This paper presents a method to calibrate the dipole sonic log data to the stress measurements derived from injection/ falloff tests. Generally stress data that is calculated from logs used for hydraulic fracture models is based on the poroelasticity model. It is often seen that independent stress tests and calculated log values differ significantly. In this paper, the strain-corrected method proposed by Blanton et al.1  is utilized to yield results that compare well with actual field data. The paper also compares methods used to calibrate stress data in the sand layers. Calibrated stress data is then used to develop a correlation between closure stress data and volume of shale for each well.

This work reviews and presents in-situ stress data gathered in various intervals in two wells in the Agency Draw field of the Uinta Basin in Uintah County, Utah. The primary targets for completion are multiple gas-productive sandstone intervals in the Blackhawk, Castlegate, Sego and Mesaverde formations. These low permeability (0.005-0.05 md) intervals cover a gross vertical extent of 1500-2000 feet. The wells in the field are completed with multiple fracture treatments, and then all perforated intervals are commingled in an effort to drain as many productive sandstone lenses as economically as possible.

Based on this correlation, stress data is input into pseudo-3D hydraulic fracture models to yield meaningful treatment designs. Post-treatment radioactive tracer survey data is then compared to the height growth predicted by the fracture model.

The paper details the importance of proper input data required for frac models and some of the pitfalls associated with incorrect assumptions.

You can access this article if you purchase or spend a download.