Microseismic monitoring in early haynesville development

Early development of the Haynesville play coincided with a growing appreciation for the value of microseismic monitoring of hydraulic fracture stimulations. Although microseismic monitoring had been in existence as a technology since the early 1970s, at least in concept (Bailey, 1973), it was not until the late 1990s that the technology began to be applied more than very rarely (Duncan and Eisner, 2010; Maxwell et al. 2010). Notwithstanding the important work on Cotton Valley completions reported by Rutledge et al. (2003), it was in the development of the Barnett Shale that the technology proved its worth in overturning the simplistic models of fracture propagation and allowing completion engineers to have a much better understanding of how effective their stimulation programs were (Maxwell et al., 2002). As operators began to complete and evaluate their Haynesville wells, many turned to microseismic monitoring to better understand how the formation responded to different pumping programs and how that response should affect a field-wide development plan.

The Haynesville posed some particular difficulties to monitoring with geophones placed at near reservoir depths in nearby wells. Being in the early stages of development, there were fewer wells-to-depth available to act as monitor wells. More importantly, the temperatures at reservoir depth were found to be close to, if not exceeding, the working temperature range of the downhole geophones. Limiting the time the geophone string spent at reservoir depth was necessary to keep from burning out the temperature-sensitive geophones. This was not a very satisfactory solution because it resulted in either not monitoring some portion of treatment or delaying the frac operation. Even then anecdotes of geophone sonde destruction were common. This problem drove the application of first a surface array approach to monitoring and then, to avoid the expense of line cutting, the development of a sparse, BuriedArray approach to monitoring well stimulations.

In what follows, we recount the design and implementation of the first of more than a dozen of these BuriedArrays (Figure 1) and then go on to discuss the data acquisition and analysis of microseismic data captured during the hydraulic fracturing of more than 80 wells. We also summarize some of the geological and engineering observations made from these data and the implications that these observations had for the development of the Haynesville.