[1307] - Large Scale Electrical Conductivity Estimation Using Quantitative Multi-Configuration Electromagnetic Induction Sensing Data

All available metadata of the dataset are listed below. Some features are available, e.g. download of dataset or additional description file.

von Hebel, C., 2013. Large Scale Electrical Conductivity Estimation Using Quantitative Multi-Configuration Electromagnetic Induction Sensing Data. PhD Report, IBG-3, Juelich. Accessed from http://www.tr32db.uni-koeln.de/data.php?dataID=1307 at 2019-04-25.
Title(s):Main Title: Large Scale Electrical Conductivity Estimation Using Quantitative Multi-Configuration Electromagnetic Induction Sensing Data
Description(s):Abstract: An electromagnetic induction (EMI) system consists of an electromagnetic field transmitting and receiving coil and measures an apparent electrical conductivity. This recorded value is a weighted average integrated over a specific sensing depth, where the sensing depth depends on both coil separation and coil orientation and increases generally with increasing coil offset. Thus, embedding one transmitter and various receivers with different offsets into a multi-configuration EMI device enable the simultaneous measurement of different depths and therefore the recording of multiple depth level information. These multi-level data could indicate the vertical variation in the subsurface electrical properties, since vertically changing electrical conductivities change the strength of the recorded signal. Of course, this is already beneficial compared to a single transmitter-receiver-pair EMI device. But more information can be gained, if not the recorded values but the layer specific conductivity contributing to the average could be known. In principle, a multi-layer inversion of the recorded ECa would tackle this goal. But current EMI systems measure only qualitative data, since a lot of external conditions influence the measurement and shift the recorded value irreproducible. Therefore the data have to be calibrated to obtain quantitative soil apparent electrical conductivities, which can be then input for a reliable inversion in order to parameterize the subsurface in terms of the electrical conductivity and corresponding layer thickness. Here we introduce a novel three-layer inversion which minimizes the misfit between measured and modeled data using the L1-norm without any smoothing and damping parameter to be respectively less biased towards outliers and to assure sharp layer boundaries. The shuffled complex evolution (SCE) global optimization inverts the data in a one-dimensional (1D) sense and assumes a horizontally layered earth. This scheme is tested on both synthetic and experimental data, where the 1D-models were perfectly reconstructed. Experimental data were acquired at the Selhausen test site sizing 160 by 60 m, which has a distinct gradient in soil texture with considerably higher gravel content in the western part and a rapid change in soil texture towards the eastern part. Here the soil is mainly composed of a loamy and sandy soil. Firstly, EMI data were acquired along two 30 m long transects at which electrical resistivity tomography (ERT) profiles were recorded as well. The ERT serves for both calibrating the EMI data and as reference, i.e. validation, of the SCE inversion results. The reconstruction of the subsurface electrical conductivity distribution was well performed such that the code was parallelized in order to invert for a large-scale model of a three-layer earth for the whole test site. The obtained electrical conductivity three-layer models were combined to form a three-dimensional (3D) image, which clearly indicates that the lateral and vertical conductivity changes of the subsurface are related to the changes in soil texture.
Responsible Party
Creator(s):Author: Christian von Hebel
Publisher:CRC/TR32 Database (TR32DB)
TR32 Topic:Soil
Related Sub-project(s):B6
Subject(s):CRC/TR32 Keywords: Soil, PhD Report
File Details
File Name:CvH_EMI_report1.docx
Data Type:Text
File Size:3151 kB (3.077 MB)
Date(s):Valid: 2013-08-13
Mime Type:application/vnd.openxmlformats-officedocument.wordprocessingml.document
Download Permission:OnlyTR32
General Access and Use Conditions:According to the TR32DB data policy agreement.
Access Limitations:According to the TR32DB data policy agreement.
North:-no map data
Measurement Region:Ellebach
Measurement Location:Selhausen
Specific Informations - Report
Report Date:13th of August, 2013
Report Type:PhD Report
Report City:Juelich
Report Institution:IBG-3
Metadata Details
Metadata Creator:Christian von Hebel
Metadata Created:2016-05-03
Metadata Last Updated:2016-05-06
Funding Phase:2
Metadata Language:English
Metadata Version:V41
Dataset Metrics
Page Visits:366
Metadata Downloads:0
Dataset Downloads:0
Dataset Activity