Abstract
Energy systems models are used to analyze long-term policies and pathways for reducing greenhouse gas (GHG) emissions. A subset of these models is bottom-up, engineering-economic models that minimize the costs of achieving exogenously specified annual reductions in GHG emissions to limit increases in global average temperature. Referred to as deep decarbonization models (DDM), their outputs include the levels and timing of energy investments that satisfy future energy demand projections across all energy sectors over multiple decades. This paper conducts an epistemological review of 42 articles on DDMs published in the last ten years. In its twopart review of the DDM literature, this paper first identifies the mathematical formulations, underlying assumptions, and constraints of the modeling framework that impact their outcomes. Next, it tabulates and discusses the possible limitations in assessing such models in terms of their epistemic values, i.e., creating new knowledge beyond common sense in capturing the possible realities; and non-epistemic values, i.e., policy relevance and usefulness of the modeling framework. This paper suggests the need for more studies and continued discussion on the potential strengths, usefulness, and limitations of the DDM framework. DDMs might be useful in identifying multiple pathways towards deep decarbonization of future energy systems, but there will always be a need to proceed with caution, strive for better models, and be more circumspect with respect to the associated policy conclusions.