Preliminary geological, geochemical and numerical study on the first EGS project in Turkey

Geothermal energy is attracting more and more attention due to its large capacity and lack of dependency on the weather. Currently, many countries have planned enhanced geothermal system (EGS) projects. In this paper the first EGS project in Turkey, which is being implemented at the license area of SDS Energy Inc., in Dikili of the İzmir province, is introduced. Extensive geological, paleostress (279 fault-slip data from 33 locations), geophysical (magnetotelluric and vertical electrical sounding at 80 and 129 locations, respectively) and geochemical studies as well as paleostress measurements have been conducted in this area within the scope of this project. In the light of all these studies, it has been determined that the Dikili region is remarkable in terms of its high thermal gradient of about 7 °C/100 m. The geothermal reservoir formation “the Kozak granodiorite” is a homogeneous, crystalline volcanic rock mass with high radiogenic heat production, and suitable for an EGS application. The analysis shows that the dominating fault system is normal, and the corresponding primary stress regime is extensional. Based on the geological, geophysical surveys and the estimated in situ stresses, numerical studies were carried out to assess the results of the hydraulic fracturing and geothermal energy production using the numerical codes FLAC3D^plus and TOUGH2MP, respectively, in the area A of the Dikili site. The simulation results show that the stimulated reservoir volume and area could reach 44.5 million m³ and 1 km², respectively, with an injection volume of 122,931 m³. Assuming the fractured zone has a height of 1000 m and a half-length of 1200 m (the distance between injection and production wells being 1000 m), an average overall geothermal capacity of 83.7 MWth in 20 years could be reached with an injection rate of 250 l/s. The injection strategy and design parameters of the reservoir stimulation and geothermal production will be further optimized with the project running.