Uncertainty analysis using experimental design and response surface techniques has been extensively used in the field of reservoir simulation. This study outlines an innovative workflow to generate multiple realizations of forward stratigraphic modelling of three Lower Cretaceous reservoirs from onshore Abu Dhabi.
Forward stratigraphic modelling is a deterministic technique that simulates basin infill providing a better understanding of vertical and lateral facies distribution and connectivity in sedimentary basins. During the course of forward modelling a variety of environmental and stratigraphic parameters are used. Due to the uncertainties of these parameters it is critical to assess their impact on the development of the basin fill. The experimental design and response surface techniques have been innovatively applied at reservoir scale to enhance the understanding of major controlling parameters on carbonate production and to produce alternative facies distribution scenarios in the study reservoirs.
The methodology used in this study was based on running multiple simulations, through varying key input parameters. The best stratigraphic models were then selected based on calibration quality and geological consistency. Calibration quality was assessed by two user defined quantitative functions called Thickness Calibration Indicator and Rock Texture Calibration Indicator.
The initial step in the workflow identified uncertain environmental parameters (e.g. eustasy, carbonate production versus depth, carbonate production versus time, wave parameters, gravity and wave transport and erosion rates) from a manually calibrated reference case and ranges of values for each parameter defined based on the knowledge of geology over the area. Latin Hypercube experimental design was then used to define a set of simulations to allow an efficient and uniform sampling of the entire uncertain domain. Sensitivity analysis was then performed on simulation responses (texture and thickness calibration indicators) using the technique of nonparametric Response Surface Modelling (RSM). The influence (quantitative and qualitative) of the impacting parameters on responses was studied to identify the most influential parameters as well as the ranges yielding good calibration indicator values.
A further set of simulations was then launched that considered the most influential parameters and their precise ranges. Non critical parameters were assigned with the constant values from the reference case model. These simulations generated a series of well calibrated models. A filtering of simulations with high calibration indicator values and good geological consistency was then performed to choose acceptable multi-realizations. Finally, thickness and texture confidence properties were mapped based on the selected multi-realizations and the reference case.
Sensitivity study on three Lower Cretaceous reservoirs from onshore Abu Dhabi successfully addressed the uncertainty associated with forward stratigraphic model input parameters. Sensitivity analysis was performed using Experimental Design and RSM. This was applied to enhance the understanding of the major controlling environmental parameters on carbonate production for individual sequence with each of the study reservoirs.