Non-Stationary Multiple-point Geostatistical Models

Strebelle, Sebastien; Zhang, Tuanfeng

doi:10.1007/978-1-4020-3610-1_24

Cited by 27 publications

(22 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Stationarity is a known problem in classical MultiplePoints Statistic studies (e.g., Caers and Zhang (2002); Strebelle and Zhang (2005); Mirowski et al (2009)). By default, the training image must show a stationary statistical distribution, a stationary orientation, and a stationary scale (Mirowski et al, 2009).…”

Section: Datamentioning

confidence: 99%

“…Here, the first requirement is not fulfilled : the deposits show a clear trend from proximal to distal. The information of distality could, therefore, be used as secondary variable, to help to account for the orientation, as proposed by Strebelle and Zhang (2005). The distality could also be discretized to split the training model into different zones to sample the corresponding information, as de Vries et al (2009) suggest.…”

Section: Datamentioning

confidence: 99%

See 1 more Smart Citation

Uncertainty management in stratigraphic well correlation and stratigraphic architectures: A training-based method

Edwards

Lallier

Caumon

et al. 2018

Computers & Geosciences

View full text Add to dashboard Cite

We discuss the sampling and the volumetric impact of stratigraphic correlation uncertainties in basins and reservoirs. From an input set of wells, we evaluate the probability for two stratigraphic units to be associated using an analog stratigraphic model. In the presence of multiple wells, this method sequentially updates a stratigraphic column defining the stratigraphic layering for each possible set of realizations. The resulting correlations are then used to create stratigraphic grids in three dimensions. We apply this method on a set of synthetic wells sampling a forward stratigraphic model built with Dionisos. To perform cross-validation of the method, we introduce a distance comparing the relative geological time of two models for each geographic position, and we compare the models in terms of volumes. Results show the ability of the method to automatically generate stratigraphic correlation scenarios, and also highlight some challenges when sampling stratigraphic uncertainties from multiple wells.

show abstract

Section: Datamentioning

confidence: 99%

Section: Datamentioning

confidence: 99%

Uncertainty management in stratigraphic well correlation and stratigraphic architectures: A training-based method

Edwards

Lallier

Caumon

et al. 2018

Computers & Geosciences

View full text Add to dashboard Cite

show abstract

“…The process itself involves modeling of facies grids with correct geometries and dimensions within the field. Facies discretization has been applied by Pyrcz et al (2012), Harding et al (2004) and Strebelle and Zhang (2004) on different hydrocarbon fields; therefore, the approach is taken to capture detailed facies associations within Mangahewa reservoir as well. It generally follows tolerable computing response with each assigned grid "nodes."…”

Section: Algorithmic Approachmentioning

confidence: 99%

Integrated 3D facies modeling of the Mangahewa Formation, Maui Gas Field, Taranaki Basin, New Zealand

Haque

Islam

Shalaby

et al. 2018

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

Abstract3D seismic data, well logs, core-based lithofacies and photographs have been combined to interpret and model the depositional facies of the Mangahewa Formation of the Maui Gas Field, Taranaki Basin, New Zealand. The primary objective of the study is to generate a robust facies model for the Middle to Late Eocene (47-37 Ma) Mangahewa Formation of the field. The facies model has included eighteen depositional facies spatially distributed over the gas field. These facies are further subgrouped into three broad depositional facies associations, namely marginal marine, shallow marine and offshore environment. We have identified that marginal marine is the most dominant facies association (64%) within the model. The model visualizes estuarine and shoreface sand geobodies dominating over other facies within the model. Both geobodies comprise over 40% of all the facies interpreted in the field. The entire modeling process involves a novel stochastic approach using unique workflow that follows 3D gridding, coding of the facies classes and multiple iterations over the interpreted facies. The model therefore realistically visualizes potential facies responsible for "good"-quality reservoir sands in the Mangahewa Formation with possible retrogradation from older to younger succession.

show abstract

“…For all these reasons, the use of indicator kriging or disjunctive kriging would not have allowed us to comprehensively test our hypothesis that "non-linearity and multi-point pattern recognition can improve the spatial mapping of complex patterns of groundwater contamination significantly". It is appropriate to mention herein that the problem of two-point statistics in ordinary kriging has also been perceived for a long time by the kriging community, and that a new approached has been developed, called "multipoint geostatistics" (Strebelle and Zhang, 2005).…”

Section: Region 1 Regionmentioning

confidence: 99%

How much does inclusion of non-linearity and multi-point pattern recognition improve the spatial mapping of complex patterns of groundwater contamination?

Chowdhury

Alouani

Hossain

2009

Nonlin. Processes Geophys.

View full text Add to dashboard Cite

Abstract.In this brief communication, we discuss the implication of the hypothesis that "non-linearity and multi-point pattern recognition can improve the spatial mapping of complex patterns of groundwater contamination". The discussion is based on our recently published work in Stochastic Environmental Research and Risk Assessment. Therein we have found that the use of a highly non-linear pattern learning technique in the form of an artificial neural network (ANN) can yield significantly superior results under the same set of constraints when compared to the more linear two-point ordinary kriging method.

show abstract

Non-Stationary Multiple-point Geostatistical Models

Cited by 27 publications

References 5 publications

Uncertainty management in stratigraphic well correlation and stratigraphic architectures: A training-based method

Uncertainty management in stratigraphic well correlation and stratigraphic architectures: A training-based method

Integrated 3D facies modeling of the Mangahewa Formation, Maui Gas Field, Taranaki Basin, New Zealand

How much does inclusion of non-linearity and multi-point pattern recognition improve the spatial mapping of complex patterns of groundwater contamination?

Contact Info

Product

Resources

About