One Dimensional Stochastic Inversion (ODiSI) Success Stories

One Dimensional Stochastic Inversion (ODiSI) is an innovative approach to seismic inversion. Initially developed by BP, ODiSI is a method that produces high-quality reservoir property estimates with associated uncertainties. Associating all captured prior knowledge about a given field, prospect or reservoir, ODiSI builds a robust and comprehensive model of the subsurface through pseudo-wells.

ODiSI: In Simple Terms

ODiSI jointly estimates facies, reservoir properties, and impedances, and their associated uncertainties, by creating a set of pseudo-wells at each seismic trace position. Rock Physics Models (RPM) are used to generate physical properties for each pseudo-well from robust prior information, and Extended Elastic Impedance (EEI) is used to derive synthetics from the pseudo-wells. The synthetic seismic traces are then statistically matched to the seismic trace, and the properties attached to the 100 best matching pseudo-wells are averaged to obtain different outputs. These outputs include net reservoir fraction, porosity, hydrocarbon pore volume, and lithofacies probabilities at each trace location.

ODiSI results provide several key advantages to geoscientists. Perhaps the most significant being that it is relatively simple to understand and employ. The most robust prior geological information extracted from well data or field analogs fuel the inversion process, giving the user complete control and enable them to QC each step of the process. 

Since its development and its later commercialization, ODiSI’s transparency and value have been recognized by several companies and applied within a range of reservoirs, prospects, and fields. In this article, we will examine just a few of the use cases in which ODiSI has successfully been applied.

Read also: BP Seismic Inversion Technology Commercialized by Cegal

ODiSI for Net-to-Gross Prediction in Siliciclastic Fields

BP has successfully applied ODiSI to numerous fields worldwide, providing estimates of the reservoir properties and associated uncertainties. The following examples show how field teams can apply ODiSI for optimized well planning, enhanced geological understanding of reservoir intervals, and improved reservoir modeling in a range of different siliciclastic fields of simple geology¹:  

• Offshore Angola: BP applied ODiSI to two relatively simple fields in offshore Angola. Considering the data quality of the seismic surveys and the analytic appropriateness of its seismic rock properties, offshore Angola proved to be an attractive field for early ODiSI testing. The field was modeled by using clean sand, shaly sand, cemented sand, and shale as its four lithofacies. Using only these four lithofacies types in a simple ODiSI parameterization was sufficient to yield good quality blind well ties and reasonable estimates of net-to-gross and net-pay thickness. ODiSI was also used in another offshore Angola field to predict net-to-gross for optimized well locations. The field had three target intervals, but initial predictions on net pay from seismic proved difficult. To overcome these challenges, the field team combined other subsurface data with new insights from the ODiSI product to predict net pay thickness to within five meters. As a result, the field team was able to move the well to a more suitable and optimal location.

• Nile Delta: BP also applied ODiSI to a Nile Delta field to successfully estimate net-to-gross. Using ODiSI, the field team generated maps of net-to-gross from two modeled reservoir intervals, showing clear geological features and high quality well tie. The findings are now actively being used to aid in well planning and to improve volumetric estimates.

• The North Sea: BP likewise applied ODiSI to a geologically-simple North Sea field. Compared to the offshore Angola example, the North Sea seismic data quality was suboptimal and had a lower resolution because a higher signal-to-noise full-stack offset was used above the contact. Although reservoir property estimations offered precise well ties, the model had layers of constant net-to-gross rather than a 3D volume. Applying ODiSI improved the history match compared to the old results and the field team now uses the ODiSI volumes of net-to-gross output as a direct input to the reservoir model. 

Although only four lithofacies types were statistically defined in the pseudo-well generation process in the above case examples, ODiSI can also be applied to more geologically complex fields. 

Read also: Effective Seismic Inversion Through Microsoft Azure Cloud

Quantifying the Extent of High-Quality Sand in an Upper Jurassic Fulmar Reservoir

Additionally, ODiSI was applied to a central North Sea field [link to content offering], this time in an attempt to quantify the extent of high-quality sand. Triassic shales and Zechstein Evaporites underlie the field, and Cretaceous lithologies overlay it, providing seismic analysis challenges, particularly with regards to inversion and Quantitative Interpretation (QI).

To quantify the extent of high-quality sand in the reservoir, E&P company Repsol Sinopec Resources UK applied ODiSI in an attempt to delineate the presence and lateral extent of the range of lithology types in the field. The ODiSI parameterization was performed on four log wells located within the field and crossing the reservoir. 

Applying ODiSI solved two notable challenges encountered in the field: the presence of relatively high impedance over- and under-burden in combination with low acoustic differentiation within the reservoir. ODiSI successfully delineated regions of high net reservoir quality sand within the field, regions consistent with blind wells. Creating a set of pseudo-wells at each seismic trace position and then matching the best pseudo-wells to the trace, ODiSI generated different 3D volumes, such as probable lithofacies and net-to-gross. These volumes presented good lateral continuity in the property and lithofacies outcome, providing strong validation of the ODiSI method and process.

Discover Blueback ODiSI