TY - THES A1 - Platz, Anna T1 - Novel pre-stack data confinement and selection for magnetotelluric data processing and its application to data of the Eastern Karoo Basin, South Africa T1 - Neue "pre-stack" Datenbeschränkungs- und Selektionskriterien für die magnetotellurische Datenbearbeitung und ihre Anwendung auf Daten aus dem östlichen Karoo-Becken in Südafrika N2 - Magnetotellurics (MT) is a geophysical method that is able to image the electrical conductivity structure of the subsurface by recording time series of natural electromagnetic (EM) field variations. During the data processing these time series are divided into small segments and for each segment spectral values are computed which are typically averaged in a statistical manner to obtain MT transfer functions. Unfortunately, the presence of man-made EM noise sources often deteriorates a significant amount of the recorded time series resulting in disturbed transfer functions. Many advanced processing techniques, e.g. robust statistics, pre-stack data selection or remote reference, have been developed to tackle this problem. The first two techniques reduce the amount of outliers and noise in the data whereas the latter approach removes noise by using data from another MT station. However, especially in populated regions the data processing is still quite challenging even with these approaches. In this thesis, I present two novel pre-stack data confinement and selection criteria for the detection of outliers and noise affected data based on (i) a distance measure of each data segment with regard to the entire sample distribution and (ii) the evaluation of the magnetic polarisation direction of all segments. The first criterion is able to remove data points that scatter around the desired MT distribution and furthermore it can, under some circumstances, even reject complete data cluster originating from noise sources. The second criterion eliminates data points caused by a strongly polarised magnetic signal. Both criteria have been successfully applied to many stations with different noise contaminations showing that they can significantly improve the transfer function estimation. The novel criteria were used to evaluate a MT data set from the Eastern Karoo Basin in South Africa. The corresponding field experiment is part of an extensive research programme to collect information of the current e.g. geological setting in this region prior to a potential shale gas exploitation. The aim was to investigate whether a three-dimensional (3D) inversion of the newly measured data fosters a more realistic mapping of physical properties of the target horizon. For this purpose, a comprehensive 3D model was derived by using all available data. In a second step, I analysed parameters of the target horizon, e.g. its conductivity, that are proxies for physical properties such as thermal maturity and porosity. N2 - Magnetotellurik (MT) kann die elektrische Leitfähigkeit des Untergrundes abbilden indem Zeitreihen von natürlichen elektromagnetischen (EM) Wechselfeldern gemessen werden. Während der Datenbearbeitung werden die Zeitreihen in Abschnitte unterteilt und für jeden Abschnitt werden Spektren berechnet, welche auf statistische Art gemittelt werden um MT Übertragungsfunktionen zu bestimmen. Unglücklicherweise beeinflusst die Anwesenheit von künstlichen EM Rauschquellen oft eine signifikante Menge der aufgezeichneten Zeitreihen. Dies führt zu gestörten bzw. falschen Übertragungsfunktionen. Mehrere Methoden wurden entwickelt um dieses Problem zu beheben, z.B. robuste Statistik, pre-stack Datenselektion oder das "remote reference" Verfahren. Die ersten beiden Techniken reduzieren den Anteil von Ausreißern und Rauschen in den Daten während das letzte Verfahren Rauschen mit Hilfe von Daten einer zusätzlichen MT Station entfernt. Trotzdem bleibt die Datenbearbeitung vor allem in besiedelten Gebieten selbst mit diesen Methoden schwierig. In dieser Arbeit präsentiere ich zwei neue pre-stack Datenselektionskriterien zur Bestimmung von Ausreißern und verrauschten Datenpunkten basierend auf (i) einem Distanzmaß unter Berücksichtigung der gesamten Datenverteilung und (ii) der Auswertung der magnetischen Polarisierungsrichtung für jeden Abschnitt. Mit Hilfe des ersten Kriteriums können Datenpunkte entfernt werden, die um die eigentliche MT Verteilung streuen. Außerdem kann es unter bestimmten Umständen sogar ganze Datencluster beseitigen, welche von Rauschquellen hervorgerufen werden. Das zweite Kriterium eliminiert Datenpunkte, welche durch ein stark polarisiertes magnetisches Signal verursacht werden. Beide Kriterien wurden erfolgreich auf viele Stationen mit unterschiedlichen Rauschverhalten angewandt. Weiterhin wurden sie genutzt um MT Daten aus dem östlichen Karoo-Becken in Südafrika auszuwerten. Das dazugehörige Feldexperiment ist Teil eines umfangreichen Forschungsprojektes, welches Informationen über die aktuelle geologische Situation in dem Gebiet sammelt, bevor eine mögliche Schiefergasförderung stattfindet. Der Fokus lag darauf zu untersuchen, ob eine 3D Inversion der neu gemessenen Daten eine realistischere Abbildung der physikalischen Eigenschaften des Zielhorizonts fördert. Zu diesem Zweck wurde ein 3D Modell mit Hilfe aller verfügbaren Daten entwickelt. Anschließend habe ich verschiedene Parameter des Zielhorizonts analysiert, welche Rückschlüsse auf die physikalischen Eigenschaften wie thermische Reife und Porosität erlauben. KW - magnetotellurics KW - Eastern Karoo Basin KW - data processing KW - Mahalanobis distance KW - magnetic polarisation direction KW - Magnetotellurik KW - Östliches Karoo-Becken KW - Datenbearbeitung KW - Mahalanobis-Distanz KW - magnetische Polarisationsrichtung Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-415087 ER - TY - JOUR A1 - Platz, Anna A1 - Weckmann, Ute T1 - An automated new pre-selection tool for noisy Magnetotelluric data using the Mahalanobis distance and magnetic