This paper addresses the sustainable design of organic Rankine cycle based geothermal binary power systems under economic and environmental criteria. A novel superstructure with multiple heat source temperatures, working fluids, and heat rejection systems is proposed. Based on the superstructure, a life cycle optimization model is formulated as a mixed-integer nonlinear fractional program (MINFP) to determine the optimal design. The nonconvex MINFP is efficiently solved by a tailored global optimization algorithm. Two case studies are considered to demonstrate the proposed modeling framework and solution algorithm. One case is based on a geothermal energy system located in California, and the other one is in New York (NY) State. The results show that the geothermal energy system in California is much more economically competitive than that in NY State. The difference in life cycle environmental impacts is less pronounced because the environmental impacts are less sensitive to geological conditions than the capital investments.