Dipole Sonic Imaging Case Study and Analysis to Estimate Structural Dips for Deviated Well

One of the key advantages of borehole sonic data, beyond measuring formation velocity, is the ability to look away from the borehole and provide high resolution images to identify structural features that may not be observed on borehole imaging logs or seismic images. Depending on the velocity of the formation and listening time of the record, radial distance of image can range from 60-250 plus feet. In addition, analysis from dipole sonic data can be used to quantify dip interpretations across a scale much larger than the borehole size. Objectives of this paper are to show analysis techniques using dipole sonic image data that provide an estimate of dips over the well interval. The goal is also to show the value of dips extracted from the data without having to image it, making the analysis faster computationally. An additional challenge of analyzing dipole sonic image data is interpreting and integrating results with supplementary data. The first part of the analysis was to look at reflection moveout of the isolated down and upgoing reflections. Signal processing techniques were applied to the data to remove noise (direct wave) and to correct for amplitude losses. By measuring moveout of the reflected events, apparent velocities were estimated from the data. This analysis helps to identify possible zones with reflections coming back from the formation, as their apparent velocity is faster than energy travelling along the borehole. Using this estimated dip information from apparent velocity, 2D velocity models in depth were built using the sonic slowness values converted to velocity. Depth migrations were performed on the data and then a statistical analysis was done to calculate dips from the imaged features. The analysis also provides the semblance of the dips as a measure of confidence. The estimated apparent velocities show strong peaks between dipping formation scenarios for 50 and 60 degrees (dip relative to well). This result was used to build a 2D velocity model, and to define the size of the model to optimize processing efficiency. Imaging quality is high and shows reflections with radial coverage up to ~60 ft away from the wellbore. The migrated image is a stack of all 13 receivers recording a dipole oriented near the horizontal plane. The dip scan plots red quivers on the data to improve identification of linear features. Estimated dip data can then be plotted with depth to estimate an average over the entire well interval. Semblance is also calculated to weigh the confidence of each pick. This case study shows that estimates of event dip can be calculated from dipole sonic data without having to image it. The dips can be measured from the apparent velocity of reflected events in the recordings. This saves time as more sophisticated imaging algorithms may have long runtimes and require extensive computing resources. Depth migrations using a 2D velocity model were output and provided geological dip information away from the borehole. True dips calculated from depth migrated image data showed the same trend as seen from first dip estimates using the velocity scan method. These dips were calculated over the well interval and can be output in a csv format to easily load into petrophysical and geological databases for future analysis.