Swimming crab (Portunus trituberculatus) is a commercially valuable species in the Yellow Sea, where recent fluctuations in resource levels have raised concerns about sustainable management. This study aimed to improve the estimation of the carapace length at 50% maturity (L50) using machine learning techniques, providing a more consistent and reproducible framework for visual maturity classification by standardizing image-based decision processes. Using geometric image augmentation (e.g., rotation, flipping, brightness adjustment), Hue–Saturation–Value (HSV) color segmentation, and algorithms, such as Extreme Gradient Boosting (XGB), Support Vector Machine (SVM), Random Forest (RF), and ensemble models, we classified the maturity of female crabs based on gonad color features. Model performance was evaluated using accuracy, AUC, and the TSS, with the ensemble model showing the highest predictive capability. The machine learning-based L50 was estimated at 64.63 mm (±1.73 mm), yielding a narrower uncertainty range than the visually derived L50 of 65.47 mm (±2.89 mm) under the same macroscopic labeling framework. These results suggest that machine learning-assisted maturity classification can enhance the precision and operational consistency of maturity estimation under a standardized framework, while biological accuracy cannot be confirmed in the absence of an independent reference, such as histological validation.
The family Ocythoidae is found throughout the temperate and subtropical waters of the world’s ocean and contains a single species, Ocythoe tuberculata. Recently, a single female specimen of the rare pelagic species O. tuberculata was collected off Uljin in the East Sea/Sea of Japan of Korea by squid jigging for the first time. The appearance of this species could be considered as evidence of the changes in species distribution of sea surface. However, further studies are required to investigate whether this phenomenon is temporary or continuous.
The waters surrounding the Five West Sea Islands of South Korea are ecologically important but challenging to survey due to their location within a strategic military zone, strong tidal currents, and significant tidal variation. To assess the fish community in this region, we conducted Korea’s first parallel investigation combining traditional bottom trawl surveys with environmental DNA (eDNA) metabarcoding. Sampling was performed at 10 stations in March, May, and August 2023, and the relationship between fish species occurrence and environmental variables (water temperature, salinity, and depth) was examined. Dominant trawl-caught species included Engraulis japonicus, Johnius grypotus, Coilia nasus, and Okamejei kenojei, each showing seasonal migration and spawning patterns associated with temperature changes. eDNA analysis detected nine additional species absent from trawl catches, such as Ilisha elongata and Thamnaconus modestus, demonstrating its sensitivity in identifying both migratory and sedentary taxa. Our findings confirm that eDNA surveys can complement traditional sampling, improving biodiversity assessment in regions with limited accessibility and complex oceanographic conditions.
To investigate the age and growth characteristics of shotted halibut, Eopsetta grigorjewi, in the West Sea of Korea, the samples were collected using gill nets throughout 2019. A total of 861 specimens (693 females and 168 males) were analyzed. The BW-TL relations were BW = 0.0034TL3.3278 (R2 = 0.8716) for females and BW = 0.0031TL3.3359 (R2 = 0.8860) for males. This analysis revealed that males exhibited a larger growth rate than females (p < 0.05). Notably, the anterior and posterior sections of the otolith displayed an elongated oval shape, exceeding the radius of the dorsal and ventral regions. By examining the correlation between radii in four directions and total length, the highest correlation was observed between the total length (TL) and the ventral radius (R). The evaluation of the relationship between the fish total length and the otolith ventral radiusresulted in an equation of TL = 14.657R + 9.1601 (R2 =0.7081) for females and TL = 15.037R + 5.0265 (R2 = 0.6992) for males (p < 0.05). The formation of annuli occurred annually, consistently in January each year, as discerned through monthly changes in marginal index (MI) values. The von Bertalanffy growth equations from the relationship between the otolith annuli radius and the total length were Lt = 83.6 (1 − exp−0.046(t+4.70)) in females (n = 693, p < 0.05) and Lt = 56.8 (1 − exp−0.078(t+2.16)) in males (n = 168, p < 0.05).
To investigate the age and growth characteristics of the pointhead flounder (Cleisthenes pinetorum) in the West Sea (Yellow Sea) of Korea, samples were collected from bottom trawl vessels throughout 2019. A total of 1116 individuals (1015 females and 101 males) were analyzed. Because specimens were obtained from commercial landings, only fish of approximately 20 cm or larger were available due to marketability constraints. The body weight (BW) − total length (TL) relationships were BW = 0.00001TL3.3443 (R2 = 0.9279) for females and BW = 0.000002TL3.2659 (R2 = 0.9347) for males. The observed sex ratio (male:female = 1:10) was strongly female-biased; however, this likely reflected the underrepresentation of smaller males in commercial catches rather than the natural population structure. Females also exhibited larger body lengths than males. Otoliths were generally round, with a slightly elongated anterior region, and measurements were taken along the longest axis from the core to the margin. The relationship between TL and otolith radius (R) was expressed as TL = 8.0342R + 5.6218 (R2 = 0.8408). Growth equations were estimated for both sexes; however, because older males were poorly represented, the male von Bertalanffy growth function was fitted with the asymptotic length fixed to 43.3 cm (110% of the observed maximum male TL). Annuli were formed annually in November, and the spawning season was identified as September–November, suggesting a close association between annulus formation and the spawning period. The von Bertalanffy growth equations were Lt = 57.66(1 − exp[−0.1865(t + 0.46)]) for females and Lt = 43.3(1 − exp[−0.2679(t + 0.5490)]) for males.
Swimming crab (Portunus trituberculatus) is a commercially valuable species in the Yellow Sea, where recent fluctuations in resource levels have raised concerns about sustainable management. This study aimed to enhance the estimation of the carapace length at 50% maturity (L₅₀) through machine learning techniques, offering a more objective alternative to traditional visual inspection. Using geometric image augmentation (e.g., rotation, flipping, brightness adjustment), Hue-Saturation-Value (HSV) color segmentation, and algorithms such as Extreme Gradient Boosting (XGB), Support Vector Machine (SVM), Random Forest (RF), and ensemble models, we classified the maturity of female crabs based on gonad color features. Model performance was evaluated using Accuracy, AUC, and TSS, with the ensemble model showing the highest predictive capability. The machine learning-based L₅₀ was estimated at 64.63 mm (±1.73 mm), which was more precise than the visually derived L₅₀ of 65.47 mm (±2.89 mm). These results suggest that machine learning techniques can serve as reliable tools for developing science-based management strategies, ultimately supporting sustainable fisheries resource management.