{"title":"Align:一款用户友好型数值地层关联应用程序","authors":"Cedric Hagen, Jessica Creveling, Peter Huybers","doi":"10.1130/gsatg575a.1","DOIUrl":null,"url":null,"abstract":"Stratigraphic correlation underpins all understanding of Earth’s history, yet few geoscientists have access to, or expertise in, numerical codes that can generate reproducible, optimal (in a least-squares framework) alignments between two stratigraphic time-series data sets. Here we introduce Align, a user-friendly computer app that makes accessible a published dynamic time warping (DTW) algorithm that, in a minute or less, catalogs a library of alignments between two time-series data sets by systematically exploring assumptions about the temporal overlap and relative sedimentation rates between the two stratigraphic sections. The Align app, written in the free, open-source R programming language, utilizes a graphical user interface (e.g., drop-down menus for data upload and sliding bars for parameter exploration) such that no coding is required. In addition to generating alignment libraries, a user can employ Align to visualize, explore, and cull each alignment library according to thresholds on Pearson’s correlation coefficient and/or temporal overlap. Here we demonstrate Align with time-series records of carbonate stable carbon isotope composition, though Align can, in principle, align any two quantitative stratigraphic time-series data sets.","PeriodicalId":35784,"journal":{"name":"GSA Today","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Align: A User-Friendly App for Numerical Stratigraphic Correlation\",\"authors\":\"Cedric Hagen, Jessica Creveling, Peter Huybers\",\"doi\":\"10.1130/gsatg575a.1\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Stratigraphic correlation underpins all understanding of Earth’s history, yet few geoscientists have access to, or expertise in, numerical codes that can generate reproducible, optimal (in a least-squares framework) alignments between two stratigraphic time-series data sets. Here we introduce Align, a user-friendly computer app that makes accessible a published dynamic time warping (DTW) algorithm that, in a minute or less, catalogs a library of alignments between two time-series data sets by systematically exploring assumptions about the temporal overlap and relative sedimentation rates between the two stratigraphic sections. The Align app, written in the free, open-source R programming language, utilizes a graphical user interface (e.g., drop-down menus for data upload and sliding bars for parameter exploration) such that no coding is required. In addition to generating alignment libraries, a user can employ Align to visualize, explore, and cull each alignment library according to thresholds on Pearson’s correlation coefficient and/or temporal overlap. Here we demonstrate Align with time-series records of carbonate stable carbon isotope composition, though Align can, in principle, align any two quantitative stratigraphic time-series data sets.\",\"PeriodicalId\":35784,\"journal\":{\"name\":\"GSA Today\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"GSA Today\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1130/gsatg575a.1\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"Earth and Planetary Sciences\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"GSA Today","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1130/gsatg575a.1","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Earth and Planetary Sciences","Score":null,"Total":0}
引用次数: 0
摘要
地层相关性是了解地球历史的基础,然而很少有地球科学家能够使用或掌握数字代码的专业知识,这些代码能够在两个地层时间序列数据集之间生成可重复的最佳(最小二乘框架)排列。在这里,我们介绍 Align,这是一个用户友好型计算机应用程序,它可以使用已发布的动态时间扭曲(DTW)算法,通过系统地探索两个地层剖面之间的时间重叠和相对沉积速率假设,在一分钟或更短的时间内为两个时间序列数据集之间的排列库编目。Align 应用程序由免费开源的 R 编程语言编写,采用图形用户界面(如用于数据上传的下拉菜单和用于参数探索的滑动条),因此无需编码。除了生成配对库之外,用户还可以利用 Align 根据皮尔逊相关系数和/或时间重叠的阈值对每个配对库进行可视化、探索和筛选。在这里,我们用碳酸盐岩稳定碳同位素组成的时间序列记录来演示 Align,尽管 Align 原则上可以对任何两个定量地层时间序列数据集进行配准。
Align: A User-Friendly App for Numerical Stratigraphic Correlation
Stratigraphic correlation underpins all understanding of Earth’s history, yet few geoscientists have access to, or expertise in, numerical codes that can generate reproducible, optimal (in a least-squares framework) alignments between two stratigraphic time-series data sets. Here we introduce Align, a user-friendly computer app that makes accessible a published dynamic time warping (DTW) algorithm that, in a minute or less, catalogs a library of alignments between two time-series data sets by systematically exploring assumptions about the temporal overlap and relative sedimentation rates between the two stratigraphic sections. The Align app, written in the free, open-source R programming language, utilizes a graphical user interface (e.g., drop-down menus for data upload and sliding bars for parameter exploration) such that no coding is required. In addition to generating alignment libraries, a user can employ Align to visualize, explore, and cull each alignment library according to thresholds on Pearson’s correlation coefficient and/or temporal overlap. Here we demonstrate Align with time-series records of carbonate stable carbon isotope composition, though Align can, in principle, align any two quantitative stratigraphic time-series data sets.