Model selection between traditional and popular methods for standardizing catch rates of target species: A case study of Japanese Spanish mackerel in the gillnet fishery

Li, Zengguang; Ye, Zhenjiang; Wu, Rong; Zhang, Chi

doi:10.1016/j.fishres.2014.08.021

Cited by 36 publications

(28 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The high variability and complex population structure increase the difficulty of stock management. Combined with the improvement in catch per unit effort (CPUE) standardization [30], a more reliable conclusion on population fluctuation could be drawn. Further investigations are also required, especially the origin of spawning components, to develop effective management policy.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Population structure of Japanese Spanish mackerel Scomberomorus niphonius in the Bohai Sea, the Yellow Sea and the East China Sea: evidence from random forests based on otolith features

Zhang

et al. 2016

Fish Sci

Self Cite

View full text Add to dashboard Cite

demographic properties, life history pattern and response to exploitation. Ignoring underlying spatial stock structure in a fishery management system can increase risk of local depletion, resulting in loss of genetic diversity [1][2][3]. Therefore, a thorough understanding of stock structure is the starting point of effective management in multi-stock fisheries [4,5].Otolith shape is subjected to a combination of genetic and environment effects and demonstrates stock-specific features [6]. It has also been deemed to be an ideal natural marker for spatial stock, and is widely used in stock identification. The introduction of multiple image analysis techniques improves the ability to visualize and compare otolith shape, such as Fourier analysis [7,8], curvature scale space [9], wavelet transform [10, 11] and partial reflection [12]. However, few efforts have been put into the improvement of classification methods, whereas machine-learning methods (e.g. random forests, artificial neural networks) have been applied in fish parasites and otolith microchemistry, and have outperformed discriminant analysis [13][14][15]. Discriminant analysis (linear, quadratic and canonical) is commonly used in population studies based on otolith shape, and multivariate normality and independence of the predictor variables are required for this method. Appropriate data transformations (log, square root, etc.) are often required to reach a normal distribution. As to multi-collinearity, there are two options: a principle component analysis or excluding variables with significant multi-collinearity. Algorithmic advantages of random forests enable researchers to eliminate the above steps, maximize the number of variables inputted for discrimination and retain the original information of each variable. However, an increase in classification accuracy has its own limits with the variables used, and many variables increase intrinsic classification accuracy but also decrease extrinsic accuracy in Abstract In this study, the population structure of Japanese Spanish mackerel Scomberomorus niphonius were investigated based on the geographical variation of otolith features, and the concept of random forests was introduced as a classifier. Samples were collected from eight main spawning grounds using commercial gill nets, covering the distribution range along the coast of China. Otolith shape was described by the shape variables and principle components of Fourier coefficients. Every possible combination of otolith variable was tested to search for an optimal variable combination. An intermediate number of variables (8 out of 13) produced the most discriminating signal, and then the eight grounds were divided into three stocks with 64.5 % global accuracy. The discriminant result provided phenotypic evidence for the common migration trajectory, and further confirmed the existence of a metapopulation in Japanese Spanish mackerel at a larger scale.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The variable with significant region/SL interaction was excluded from subsequent analysis, while the remaining variables that correlated significantly with FL were standardized using their respective common within-group slope [30], followed by the expression:…”

Section: Otolith Shape and Statistical Analysismentioning

confidence: 99%

Population structure of Japanese Spanish mackerel Scomberomorus niphonius in the Bohai Sea, the Yellow Sea and the East China Sea: evidence from random forests based on otolith features

Zhang

et al. 2016

Fish Sci

Self Cite

View full text Add to dashboard Cite

show abstract

“…Among the three modelling methods, RFs provided the best predictability and stable predictions over years but had a lower R 2 compared to ANNs. Actually, the relative predictive capacity of ANNs and RFs varied greatly among studies with respect to different objectives and circumstances of their applications [ 35 , 43 , 64 ]. For instance, some studies suggested that RFs had advantages over ANNs in relation to avoiding overfitting [ 65 ] and simple adjustment to parameters [ 35 ], whereas ANNs could be adaptively trained to solve more complex ecological relationships [ 25 , 27 ].…”

Section: Discussionmentioning

confidence: 99%

“…Actually, the relative predictive capacity of ANNs and RFs varied greatly among studies with respect to different objectives and circumstances of their applications [ 35 , 43 , 64 ]. For instance, some studies suggested that RFs had advantages over ANNs in relation to avoiding overfitting [ 65 ] and simple adjustment to parameters [ 35 ], whereas ANNs could be adaptively trained to solve more complex ecological relationships [ 25 , 27 ]. For modelling response curves, the simple patterns provided by GAM appeared to be more reasonable, whereas the complex relationships identified by ML methods did not necessarily mean they were unrealistic, because species-environment responses often tend to be complex, even after accounting for interactions between variables [ 21 ].…”

