Authors: 
Takashi Uenaka, Naohisa Sakamoto, Koji Koyamada
Abstract: 
In this study, we propose a novel integrated visualization system that enables interactive visual development of a Habitat Suitability Index (HSI) model for predicting the locations of fishing grounds. Our system enables the interactive selection of variables that are highly correlated to fish catches recorded by fishermen on their vessels. We illustrate the use of this system using a real-world simulation of the Pacific Ocean. Vortex structures, such as those used for fishing grounds exploration, are crucial for exploring fishing ground because they can drive nutrients from the ocean floor to the sea surface. The ability to locate fishing grounds therefore depends on accurate ocean forecasting systems for calculating ocean model variables such as temperature, salinity, velocity etc. Currently, these forecasts are based on multiple spatio-temporal simulations that produce multidimensional and multivariate results. Critical points in the ocean model are good indicators of the likelihood of finding microscale vortices, and our visualization approach enables the interactive exploration and analysis of these critical points. The proposed system makes it possible to analyze the vortex structure in the spatial domain, and explore fishing grounds on the sea surface in detail.