Henderson  Petrophysics

Reservoir Fluid Contacts From Formation Pressure Data -
Methods that Include Analysis of Uncertainty

Introduction:

Modern wireline logging tools such as Schlumberger's MDT tool are now routinely used to quickly acquire a very accurate survey of reservoir formation pressure.

In reservoirs with reasonable permeability, that are not volumetrically restricted, the tools record the pressure of the continuous fluid phase in the formation. The formation pressure gradient (slope of line connecting pressure measurements) is proportional to the in-situ fluid density.

One of the many invaluable uses of the formation pressure data is the determination of fluid contacts in reservoirs where the pressure has not been disturbed by production operations.

In the following sections we describe three methods of using the formation pressure data to determine fluid contacts. The first method is the traditional graphical or algebraic method of determining the intersection of two straight lines. The second and third methods involve data visualization and normalization procedures to evaluate the statistics of the Free Fluid Level estimate.

Definitions

Free Fluid Level is the generic term for fluid contacts without the perturbations of reservoir capillary effects. In simpler words, this would be the elevation of the fluid contact if you were to remove all of the rock material. The FFL in gas-water and oil-water systems is usually referred to as the Free Water Level (FWL). In water-wet reservoirs the FWL is always below the lowest occurrence of hydrocarbons. In oil-wet reservoirs the FWL is located at the top of the transition zone.

Software for Estimating Reservoir Fluid Contacts

The Henderson Petrophysics Free Downloads section contains an example EXCEL Spreadsheet and associated documentation for processing formation pressure data to determine formation fluid pressure gradients and fluid contacts.

The Spreadsheet contains Visual Basic for Applications Macros that activate the analysis modules and process data. This Spreadsheet is intented only for advanced EXCEL users who are familiar with processing of formation pressure data. Please look at the ReadMeFirst text file before attempting to open the Spreadsheet. This file discusses issues arising from the use of Spreadsheets with embedded Macros.

The Data Model

Figure 1 shows formation pressure data from a contiguous oil-water system. Subsea depth is plotted on the y-axis and formation pressure is on the x-axis.

	Green triangles: Oil zone formation pressure data Green line: 'Best-fit' linear line through oil zone data Blue circles:Water zone formation pressure data Blue line: 'Best-fit' linear line through water zone data
Figure 1: Formation Pressure Data From Oil and Water Zones

The analysis models presented below use the following nomenclature;

In the oil zone;

• Po is formation pressure at depth Do,
• OG is the oil fluid pressure gradient. Units are psi/m or psi/ft,
• bo is the x-intercept of the formation pressure trend line,

In the water zone;

• Pw is formation pressure at depth Dw,
• WG is the water fluid pressure gradient,
• bw is the x-intercept of the formation pressure trend line.

Method 1 - Intersection of Linear Fluid Pressure Trends

This is the traditional method of algebraically determining a Free Fluid Level.

A straight line through oil column data points has the general form;

(1)

A straight line through water zone data points has the general form;

(2)

These two lines intersect at the Oil-water Free Fluid Depth (Free Water Level), with;

Setting equation 1 equal to equation 2 and solving for pressure yields Pfwl, the formation pressure at the oil-water interface;

(3)

The depth of the Free Water Level is calculated by entering this pressure value into either equation 1 or equation 2.

The above method is easy to implement. However, there is no information on the analysis uncertainty. It is often difficult to identify and remove noisy data or 'outliers' that can result in incorrect estimates of the fluid contact.

Method 2 - Fluid Contact Determination Using Datum Pressure Data

With this method the Free Fluid Level is calculated using formation pressure data that has been projected to a datum (common) depth, Dd.

Figure 2 shows an example of formation pressure data from an oil-water system that has been projected to a datum elevation above the reservoir. Perfect data would project to two pressure data points (one for each of the oil and water trends). The data in figure 2 is of exceptonal quality but there are a few data points that are clearly anomalous. For example, the middle point and second from the right in the oil zone data are off-trend enough to significantly increase the uncertainty (variance) of the calculated fluid contact.

