干旱区生态环境知识资源中心(Arid): Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping

Arid

DOI	10.5194/gmd-14-3421-2021
	Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping
	Jiang, Zhenjiao; Mallants, Dirk; Gao, Lei; Munday, Tim; Mariethoz, Gregoire; Peeters, Luk
通讯作者	Jiang, ZJ (corresponding author), Jilin Univ, Coll Environm & Resources, Key Lab Groundwater Resources & Environm, Minist Educ, Changchun 130021, Peoples R China. ; Jiang, ZJ (corresponding author), CSIRO Land & Water, Locked Bag 2, Glen Osmond, SA 5064, Australia.
来源期刊	GEOSCIENTIFIC MODEL DEVELOPMENT
ISSN	1991-959X
EISSN	1991-9603
出版年	2021
卷号	14 期号:6 页码:3421-3435
英文摘要	This study introduces an efficient deep-learning model based on convolutional neural networks with joint autoencoder and adversarial structures for 3D subsurface mapping from 2D surface observations. The method was applied to delineate paleovalleys in an Australian desert landscape. The neural network was trained on a 6400 km(2) domain by using a land surface topography as 2D input and an airborne electromagnetic (AEM)-derived probability map of paleovalley presence as 3D output. The trained neural network has a squared error < 0.10 across 99% of the training domain and produces a squared error < 0.10 across 93% of the validation domain, demonstrating that it is reliable in reconstructing 3D paleovalley patterns beyond the training area. Due to its generic structure, the neural network structure designed in this study and the training algorithm have broad application potential to construct 3D geological features (e.g., ore bodies, aquifer) from 2D land surface observations.
类型	Article
语种	英语
开放获取类型	gold, Green Submitted
收录类别	SCI-E
WOS记录号	WOS:000661359800001
WOS关键词	VALLEYS ; ELECTROMAGNETICS ; CLASSIFICATION ; HETEROGENEITY ; REFLECTION ; SIMULATION ; AQUIFER ; RIVER
WOS类目	Geosciences, Multidisciplinary
WOS研究方向	Geology
来源机构	Commonwealth Scientific and Industrial Research Organisation
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/350432
作者单位	[Jiang, Zhenjiao] Jilin Univ, Coll Environm & Resources, Key Lab Groundwater Resources & Environm, Minist Educ, Changchun 130021, Peoples R China; [Jiang, Zhenjiao; Mallants, Dirk; Gao, Lei] CSIRO Land & Water, Locked Bag 2, Glen Osmond, SA 5064, Australia; [Munday, Tim; Peeters, Luk] CSIRO Mineral Resources, Locked Bag 2, Glen Osmond, SA 5064, Australia; [Mariethoz, Gregoire] Univ Lausanne, Fac Geosci & Environm, Inst Earth Surface Dynam, Lausanne, Switzerland
推荐引用方式 GB/T 7714	Jiang, Zhenjiao,Mallants, Dirk,Gao, Lei,et al. Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping[J]. Commonwealth Scientific and Industrial Research Organisation,2021,14(6):3421-3435.
APA	Jiang, Zhenjiao,Mallants, Dirk,Gao, Lei,Munday, Tim,Mariethoz, Gregoire,&Peeters, Luk.(2021).Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping.GEOSCIENTIFIC MODEL DEVELOPMENT,14(6),3421-3435.
MLA	Jiang, Zhenjiao,et al."Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping".GEOSCIENTIFIC MODEL DEVELOPMENT 14.6(2021):3421-3435.