Classifying Goliath Grouper (Epinephelus itajara) Behaviors from a Novel, Multi-Sensor Tag

Inertial measurement unit sensors (IMU; i.e., accelerometer, gyroscope and magnetometer combinations) are frequently fitted to animals to better understand their activity patterns and energy expenditure. Capable of recording hundreds of data points a second, these sensors can quickly produce large datasets that require methods to automate behavioral classification. Here, we describe behaviors derived from a custom-built multi-sensor bio-logging tag attached to Atlantic Goliath grouper (Epinephelus itajara) within a simulated ecosystem. We then compared the performance of two commonly applied machine learning approaches (random forest and support vector machine) to a deep learning approach (convolutional neural network, or CNN) for classifying IMU data from this tag. CNNs are frequently used to recognize activities from IMU data obtained from humans but are less commonly considered for other animals. Thirteen behavioral classes were identified during ethogram development, nine of which were classified. For the conventional machine learning approaches, 187 summary statistics were extracted from the data, including time and frequency domain features. The CNN was fed absolute values obtained from fast Fourier transformations of the raw tri-axial accelerometer, gyroscope and magnetometer channels, with a frequency resolution of 512 data points. Five metrics were used to assess classifier performance; the deep learning approach performed better across all metrics (Sensitivity = 0.962; Specificity = 0.996; F1-score = 0.962; Matthew’s Correlation Coefficient = 0.959; Cohen’s Kappa = 0.833) than both conventional machine learning approaches. Generally, the random forest performed better than the support vector machine. In some instances, a conventional learning approach yielded a higher performance metric for particular classes (e.g., the random forest had a F1-score of 0.971 for backward swimming compared to 0.955 for the CNN). Deep learning approaches could potentially improve behavioral classification from IMU data, beyond that obtained from conventional machine learning methods.

USO DE MÍDIAS DIGITAIS APLICADO À ESTUDOS DE CONSERVAÇÃO DO MERO Epinephelus itajara NO LITORAL AMAZÔNICO BRASILEIRO = USE OF DIGITAL MEDIA APPLIED TO CONSERVATION STUDIES OF THE ATLANTIC GOLIATH GROUPER Epinephelus itajara (LICHTENSTEIN, 1822) IN THE BRAZILIAN AMAZON COAST

O mero Epinephelus itajara é uma espécie ameaçada atualmente classificada como Vulnerável (VU) de acordo com a Lista Vermelha da União Internacional para a Conservação da Natureza (IUCN) e Criticamente em Perigo (CR) na Lista Vermelha Brasileira do Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio). Uma pesquisa utilizando as mídias digitais como base para a obtenção de dados sobre essa espécie mostrou que existe uma pesca contínua desse animal no litoral do Estado do Maranhão, mesmo protegido pela legislação nacional. Além disso, as mídias digitais têm um grande potencial para serem utilizadas como uma ferramenta de inspeção ambiental e uma importante estratégia para identificar áreas de captura e locais de desembarque de espécies ameaçadas, como o Epinephelus itajara.ABSTRACTThe Atlantic goliath grouper Epinephelus itajara is one of the largest species in the Serranidae family. E. itajara is an endangered species currently categorized as Vulnerable (VU) according to the International Union for Conservation of Nature’s Red List (IUCN) and Critically Endangered (CR) on the Brazilian Red List of the Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio). A survey using digital media as a basis for obtaining data on this species showed that there is a continuous fishing of this animal across the coast of the Maranhão State, even protected by national legislation. In addition, digital media have great potential to be used as an environmental inspection tool and an important strategy for identifying capture areas and landing places of threatened species such as Epinephelus itajara.

Nonlethal stable isotope analysis reveals consistent trophic growth of juvenile Atlantic goliath grouper Epinephelus itajara in Brazilian estuaries

Atlantic goliath grouper Epinephelus itajara (Lichtenstein, 1822) are classified as vulnerable by the IUCN and have decreasing local populations throughout their distribution due to overfishing and habitat degradation. Due to their protected status, basic life history information that can inform management and conservation is lacking for some local populations, including in Brazil. In the present study, we examined how δ15N of juvenile Atlantic goliath grouper fin rays, a nonlethal method, varied with total length across four estuaries in Brazil. A total of 100 juvenile Atlantic goliath grouper (total length range: 95–505 mm) were analyzed, and we observed positive relationships between δ15N and fish lengths (i.e., evidence of trophic growth). Among-estuarine slopes did not differ, suggesting trophic growth was consistent among sites despite different δ15N values between the northernmost site and a group of southern sites, possibly reflecting different isotopic baselines. This study is the first effort to provide useful insights into the trophic ecology of juvenile Atlantic goliath grouper in Brazil, which could help address knowledge gaps and conserve this endangered species. The nonlethal methodology employed in this study could be used to advance our understanding of the trophic ecology of other vulnerable and endangered marine fishes and help inform conservation and management practices.

Sanitary conditions of the Goliath grouper Epinephelus itajara in captivity on estuarine environment: case report

ABSTRACT This study aimed to report the sanitary conditions through the hematological analysis of grouper E. itajara reared in captivity on estuarine conditions. Seven Goliath groupers (1,881.5±1,246.03g) were captured and kept in two tanks located on estuary. After 20 days, fish were collected for morphologic and hemato-physiologic evaluation. Two fish had clinical signs such as hemorrhagic spots and loss of scale due to agonistic behavior. Blood samples were collected, and the hematological parameters (biochemical, erythrogram and leukogram) were determined. Blood cells were characterized by their size, color and shape. Univariate statistic and principal components analysis were used to identify a hematological standard between fish with or without clinical signs. Four leukocyte types were found: lymphocyte, monocyte, neutrophil and basophil. Regardless of the clinical signs the cell morphology did not present any difference among the fish. However, there is a significant correlation between erythrocyte and lactate on fish with clinical signs. Thus, agonistics encountered among the fish is a stressing factor in captivity conditions making it necessary to have adequate management related to the size of fish and stocking density.