An integrated approach for modeling depositional facies and diagenetic trends to capture heterogeneities in a lower cretaceous carbonate reservoir

  • Mina Sameh Yousef Salib

Student thesis: Master's Thesis


Carbonate reservoirs can be complex due to diagenetic overprint of original depositional facies. This overprint affects the porosity and permeability heterogeneity and as a result, needs to be incorporated in the model. Reservoir property modeling routinely relies heavily on geostatistical propagation of well data, without honoring geological depositional and/or diagenetic conceptual models. The resulting grids can be unrealistic and produce unreliable predictions. Nevertheless, when used, facies don't tend to have any relationship with petrophysical properties, if classified using the classical approaches which are based on texture (e.g. Dunham classification). The main objective of this study was to integrate the depositional and diagenetic patterns within the sequence stratigraphic framework and be able to model it successfully to establish the link between geology and petrophysics. Having a suitable facies classification scheme was key input to the workflow implemented in this study. Diverting from the usual Dunham (1962) facies classification scheme has provided the advantage and freedom of being able to incorporate more information about the rock final state and texture into its nomenclature. This, in turn, resulted in the possibility of assessing the reservoir quality through the ‘diagenetic facies packages', the name of which directly indicates the predicted quality. On a porosity-permeability crossplot such facies are expected to display fairly distinct clusters/clouds, according to their diagenetic class. The proposed workflow for this project started with the descriptive phase which involved core description and sequence stratigraphy analysis. This was followed by the interpretive phase which involved analyzing the results for model input in the form of discrete depositional facies logs, diagenetic facies logs, and generating conceptual sedimentological trend maps. Finally, in the modeling phase, all the gathered information was incorporated to be appropriately captured in the static model. Hierarchical facies modeling was carried out in which large scale depositional facies form the main framework within which the diagenetic facies are distributed stochastically, simulating the possible local variations within each depositional facies region. Being sensitive to diagenetic overprint these variations give an insight into the reservoir quality fluctuations. The porosity model was then built, conditioned to the generated diagenetic model constraining the porosity ranges to each particular diagenetic facies, thus capturing the reservoir quality variations. Permeability was then conditioned to the diagenetic facies and populated using bivariate distribution with porosity being the secondary variable. This relationship was captured through poro-perm crossplots per diagenetic facies per zone. Blind tests were carried out to validate the model and showed promising results in terms of facies and porosity predictability. Having sparse data points due to low number of wells has introduced some uncertainties in this model as it was being built. The boundaries of the depositional packages used for sedimentological mapping are dependent on the available well data and can be enhanced as more wells are drilled. Higher uncertainty is associated with these polygons towards the flanks of the field as there is no well control downflank. Also, for a more proper variogram analysis (due to large number of diagenetic facies classified), it is advisable to incorporate a larger number of wells in the future. Overall, the model is believed to be more representative of the geological understanding as it combines sedimentological framework and diagenetic overprint which provides a robust framework for predicting reservoir architecture and can potentially lead to a better field development planning and improved decision making.
Date of Award2016
Original languageAmerican English
SupervisorJorge Salgado Gomes (Supervisor)


  • Applied sciences
  • Depositional
  • Diagenetic
  • Facies modeling
  • Trend modeling
  • Petroleum engineering
  • 0765:Petroleum engineering

Cite this