Learning curves for environmental technology and their importance for climate policy analysis

We seek to improve the ability of integrated assessment (IA) models to incorporate changes in CO2 capture and sequestration (CCS) technology cost and performance over time. This paper presents results of research that examines past experience in controlling other major power plant emissions that might serve as a reasonable guide to future rates of technological progress in CCS systems. In particular, we focus on US and worldwide experience with sulfur dioxide (SO2) and nitrogen oxide (NOx) control technologies over the past 30 years, and derive empirical learning rates for these technologies. Applying these rates to CCS costs in a large-scale IA model shows that the cost of achieving a climate stabilization target are significantly lower relative to scenarios with no learning for CCS technologies.