3D object-oriented image analysis in 3D geophysical modelling: Analysing the central part of the East African Rift System

Fadel, Islam; Meijde, M. van der; Kerle, N.; Lauritsen, Nynne Louise Berthou

doi:10.1016/j.jag.2013.11.004

Cited by 10 publications

(8 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This might be particularly useful for the inversion of potential field data where lateral variations are often reasonably well modelled but the depth resolution is very poor, thereby leaving a large degree of uncertainty on the final model in both object boundaries and property characteristics. A case study implementation for reducing non-uniqueness in satellite gravity inversions is shown by Fadel et al (2013). This also might be particularly relevant for recent satellite gravity studies which use seismic tomography models (for a review, see van der Meijde et al 2013b).…”

Section: Discussion a N D C O N C L U S I O N Smentioning

confidence: 96%

3-D object-oriented image analysis of geophysical data

Fadel

Kerle

Meijde

2014

Geophysical Journal International

Self Cite

View full text Add to dashboard Cite

Geophysical data are the main source of information about the subsurface. Geophysical techniques are, however, highly non-unique in determining specific physical parameters and boundaries of subsurface objects. To obtain actual physical information, an inversion process is often applied, in which measurements at or above the Earth surface are inverted into a 2-or 3-D subsurface spatial distribution of the physical property. Interpreting these models into structural objects, related to physical processes, requires a priori knowledge and expert analysis which is susceptible to subjective choices and is therefore often non-repeatable. In this research, we implemented a recently introduced object-based approach to interpret the 3-D inversion results of a single geophysical technique using the available a priori information and the physical and geometrical characteristics of the interpreted objects. The introduced methodology is semi-automatic and repeatable, and allows the extraction of subsurface structures using 3-D object-oriented image analysis (3-D OOA) in an objective knowledge-based classification scheme. The approach allows for a semi-objective setting of thresholds that can be tested and, if necessary, changed in a very fast and efficient way. These changes require only changing the thresholds used in a so-called ruleset, which is composed of algorithms that extract objects from a 3-D data cube. The approach is tested on a synthetic model, which is based on a priori knowledge on objects present in the study area (Tanzania). Object characteristics and thresholds were well defined in a 3-D histogram of velocity versus depth, and objects were fully retrieved. The real model results showed how 3-D OOA can deal with realistic 3-D subsurface conditions in which the boundaries become fuzzy, the object extensions become unclear and the model characteristics vary with depth due to the different physical conditions. As expected, the 3-D histogram of the real data was substantially more complex. Still, the 3-D OOA-derived objects were extracted based on their velocity and their depth location. Spatially defined boundaries, based on physical variations, can improve the modelling with spatially dependent parameter information. With 3-D OOA, the non-uniqueness on the location of objects and their physical properties can be potentially significantly reduced.

show abstract

Section: Discussion a N D C O N C L U S I O N Smentioning

confidence: 96%

3-D object-oriented image analysis of geophysical data

Fadel

Kerle

Meijde

2014

Geophysical Journal International

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, using vote maps to extract features from a collection of seismic tomographies is the most basic way to do this, and a more refined method (e.g. Fadel et al, 2015) could lead to better constrained volumes. In particular it might benefit an inversion if uncertainties could be placed on the size and position of the volumetric features to allow them to change during inversion.…”

Section: Discussionmentioning

confidence: 99%

“…The regularization parameters β and γ exert critical control over the resulting density structure and the achievable data fit. The most popular methods to determine such parameters are L curves and cross-validation (Farquharson and Oldenburg, 2004). Here we use cross-validation to estimate the regularization parameter.…”

Section: Cross-validation Inversion Approachmentioning

confidence: 99%

Increased density of large low-velocity provinces recovered by seismologically constrained gravity inversion

2020

View full text Add to dashboard Cite

Abstract. The nature and origin of the two large low-velocity provinces (LLVPs) in the lowest part of the mantle remain controversial. These structures have been interpreted as a purely thermal feature, accumulation of subducted oceanic lithosphere or a primordial zone of iron enrichment. Information regarding the density of the LLVPs would help to constrain a possible explanation. In this work, we perform a density inversion for the entire mantle, by constraining the geometry of potential density anomalies using tomographic vote maps. Vote maps describe the geometry of potential density anomalies according to their agreement with multiple seismic tomographies, hence not depending on a single representation. We use linear inversion and determine the regularization parameters using cross-validation. Two different input fields are used to study the sensitivity of the mantle density results to the treatment of the lithosphere. We find the best data fit is achieved if we assume that the lithosphere is in isostatic balance. The estimated densities obtained for the LLVPs are systematically positive density anomalies for the LLVPs in the lower 800–1000 km of the mantle, which would indicate a chemical component for the origin of the LLVPs. Both iron-enrichment and a mid-oceanic ridge basalt (MORB) contribution are in accordance with our data, but the required superadiabatic temperature anomalies for MORB would be close to 1000 K.

show abstract

“…For these models alternative analysis approaches might be more relevant. Recent developments use object oriented segmentation approaches for upper mantle seismic tomography models to obtain a priori information on subsurface tectonical objects (Fadel et al, 2014(Fadel et al, , 2015. With this approach no generic starting values (density, thickness) for the whole continent have to be selected (Tugume et al, 2013;van der Meijde et al, 2013a).…”

Section: The Way Forwardmentioning

confidence: 99%