	Plot of formation pressure data that has been projected to a Datum (common) depth Green triangles: Oil zone data Blue circles: Water zone data
Figure 2: Formation pressure data projected to a common (datum) depth

For the oil zone data points datum pressure is calculated using;

(4)

For water zone data points datum pressure is calculated using;

(5)

The Free Water Level (Free Fluid Level in an oil-water system) is defined as the depth where the two Datum fluid pressure values are equal. This is calculated by subtracting the Datum Pressure values and dividing by the difference in fluid pressure gradients;

(6)

One advantage of this method is that the FFL can be calculated using either the average or median (middle) values for the Datum pressure. The median value is preferred because it is less affected by anomalous data points than the average value.

With this method it is also very easy to identify data points that contribute to uncertainty in the FFL estimate. For example, looking at the oil zone datum pressure values in Figure 2, the ninth data point from the left and the second point from the right could be removed from the data set to reduce uncertainty in the FFL estimate. In the water data, the highest and lowest points could be removed from the calculations.

In summary, this method is very good for data visualization and quality control and offers some insight into contributions to FFL uncertainty.

Method 3 - Fluid Contacts Using Data Intersection Points

This method uses all of the pressure data points to estimate the Free Fluid Level. For illustration purposes this example uses data from an oil-water system. Any combination of data points in the oil and water zone can be used to calculate the FWL.

For data in the oil zone pressure at the Free Water level is calculated using;

(7)

For data in the water zone pressure at the Free Water level is calculated using;

(8)

At the Free Water Level Equation 7 and Equation 8 are equal. Combining Equations 7 and 8 the Free Water Level is calculated using

(9)

Since this method uses all of the available pressure data it is easy to evaluate the uncertainty in the FWL estimate. Figure 3 shows the frequency distribution of the Free Water Level estimates. It is very easy to calculate the average, median(middle) and uncertainty (Standard Deviation) values for the FWL estimate.

	Histogram of Free Water level calculated using filtered data: Mobility > 10, and, Elimination of 'Outliers' Average: 794.89 mSS Median: 794.78 mSS St. dev.: 0.73 m
Figure 3:Frequency Histogram of Free Water Level estimates

Reducing Analysis Uncertainty

We have seen how formation pressure data can be used to estimate Free Fluid Level and also to determine the uncertainty of the estimate.

By eliminating unreliable data we can reduce the uncertainty of the FFL estimate. Figure 4 is a Datum Pressure Plot with arrows identifying anomalous data points that can be eliminated from the analysis to improve the accuracy and uncertainty of the FWL estimate.

	Datum Pressure Plot with Arrows identifying anomalous data.
Figure 4:Datum Pressure Plot With Identification of Anomalous Data

Limitations for Data Analysis

Very low permeability formations may be "supercharged". Drilling muds are designed to develop an impermeable mudcake that isolates the reservoir from the higher pressure hydrostaic mud column. Mud filtrate invades the formation while the mudcake is being developed, or after it is damaged during drilling. In permeable reservoirs this fluid and associated pressure differential is quickly diffused into the formation. However, in very low permeability formations the mud filtrate cannot dissipate and pressure measured near the wellbore will be intermediate between the mud hydrostatic pressure and the formation pressure.

Pretests in supercharged formations are characterised by slowly increasing pressure that is higher than the reservoir pressure but never stabilises. Pressure measurements in supercharged formations are not suitable for inclusion in fluid contact calculations.

Sometimes formation pressure measurements appear to be valid based on attainment of stabilised pressure and estimates of drawdown mobility (permeability divided by fluid viscosity). However, the stabilised pressure is slightly higher than expected (see, for example, the middle oil-zone data point in figure 2). One possible explanation for this phenomenon is that the samples zone is very small and is isolated from the larger reservoir unit. An example would be a sand lense with impermeable clay drape structures. This small rock volume is slightly pressure charged by addition of the mud filtrate.

Production of reservoir fluids will always result in some reduction of reservoir pressure. It is very unlikely that there will be a uniform pressure reduction in all of the reservoir units. Therefore, pressure data from depleted reservoirs is unlikely to be suitable for determination of fluid contacts.

Downloads

Click on this link to download the demonstration EXCEL spreadsheet and related documentation

We have prepared a PowerPoint presentation dealing with fluid contact determination using formation pressure data. If you would like a copy just sent us an email (click on the icon below) and type "FFL PowerPoint" in the subject line.