Datapages, Inc.Print this page

Forecasting More Realistic Reservoir Performance - Observations from Studies of the Wafra First Eocene Carbonate Reservoir and Other Reservoirs

Meddaugh, William *1; Champenoy, Nicole 1; Osterloh, W. Terry 1
(1) Chevron, Houston, TX.

Reservoir forecasts tend to be optimistic. Forecasts for EOR projects tend to be particularly optimistic. Sources of the optimism can be divided into several broad technical categories including: (1) Data - quantity, quality, appropriateness; (2) Static Modeling - model detail/complexity, particularly for permeability contrasts (barriers and baffles to flow), model parameters, and model algorithms; and, (3) Dynamic Modeling - model detail/complexity (e.g. grid size), upscaling, well location optimization. In addition, hard to quantify human factors also tend to drive projects towards optimistic forecasts.

The Partitioned Zone (PZ) Wafra First Eocene steamflood pilots, both the single pattern Small Scale Test (6 wells in a 1.25 acre area) and the 16-pattern Large Scale Pilot (60 wells in a 40 acre area) provide unique data with which to assess the impact of a number of the above factors listed above on forecast quality for EOR development. Among the most critical modeling parameters are the areal grid size, particularly for dynamic models. Models with a small number of grid cells (e.g. less than 5-10) between producer and injector wells tend to optimistic compared to models with many grid cells (e.g. more than 10-15) between producer and injector wells. Also important is the semivariogram model range, particularly for waterflooding. Forecasts for waterflooding generated from models constructed with large semivariogram ranges tend to be more conservative than models constructed using small semivariogram ranges. Forecasts for steamflooding are relatively insensitive to the semivariogram parameter.

One of the most significant sources of optimistic forecasts is optimistic estimates of the original or net/targeted in place hydrocarbon volume. Some bias is due to sampling, particularly for green-field developments, and this bias can be reduced statistically or by use of appropriate uncertainty-based workflows together with a reasonable uncertainty assessment that includes the available data and an appropriate suite of analogs. A less appreciated source of potentially significant forecast optimism is the use of stochastic models for manual or automated well location optimization.


AAPG Search and Discovery Article #90141©2012, GEO-2012, 10th Middle East Geosciences Conference and Exhibition, 4-7 March 2012, Manama, Bahrain