Formation properties such as porosity and fluid saturations control the volume of hydrocarbons that may be stored in a reservoir. Both formation properties such as permeability and fluid saturations and reservoir fluid properties such as viscosity affect producibility, recovery and economics. This information is also required to adequately design and plan production facilities for a reservoir.

Nuclear magnetic resonance (NMR) log data can be used to identify porous and permeable

Figure 1. MREX continuous viscosity profile. (Graphic courtesy of Baker Atlas)
zones in addition to identifying reservoir fluids and determining hydrocarbon viscosity. Wireline formation tester (WFT) operations are time-consuming, but optimizing the WFT program can greatly reduce the total time required to acquire measurements. WFT program optimization can be achieved through careful pre-job planning and by surveying only those intervals of the reservoir that are porous and permeable and have sufficient hydrocarbon saturation. Through proper planning, data acquisition and integration, WFT and NMR data prove complementary and aid in the assessment of reservoir producibility.

WFT optimization
While WFT is much faster compared to more conventional pressure-transient testing methods, WFT surveys, particularly fluid sampling, are still time-consuming. Optimizing the WFT data acquisition program to test and sample only desirable reservoir intervals can save hours of costly rig time.

Pre-job modeling and planning is essential to determine the most efficient and cost-effective WFT program. Typically, the pre-job modeling and planning phase of a project consists of reservoir fluid properties prediction, pressure test modeling, sample clean-up, nitrogen pre-charge requirement and instrument conveyance modeling. These steps ensure the optimum configuration of the Reservoir Characterization Instrument (RCI) hardware to meet the operator’s objectives. Pre-job modeling also provides an indication of the time required to satisfy the operator’s objectives so operations are not rushed and so pressure and sample integrity is not compromised.

Sampling
To obtain representative reservoir fluid samples, the RCI uses the SampleView fluid characterization module to monitor the fluids being withdrawn from the formation. The current generation of this tool monitors clean-up and determines formation fluid types. The newest-generation instrument, currently under field-test, incorporates real-time measurements of density, viscosity, API gravity and gas-oil ratio (GOR) in addition to the current optical measurements provided. While these new measurements will be useful for monitoring the clean-up of fluid from mud filtrate to native reservoir fluid, they have also proven to be critical information for assessing reservoir compartmentalization.

Once the various sensors indicate that the fluid being withdrawn from the formation is clean reservoir fluid, the next critical step in sampling operations is to maintain the native reservoir fluid at in-situ conditions during the entire operation. To prevent altering the reservoir fluid during pumping, the pump rate must be controlled to manage the pressure drawdown. The optimum pump rate is dependent upon formation permeability as well as reservoir fluid viscosity and saturation pressure. These parameters are assessed during the pre-job modeling phase of the project.

Optimize WFT acquisition
Formation permeability and reservoir fluid properties are desired outputs of the WFT acquisition program. However, they are also required inputs to optimize the WFT program and to ensure the quality of the collected fluid samples. Because of this, it is useful to consider how these parameters may be estimated in advance of the WFT program.
In appraisal or development fields, operators may already have adequate knowledge of formation permeability and reservoir fluid properties. However, in these cases the objectives of the WFT program are often markedly different than the objectives in an exploration program where operators usually do not have a firm handle on permeability and fluid properties.

It has been demonstrated that NMR logging is useful for identifying porous and permeable reservoir intervals and thus is useful for selecting intervals to be tested with WFT instruments. While some may argue that permeability from an NMR log is model-based and may be more qualitative than quantitative, NMR permeability generally provides at least an indication of permeability that is of the correct order of magnitude. In addition, through further tuning of the permeability model on a field level, NMR permeability can prove to be quite accurate. Estimates of permeability from NMR services are currently available from both wireline and logging-while-drilling (LWD) NMR logging tools.

Where reservoir fluid properties are not well known, the PVTMOD service provides a prediction of expected fluid properties based on well location, reservoir depth, reservoir temperature and reservoir fluid density. This service provides in-situ estimates of critical pressure, volume and temperature (PVT) parameters such as saturation pressure and viscosity. These properties are used during both the pre-job planning phases and real time during sampling operations to maintain the collected sample as a representative fluid sample.

