The application of a neural network to map clay zones in crystalline rock

Meller, Carola; Genter, Albert; Köhl, Thomas

doi:10.1093/gji/ggt423

Cited by 5 publications

(7 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GPK4 shows the highest calcite content of up to nearly 19 wt % and is the borehole with the lowest inflow or outflow rate (2 L s À1 ) during the 2005 circulation test (GPK2: 13 L s À1 outflow, GPK3: 15 L s À1 inflow). A generally high clay content is observed on the synthetic clay content logs (SCCL, Meller, Genter, et al, 2014) for this well.…”

Section: Synthesis Of Results With Complementary Calcimetry Investigamentioning

confidence: 99%

Is There a Link Between Mineralogy, Petrophysics, and the Hydraulic and Seismic Behaviors of the Soultz‐sous‐Forêts Granite During Stimulation? A Review and Reinterpretation of Petro‐Hydromechanical Data Toward a Better Understanding of Induced Seismicity and Fluid Flow

Meller

Ledésert

2017

JGR Solid Earth

Self Cite

View full text Add to dashboard Cite

In the framework of the European Soultz‐sous‐Forêts enhanced geothermal system (EGS) in Alsace, France, 20 years of scientific and preindustrial tests had to be performed before the site began production of electricity in 2008. Stimulation tests were designed to enhance the permeability because most of the numerous natural fractures that crosscut the granite body were sealed by secondary minerals that crystallized as an effect of the circulation of local hot brines. The deep‐seated granitic reservoir is located between 4,500 and 5,000 m depths. Hydraulic stimulations were conducted in the four deep wells (GPK1, GPK2, GPK3, and GPK4) inducing different microseismic event patterns, which cannot be explained by tectonic structures alone. In the present work, we provide a review of the hydraulic tests and reinterpret them in the light of mineralogical data obtained along the boreholes. A clear relationship appears between mineralogy (mainly clay and calcite content) and the petrophysical, mechanical, and hydraulic behaviors of the rock mass. High calcite contents are correlated with an abundance of clay minerals, low Young's modulus, low magnetic susceptibility, and variation in spectral gamma ray. Microearthquakes are generated in the fresh granite zones, while clay and calcite‐rich zones, linked with hydrothermal alteration, might behave aseismically during hydraulic stimulations. These findings highlight the importance of a detailed knowledge of the petrography of a reservoir to conduct an effective stimulation while keeping the seismic hazard at a minimum.

show abstract

Section: Synthesis Of Results With Complementary Calcimetry Investigamentioning

confidence: 99%

Is There a Link Between Mineralogy, Petrophysics, and the Hydraulic and Seismic Behaviors of the Soultz‐sous‐Forêts Granite During Stimulation? A Review and Reinterpretation of Petro‐Hydromechanical Data Toward a Better Understanding of Induced Seismicity and Fluid Flow

Meller

Ledésert

2017

JGR Solid Earth

Self Cite

View full text Add to dashboard Cite

show abstract

“…Increased clay content is seen at the bottom of the wells below 4,600 m at the transition between the porphyritic and the two-mica granite. For details of this neural network method, the SCCLs, and the calibration of the logs by magnetic mineralogical investigations refer to Meller et al (2014a;2014b) (Figure 3).…”

Section: Methodsmentioning

confidence: 99%

“…The basis of the analyses is synthetic clay logs, which are created from spectral gamma ray logs using a technique introduced by Meller et al (2014a). These logs are indicative of the occurrence of clay-bearing fractures along the boreholes.…”

Section: The Significance Of Clay For Geothermal Projectsmentioning

confidence: 99%

The significance of hydrothermal alteration zones for the mechanical behavior of a geothermal reservoir

Meller

Köhl

2014

Geotherm Energy

Self Cite

View full text Add to dashboard Cite

show abstract

“…Deep learning approaches based on convolutional neural networks with sharing weights were established in 2006 (Gu et al, 2017), and are now well accepted in the field of visual recognition, speech recognition and language processes. They provide efficient high-dimensional interpolators that cope with multiple scales and heterogeneous information (Marcais and de Dreuzy, 2017), and have been applied in geoscience for earthquake detection based on seismic monitoring (Perol et al, 2018), object and disaster recognition from remote-sensing data (Längkvist et al, 2016;Amit et al, 2016), and mineral prospectivity evaluation by the fusing of different geophysical datasets (Granek, 2016;Meller et al, 2013). Furthermore, a super-resolution convolutional neural network (SRCNN) approach composed merely of convolutional layers was established to directly capture the relationship between low-and high-resolution images (Dong et al, 2016).…”

Section: Introductionmentioning

confidence: 99%

High-resolution paleovalley classification from airborne electromagnetic imaging and deep neural network training using digital elevation model data

Jiang

Mallants

Peeters

et al. 2019

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Paleovalleys are buried ancient river valleys that often form productive aquifers, especially in the semiarid and arid areas of Australia. Delineating their extent and hydrostratigraphy is however a challenging task in groundwater system characterization. This study developed a methodology based on the deep learning super-resolution convolutional neural network (SRCNN) approach, to convert electrical conductivity (EC) estimates from an airborne electromagnetic (AEM) survey in South Australia to a high-resolution binary paleovalley map. The SRCNN was trained and tested with a synthetic training dataset, where valleys were generated from readily available digital elevation model (DEM) data from the AEM survey area. Electrical conductivities typical of valley sediments were generated by Archie's law, and subsequently blurred by down-sampling and bicubic interpolation to represent noise from the AEM survey, inversion and interpolation. After a model training step, the SRCNN successfully removed such noise, and reclassified the low-resolution, converted unimodal but skewed EC values into a high-resolution paleovalley index following a bimodal distribution. The latter allows us to distinguish valley from non-valley pixels. Furthermore, a realistic spatial connectivity structure of the paleovalley was predicted when compared with borehole lithology logs and a valley bottom flatness indicator. Overall the methodology permitted us to better constrain the three-dimensional paleovalley geometry from AEM images that are becoming more widely available for groundwater prospecting.

show abstract

The application of a neural network to map clay zones in crystalline rock

Cited by 5 publications

References 31 publications

Is There a Link Between Mineralogy, Petrophysics, and the Hydraulic and Seismic Behaviors of the Soultz‐sous‐Forêts Granite During Stimulation? A Review and Reinterpretation of Petro‐Hydromechanical Data Toward a Better Understanding of Induced Seismicity and Fluid Flow

Is There a Link Between Mineralogy, Petrophysics, and the Hydraulic and Seismic Behaviors of the Soultz‐sous‐Forêts Granite During Stimulation? A Review and Reinterpretation of Petro‐Hydromechanical Data Toward a Better Understanding of Induced Seismicity and Fluid Flow

The significance of hydrothermal alteration zones for the mechanical behavior of a geothermal reservoir

High-resolution paleovalley classification from airborne electromagnetic imaging and deep neural network training using digital elevation model data

Contact Info

Product

Resources

About