Non-Invasive Fish Biometrics for Enhancing Precision and Understanding of Aquaculture Farming through Statistical Morphology Analysis and Machine Learning.

Aquaculture requires precise non-invasive methods for biomass estimation. This research validates a novel computer vision methodology that uses a signature function-based feature extraction algorithm combining statistical morphological analysis of the size and shape of fish and machine learning to improve the accuracy of biomass estimation in fishponds and is specifically applied to tilapia (Oreochromis niloticus). These features that are automatically extracted from images are put to the test against previously manually extracted features by comparing the results when applied to three common machine learning methods under two different lighting conditions. The dataset for this analysis encompasses 129 tilapia samples. The results give promising outcomes since the multilayer perceptron model shows robust performance, consistently demonstrating superior accuracy across different features and lighting conditions. The interpretable nature of the model, rooted in the statistical features of the signature function, could provide insights into the morphological and allometric changes at different developmental stages. A comparative analysis against existing literature underscores the competitiveness of the proposed methodology, pointing to advancements in precision, interpretability, and species versatility. This research contributes significantly to the field, accelerating the quest for non-invasive fish biometrics that can be generalized across various aquaculture species in different stages of development. In combination with detection, tracking, and posture recognition, deep learning methodologies such as the one provided in the latest studies could generate a powerful method for real-time fish morphology development, biomass estimation, and welfare monitoring, which are crucial for the effective management of fish farms.

Integration of Morphometrics and Machine Learning Enables Accurate Distinction between Wild and Farmed Common Carp.

Morphology and feature selection are key approaches to address several issues in fisheries science and stock management, such as the hypothesis of admixture of Caspian common carp (Cyprinus carpio) and farmed carp stocks in Iran. The present study was performed to investigate the population classification of common carp in the southern Caspian basin using data mining algorithms to find the most important characteristic(s) differing between Iranian and farmed common carp. A total of 74 individuals were collected from three locations within the southern Caspian basin and from one farm between November 2015 and April 2016. A dataset of 26 traditional morphometric (TMM) attributes and a dataset of 14 geometric landmark points were constructed and then subjected to various machine learning methods. In general, the machine learning methods had a higher prediction rate with TMM datasets. The highest decision tree accuracy of 77% was obtained by rule and decision tree parallel algorithms, and "head height on eye area" was selected as the best marker to distinguish between wild and farmed common carp. Various machine learning algorithms were evaluated, and we found that the linear discriminant was the best method, with 81.1% accuracy. The results obtained from this novel approach indicate that Darwin's domestication syndrome is observed in common carp. Moreover, they pave the way for automated detection of farmed fish, which will be most beneficial to detect escapees and improve restocking programs.

Traits, landmarks and outlines: Three congruent sides of a tale on coral reef fish morphology.

Quantifying the morphology of organisms remains fundamental in ecology given the form-function relationship. Morphology is quantifiable in traits, landmarks, and outlines, and the choice of approach may influence ecological conclusions to an unknown extent. Here, we apply these three approaches to 111 individual coral reef fish of 40 species common in Micronesia. We investigate the major dimensions of morphological variability among individuals, families, and predefined feeding functional groups. We find that although the approaches are complementary, they coincide in capturing elongation as the main dimension of variability. Furthermore, the choice of approach led to different interpretations regarding the degree of morphological differentiation among taxonomic and feeding functional groups. We also use each morphology dataset to compute community-scale morphological diversity on Palauan reefs and investigate how the choice of dataset affects the detection of differences among sites and wave exposure levels. The exact ranking of sites from highest to lowest morphological diversity was sensitive to the approach used, but not the broad spatial pattern of morphological diversity. Conclusions regarding the effect of wave exposure on morphological diversity were robust to the approach used. Biodiversity hotspots (e.g., areas of exceptionally high diversity and/or endemism) are considered important conservation targets but their location may depend on the biodiversity metric used. In the same vein, our results caution against labelling particular sites as morphological diversity hotspots when metrics consider only a single aspect of morphology.

Investigating the morphological and genetic divergence of arctic char (Salvelinus alpinus) populations in lakes of arctic Alaska.

Polymorphism facilitates coexistence of divergent morphs (e.g., phenotypes) of the same species by minimizing intraspecific competition, especially when resources are limiting. Arctic char (Salvelinus sp.) are a Holarctic fish often forming morphologically, and sometimes genetically, divergent morphs. In this study, we assessed the morphological and genetic diversity and divergence of 263 individuals from seven populations of arctic char with varying length-frequency distributions across two distinct groups of lakes in northern Alaska. Despite close geographic proximity, each lake group occurs on landscapes with different glacial ages and surface water connectivity, and thus was likely colonized by fishes at different times. Across lakes, a continuum of physical (e.g., lake area, maximum depth) and biological characteristics (e.g., primary productivity, fish density) exists, likely contributing to characteristics of present-day char populations. Although some lakes exhibit bimodal size distributions, using model-based clustering of morphometric traits corrected for allometry, we did not detect morphological differences within and across char populations. Genomic analyses using 15,934 SNPs obtained from genotyping by sequencing demonstrated differences among lake groups related to historical biogeography, but within lake groups and within individual lakes, genetic differentiation was not related to total body length. We used PERMANOVA to identify environmental and biological factors related to observed char size structure. Significant predictors included water transparency (i.e., a primary productivity proxy), char density (fish·ha-1), and lake group. Larger char occurred in lakes with greater primary production and lower char densities, suggesting less intraspecific competition and resource limitation. Thus, char populations in more productive and connected lakes may prove more stable to environmental changes, relative to food-limited and closed lakes, if lake productivity increases concomitantly. Our findings provide some of the first descriptions of genomic characteristics of char populations in arctic Alaska, and offer important consideration for the persistence of these populations for subsistence and conservation.

