A machine learning approach to tracking crustal thickness variations in the eastern North China Craton

作     者：Shaohao Zou Xilian Chen Deru Xu Matthew JBrzozowski Feng Lai Yubing Bian Zhilin Wang Teng Deng Shaohao Zou;Xilian Chen;Deru Xu;Matthew J.Brzozowski;Feng Lai;Yubing Bian;Zhilin Wang;Teng Deng

作者机构：State Key Laboratory of Nuclear Resources and EnvironmentEast China University of TechnologyNanchang 330013China Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment MonitoringMinistry of EducationSchool of Geosciences and Info-PhysicsCentral South UniversityChangsha 410083China State Key Laboratory for Mineral Deposits ResearchSchool of Earth Sciences and EngineeringNanjing UniversityNanjing 210023China 

出 版 物：《Geoscience Frontiers》 (地学前缘（英文版）)

年 卷 期：2021年第12卷第5期

页      面：215-223页

核心收录：

学科分类：070904[理学-构造地质学] 0709[理学-地质学] 07[理学] 08[工学] 081601[工学-大地测量学与测量工程] 0816[工学-测绘科学与技术] 

基　　金：co-funded by the National Natural Science Foundation of China(Grant Nos.42002089,41930428) the National Key R&D Program of China(Grant Nos.2016YFC0600401 and 2017YFC0602302) by Open Research Fund Program of Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring(Central South University) Ministry of Education(Grant Nos.2020YSJS02,2020YSJS01) 

主　　题：Machine learning Geochemical database Crustal thickness Eastern North China Craton 

摘      要：The variation of crustal thickness is a critical index to reveal how the continental crust evolved over its four billion ***,ratios of whole-rock trace elements,such as Sr/Y,(La/Yb)n and Ce/Y,are used to characterize crustal ***,sometimes confusing results are obtained since there is no enough filtered ***,a state-of-the-art approach,based on a machine-learning algorithm,is proposed to predict crustal thickness using global major-and trace-element geochemical data of intermediate arc rocks and intraplate basalts,and their corresponding crustal *** the validation processes,the root-mean-square error(RMSE)and the coefficient of determination(R2)score were used to evaluate the performance of the machine learning algorithm based on the learning dataset which has never been used during the training *** results demonstrate that the machine learning algorithm is more reliable in predicting crustal thickness than the conventional *** trained model predicts that the crustal thickness of the eastern North China Craton(ENCC)was-45 km from the Late Triassic to the Early Cretaceous,but-35 km from the Early Cretaceous,which corresponds to the paleo-elevation of 3.0±1.5 km at Early Mesozoic,and decease to the present-day elevation in the *** estimates are generally consistent with the previous studies on xenoliths from the lower crust and on the paleoenvironment of the coastal mountain of the ENCC,which indicates that the lower crust of the ENCC was delaminated abruptly at the Early Cretaceous.