Turkish geothermal fields as natural analogues of CO2 storage sites: Gas geochemistry and implications for CO2 trapping mechanisms

摘要

The CO2-He gas systematics of Turkish geothermal systems are modeled to compare and quantify various CO2 trapping mechanisms involved in deep aquifers. We utilize data from geothermal fields from three different tectonic provinces of Turkey: North Anatolian Fault Zone (NAFZ), East Anatolian Fault Zone (EAFZ), and western Anatolia. Our modeling approach distinguishes between possible subsurface processes that control CO2 gas contents in the geothermal systems mixing, degassing, dissolution, and calcite precipitation. The approach is predicated on selecting pristine gas compositions, which is achieved by eliminating those samples likely to reflect crust-mantle mixing (via CO2/He-3 vs. R/RA relationships), and shallow-level degassing (via CO2 vs. delta C-13 relationships). Remaining samples are then used to discriminate between dissolution and calcite precipitation processes using delta C-13 vs. CO2/He-3 relationships. Based upon temperature- and pH-dependent fractionation between CO2 and He gases, and carbon isotopes (delta C-13), quantitative estimates from Rayleigh fractionation models suggest dissolution as the major CO2 trapping mechanism, accounting for up to 95% of the variations in gas compositions. Calcite precipitation accompanies dissolution at high temperatures Characteristic of reservoir conditions, and can stabilize up to 80% of the emplaced CO2 in some instances. The relative contribution of calcite precipitation to CO2 stabilization, however, is generally minor (<25%) compared to dissolution (>75%) and appears to change as a complex function of the nature and duration of water-rock interaction along with temperature. As shown by this study, CO2-He systematics provide a promising tool to predict and monitor the behaviour of CO2 in storage systems like geothermal fields which have recently gained interest as prospective storage sites for enhanced geothermal heat recovery projects. (C) 2016 Elsevier Ltd. All rights reserved.