Size-dependent escape risk of lumpfish (Cyclopterus lumpus) from salmonid farm nets.

In the last decade, the salmon aquaculture industry has considerably increased the use of lumpfish juveniles as cleaner fish. Potential escape of reared lumpfish into the wild may spread diseases or genetically contaminate wild stocks. The guidelines for minimum sizes of cleaner fish to use in aquaculture cages are currently based on simple mesh penetration tests. However, these guidelines do not consider the potential compressibility of fish or changes in mesh state due to factors such as sea conditions and maintenance operations. This study shows that the industry-recommended minimum stocking sizes for a given mesh size may result in escape risk and that ignoring fish compressibility and mesh state can lead to underestimation of the lumpfish sizes that are able to escape. Our results can be used to develop new guidelines that will contribute to reduced escape of lumpfish from salmonid farms and lessen the potential environmental consequences.

A century of morphological variation in Cyprinidae fishes.

BACKGROUND: Aquatic habitats have been altered over the past century due to a variety of anthropogenic influences. Ecomorphology is an area of aquatic ecology that can both directly and indirectly assess the effects of habitat alterations on organisms. However, few studies have explored long term trends in morphological variation. Long term changes in morphology can potentially impact niche and ultimately contribute to organismal success and the ecosystem. Therefore, in this study we assessed long term morphological variation with body size, sex, time, and hydrology using museum collections of five species of Cyprinidae (Minnows) from lentic and lotic systems over the past 100 years to gain insight into long term patterns in morphology. RESULTS: Variation in Cyprinidae morphology tended to relate to: body size-indicating strong allometric growth patterns with robustness of larger individuals; sex-indicating a level of fecundity selection for deeper bodies in females compared with males; and year-indirectly suggesting responses to habitat changes over the past century. In lotic ecosystems, Cyprinidae morphology tended to be more fusiform in conjunction with lower mean annual discharge or higher variation in discharge. In lentic ecosystems, change in morphology was observed but no historic habitat variables were available to discern potential mechanisms. Interestingly, not all species responded in the same magnitude or directionality. CONCLUSIONS: Long term changes in morphological variation provide a link to exploring functional relationships between taxa and their environment and have implications for understanding ecosystem attributes, community assembly patterns, and conservation.

Warming alters the body shape of European perch Perca fluviatilis.

The consequences of elevated temperature on body shape were investigated by comparing European perch Perca fluviatilis from the Forsmark area of the Baltic Sea to P. fluviatilis from a nearby Biotest enclosure. The Biotest is a man-made enclosure within the Baltic Sea that has received warm water from a nuclear power plant since 1980, resulting in temperatures that are elevated 5-10 °C relative to the surrounding Baltic Sea. Sampled fish ranged from young-of-the-year to 14 years. Geometric morphometrics and multivariate statistical analysis revealed significant morphological differences between individuals of P. fluviatilis from these two habitats. Most importantly, relative shape changed with size, with small individuals of P. fluviatilis from Biotest being characterized by a deeper body shape and a larger caudal peduncle than the smaller Baltic individuals. In large specimens, smaller differences were found with Biotest individuals being more slender than Baltic individuals. These results show that, in order to have a full understanding of the biological effects of elevated temperatures, studies that cover the entire size range of organisms will be important. Apart from the direct influence of temperature on growth rate and body shape, other ecological factors affected by temperature are discussed as possible contributors to the observed differences between the two populations.

Phylogeography and demographic history of two widespread Indo-Pacific mudskippers (Gobiidae: Periophthalmus).

This study provides a first description of the phylogeographic patterns and evolutionary history of two species of the mudskipper genus Periophthalmus. These amphibious gobies are distributed throughout the whole Indo-Pacific region and Atlantic coast of Africa, in peritidal habitats of soft-bottom coastal ecosystems. Three sequence datasets of two widely distributed species, Periophthalmus argentilineatus and P. kalolo, were obtained by amplifying and sequencing two mtDNA markers (D-loop and 16S rDNA) and the nDNA rag1 region. The three datasets were then used to perform phylogeographic, demographic and population genetic analyses. Our results indicate that tectonic events and past climatic oscillations strongly contributed to shape present genetic differentiation, phylogeographic and demographic patterns. We found support for the monophyly of P. kalolo, and only shallow genetic differentiation between East-African and Indo-Malayan populations of this species. However, our collections of the morphospecies P. argentilineatus include three molecularly distinct lineages, one of them more closely related to P. kalolo. The presence of Miocenic timings for the most recent common ancestors of some of these morphologically similar clades, suggests the presence of strong stabilising selection in mudskippers' habitats. At population level, demographic analyses and palaeoecological records of mangrove ecosystems suggest that Pleistocene bottlenecks and expansion plus secondary contact events of the studied species were associated with recurrent sea transgressions during interglacials, and sea regressions or stable regimes during glacials, respectively.