Environmental variables affecting Sargassum distribution in the East China Sea and the Yellow Sea

Autor: Jisun Shin, Jang-Geun Choi, So-Hyun Kim, Boo-Keun Khim, Young-Heon Jo
Jazyk: angličtina
Rok vydání: 2022
Předmět:
Zdroj: Frontiers in Marine Science, Vol 9 (2022)
Druh dokumentu: article
ISSN: 2296-7745
DOI: 10.3389/fmars.2022.1055339
Popis: Floating Sargassum horneri has flowed into Jeju Island and the coast of the Korean Peninsula every year between February and May since 2015, causing considerable damage to aqua-farming sites and navigation. This study aimed to address the relationship between Sargassum distribution in the Yellow Sea (YS) and the East China Sea (ECS) and environmental variables for determining Sargassum distribution toward the Korean Peninsula. From feature importance ranking, we found that sea surface temperature (SST) is the most influential environmental variable in Sargassum distribution. From variables such as sea surface height (SSH), eastward seawater velocity (uo), and northward seawater velocity (vo), it was observed that Sargassum patches were not distributed in the southeast below 29 °N. Subsequently, we employed bagged tress models to evaluate the specific sensitivity of each environmental variable to Sargassum distribution. This model showed the best quantitative and qualitative performance when trained with physical and geographical variables. When estimating expanded areas of Sargassum distribution over time with the change in SST, a sider distribution range of Sargassum patches than usual and an early inflow into the Korean Peninsula were observed when the SST increased from the original. In addition, we found that the tolerable and favorable SST for Sargassum was 12–20 and 18°C, respectively. These results will enhance the understanding of the relationship between environmental variables and Sargassum distribution and provide valuable data for establishing a pre-disaster system for Sargassum blooms flowing toward the Korean Peninsula.
Databáze: Directory of Open Access Journals