Detection of a Formation Fracture in a Waterflooding Experiment

Abstract

Summary A procedure to detect and to evaluate fracturing during waterflooding is described. The approach requires (1) use of a radial-flow analysis to detect changes in fluid transmissibility, (2) determination of the in-situ stress changes, caused by pore pressure buildup and temperature decrease, and comparison of the modified stresses with the bottomhole pressure (BHP), and (3) modeling of the fracture by means of a pressure (BHP), and (3) modeling of the fracture by means of a three-dimensional(3D) hydraulic fracture simulator. This procedure is applied to 30-day waterflooding injection into a limestone oil reservoir located in an offshore well within the Idd el Shargi reservoir (Qatar) in which fracture occurrence was suspected. Both the radial-flow analysis and the quantification of stress changes indicated the occurrence of fracture. Finally, the resulting fracture geometry was delimited by simulation of the fracturing process. Introduction Injection of water or other fluids into a reservoir for relatively long periods of time can alter the in-situ state of stress by decreasing the formation temperature and increasing pore pressure. Depending on the injection history and on the injected fluid and reservoir temperature, these changes can have an effect on the in-situ state of stresses that is sufficient to give rise to fracture onset. Perkins and Gonzales and Marx and Langenheim used an energy balance method to determine the reservoir temperature profile and the elasticity theory to quantify the resulting in-situ stress changes. Similarly, the elasticity theory was used to quantify stress changes caused by the pore pressure buildup. With the assumption of a linear elastic model, the stress changes calculated can be superposed on the undisturbed state of stresses to determine the modified closure stresses. Comparison of these stresses with the bottomhole fluid pressures may be used to estimate fracture onset. Alternatively, a radial-flow analysis can evaluate changes in formation transmissibility, kh/mu, relative to the value at the initial stages of injection. Changes in kh/mu would indicate changes in reservoir conditions caused by formation fracturing or formation damage. This work describes the detection of a formation fracture in a 30-day waterflooding experiment. This approach (1) uses a radial-flow analysis to determine changes in fluid transmissibility, (2) determines the changes in insitu stresses caused by changes in pore pressure and temperature and compares the modified stresses with the bottomhole fluid pressures, and (3) models the fracture geometry resulting from the injection of fluid after fracture onset. Description of the Waterflooding Experiment. The waterflooding test consisted of a 30-day filtered seawater injection into two perforated formations defined as Zones A and B of an oil-bearing limestone reservoir located offshore near the west coast of the Persian Gulf (Fig. 1). Zone A is 103 ft [31.4 m] thick while Zone B is 63 ft [19.2 m] thick. The average porosities for both zones were approximately equal ( =27%), while the average permeability for Zone A (k=2.21 md) was higher than for permeability for Zone A (k=2.21 md) was higher than for Zone B (k=0.87 md). Before injection, in-situ stresses were measured at several depths along the wellbore, as shown in Figs. 1 and 2. The injection test was conducted at three levels of constant tubing-head pressures (THP). The initial THP of 990 psig [6826 kpa] was increased to an intermediate step of psig [6826 kpa] was increased to an intermediate step of 1,240 psig [8550 kPa], and finally to a pressure of 1,490 psig [10 273 kpa]. Plots of the recorded injection rates psig [10 273 kpa]. Plots of the recorded injection rates and THP vs. time are displayed in Fig. Recordings of BHP's indicated negligible friction drop. Except for some interruptions (i.e., acid wash and pump malfunctions), pumping was continuous. The injected seawater pumping was continuous. The injected seawater temperature (110 degrees F [43 degrees C]) was approximately 51 degrees F [11 degrees C] lower than reservoir temperature (161 degrees F [72 degrees C]). A comparison of injectivity indices calculated from spinner surveys before test completion with expected injectivity indices obtained from kh values indicated 70% greater-than-expected injectivity for Zone B and 15% smaller-than-expected injectivity for Zone A. These differences led us to suspect the occurrence of a fracture growing predominantly into Zone B and extending into Zone A where the smallest closure stresses were measured. Flow Rate Analysis. The system of concern is a single well with an inner moving seawater bank displacing an outer oil bank. In this system, a radial-flowrate analysis would provide an indication of changes in the kh/mu with changes in the reservoir conditions. JPT P. 1113