Multivariate mapping of seabed grain size parameters in the Bay of Fundy using convolutional neural networks

IF 2.6 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Marine Geology Pub Date : 2024-05-03 DOI:10.1016/j.margeo.2024.107299
Benjamin Misiuk , Yan Liang Tan , Michael Z. Li , Thomas Trappenberg , Ahmadreza Alleosfour , Ian W. Church , Vicki Ferrini , Craig J. Brown
{"title":"Multivariate mapping of seabed grain size parameters in the Bay of Fundy using convolutional neural networks","authors":"Benjamin Misiuk ,&nbsp;Yan Liang Tan ,&nbsp;Michael Z. Li ,&nbsp;Thomas Trappenberg ,&nbsp;Ahmadreza Alleosfour ,&nbsp;Ian W. Church ,&nbsp;Vicki Ferrini ,&nbsp;Craig J. Brown","doi":"10.1016/j.margeo.2024.107299","DOIUrl":null,"url":null,"abstract":"<div><p>High-resolution seabed sediment information is critical for a range of marine spatial planning applications in multi-use shelf environments. To establish this information for the Bay of Fundy, Canada, legacy seabed sediment measurements were obtained from regional data compilations, and eight parameters describing the grain size were modelled across the extent of the bay using high resolution acoustic seafloor mapping and oceanographic datasets. This was achieved using a purpose-made convolutional neural network configured for geospatial modelling of multivariate grain size parameters. Shared information between the response parameters enabled model training with partially complete observations from the varied legacy data sources, and an explicit multiscale model architecture ensured that environmental predictors were implemented at appropriate scales for modelling each parameter. This avoids typical exhaustive exploration and selection of scale-specific predictor sets that often precede model building. Compositional grain size parameters were additionally accommodated using appropriate output activation functions, providing an efficient alternative to compositional data transformation and imputation. Results agreed well with our current understanding of the surficial geology of the bay, and cross-validation was used to quantitatively evaluate map predictions. Of the eight predicted parameters, the mean grain size and mud (clay and silt) fractions were predicted with high accuracy (&gt; 50% variance explained); the accuracy of grain size skewness was comparatively low (24% variance explained). Exploration of variable importance suggested that compiled acoustic backscatter was the most important environmental variable for predicting the grain size, but that geographic information describing the latitude and longitude within the bay was also highly useful. We hypothesize an interaction between these variables that enables location-specific prediction. Data layers of predicted grain size parameter values are made available for further sedimentological and ecological exploration, and for marine spatial planning activities within the bay.</p></div>","PeriodicalId":18229,"journal":{"name":"Marine Geology","volume":null,"pages":null},"PeriodicalIF":2.6000,"publicationDate":"2024-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0025322724000835/pdfft?md5=2d1ef359740aedc645e7944a218aafd9&pid=1-s2.0-S0025322724000835-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Marine Geology","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0025322724000835","RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"GEOSCIENCES, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

High-resolution seabed sediment information is critical for a range of marine spatial planning applications in multi-use shelf environments. To establish this information for the Bay of Fundy, Canada, legacy seabed sediment measurements were obtained from regional data compilations, and eight parameters describing the grain size were modelled across the extent of the bay using high resolution acoustic seafloor mapping and oceanographic datasets. This was achieved using a purpose-made convolutional neural network configured for geospatial modelling of multivariate grain size parameters. Shared information between the response parameters enabled model training with partially complete observations from the varied legacy data sources, and an explicit multiscale model architecture ensured that environmental predictors were implemented at appropriate scales for modelling each parameter. This avoids typical exhaustive exploration and selection of scale-specific predictor sets that often precede model building. Compositional grain size parameters were additionally accommodated using appropriate output activation functions, providing an efficient alternative to compositional data transformation and imputation. Results agreed well with our current understanding of the surficial geology of the bay, and cross-validation was used to quantitatively evaluate map predictions. Of the eight predicted parameters, the mean grain size and mud (clay and silt) fractions were predicted with high accuracy (> 50% variance explained); the accuracy of grain size skewness was comparatively low (24% variance explained). Exploration of variable importance suggested that compiled acoustic backscatter was the most important environmental variable for predicting the grain size, but that geographic information describing the latitude and longitude within the bay was also highly useful. We hypothesize an interaction between these variables that enables location-specific prediction. Data layers of predicted grain size parameter values are made available for further sedimentological and ecological exploration, and for marine spatial planning activities within the bay.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
利用卷积神经网络多变量绘制芬迪湾海底粒度参数图
高分辨率的海底沉积物信息对于多用途陆架环境中的一系列海洋空间规划应用至关重要。为建立加拿大芬迪湾的这一信息,从区域数据汇编中获得了遗留的海底沉积物测量数据,并利用高分辨率声学海底测绘和海洋学数据集,对整个海湾范围内描述粒度的八个参数进行了建模。这是利用专门为多变量粒度参数地理空间建模而配置的卷积神经网络实现的。响应参数之间的共享信息使模型训练能够使用来自不同传统数据源的部分完整观测数据,而明确的多尺度模型架构则确保了环境预测因子能够以适当的尺度对每个参数进行建模。这就避免了通常在建立模型之前对特定尺度预测集进行详尽的探索和选择。此外,还使用适当的输出激活函数来处理组成粒度参数,为组成数据转换和估算提供了一种高效的替代方法。结果与我们目前对海湾表层地质的理解非常吻合,交叉验证用于对地图预测进行定量评估。在八个预测参数中,平均粒度和泥(粘土和粉土)组分的预测准确率较高(50% 的方差解释率);粒度偏度的准确率相对较低(24% 的方差解释率)。对变量重要性的探讨表明,汇编的声学反向散射是预测粒度的最重要环境变量,但描述海湾内经纬度的地理信息也非常有用。我们假设这些变量之间存在相互作用,从而可以进行特定地点的预测。预测粒度参数值的数据层可用于进一步的沉积学和生态学探索,以及海湾内的海洋空间规划活动。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Marine Geology
Marine Geology 地学-地球科学综合
CiteScore
6.10
自引率
6.90%
发文量
175
审稿时长
21.9 weeks
期刊介绍: Marine Geology is the premier international journal on marine geological processes in the broadest sense. We seek papers that are comprehensive, interdisciplinary and synthetic that will be lasting contributions to the field. Although most papers are based on regional studies, they must demonstrate new findings of international significance. We accept papers on subjects as diverse as seafloor hydrothermal systems, beach dynamics, early diagenesis, microbiological studies in sediments, palaeoclimate studies and geophysical studies of the seabed. We encourage papers that address emerging new fields, for example the influence of anthropogenic processes on coastal/marine geology and coastal/marine geoarchaeology. We insist that the papers are concerned with the marine realm and that they deal with geology: with rocks, sediments, and physical and chemical processes affecting them. Papers should address scientific hypotheses: highly descriptive data compilations or papers that deal only with marine management and risk assessment should be submitted to other journals. Papers on laboratory or modelling studies must demonstrate direct relevance to marine processes or deposits. The primary criteria for acceptance of papers is that the science is of high quality, novel, significant, and of broad international interest.
期刊最新文献
Editorial Board Evolution process of chemical weathering and sediment sources in the Makran Continental margin since the Younger Dryas How do morphological characteristics affect tidal asymmetry in the Radial Sand Ridges? Double tombolo formation by regressive barrier widening and landside submergence: The case of Orbetello, Italy Formation of vertical columnar seismic structures and seafloor depressions by groundwater discharge in the drowned Miami Terrace platform and overlying deep-water carbonates, southeastern Florida
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1