field constraints JF - Geophysical journal international N2 - In Magnetotellurics (MT) natural electromagnetic field variations are recorded to study the electrical conductivity structure of the subsurface. Thereby long time-series of electromagnetic data are subdivided into smaller segments, which are Fourier transformed and typically averaged in a statistically robust manner to obtain MT transfer functions. Unfortunately, nowadays the presence of man-made electromagnetic noise sources often deteriorates a significant fraction of the recorded time-series by overprinting the desired natural field variations. Available approaches to obtain undisturbed and high quality MT results include, for example robust statistics, remote reference or multi-station analyses which aim at the removal of outliers or uncorrelated noise. However, we have observed that intermittent noise often affects a certain time span resulting in a second cluster of transfer functions in addition to the expected true MT distribution. In this paper, we present a novel criterion for the detection and pre-selection of EM noise in form of outliers or additional clusters based on a distance measure of each data segment with regard to the centre of the data distribution. For this purpose, we utilize the Mahalanobis distance (MD) which computes the distance between two multivariate points considering the covariance matrix of the data that quantifies the shape and the size of multivariate data distributions. As the MD considers the covariance matrix, it corrects not only for different variances but also for any correlation between the data. The computation of both, the mean value and covariance matrix, is susceptible to ouliers (e.g. noise) and requires a statistically robust estimation. We tested several robust estimators, for example median absolute deviation or minimum covariance determinant algorithm and finally implemented an automatic criterion using a deterministic minimum covariance determinant algorithm. We will present results using MT data from various field experiments all over the world, which illustrate successfull data improvement. This approach is able to remove scattered data points as well as to reject complete data cluster originating from noise sources. However, like all purely statistical algorithms the criterion is limited to cases where the majority of the recorded data is well-behaved, that is noise content is below 50 per cent. If the majority of data points originates from noise sources, the new criterion will fail if used in an automatic way. In these cases, additional input by the user either manually or in an automated fashion can be utilized. We therefore suggest to use an add-on criterion to back the MD selection and subsequent robust stacking in form of a physically motivated constraint based on the magnetic incidence direction. This property indicates whether the magnetic field originates from various sources in the far field or from a strong and well defined source in the near field. KW - Magnetotellurics KW - Statistical methods KW - Time-series analysis Y1 - 2019 U6 - https://doi.org/10.1093/gji/ggz197 SN - 0956-540X SN - 1365-246X VL - 218 IS - 3 SP - 1853 EP - 1872 PB - Oxford Univ. Press CY - Oxford ER - TY - JOUR A1 - Platz, Anna A1 - Weckmann, Ute A1 - Pek, Josef A1 - Kovacikova, Svetlana A1 - Klanica, Radek A1 - Mair, Johannes A1 - Aleid, Basel T1 - 3D imaging of the subsurface electrical resistivity structure in West Bohemia/Upper Palatinate covering mofettes and quaternary volcanic structures by using magnetotellurics JF - Tectonophysics : international journal of geotectonics and the geology and physics of the interior of the earth N2 - The region of West Bohemia and Upper Palatinate belongs to the West Bohemian Massif. The study area is situated at the junction of three different Variscan tectonic units and hosts the ENE-WSW trending Ohre Rift as well as many different fault systems. The entire region is characterized by ongoing magmatic processes in the intra-continental lithospheric mantle expressed by a series of phenomena, including e.g. the occurrence of repeated earthquake swarms and massive degassing of mantle derived CO2 in form of mineral springs and mofettes. Ongoing active tectonics is mainly manifested by Cenozoic volcanism represented by different Quaternary volcanic structures. All these phenomena make the Ohre Rift a unique target area for European intra-continental geo-scientific research. With magnetotelluric (MT) measurements we image the subsurface distribution of the electrical resistivity and map possible fluid pathways. Two-dimensional (2D) inversion results by Munoz et al. (2018) reveal a conductive channel in the vicinity of the earthquake swarm region that extends from the lower crust to the surface forming a pathway for fluids into the region of the mofettes. A second conductive channel is present in the south of their model; however, their 2D inversions allow ambiguous interpretations of this feature. Therefore, we conducted a large 3D MT field experiment extending the study area towards the south. The 3D inversion result matches well with the known geology imaging different fluid/magma reservoirs at crust-mantle depth and mapping possible fluid pathways from the reservoirs to the surface feeding known mofettes and spas. A comparison of 3D and 2D inversion results suggests that the 2D inversion results are considerably characterized by 3D and off-profile structures. In this context, the new results advocate for the swarm earthquakes being located in the resistive host rock surrounding the conductive channels; a finding in line with observations e.g. at the San Andreas Fault, California. KW - Magnetotellurics KW - Ohre Rift KW - Conductive channel KW - Fluid/magma reservoir KW - Earthquake swarm Y1 - 2022 U6 - https://doi.org/10.1016/j.tecto.2022.229353 SN - 0040-1951 SN - 1879-3266 VL - 833 PB - Elsevier CY - Amsterdam ER -