A operational search and rescue modeling system for the regional seas of Korea

Abstract

An operational search and rescue (SAR) modeling system has been developed to predict tracks of victims or debris from aircraft crashes or shipwrecks. The system employs a reliable trajectory model which is incorporated with a Monte Carlo ensemble technique to estimate the leeway (motion relative to the wind) of the drifting objects. Wind and current fields forecasted from the Korea Operational Oceanographic System of the Korea Ocean Research and Development Institute are used as forcing input to the SAR modeling system. The advantage of the system is to forecast immediately search tracks and areas up to 72-hour into the future while operating. To assess the performance of a SAR model, several drifting buoy experiments were conducted in the coast of Korea. The predicted trajectories are in good agreement with buoy trajectories observed on May 12, 2011. A circle assessment method is used for analyzing trajectory data and quantifies the model’s capability to reproduce the field drifter path. This method provides diagnosticanalysis with regard to the improvement of model performance by introducing two circles: target circle and error circle. It presents that predicted trajectories with wind forcing remains in an error circle longer than those without wind forcing. It suggeststhat forcing fields play a significant role in improving the behavior of the SAR model. In practice, the SAR system is applied to support the Korea Coast Guard in a cargo plane crash on 28 July 2011 and the prediction is reasonably well compared with the locations of the recovered wreckage.