This paper presents an overall summary of the in-situ stress data obtained through the Gas Research Institute (GRI) Tight Gas Sands Research Program. Data of this type can be invaluable to anyone designing a hydraulic fracture treatment in a complex, multi-layered reservoir. Correlations are presented that allow the in-situ stress gradient to be calculated from openhole log data as a function of reservoir pressure.
Results from thirty-four (34) stress tests in five wells in the Travis Peak formation of east Texas are presented. Twenty of these tests were conducted as part of the 1986 GRI Staged Field Experiment in the Waskom (Travis Peak) Field in Harrison County, Texas. The primary purpose of these tests was to determine the in-situ stresses of the various lithologies in this multi-layered, complex reservoir. Knowledge of the vertical stress profile can help to make estimates of vertical fracture growth as a function of excess pressure in the fracture.
Formation pressures were obtained in many zones with either wireline formation testers or pressure buildup tests. These estimates of reservoir pressure were then correlated with the actual measured values of in-situ stress to determine the effects of pressure depletion on the in-situ stress. Calculated values of in-situ stress from logs were calibrated using the measured values of reservoir pressure and in-situ stress.
An important area of research today concerns the study of factors which control the vertical growth of a hydraulic fracture. When large fracture treatments are pumped, vertical height growth is likely to occur as the excess pressure in the fracture increases. In many cases, uncontrolled height growth is the main cause of poor well performance, because the desired propped fracture length was not achieved. Thus, it is imperative that we be able to predict accurately the magnitude of the vertical growth of a fracture. This will allow us to more accurately predict the fracture length and width.
It bas been well established that the in-situ stress contrasts between the individual rock layers control fracture propagation. Techniques exist which can be used to estimate the elastic rock properties and the in-situ stresses from logs and cores. These methods may not always fully represent in-situ conditions. Injection tests are by far the most reliable sources of actual in-situ stress data. However, these can be expensive, time consuming, and difficult to interpret.
There have been several studies published which detail the methods and results of predicting in-situ stress. These studies usually tend to concentrate only on the producing reservoir and not the rock layers above and below the formation which is to be fracture treated. They typically have not dealt with multi-layered complex reservoirs in which the higher stress intervals that might act as barriers to vertical fracture growth are relatively thin. By measuring the in-situ stresses of sandstones, siltstones, and shales, a more comprehensive stress profile can be obtained which will allow for a more accurate prediction of the vertical fracture growth.
The major objective of this study was to analyze multiple in-situ stress tests from several wells in the Travis Peak formation of east Texas in order to determine the in-situ stresses of the various levers of rock. Various lithologies within the Travis Peak were selected for study.