Operational ocean prediction experiments for smart coastal fishing

• We newly exhibit synergistic approach of combining high-density observation and high-resolution prediction system.
• We suggest new way of collecting large amount of flow field data by using fishing boats.
• A large number of temperature profiles is also observed by sensors attached with trawl net.
• New coastal ocean model with ∼1.5 km grid spacing is developed for the simulation of submesoscale phenomena.
• The synergistic approach can bring fishermen and oceanographers further insights for the ocean dynamics and smart fishing.

This paper describes a new combination of in situ, high-density observations gathered by fishermen, and a real-time, high-resolution (approx. 1.5 km) prediction model developed toward more efficient fishing. Flow field data can be successfully collected by observations from acoustic Doppler current profilers installed on commercial fishing boats, which uncover sub-mesoscale structures such as small (approx. 10 km) eddies in the eastern boundary current region of the Japan/East Sea. Frequent vertical temperature profiles observed by sensors attached to casting trawl nets indicate fine feature of summertime upwelling area associated with fishing grounds. These observational assets back up routine observations conducted by using stationary buoys, research vessels, commercial passenger lines, and tide gauges. These assets enable evaluation of system predictability and improvement through calibration of physical model parameters in addition to data assimilation using low-resolution remote-sensing satellites. Our prediction system is automated with high-end computers and enables better understanding of sub-mesoscale phenomena for more accurate determination of fishing conditions. High-resolution forecasts of hazardous rapid currents can be delivered via mobile phone to prevent damage to nets.