Cheng-Ting Huang

Accurate biomass estimation is essential for fisheries management and aquaculture production. Conventional weight estimation typically relies on the length–weight power fitting (PF) method, which uses body length as a single predictor and is often constrained by biological variability, allometric growth, and incomplete sampling across growth stages, leading to estimation bias. To address these limitations, this study developed a multivariate weight estimation framework, termed the cluster–discriminant (CluDis) method. The approach integrates cluster analysis to group fish according to morphological characteristics, establishes group-specific linear regression models, and applies discriminant analysis to identify the most appropriate model for individual weight prediction.The method was validated using two knifejaw congeners (Oplegnathus fasciatus, n = 325; O. punctatus, n = 857) and their hybrids (n = 1,656), representing economically important aquaculture species. Substantial interspecific and intraspecific variation in morphometric traits supported the need for cluster-based modeling. Comparative analysis showed that the CluDis method significantly outperformed the traditional PF approach (p < 0.001), achieving estimation accuracies exceeding 91% across species and improving estimation efficiency by 32–49%. Moreover, CluDis reduced weight discrepancies and maintained prediction errors largely within a 10% margin, demonstrating greater robustness to biological variability and partial growth-stage data.These findings indicate that integrating multi-trait morphological information with automated model selection enhances weight estimation precision and provides a practical statistical tool for biomass monitoring and production management in aquaculture systems.

Aquaculture faces rising climate-change risks, while photovoltaic power generation requires substantial land resources, underscoring the need for multifunctional land use in the energy transition. Aquavoltaics—combining aquaculture with solar power—has emerged, and its success depends on stakeholder cooperation. Using white shrimp aquaculture in Taiwan as a case study, this study examines four cooperation schemes—a sole-investment scheme, a photovoltaic-led leasing scheme, a landowner lease-back scheme, and a separated aquaculture and photovoltaic operation scheme—across six aquavoltaic types. Cost–benefit and data envelopment analyses assess the economic returns and operational efficiency of aquaculture operators and photovoltaic companies under each scheme. The results show that all schemes offer profit potential but differ in efficiency and income distribution. The landowner lease-back scheme is most efficient for aquaculture operators, while photovoltaic companies operate near the efficiency frontier in all schemes. The findings highlight that cooperation design shapes the economic feasibility, efficiency, and livelihood impact of aquavoltaic systems, providing a useful reference for future policy and model design.

Accurate biomass estimation is fundamental to scientific fisheries management and the advancement of intelligent aquaculture. Traditional weight estimation often relies on the power fitting (PF) method, which uses a single parameter—body length—to predict weight. However, the PF method is frequently limited by biological variability, allometric growth patterns, and environmental influences, leading to significant estimation bias. Based on multivariate weight-estimation models, the present study developed a new fish weight estimation method, referred to as the CluDis method. This method combines cluster analysis (Clu) to group fish based on body morphological characteristics, simultaneously establishes a linear regression weight estimation model for each group, and then uses discriminant analysis (Dis) to screen and identify the optimal linear regression weight estimation model for each group. The accuracy of the CluDis method was tested and compared to assess whether it improves estimation efficiency. This study utilized two knifejaw congeners (Oplegnathus fasciatus and O. punctatus) (n = 325 and 857, respectively) and hybrid knifejaw (n = 1656), three economically important aquaculture fish species, for measurement and estimation. Morphometric analysis revealed substantial interspecific and intraspecific variation, with weight coefficient of variation (CV) values reaching 74.7%, necessitating species-specific modeling. The CluDis method achieved high estimation accuracies of 93.26% for O. fasciatus, 92.95% for O. punctatus, and 91.47% for hybrids. The weight estimation differences for O. fasciatus, O. punctatus, and hybrid knifejaw were reduced by 3.24%, 6.75%, and 3.57%; the estimation efficiency improved by 32.5%, 48.99%, and 29.5%, respectively. Comparative analysis demonstrated that the CluDis method significantly outperformed the traditional PF method (p < 0.001). Specifically, CluDis reduced weight discrepancies and improved accuracy by 32.5%-48.9% across the studied species. Unlike the PF method, which is sensitive to data gaps and outliers, the CluDis method showed greater tolerance for partial data across growth stages and maintained prediction errors within a 10% margin. The results of the present study demonstrate that the CluDis method is superior to previous methods in estimating fish body weight. These findings suggest that integrating multi-trait morphological measurements with automated screening mechanisms provides a superior alternative for biomass monitoring. Furthermore, the CluDis method holds significant potential for integration into machine learning and computer vision systems, enabling non-invasive, real-time weight estimation in aquaculture environments without inducing handling stress. This research provides a critical technical foundation for sustainable fisheries management and the realization of automated, high-precision smart farming.

A thorough investigation into the effects of crustacean eyestalk ablation across different genders is crucial for optimizing aquaculture practices. This research was conducted from June to September 2023, with two separate experiments to assess how eyestalk removal and the gender of mud crabs ( Scylla paramamosain ) affected the performance of raising double-skinned crabs using intermolt and mature crabs. Both experiments utilized a completely randomized design with three replications in rectangular tanks (a bottom area of 0.3 m²). Crabs were divided into four eyestalk ablation groups (no ablation, right ablation, left ablation, and bilateral ablation) in experiment 1, while experiment 2 focused on right eyestalk ablation crabs and categorized them into three gender groups (males, females, and mixed-sex). The animals were stocked at a density of 20 individuals m ⁻² and were fed trash fish at a rate of 5% of their wet body weight throughout the experiments. The results showed that eyestalk ablation did not significantly affect overall yields. However, it improved the specific growth rate (SGR), the time needed to form double-skinned crabs (duration), and the feed conversion ratio (FCR). The right ablation group showed the most notable improvements. The bilateral ablation group showed improvements in FCR but had the fewest benefits in SGR, duration, and survival rate (SR). Additionally, gender factors had a minor effect on duration, with the shortest time recorded for the female group. Both male and mixed-sex groups showed improvements in FCR but experienced a decline in SR, with a significant reduction in the male group. Based on these findings, it is recommended that right eyestalk ablation in intermolt and mature female mud crabs is the most effective strategy for producing double-skinned crabs under the conditions studied.

