prey density are examined and observational methods to evaluate the model parameters are discussed.     

Assessing the impact of discretisation assumptions in a length-structured population growth model

Most of the traditional assessment models are age-structured. However, many biological and exploitation processes are more length-dependent than age-dependent, and the required length–age conversion of available data is often not reliable. Consequently, length-structured or age–length structured models have undergone considerable development in recent years. The growth transition matrix used to model the mean growth and growth variability of the population, is of primary importance in a length-structured matrix model. Building this growth transition matrix is not trivial and it is necessary to assess the impact that various assumptions may have to identify robust model structures. In this study, we assess the effects of (1) time and length discretisation, (2) the distribution of individuals within length classes and (3) the statistical distribution used to describe growth variability, by fitting a growth matrix model to individual quasi-continuous simulated growth data. The study quantitatively demonstrates that the choice of the time step and of length class width is the key point when building a length-structured population growth model. The use of a gamma distribution for the growth increments and/or a uniform distribution of individuals within length classes were found to make the model more robust.     

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.     

改进YOLO v4模型在鱼类目标检测上的应用研究

鱼类目标检测对渔业精准养殖、生产自动化、资源调查及鱼行为的研究等具有重要的意义.为了能快速准确地得到鱼类目标的位置和所属类别,提出了一种改进YOLO v4模型的鱼类目标检测方法,在CIoU(Complete Intersection over Union)损失函数基础上构建了新的损失项,改进的损失函数使真实框与相交框呈...     

Population estimation methods to quantify temporal variation in fish accumulations downstream of a weir

Abstract  Temporal changes in the population sizes of seven migratory species were assessed over a 2-year period prior to the construction of a fishway on the Murrumbidgee River, Australia. Mark-recapture (Chapman–Petersen) and Jackknife removal methods were used at 3-month intervals to determine the composition and sizes of migratory fish populations. No difference in population size estimate was detected between the two methods. However, the Jackknife removal estimator generally provided less variable estimates for a greater number of species. Population sizes were greatest during summer and autumn when bony herring, Nematalosa erebi Günther, and common carp, Cyprinus carpio L., dominated the migratory community. Both estimators provided data essential to further the development of an appropriate fish passage facility and could be applied to other sites where the size of migratory fish populations is of interest.     

Coupling of an individual-based population dynamic model of Calanus finmarchicus to a circulation model for the Georges Bank region

An individual-based life history and population dynamic model for the winter–spring dominant copepod of the subarctic North Atlantic, Calanus finmarchicus , is coupled with a regional model of advection for the Gulf of Maine and Georges Bank. Large numbers of vectors, each representing individual copepods with elements for age, stage, ovarian status and other population dynamic variables, are carried in a computation through hourly time steps. Each vector is updated at each time step according to development rate and reproductive functions derived from experimental data. Newly spawned eggs are each assigned new vectors as needed. All vectors are subject to random mortality. Thus, both life history progression and population dynamics of C. finmarchicus are represented for the temperatures in the Gulf of Maine–Georges Bank region in the active season. All vectors include elements representing depth, latitude and longitude. This allows coupling of the population dynamics to the tide- and wind-driven Dartmouth model of New England regional circulation. Summary data from the physical model are used to advance vectors from resting-stock locations in Gulf of Maine basins through two generations to sites of readiness for return to rest. Supply of Calanus stock to Georges Bank comes from all of the gulf and from the Scotian Shelf. The top of the bank is stocked from western gulf basins; the North-east Peak is stocked from Georges Basin and the Scotian Shelf. All sources contribute to stock that accumulates in the SCOPEX gyre off the north-west shoulder of Georges Bank, explaining the high abundance recurrently seen in that region. There is some return of resting stock to Wilkinson Basin in the western gulf, but other basins must mostly be restocked from upstream sources to the north-east.     

