INJECTION WELL TESTING

By Darrel Dunn, Ph.D., Hydrogeologist - Geologist

(This is a technical page on injection well testing. Click link to see a nontechnical page on injection well testing.)

Introduction to Injection Well Testing Web Page

The purpose of this web page is to describe some techniques that may be applied to testing injection wells to determine the properties of the well and the formation receiving the injectate. The petroleum industry uses a large proportion of all injection wells if one includes wells for disposal of produced water and injection wells used in enhanced oil and gas recovery operations. Consequently, much injection well technology has been developed by petroleum engineers and applied to thousands of wells. The petroleum engineering literature on injection well testing is extensive. This web page is a limited treatment of the subject. It does not include all tests applied to wastewater injection wells. It does not include tests exclusively related to injection of fluids into oil and gas reservoirs for enhanced recovery, and it does not include injection tests exclusively used for pre-frac testing of oil and gas wells.

Step-Rate Injectivity Test

The step-rate injectivity test (aka step-rate test, injectivity test, fracture step rate test, step rate injection test) is a common test used to estimate the threshold pressure at which fractures are opened in the injected formation and/or the overlying or underlying beds. Some agencies that regulate injection wells require it. This threshold pressure is sometimes called the "fracture pressure." The test involves injecting fluid at a series of increasing rates with each rate preferably lasting the same length of time. The fluid pressure at the formation-face (usually the bottom-hole pressure) is recorded and plotted on a graph of injection pressure at the end of each step versus injection rate. Ideally, the points on the graph corresponding to low injection rates well plot as a straight line, and points corresponding to higher injection rates will plot as another straight line with a smaller slope. Figure 2 is an example of a simple 4-step, sparse-data step-rate injectivity test plot corresponding to the injection test graphed in Figure 1.

Figure 1. Example of a sparse-data step-rate injectivity test. Data from Erlougher (1977).

Figure 2. Step-rate injectivity test graph.

Some step-rate injectivity tests are not as simple as the one depicted in Figures 1 and 2. In a simple case where skin effects and/or wellbore storage are not significant, the points below the fracture pressure conform to the inverse of the Jacob approximation of the Theis equation expressed in terms of gauge pressure rather than drawdown. Consequently, an estimate of transmissiviy may be obtained from the slope of the line fitting these points if the steps are of equal duration. Transmissivity may be converted to intrinsic permeability (darcys) by supplying values for dynamic viscosity and specific weight of the injectate.

Specifications for conducting step-rate injectivity tests vary. A comprehensive specification would maximize the information that could be obtained from the test. Items might include the following:

Allow the pressure in the injection formation to recover to a steady pressure before starting the test.
Install automatic pressure transducer(s) capable of recording rapid readings (0.2 seconds) to within 0.01 feet. The readings should be accessible during the test.
Install a transducer at the injection formation level, and if positive pressures are expected at the top of the wellbore, install a transducer there as well. (Tandem transducers at the injection level can provide a backup in case one fails. If it is impractical to install a transducer at the injection level, accurate data must be available of injectate specific weight and friction losses in the well.)
Install an accurate flow meter and a backup flow meter on the injection pipe.
Use an injectate that is compatible with the formation water or one that represents the intended long-term injectate depending on the purpose of the test..
Use 60-minute injection steps to maximize information obtained. Shorter steps may provide adequate fracture pressures..
Maintain a constant injection rate for each step and record the rate frequently enough to identify deviations of 5 percent.
Program a pressure reading schedule of 0.2 seconds for 2 seconds, 1.0 seconds for 2 to 20 seconds, 5.0 seconds for 20 to 120 seconds, 30 seconds for 2 to 10 minutes, and 2.0 minutes for 10 to 60 minutes for each step of the test.
Estimate a maximum injection rate appropriate for the purpose of the test, which might be to determine the fracture pressure, or to verify that the maximum injection pressure is below the fracture pressure.
Plan 15% rate increases to the maximum injection rate, with at least three rates below the expected fracture pressure.
Monitor the pressure during the step-test to determine when the fracture pressure is exceeded.

Local information may be used for the preliminary estimate of fracture pressure. In the absence of local information, one could estimate the fracture pressure based on a fracture gradient of 1.0 psi/ft for depths less than 2000 feet, and ranging down to 0.4 psi/ft for greater depths. Fractures approximately normal to the least principal stress open first. At shallow depths the least principal stress tends to be vertical, but it tends to be horizontal at greater depths. However, stress fields vary considerably, especially between cratonic areas, sedimentary basins, and tectonically active and inactive areas.

Interpretation of step-rate data my be more complex than suggested by Figure 2. Damage to the casing cement bond can cause an inflection point that is not the fracture pressure. Variable fracture geometry and orientation can conceivably produce a curve with no clear inflection point. A pre-existing open fracture might result in a graph with no inflection point or a slight curve.

Injection and Falloff Tests

After an injection rate that will not induce fracturing is determined, a variety of test procedures are available to estimate the injection formation parameters. Some of the procedures are the same as ones described in my aquifer testing web pages on step testing, unconfined aquifer testing, leaky aquifer testing, and fractured aquifer testing. However, when these procedures are used for injection tests the discharge is negative and is called the injection rate; and the procedures only apply for injection pressures less than the fracture pressure. In injection tests the drawdown is negative and is often expressed in terms of pressure increase. Furthermore, hydraulic conductivity may be expressed in terms of intrinsic permeability (darcys) of the injection formation, and in terms of specific weight and dynamic viscosity of the injectate or various equivalent measures of the injectate properties. Tests involving recovery after a well is shut-in are often called falloff tests.