Bandung – Researchers Gilar Budi Pratama and Lady Ayu Sri Wijayanti from Universitas Padjadjaran have developed a web-based platform called FishPost (Fishing Prediction and Oceanographic Spatial Tool) to predict potential skipjack tuna fishing grounds using habitat suitability modeling based on oceanographic parameters. The system transforms scientific habitat models into an operational decision-support tool for fisheries users.
Skipjack tuna is one of Indonesia’s key capture fisheries commodities, but its spatial distribution is strongly influenced by marine environmental conditions such as sea surface temperature, sea surface height, salinity, and ocean current velocity. Variations in these parameters affect ocean productivity, prey availability, and migration patterns of pelagic fish species.
Traditionally, fishing ground prediction models have been presented only as static spatial maps in academic publications, limiting their accessibility for field users. This study addresses that limitation by integrating habitat suitability modeling outputs into an interactive web-based spatial system that can be used directly by fishers and fisheries managers.
FishPost was developed using a Habitat Suitability Index (HSI) approach derived from a Generalized Additive Model (GAM), which captures nonlinear relationships between environmental variables and skipjack tuna presence probability. The resulting suitability values range from 0 to 1, with higher values indicating more favorable habitat conditions.
The system integrates monthly habitat suitability layers covering Indonesian Fisheries Management Areas 573, 713, and 714. Users can explore spatial predictions interactively through map visualization tools that display potential fishing zones across different months of the year.
FishPost provides two evaluation methods for identifying fishing zone potential. The first method uses geographic coordinate inputs to extract Habitat Suitability Index values from spatial raster layers. The second method evaluates environmental parameter inputs such as sea surface temperature, sea surface height, salinity, and current velocity to determine habitat suitability conditions directly.
According to the spatial distribution results presented on page 7 of the study, offshore regions generally show higher suitability values above 0.7, indicating favorable habitat conditions for skipjack tuna, while several coastal areas display lower suitability values below 0.4.
The predictive performance of the GAM-based model used in FishPost also demonstrated strong accuracy. Average Area Under the Curve (AUC) values reached 0.8806 in Fisheries Management Area 573 and 0.845 in Areas 713–714, while True Skill Statistic (TSS) values reached 0.6717 and 0.660 respectively, confirming reliable classification capability between suitable and unsuitable habitats.
By transforming habitat modeling outputs into a web-based system, FishPost bridges the gap between scientific research and operational fisheries applications. The platform enables users to evaluate fishing potential both spatially and environmentally through an accessible digital interface.
For fishers, this system can reduce search time for fishing locations and lower fuel consumption during operations. For fisheries managers, FishPost offers a practical decision-support tool for improving resource management strategies and optimizing fishing efficiency.
The development of FishPost demonstrates how oceanographic data digitalization can enhance the practical use of scientific modeling in capture fisheries. Future improvements are expected to integrate real-time environmental datasets to support more dynamic and daily fishing ground predictions.
0 Komentar