Evaluating Methods to Differentiate Natural and Induced Seismicity

Abstract

Flowback and produced water generated from hydraulic fracturing and oil and gas production have led to increased wastewater disposal through underground injection wells. Several recent studies have suggested recently felt seismic events may be associated with underground injection wells in Arkansas, Ohio, Texas and Oklahoma, among others. However, in many cases, such as in Oklahoma, there is debate about whether the earthquakes were the result of fluid injection (Keranan et al. 2013), natural tectonic processes (Oklahoma Geological Survey, 2013), or were related to remote events (van der Elst et al. 2013). Moreover, it is unclear why earthquakes have occurred near some injection wells but not others, with apparently similar Geology, target reservoirs, and injection rates (e.g. Frohlich, 2012). In instances where injection occurred near a fault (e.g. Rangely, CO), the timing and distribution of seismic events was well correlated to fluid volumes, and the interaction between injection and induced seismicity was easily resolved. In other cases (e.g. Oklahoma, Texas), it appears more difficult to interpret whether a particular injection well was related to observed seismic events. Therefore, it is not always a straight forward process to differentiate between natural and induced seismicity. In this presentation we explore different approaches to distinguish between natural and induced seismic events, including recently developed quantitative approaches, such as an analysis of frequency magnitude distributions. We draw examples from a number of datasets compiled from different types of injection activities that span a range of operational characteristics and geological settings.

AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014