Quantification and Unsupervised Clustering Analysis of Morphological Characteristics of Seamounts in South China Sea Basin
Deng Dazhen
Zhao Yanghui
Bryan Riel
Gao Jinyao
Fang Yinxia
Abstract：Morphological differences are evident among submarine volcanoes formed by varying eruption patterns.However,their interrelationship remains elusive due to methodological constraints.This study innovatively employs machine learning clustering analysis on high-resolution multibeam bathymetry data to quantitatively evaluate the morphological parameters of seamounts in the South China Sea Basin.The analysis discerns three distinct seamount types.Type Ⅰ:large,isolated seamounts characterized by significant volume,steep slopes,and rounded bases.Type Ⅱ:large,linear seamounts with substantial volume,gentle slopes,and elongated bases.Type 111:smaller seamounts with limited volume,gentle slopes,and elliptical bases.Type Ⅰ and Ⅱ seamounts are primarily found in the mid-ocean ridge zone of the eastern sub-basin.The pronounced morphology of Type Ⅰ suggests an active and rapid eruption regime,whereas Type Ⅱ's subdued form indicates slower volcanic activity with more fluidic lava flows.Conversely,Type Ⅲ,situated along the transform faults and distant from the mid-ocean ridge of the eastern sub-basin,signifies less intense volcanic activities.This research establishes a foundational understanding that seamount formations under distinct tectonic backgrounds follow general morphological patterns.The novel clustering approach proposed here offers fresh perspectives for probing volcanic eruption patterns,especially when extensive petrological data is not available.
Keywords：seamount morphologyvolcanic eruption patternSouth China Sea Basincluster analysismachine learningmarine geology
Publication Date：2025-01-14
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：17( 217-233 )