Identification and Characterization of Sonic Dual Arrival Events in High-Angle and Horizontal Wells for Enhanced Formation Evaluation and Well Placement

It is often assumed that reservoirs are homogenous in nature when conducting petrophysical evaluation of vertical pilot wells. However, this may not be the case with high-angle or horizontal (HAHZ) wells. Dual arrival events are well known to occur in sonic logs acquired in HAHZ wells where the compressional P-waves and shear S-waves propagate within the layer containing the tool sonde and refract along nearby shoulder beds. Here we present methods to process these dual arrival events to distinguish the tool layer slowness and provide a curtain section of formation slowness along the well track. Physics modeling shows how the shoulder bed and tool layer arrivals present themselves to an acoustic receiver array as functions of measured depth, source-receiver offset, and nominal receiver azimuth. Built around a series of tau-P transforms in the common offset, common shot, and common ring gather domains, an automated time pick is presented. A labeling procedure then distinguishes between the tool layer and shoulder bed arrivals based on their apparent slowness and creates a log of tool layer P and S slowness. Finally, a ray tracing inversion is used to invert the arrival times associated with the shoulder bed arrivals for a locally two-layer model of formation P or S slowness. Results of the physics modeling show that shoulder bed events have arrival times that vary with distance to the shoulder bed, have an apparent slowness that is faster than that of the tool layer arrival, and exhibit a sinusoidal moveout as a function of nominal receiver azimuth that indicates whether the shoulder bed is above or below the well track. The automated time pick is a generalization of the slowness time coherence workflow of Kimball and Marzetta. The ray tracing inversion determines the distance to and relative dip of the shoulder bed and distinguishes whether the shoulder bed arrival event is a reflection or refraction. The ray tracing inversion results are then woven together into a curtain plot section of formation slowness along the well track. This workflow of dual arrivals is demonstrated with both modeled waveforms and field data examples. The above workflow produces logs of tool layer P and S slowness that are critical for accurate formation evaluation answers such as porosity, compressional and shear wave velocity ratio, and mechanical properties to derive stress estimates particularly when the well trajectory is close to a faster shoulder bed. Moreover, the resulting curtain plot section of P and S slowness is an acoustics analog of curtain plot sections of horizontal resistivity derived from logging-while-drilling electromagnetics measurements and so opens new avenues for increasing the use of acoustics for formation evaluation, completion operations, and well placement in HAHZ wells and geologically complex environments.