자기 조직화 지도 기법을 이용한 적조 분포와 해양 인자의 관계

Relationship between the red tide distribution and ocean factors by using self-organizing map technique

The red tide occurs almost every year around in the South Sea of Korea. Once it occurs, it causes serious fisheries damages. Sea water temperature is an important factor of causing the red tide. The Self-Organizing Map(SOM) is useful algorithm for pattern recognition and clustering methods for climate change when used large datasets. In this study, it is investigated the relations of SST(Sea Surface Temperature) and SLA(Sea Level Anomaly) when red tide occurred, how SST and SLA pattern changes by applying SOM technique. SST data and SLA data were July 1st to October 31st in 2013. Applying SOM algorithm, hexagonal 3 x 3 size map were used. It is able to distinguish the patterns to the red tide occurrence and disappearance. However, because of the limited study period to 2013, it is necessary to apply them to other periods in which red tide occurred, and it is required to apply the other environmental data. When the red tide disappeared, SST gradually increased to reach the high temperature, SLA was less then 0 cm which is negative anomaly.