| Title | Building the Fracture Network Model for Okuaizu Geothermal Field Based on Microseismic Data Analysis |
|---|---|
| Authors | Dian DARISMA, Yusuke MUKUHIRA, Naoki AOYAGI, Kyosuke OKAMOTO, Takuya ISHIBASHI, Hiroshi ASANUMA, Takatoshi ITO |
| Year | 2023 |
| Conference | Stanford Geothermal Workshop |
| Keywords | microseismicity, hypocenter location, focal mechanism, fracture, fracture network model |
| Abstract | In a geothermal field, microseismicity is used to monitor subsurface conditions, especially in a geothermal reservoir. Hypocenter location can provide us with information about existing fractures as well as permeable zones. Therefore, the fracture network model based on real data sets benefits reservoir modeling. This study uses microseismic clustering analysis to build the fracture network model for Okuaizu Geothermal Field in Japan. The microseismic monitoring has been operated using nine seismic stations, consisting of five surface stations and four borehole stations. However, the previous study focused on the triggering process during injection, and the fracture system of this field has not been well understood. In this study, we analyzed microseismic data to get a more precise hypocenter location using the extended period of data during the reinjection process from 2019 to 2021. First, we located the absolute locations for 8,199 events with manually picked P- and S- wave arrival information with a homogeneous seismic velocity model. The more precise 4,235 locations were relocated using Growcust. Most of the seismic activity triggered during reinjection was concentrated at the bottom of the reinjection well. We limited the relocation results using cross correlation (CC) above 0.6 and filtered using a frequency of 5-20 Hz. Using a hierarchical clustering algorithm, Growclust, we also successfully identified some clusters in the northeast and southern parts of the injection well, which have the same trend as the NNE-SSW fault. The cluster results also correlate with microseismicity distribution by time. |