About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

GCAGS Transactions

Abstract


Gulf Coast Association of Geological Societies Transactions Vol. 58 (2008), Pages 755-758

EXTENDED ABSTRACT: Calibration of Predicted Pore Pressure: Perceptions and Pitfalls

Selim Shaker

11221 Gladewick Dr., Houston, Texas 77077

EXTENDED ABSTRACT

Transforming petrophysical properties into pore pressure (PP) is a very challenging process to which unscientific calibration templates are often applied.

The assumption that measured pore pressure in permeable beds (sand) is equal to predicted pore pressure in low permeable beds (shale, mud) does not hold true in most cases. If it were true, drilling problems and hydrocarbon trapping risk would be negligible.

Before drilling, seismic interval velocity extracted from root mean square (RMS) velocity is widely used for this transformation purpose. Velocity picks from normal move out (NMO) gathers should be checked against semblance for quality control. Sequence stratigraphy can help in picking the shale velocity intervals. During drilling, a bundle of data becomes available for PP prediction (PPP), measurements, and calibration. Logging while drilling (LWD), measurement while drilling (MWD), conventional logs, mud log, and direct PP measurements in the sand (RFT [Repeat Formation Tester], MDT [Modular Formation Dynamics Tester], DFT [Drilling Formation Tester], FPWD [Formation Pressure While Drilling), and engineering drilling records are used to establish the subsurface geopressure profile to reach the target objective in a safe and economically feasible manner.

Seismic velocity of a proposed wildcat location and data derived from offset wells, such as sonic and resistivity, are utilized to predict the subsurface geopressure profile. A pre-drilling geopressure model is built based on the variables of velocity picks, normal compaction trends slope, and effective stress exponents.

The prediction of pore pressure is primarily established based on the divergence of the petrophysical measurements from the normal compaction trend. In the transition zone between the hydrostatically pressured and geopressured systems, formation water is expelled gradually from sediments due to pressure gradient drop from deeper to shallower depth. In this transition zone velocity, resistivity, and density increase downward concurrent with the rate of the dewatering process. The Normal Compaction Trend (NCT) represents the optimum fitted linear trend of these measured data in the low permeable beds in this transition zone. Inversely, in the geopressured system (where water is no longer capable of escaping) velocity, density, and resistivity measurements decrease in the low permeable beds.

The effective stress model of transforming the petrophysical measurement (e.g., sonic slowness) to pore pressure in the fine clastic (shale, mud, fine silt) beds is based on:

PP = PS – ES

PPz = OBz– (OBz– Pnz) * (ΔTn/ΔTo)^X

where PP = predicted pore pressure, PS = principal stress = overburden (OB) in case of passive structure areas, ES = effective stress, Z = depth to point of measurement, Pn = the normal pressure at depth Z, ΔTn = the assumed normal sonic slowness at depth Z (calculated from the NCT), ΔTo = the observed (measured) sonic slowness at depth Z, and X = PP transformation exponent (variable with age/basin location).

Therefore, the keystone for this prediction practice is the value of ΔTn /ΔTo, which is mainly conveyed as a result of establishing the slope on the NCT (Fig. 1).

During and post-drilling, a calibration fitting process should be applied to the modeled values of shale beds. However, some of these calibrating methods are based on illogical assumptions, such as:

  • Predicting pore pressure in sand using the same effective stress model designed for shale (Fig. 2).
  • Assuming measured pore pressure in sand is equal to predicted pore pressure in shale (Fig. 2).
  • Applying universal effective stress exponent.
  • Breaking of the normal compaction trend to selective segments, due to faulting or other reasons, to fit the predicted pore pressure with the drilling records and the measured pressure data during and post drilling is another interpretation pitfall. This leads to incorrect adjustments of the .Tn/.To values for the purpose of calibration and consequently compromises the effective stress transformation model (Fig. 3).
  • Using pressure-depth plots in PPG-MWE (pounds per gallon of mud weight equivalent) for calibrating the predicted values to the measured ones (Fig. 4).

 

Figure 1. The effective stress transformation model from velocity (on the left panel) to pore pressure(on the right panel). Notice the importance of defining the slope on the normal compaction trend (NCT) and the exponent X.

 

Figure 2. Pitfalls of predicting pressure in sand beds in Kerr McGee well #1 in East Breaks Block 602 (Nansen Field). The left panel tries to match the predicted pressure (PPP psi shale) with the measured pressure (PP psi sand) for the interval between 11300-11500 ft (3444-3505 m). The result is much higher PPP (1500 psi) than mud weight (MW), which is not possible. On the right panel, matching is done on the interval between 10600-10900 ft (3231-3322 m), and excessive mud weight overbalance (1700 psi) is observed at the lower zone between 11300-11500 ft (3444-3505 m).

 

Figure 3. Two examples of breaking the normal compaction trend for the purpose of calibration.

Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Watermarked PDF Document: $14
Open PDF Document: $24