Section: Discussionmentioning

confidence: 99%

“…In this study, we conducted a bottom-trawl crab survey and collected relevant environmental variables including temperature, salinity, depth, and sediment type. SDMs of three Portunidae crabs were developed using three modelling methods, including one traditional generalized additive model (GAM), and two ML approaches, random forest (RF) and artificial neural network (ANN) [ 33 , 34 , 35 ]. In order to understand these crabs’ spatial distributions and to assess the reliability of their distribution models, we identified the effects of environmental factors on three crabs and compared the performances among these models regarding fitting capability, predictability, and model stability.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modelling the spatial distribution of three Portunidae crabs in Haizhou Bay, China

Luan

Zhang

et al. 2018

PLoS ONE

View full text Add to dashboard Cite

Crab species are economically and ecologically important in coastal ecosystems, and their spatial distributions are pivotal for conservation and fisheries management. This study was focused on modelling the spatial distributions of three Portunidae crabs (Charybdis bimaculata, Charybdis japonica, and Portunus trituberculatus) in Haizhou Bay, China. We applied three analytical approaches (Generalized additive model (GAM), random forest (RF), and artificial neural network (ANN)) to spring and fall bottom trawl survey data (2011, 2013–2016) to develop and compare species distribution models (SDMs). Model predictability was evaluated using cross-validation based on the observed species distribution. Results showed that sea bottom temperature (SBT), sea bottom salinity (SBS), and sediment type were the most important factors affecting crab distributions. The relative importance of candidate variables was not consistent among species, season, or model. In general, we found ANNs to have less stability than both RFs and GAMs. GAMs overall yielded the least complex response curve structure. C. japonica was more pronounced in southwestern portion of Haizhou Bay, and C. bimaculata tended to stay in offshore areas. P. trituberculatus was the least region-specific and exhibited substantial annual variations in abundance. The comparison of multiple SDMs was informative to understand species responses to environmental factors and predict species distributions. This study contributes to better understanding the environmental niches of crabs and demonstrates best practices for the application of SDMs for management and conservation planning.

show abstract

The importance of temporal scale in distribution modeling of migratory Caspian Kutum, Rutilus frisii

Moëzzi,

Poorbagher,

Eagderi

et al. 2024

Ecology and Evolution

View full text Add to dashboard Cite

The choice of temporal resolution has high importance in ecological modeling, which can greatly affect the identification of the main drivers of an organism's distribution, considering the spatiotemporal dynamism of environmental predictors as well as organisms’ abundance. The present study aimed to identify the spatiotemporal distribution patterns of Caspian Kutum, Rutilus frisii, along the southern coast of the Caspian Sea, north of Iran, evaluating multiple temporal resolutions of data. The boosted regression trees (BRT) method was used to model fish catch distribution using a set of environmental predictors. Three temporal scales of data, including seasonal, sub‐seasonal, and monthly time frames over the catch season (October–April), were considered in our modeling analyses. The monthly models, utilizing more detailed data scales, exhibited the highest potential in identifying the overall distribution patterns of the fish, compared to temporally‐coarse BRT models. The best models were the BRTs fitted using data from March and April, which represented the final months of the catch season with the highest catch levels. In the monthly models, the main determinants of the Kutum's aggregation points were found to be dynamic variables including sea surface temperature, particulate organic and inorganic carbon, as opposed to static topographic parameters such as distance to river inlets. Seasonal and sub‐seasonal models identified particulate inorganic matter and distance to river inlets as the predictors with the highest influence on fish distribution. The geographical distributions of fish biomass hotspots revealed the presence of a stable number of fish aggregation hotspot points along the eastern coast, while some cold‐spot points were identified along the central and western coasts of the Caspian Sea. Our findings indicate that utilizing fine time scales in modeling analyses can result in a more reliable explanation and prediction of fish distribution dynamics. The investigated approach allows for the identification of intra‐seasonal fluctuations in environmental conditions, particularly dynamic parameters, and their relationship with fish aggregation.

show abstract

Model selection between traditional and popular methods for standardizing catch rates of target species: A case study of Japanese Spanish mackerel in the gillnet fishery

Cited by 36 publications

References 42 publications

Population structure of Japanese Spanish mackerel Scomberomorus niphonius in the Bohai Sea, the Yellow Sea and the East China Sea: evidence from random forests based on otolith features

Population structure of Japanese Spanish mackerel Scomberomorus niphonius in the Bohai Sea, the Yellow Sea and the East China Sea: evidence from random forests based on otolith features

Modelling the spatial distribution of three Portunidae crabs in Haizhou Bay, China

The importance of temporal scale in distribution modeling of migratory Caspian Kutum, Rutilus frisii

Contact Info

Product

Resources

About