Correlations have been published showing that NMR measurements of relaxation time (T2) and diffusivity are both related to viscosity. Thus, the NMR data can provide an estimate of fluid viscosity, which may be used in the modeling to optimize the WFT sampling rate. Case studies have also been presented where a continuous viscosity profile provided by NMR logging, as shown in Figure 1, was used to identify reservoir intervals with different fluid properties so that samples of the different fluids could be collected.

One of the most critical properties needed for reservoir fluid properties prediction with the service is reservoir fluid density. The density is required to estimate the saturation pressure of the reservoir fluid in order to maintain the reservoir fluid as a single phase during the entire sampling operation. Typically, fluid density may be obtained from pressure gradients acquired during the precision pressure-testing portion of the RCI operation. Pressure gradients may also be obtained from LWD pressure measurements.

However, many times the pressure gradient plots are difficult to interpret. It may be difficult
Figure 2. GOR estimate from MREX 2D NMR. (Image provided courtesy of Jiansheng Chen)
to distinguish, from pressure measurements alone, if there is one large reservoir with a single gradient or many smaller reservoir compartments requiring fitting a gradient to each compartment. Again, integration with NMR data provides an alternate source of fluid properties needed to optimize the RCI sampling program. In a study of the relationship between NMR T2 and viscosity, it was reported that NMR relaxation time correlates better with specific gravity than with viscosity. Since the NMR log is acquired as a continuous measurement over the reservoir, the specific gravity can be converted into a continuous profile of reservoir fluid density.

Optimize NMR interpretation
Recent advances in NMR acquisition and 2-D NMR interpretation methods have resulted in the capability to distinguish between oil, condensate and gas. In addition, it has long been recognized that the NMR T2 is related to fluid properties such as viscosity, and more recently, specific gravity or fluid density. Most recently, estimating GOR from the 2-D NMR maps obtained with the MR Explorer (MREX) service, as shown in Figure 2, is being tested.

However, many times these interpretations are performed for exploration wells where reservoir fluid properties are not well known in advance. Because NMR-based fluid properties depend on correlations, there is no independent verification of the estimates and no way of determining the accuracy of the estimates until many months later when RCI fluid samples are analyzed or well tests are performed.

With the newest-generation fluid characterization instrument, fluid properties such as density, viscosity, API gravity and GOR will be delivered in real time at the wellsite. While these measurements will be highly precise and accurate, they are delivered only at discrete points in the well rather than continuously along the well bore. Still, these measured, in-situ fluid properties will be useful for tuning the NMR-based correlations, which will then provide accurate fluid properties as a continuous profile throughout the reservoir.

WFT tools also provide high-resolution and accurate mobility measurements. Once the viscosity of the fluid is determined, either as a real-time measurement or through analysis of fluid samples, this is then converted to high-resolution and accurate permeability. Again, however, these measurements are obtained only at discrete points in the well rather than continuously along the well bore. As with the fluid properties, the permeability can be used to calibrate the permeability model for the field, which is also provided as a continuous measurement over the entire reservoir interval.

Some caution must be exercised when calibrating the NMR permeability model in this fashion. An often-asked question is, “What permeability does the NMR permeability model represent?” When the model is calibrated to another source of permeability, the model represents the type of permeability it was calibrated to. When calibrated to WFT permeability, one must take into account how the pressure testing was performed and what fluids were present near the well bore at the time of pressure testing in order to properly understand the calibrated NMR permeability model. If there is invasion and pressure testing is performed without pumping a substantial volume of fluid from the formation, the RCI-determined permeability will represent effective permeability to the mud filtrate at the invaded zone saturations. However,
if pressure testing is performed after pumping a substantial volume of fluid from the formation, such as a pressure test obtained immediately after collecting a sample, the RCI permeability will represent effective permeability to the native reservoir fluid at native reservoir saturation. The latter permeability should prove to be more useful to the operator in determining reservoir producibility. Typically, though, most pressure testing is performed before any pumping occurs, and many times post-sampling pressure tests are not performed.
Through proper planning, data acquisition and interpretation, WFT and NMR data prove complementary and provide robust assessment of reservoir fluid type, hydrocarbon properties and permeability. This information is valuable for assessing the producibility of the reservoir as well as for planning production facilities.

Comments