Snakehead (Channa lucius Cuvier, 1831) is a species of potential aquaculture interest in the Mekong Delta, Vietnam. However, their optimum environmental conditions have yet to be determined. This study aims to study the hatching success and larval and fry growth of C. lucius at six pH levels (5.5, 6.0, 6.5, 7.0, 7.5, and 8.0). Two consecutive experiments were conducted. Experiment 1 focused on incubating eggs and nursing the yolk sac larvae, and was carried out in aquariums. The monitoring included incubation time (IT), hatching rate (HR), and deformation rate of the newly hatched larvae (DR). Larvae were also collected daily during nursing for growth determination. In Experiment 2, four-day-old fry after hatching from experiment 1 were reared in plastic tanks with live feeds for 30 days. Growth was checked every ten days, while survival rate (SR) and coefficient of variation (CV) were determined at the end of the experiment. pH 5.5–8.0 was favorable for incubation, larvae, and fry, as assessed through good indicators of IT, HR, DR, and larval length and weight growths, as well as length and weight growths, SR, and CV of fry. Furthermore, the lowest DR (0.33%) was at pH 5.5, while except for pH 7.0, the shortest IT (41.1 h) significantly differed from that at higher pH levels (p < 0.05), and the highest larval growth parameters were at pH 5.5–6.0; meanwhile, pH 5.5–6.5 supported a better life for fry, as showed by the significantly higher growth parameters (p < 0.05) and SR and CV improvements in fry after 30 rearing days. Overall, pH 5.5–8.0 was suitable for the early stages of C. lucius, in which pH 5.5–6.0 was better for incubation and larval growth and pH 5.5–6.5 was better for fry development.

Grouper aquaculture is a rapidly growing and dynamic industry within the Taiwan aquaculture sector that serves as an important source of income generation. Farming fishes at a better location (FFABL) is a promising production strategy. This study evaluated the production and economic performance of grouper farming in Taiwan. This investigation employed two steps: (1) a cost–benefit analysis to determine the main component affecting profitability; and (2) multivariate statistics to clarify the relationship between biological and economic factors. The findings suggest that the best choice of species is hybrid giant grouper that generates and highlights a positive income, viable benefit–cost ratio and profit rate. The study found the feed management of hybrid giant groupers requires more consideration due to its low feed conversion ratio. The producers need to replace trash fish in commercial feed during the production cycle to ensure better feed management. Finally, this paper argues that feed is the main constraint for grouper farming, which is sensitive to any change in feed productivity, and this constraint must be solved to sustain the grouper industry in Taiwan.

This study aimed to investigate the growth and spawning characteristics of wild Channa lucius (Cuvier, 1831) populations in the Mekong Delta, Vietnam. The study was carried out from March 2011 to February 2012. The length-weight relationship of C. lucius was determined according to the equations W = 0.0044L(3.2494), R-2 = 0.9157 (female fish) and W = 0.0047L(3.2178), R-2 = 0.9434 (male fish). The length and weight of the body for female fish ranged from 16.3 to 36.0 cm and 49.0 to 550.0 g, and for male fish, from 14.6 to 38.2 cm and 46.0 to 639.0 g, respectively. Female and male fish had almost similar lengths at first maturity (21.3958 cm for females and 21.3952 cm for males). The condition factor of female fish ranged from 0.842 to 0.864 and for male fish ranged from 0.848 to 0.874. The gonadosomatic index of female fish was highest in June (1.68%) and lowest in November (0.69%); that of male fish was highest in June (0.53%) and lowest in September (0.15%). The sexual maturity rate (SMR) of female fish peaked in May, with SMRs of 33.3% (stage III) and 58.3% (stage IV). In conclusion, C. lucius has a positive allometric range (b > 3), reaching sexual maturity when its length reaches 21.39 cm for both sexes, and the spawning season begins in December but peaks in May and June.

Onshore farming for premium aquaculture is under scrutiny and criticism partially due to possible causes of adverse environmental impacts on other resource users and the surrounding environment. The best alternative to preventing or minimizing these impacts is to utilize open seawater by large submersible cage culture. The current operation in Taiwan has demonstrated that the culture operation is technically feasible but economically demanding because of high capital and operating costs. Therefore, this study conducted an economic analysis of the expansion of large submersible cage culture by selecting two premium species of snubnose pompano (Trachinotus anak) and cobia (Rachycentron canadum) and examined the profitability of large submersible cage culture investment. This study found that the current operation of four-unit cages highlighted a negative net present value and internal rate of return with a payback period of over ten and six years, respectively. Large submersible cage culture can be financially profitable when its operation unit expands from 8 to 24 units. Increasing unit cages to eight incurred a gross margin of 17.09%, BCR 1.21, with a payback period of 5.36 years. Expanding the operation to 24-unit cages was a potentially lucrative investment with a gross margin of 18.51%, BCR 1.23, PI 2.15, internal rate of return of 20.84%, and a payback period of 3.55 years. Sensitivity analyses revealed that market price and survival rate significantly impact the profitability of large submersible cage culture. Finally, it is suggested that producers could invest in 8-unit cages and maintaining the survival rate of snubnose pompano and cobia at 80% and 40%, respectively.