The Statistical Reservoir Model: calibrating faults and fractures, and predicting reservoir response to water flood

Abstract

This paper describes the new concept of a ‘Statistical Reservoir Model’ to determine significant well-pair correlations. We solve this conceptual problem using a predictive error filter, combined with Bayesian methods that identify those well pairs that are related to each other with statistical significance, for the Gullfaks reservoir in the North Sea. Significant, long-range, correlations in the whole field are found at an optimal time lag of one month. The correlation function for significantly-correlated well pairs, after normalization for the distribution of available wells, shows a long-range power-law decay that is consistent with a critical-point response at the reservoir scale. A principal component analysis shows a strong correlation with the location and orientation of faults that intersect the main producing horizon. A predictive experiment shows that the model performs very well both in history matching and predictive mode on a time scale of about one month.

Modelling of Depletion-Induced Microseismic Events by Coupled Reservoir Simulation: Application to Valhall Field

“…14 The 3D coupled reservoir simulations have demonstrated that depletion-induced reservoir compaction has a significant impact on the deformation of the overburden related to stress change, displacements, strains, and microseismic events. This depletionrelated deformation also featured long-range, stress-and fault-related characteristics that are also prevalent in microseismic data and correlations of flow rate fluctuations (Main et al 2007). Field test data indicate that passive seismic monitoring has the potential to provide valuable engineering information for reservoir management, such as well placement planning, well design, waterflood schedule, hydraulic fracturing monitoring and drill-cuttings re-injection (DCRI), and in more applications as presented by Kristiansen (2009).…”

The challenges and impact of compartmentalization in reservoir appraisal and development

“…Møller-Pedersen & Koestler 1997;Coward et al 1998;Jones et al 1998;Nieuwland 2003;McClay 2004;Swennen et al 2004;Boult & Kaldi 2005;Shaw 2005;Sorkhabi & Tsuji 2005;, Jolley et al 2007aLonergan et al 2007;Wibberley et al 2008a). In understanding the uncertainties and risks from compartmentalization these largely lie in the areas of: reservoir scale structural analysis Dee et al 2005;Fossen & Bale 2007); fault and fracture growth, fault zone architecture and characterization Wibberley et al 2008b); fault seal prediction (Bretan et al 2003;Dee et al 2007); the incorporation of faults and fault zone properties in reservoir models and simulation (Manzocchi et al 1999(Manzocchi et al , 2002(Manzocchi et al , 2008Harris et al 2005Harris et al , 2007Childs et al 2007;Fisher & Jolley 2007;Zijlstra et al 2007); integrated fluid description (Smalley & England 1994;Larter & Aplin 1995;Smalley et al 2004); the use of geomechanical data along with field and inter-well scale stress and strain analysis and modelling -incorporating the use of micro-seismicity for surveillance (Heffer 2002;Rutledge et al 2004;Sanderson & Zhang 2004;Main et al 2007;Zhang et al 2007;Osorio et al 2008). …”

“…The advent of high performance computing capability enables advances in numerical geomechanical modelling techniques together with rock property studies (Koutsabeloulis & Hope 1998;Nieuwland 2003;Zhang et al 2007). Resulting numerical stress and strain simulation and forward prediction capabilities offer the opportunity to better model the spatial distribution of the processes controlling the formation of faults and fractures (McClay et al 2002;Main et al 2007;Wilkins 2007). Application of these approaches at a scale an order of magnitude or more below seismic resolution, coupled with forward modelling of strain from restored reconstructions of 3D structural restorations, enables an understanding of subseismic fault predictions to be determined with an understanding of the related uncertainty range.…”

“…Application of these approaches at a scale an order of magnitude or more below seismic resolution, coupled with forward modelling of strain from restored reconstructions of 3D structural restorations, enables an understanding of subseismic fault predictions to be determined with an understanding of the related uncertainty range. With advances in non-orthogonal grid manipulation techniques and the ability to incorporate local grid refinement in reservoir engineering simulation software, we have an opportunity to integrate these geomechanical predictions with multiple sedimentological, geochemical, fault and fracture property realizations (Harris et al 2007;Jolley et al 2007b;Main et al 2007;Mäkel 2007;Zhang et al 2007;Manzocchi et al 2008). When the modelling system is linked to real-time (down hole) reservoir monitoring data, we will be able to make live updates of our models and forward prediction of primary, secondary and tertiary recovery behaviour and sweep efficiency (Reddick 2006).…”

“…We have begun to place greater emphasis on understanding what production data is telling us about fluid communication, reservoir geology and the subsequent changes in reservoir properties that occur over production time-scales. Production and injection data that historically provided information on fluid flow rate are now recognized as containing dynamic behaviour signals (Heffer et al 1995;Heffer 2002;Main et al 2007). As an oil field matures it can often display increasing mechanically influenced dynamic responses.…”