We perform an ex-ante life cycle assessment, integrating cradle-to-gate greenhouse gas emissions into an off-grid energy system model. By applying a multi-objective optimization, minimizing both costs and carbon dioxide equivalent (CO 2 e) emissions, we find the Pareto boundary between these two goals. As a case study, we chose the power supply to an astronomical observatory in Chile, using mostly solar power and energy storage in batteries and hydrogen. We compare the ex-ante study to a prior ex-post life cycle assessment, and furthermore dive into sensitivities of our model regarding component lifetimes and costs. We find (i) low-hanging fruit in lowering emissions with small cost increases, (ii) ex-ante life cycle assessments' possibility to find less CO 2 e intensive power systems compared to ex-post conducted studies, (iii) a pronounced sensitivity of the optimization model on assumed lifetimes of energy storage components. This study shows the importance of including life cycle CO 2 e emissions into the optimization objective of energy system models. This method uncovers the environmental and economic trade-offs associated with high shares of renewable energies in off-grid energy systems. • We integrate life cycle assessment (LCA) into an off-grid energy system optimization. • The applied case study powers a remote astronomical telescope in Chile. • A multi-objective optimization finds Pareto curves at varying costs and lifetimes. • Low-hanging fruit shows low additional system costs to reduce CO 2 e impacts. • In contrast, a separate LCA did not find such trade-offs in costs and CO 2 e emissions. [ABSTRACT FROM AUTHOR]