Efficiency of a dual density studded fish pass designed to mitigate for impeded upstream passage of juvenile European eels (Anguilla anguilla) at a model Crump weir

This study demonstrated that juvenile (glass) eels used a specific substrate (eel tiles) to circumvent a model Crump weir under an experimental setting. Upstream passage efficiency was 0 and 67% for the unmodified (no studded eel tiles on the downstream face; control) and modified (with studded eel tiles on the downstream face; treatment) set‐ups, respectively, and was greater for a small (59%) compared to large (41%) stud configuration. Eels were active and motivated to ascend the weir during both control and treatment set‐ups. Approach and attempt rates were elevated during the first few minutes of the treatment compared to control trials. Eels were edge‐oriented under both set‐ups and ascended the weir through the tiles during single burst swimming events (reaching estimated speeds of 68.5 cm·s⁻¹). Eel tiles may provide a cost‐effective solution for mitigating impacts of anthropogenic barriers to juvenile eel migration. Further research is required to determine passage efficiencies under higher flows, for a greater size range of eel, and for other migratory anguilliform fish (e.g. lamprey, Lampretra spp. and Petromyzon marinus L.). The performance of eel tiles should be validated through robust field studies.     

Application of the Tweedie distribution to zero-catch data in CPUE analysis

We focus on the zero-catch problem of CPUE (catch per unit effort) standardization. Because the traditional CPUE model with a log-normal error structure cannot be applied in this case, three methods have often been utilized as follows:

(1) Ad hoc method adds a small constant value to all response variables.

(2) Catch model with a Poisson or negative-binomial (NB) error structure.

(3) Delta-type two-step method such as the delta-normal model (after estimating the ratio of zero-catch using a logit or probit model, a model such as CPUE log-normal or Catch-Poisson is applied to CPUE without zero-data).

However, there are some statistical problems with each of these methods.In this paper, we carried out the CPUE standardization mainly using the Tweedie distribution model based on the actual by-catch data (silky shark, Carcharhimus falciformis, in the North Pacific Ocean caught by Japanese training vessels) including many observations with zero-catch (>2/3rd) and tuna fishery data as a target (yellowfin tuna, Thunnus albacares, in the Indian Ocean caught by Japanese commercial vessels) where the ratio of zero-catch is not so high (<1/3rd). The Tweedie model is an extension of compound Poisson model derived from the stochastic process where the weight of the counted objects (i.e., number of fish) has a gamma distribution and has an advantage of handling the zero-catch data in a unified way.We also compared four candidate models, the Catch-NB model, ad hoc method, Delta-lognormal model (delta-type two-step method) and Tweedie distribution, through CPUE analyses of actual fishery data in terms of the statistical performance. Square error and Pearson's correlation coefficient were calculated based on the observed CPUE and the corresponding predicted CPUE using the n-fold cross-validation.As a result, the differences in the trend of CPUE between years and model performance between the ad hoc method and Tweedie model were found to be not so large in the example of yellowfin tuna (target species). However, the statistical performance of Tweedie distribution is rather better than Delta-lognormal model, the Catch-NB distribution and ad hoc method in the example of silky shark (by-catch species). Standardized CPUE year trend of ad hoc method was found to be quite different from that of the Tweedie distribution and other two models. Model performance of the Tweedie distribution is good judging from the 5-fold cross-validation using the fishery data if including many zero-catch data such as by-catch species.     

基于模糊神经网络的池塘溶解氧预测模型

在分析了池塘溶解氧影响因素的基础上,利用模糊神经网络良好的非线性逼近能力建立了池塘溶解氧的模糊神经网络预测模型。神经网络模型如采用常规的BP或其它梯度算法,常导致训练时间较长且易陷入局部极小点,本实验采用快速的粒子群优化算法对模糊神经网络进行训练,收敛速度明显加快。实验结果表明采用该方法预报溶解氧的预测精度较常规BP递推算法的预测精度明显提高,所采用的模型能对溶解氧进行可靠的预测,该方法为研制开发智能水质检测仪以及工厂化养殖工作奠定了基础,对实际生产具有一定的指导意义。     

基于统计模型的养殖鱼塘溶解氧预报技术研究

根据武汉市2009-2011年的养殖鱼塘溶解氧数据和对应的气象要素,通过多元逐步回归分析构建鱼塘的溶解氧含量预报预测模型,并在武汉和荆州进行了验证。结果表明:一般鱼塘溶解氧在5-9月期间较低,低值一般出现在清晨6-7时,且较低值出现与天气相关,即闷热型天气突降大雨易导致溶解氧值降低。溶解氧量预报预测模型R2均在0.7以上,拟合效果良好。