A Comprehensive Evaluation of Machine Learning on Coral Trace Element Paleothermometers for Sea Surface Temperature Reconstruction
文献类型：期刊论文
| 作者 | Wei, Yuxuan1,2; Deng, Wenfeng1; Chen, Xuefei1; Wei, Gangjian1 |
| 刊名 | PALEOCEANOGRAPHY AND PALEOCLIMATOLOGY
 |
| 出版日期 | 2024-10-01 |
| 卷号 | 39期号:10页码:19 |
| 关键词 | machine learning sea surface temperature coral paleothermometer non-SST effect Sr/Ca Li/Mg |
| ISSN号 | 2572-4517 |
| DOI | 10.1029/2024PA004885 |
| 英文摘要 | This research introduces a novel approach to reconstruct sea surface temperature (SST) by developing a universal coral thermometer using machine learning (ML) algorithms on monthly resolved Porites coral proxies and SST data. A total of 1,202 data sets from 19 corals, covering SSTs ranging from 21.5 to 31.5 degrees C, with proxies including Sr/Ca, Mg/Ca, Li/Mg, U/Ca, and B/Ca ratios were analyzed. The data were divided into four sub-datasets by regional and taxon constraints. An exhaustive analysis was conducted, training 1,612 models using various proxy combinations and ML strategies to assess the impact of the non-SST effect on the universality of ML models. The results indicated that the non-SST effect is more significantly attributed to regional variations than to taxon differences, underscoring the importance of regional factors in Porites coral proxy-based SST reconstructions. Sr/Ca and Li/Mg proxies were identified as the most indicative of SST, showing clearer relationships with temperature than other proxies. Non-linear approaches achieved a Root Mean Square Error (RMSE) of less than 0.90 degrees C, which further decreased to 0.72 degrees C upon incorporating specific regional and taxon constraints. In an independent test set focusing exclusively on Li/Mg and Sr/Ca proxies, the tree-based algorithms particularly excelled, achieving an average RMSE improvement of at least 0.52 degrees C over the Universal Multi-Trace Element Calibration Scheme and the Li/Mg empirical equation. This research underscores the potential of applying ML to coral-based SST reconstructions, especially highlighting the effectiveness of tree-based algorithms and the suitability of Sr/Ca and Li/Mg proxies for accurate temperature estimations. A universal coral thermometer was developed using machine learning (ML) on Porites coral proxies, enhancing paleo sea surface temperature (SST) reconstruction accuracy Li/Mg and Sr/Ca proxies were identified as the most effective for SST reconstructions, showcasing clearer relationships with SST Tree-based ML models outperformed others in predicting SST, advocating their use when Sr/Ca and Li/Mg proxies are available |
| WOS研究方向 | Geology ; Oceanography ; Paleontology |
| 语种 | 英语 |
| WOS记录号 | WOS:001321813800001 |
| 源URL | [http://ir.gig.ac.cn/handle/344008/81194]  |
| 专题 | 同位素地球化学国家重点实验室 |
| 通讯作者 | Deng, Wenfeng |
| 作者单位 | 1.Chinese Acad Sci, Guangzhou Inst Geochem, CAS Ctr Excellence Deep Earth Sci, State Key Lab Isotope Geochem, Guangzhou, Peoples R China 2.Univ Chinese Acad Sci, Coll Earth & Planetary Sci, Beijing, Peoples R China |
| 推荐引用方式 GB/T 7714 | Wei, Yuxuan,Deng, Wenfeng,Chen, Xuefei,et al. A Comprehensive Evaluation of Machine Learning on Coral Trace Element Paleothermometers for Sea Surface Temperature Reconstruction[J]. PALEOCEANOGRAPHY AND PALEOCLIMATOLOGY,2024,39(10):19. |
| APA | Wei, Yuxuan,Deng, Wenfeng,Chen, Xuefei,&Wei, Gangjian.(2024).A Comprehensive Evaluation of Machine Learning on Coral Trace Element Paleothermometers for Sea Surface Temperature Reconstruction.PALEOCEANOGRAPHY AND PALEOCLIMATOLOGY,39(10),19. |
| MLA | Wei, Yuxuan,et al."A Comprehensive Evaluation of Machine Learning on Coral Trace Element Paleothermometers for Sea Surface Temperature Reconstruction".PALEOCEANOGRAPHY AND PALEOCLIMATOLOGY 39.10(2024):19. |
入库方式： OAI收割
来源：广州地球化学研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。