[S-SS12] Statistical seismology and underlying physical processes

[SSS12-P06] A study of analytical condition for application of the ETAS model to microseismicity in geothermal areas

Naoki Aoyagi¹, *Kyosuke Okamoto¹, Hiroshi Asanuma¹, Yusuke Mukuhira² (1.National Institute of Advanced Industrial Science and Technology, 2.Tohoku University)

Keywords:Microseismicty, Geothermal field, ETAS model

In geothermal areas, water injection into geothermal reservoirs is conducted to recover and stabilize steam production. The spatiotemporal distribution of induced microseismicity induced by this process can be utilized to indirectly capture fluid flow activity and evolution of the geothermal reservoir. However, distinguishing between microseismic events caused by fluid flow associated with water injection and other events is essential. The Epidemic-Type Aftershock Sequence (ETAS) model is employed as a one statistical approach to clarify the tendency of microseismic events triggered by fluid flow among the detected events. The ETAS analysis models an observed microseismic events as an independent events following a Poisson process (µ) and superposition of aftershock sequences governed by a modified Omori law. In geothermal areas, where fluid activity is significant, µ can be considered an approximation of the probability of fluid-induced microseismicity. It is crucial to analyze the spatiotemporal variations in ETAS parameters during water injection. Furthermore, the ETAS analysis requires a sufficient number of seismic events to ensure statistical stability. However, the seismic catalog must be divided into multiple subsets in spatiotemporal ETAS analysis, potentially leading to challenges in stable parameter estimation. Additionally, in geothermal areas, where fluid activity is intense, the ETAS parameters of microseismicity in geothermal areas may significantly differ from those of tectonic earthquakes. Therefore, a specialized investigation focusing on microseismicity in geothermal areas is necessary. In this study, we aim to evaluate the minimum threshold of seismic events required for reliable ETAS analysis using a synthetic catalog that simulates the seismicity of the Okuaizu geothermal field in Fukushima, Japan. First, the ETAS analysis was performed on the synthetic catalog while systematically varying the length of the analysis window to ensure a fixed number of seismic events (16 different conditions in total). Each condition was analyzed 1000 times. For each condition, we confirmed that the distribution of the 1000 estimated ETAS parameters followed a normal distribution and calculated their mean and standard deviation. The results indicated that when the number of seismic events within the analysis window exceeded 100, the estimated values of µ followed a normal distribution. Similarly, under the same conditions, the true µ value used to generate the synthetic catalog was found to lie within the ±1σ range of the estimated values. These findings suggest that for ETAS analysis of microseismicity in geothermal areas, using at least 100 seismic events enables high-precision parameter estimation. Future studies will apply this knowledge to investigate the spatiotemporal variations in ETAS parameters within geothermal areas.