Kv/Kh in Heterogeneous Reservoirs of a Brown Field of the Niger Delta of Nigeria – Valuable Uncertain Parameter for Assisted History Matching

Abstract

It is commonly practice in reservoir modelling to use constant kv/kh values to define vertical permeability in numerical reservoir models. This is often in response to paucity of core data and need to reduce simulation run time. That traditional approach works fine in fairly homogeneous reservoirs; however, in reservoirs with heterogeneity, due to more complex depositional environment like shallow marine deposits, it is not necessarily a good practice to assume single averaged kv/kh values to represent vertical connectivity for such reservoirs. This paper highlights the importance of understanding reservoirs' vertical connectivity by properly defining kv/kh as a function of lithofacies and evaluating its impact on history match and prediction outcomes on one case study for a heterogeneous brown field reservoir with complex depositional environment in the Niger Delta. Core data, wellsite cuttings data along with available production, and pressure data were used to calibrate the petrophysical and numerical dynamic models. The study also considers substantially a multi-disciplinary approach to uncertainty management, using Experimental Design (ED) methods to understand and validate kv/kh ranges as a function of lithofacies. This methodology was successfully implemented across several oil reservoirs (D2000-D7000 EXHJK) in a recently concluded field project in the Niger delta of Nigeria. As an example, this paper discusses the results, and recommendations for one reservoir (D5000X) selected out of this project study. The main objective from this project is the importance of understanding vertical connectivity in a complex reservoir by capturing heterogeneity and major geological features in the integrated 3D dynamic model aiming to achieve acceptable history match at reservoir level, but also more importantly on a well-by-well basis. The robustness of the integrated dynamic model provided higher level of confidence for prediction, and resulted with increased certainty for the forecasts by a factor of 3 (reduced solution space) after implementing variable kv/kh as function of facies.

Analysis of Vertical Permeability and Its Influence on CO2 Enhanced Oil Recovery and Storage in a Carbonate Reservoir

“…This is because a long injection duration causes more CO 2 to be affected by buoyant flow, and the influence of k z on CO 2 displacing oil is thereby enhanced. Campero et al (2014) found that using constant k z /k x to populate vertical permeability works in homogeneous reservoirs but fails in heterogeneous reservoirs with complex depositional environments (related to autocorrelation length). This is consistent with our findings that the oil production and CO 2 retention metrics change as the horizontal autocorrelation length increases.…”

“…Unfortunately, most flow simulations that consider k z typically treated the k z /k x ratio as a sensitivity parameter and examined its influence on defined metrics (e.g., Kerans et al 1994;Ren and Duncan 2021;Abdelaal et al 2021). Campero et al (2014) made an analysis of k z /k x based on the geological data sets collected in a field and evaluated the influence of k z /k x ratio on history match and prediction during waterflooding. However, all these studies avoided a fundamental question: how good is using a constant k z /k x ratio to populate vertical permeability for flow prediction as compared to having variable k z ?…”