Incorporating risk in geothermal field development planning: applying decision-theoretic agents

Abstract Geothermal field development requires decisions about data acquisition and the selection of an optimal development scenario (DS) under uncertainty. Effective planning must address two distinct types of risk: systematic risk, which affects the broader market, and unsystematic (idiosyncratic) risk, which is asset-specific. The inherent complexity of these decisions often leads to prolonged appraisal phases and suboptimal outcomes—particularly when it comes to incorporating risk into planning. This study applies decision-theoretic agents to assess the impact of explicitly incorporating both systematic and unsystematic risk into the development of low-enthalpy geothermal projects. We explore common practices, such as embedding unsystematic risk into discount rates—evaluating three rates: 7% (cost of capital), 10% (3% risk premium), and 13% (6% risk premium)—to examine their influence on project economics, data acquisition strategies, and DS selection. Our results show that including unsystematic risk in discount rates significantly reduces Expected Monetary Value (EMV), increases exploration costs, and impairs the ability to choose the most appropriate DS or to walk away when warranted. Crucially, this approach fails to meaningfully reduce expected losses. To overcome these limitations, we propose a more robust framework that integrates physical numerical models with probabilistic economic evaluations. This approach enables more accurate handling of both systematic and unsystematic risks, leading to better-informed decisions in geothermal field development.