Formation Testing: Part II of III
Posted by D Nathan Meehan April 9, 2011

In the prior blog entry we discussed the tools and interpretation theory for formation testers.  In this entry we will talk about using FTs for a unique aspect of reservoir characterization that can only be done commercially during the drilling phase of a well and before its completion.  The next blog entry will discuss fluid identification applications and other uses of FTs.

Example use of gradients

In a reservoir that is hydrostatic[1] the pressures in the continuous phases vary by depth based on the density of the fluid in the continuous phase. In an oil reservoir, oil density is primarily a function of the gravity of the oil, the amount of dissolved gas and the pressure. Compositional variations in oil density and dissolved gas are common over large areas and in very thick reservoirs.  Nonetheless, it is more common to have an approximately constant pressure gradient over the thickness of an oil reservoir among all wells that are hydraulically continuous. Similarly, water gradients are usually even more constant. Except for very heavy oils (whose density approaches that of water) it is usually possible to distinguish oil and water bearing formations by obtaining multiple formation pressures at different depths in the reservoir. If there is no clear oil/water contact in a wellbore, the use of these gradients can often identify an oil/water contact depth.  Natural gas gradients are typically much less than liquid gradients and can serve a similar purpose.  In thick reservoirs the density of oil usually varies with changing depth. This is quite common and it is possible to detect varying density (compositional grading) from a depth vs. pressure plot using FT data.

The geology of the reservoir is very important when considering gradient analysis, prior knowledge of its structure can serve as lead indicators to what the gradient trends will look like and is demonstrated by the diagram of the single well producing multiple zones from the same field but not necessarily the same HC deposit or communicating reservoirs. A discontinuity in the pressures obtained from formation tests can be used to identify probable hydraulic isolation among reservoirs due to layering, faulting or other geological heterogeneity.

Similarly, varying free water levels will result in different gradient interpretations. In the following example the first illustration shows Well A encountering an oil sand and a water sand. Because the top layer has no water contact, the reservoir engineer could conclude that if the reservoirs were in hydraulic communication that the FWL in the top layer could be at the subsea depth associated with the water encountered in the lower, water-bearing layer. Because the gradient slopes intersect at a subsea depth greater than the known oil levels, the layers are hydraulically discontinuous and such a conclusion cannot be reached.

The two oil layers in the second Well A example are shown to be hydraulically separated and will behave independently during production.

Similarly, pressure gradients and magnitudes shift during production, in transition zones and due to injection. One of the most powerful indicators of bypassed oil in a mature waterflood is a low pressure layer in an infill well. While pressure reductions can be communicated over time in permeable reservoirs, the increase in pressure due to fluid injection requires good hydraulic communication and sufficient volumetric sweep. Low pressure layers identify poorly swept zones that can often result in significant increases in production and recovery when flooded. Pressure barriers with layers can also be identified using FTs.

In the following example, a series of formation tests were obtained for three adjacent wells in a sand that was indicated from open hole logs to be oil-bearing in Well 1 and water bearing in Wells 2 and 3. Geological mapping suggests that they are in a continuous reservoir and that there should be an oil-water in contact Wells 2 and 3 and below the bottom of Well 1. Unfortunately, no fluid samples were obtained with any of the formation tests.

Use the following data to answer (if possible) the following questions.

  • Which wells are in hydraulic communication?
  • What contacts are present in any of the wells?
  • What is the in-situ density of the oil? Of the water?
RCI Pressure
Subsea Depth, ft Well 1 Well 2 Well 3
-6123.0 3431.6
-6125.0 3431.7
-6130.9 3433.9
-6137.0 3436.2
-6142.0 3438.1
-6158.0 3444.1
-6182.0 3453.7
-6190.0 3457.3
-6201.0 3462.3
-6256.0 3505.1
-6275.0 3513.7
-6278.0 3515.1


Graphing the pressure points as a function of depth yields a figure as follows:

The gradient between the Well 1 top data point at -6123 is much less than that at -6125 and below. That suggests that this top point is potentially in a gas cap. With only one data point, graphical methods could not be used to determine the gas gradient. However, if gas density is estimated (perhaps from nearby wells) and with the known values for pressure and temperature, an approximate gas-oil contact of -6124 ft can be obtained. The pressure gradient of the water bearing sands is approximately 0.4546 psi/ft. The density of fresh water is 0.433 psi/ft, so the in-situ density of the salt water is 0.4546/0.433 or 1.05. The oil gradient suggests an in-situ density of 0.87 g/cm3. Specific and API gravity are normally quoted at standard conditions so both of these densities would need to be corrected for temperature, pressure and gas saturation to indicate surface values.

Examining the gradients, it appears that Well 3 is hydraulically isolated from Wells 1 and 2. This has a relatively high degree of certainty. Wells 1 and 2 appear to be hydraulically connected and simultaneously solving the gradients measured in the oil and water zones suggests an oil-water contact of  -6179 ft. The closer the gradients are in the oil and water zones, the more difficult it is to be accurate about the value for the OWC. Additionally, the further apart the data points are (vertically or areally) the greater the chance that other errors are possible in the OWC determination. While caution is warranted, the use of formation test data to interpolate a best technical estimate for OWC is normal practice.  Since the capillary pressure effect may be important in pressure measurements, the calculated OWC may be shifted up or down based on the wettability conditions. Therefore, it can be misleading to rely on the pressure data alone to determine the OWC. It needs to be integrated with other log data to make a reliable determination of OWC.

Want to know more?

Check out Baker Hughes technology including the Reservoir Characterization eXplorer and connect with our technical representatives.

The Baker Hughes Reservoir Characterization eXplorer™ (RCX™) service acquires pressure data and samples in extreme conditions or hostile environments. Despite the challenges, our RCX service delivers high-quality data and service performance in HP/HT operations. Reliable formation testing and sampling occurs up to 25,000 psi [172.4 MPa] and 375 deg. F [191 deg. C] for 20 hours.

The RCX service offers critical information so you can assess the commercial value of your assets and make important decisions about facility design and recovery strategy. Our RCX service ensures optimum recovery and maximum return on investment.

RCX capabilities include

  • high-quality Precision Pressures™
  • representative single-phase samples
  • cost savings from improved reliability in challenging wells
  • time savings from wireline-conveyed deployment in high-angled wells up to 75-deg. deviation when combined with RCI flywheels.

[1] As opposed to hydrodynamic. Hydrodynamic reservoirs exist, particularly when associated with aquifers that outcrop to the surface or an ocean or lake bottom and large height differences exist. Other hydrodynamic conditions may exist when two reservoirs share a common aquifer and large relative fluid withdrawal occurs in one of the reservoirs. Tilted water-oil tables and unusual trapping characteristics may occur in such situations.

0 responses | Add Yours


You must be logged in to post a comment.

We’re here to help
D Nathan Meehan