The carbon abatement potential of high penetration intermittent renewables† Elaine K. Hart * and Mark Z. Jacobson Received 18th December 2011, Accepted 23rd February 2012 DOI: 10.1039/c2ee03490e The carbon abatement potentials of wind turbines, photovoltaics, and concentrating solar power plants were investigated using dispatch simulations over California with 2005–06 meteorological and load data. A parameterization of the simulation results is presented that provides approximations of both low- penetration carbon abatement rates and maximum carbon abatement potentials based on the temporal characteristics of the resource and the load. The results suggest that shallow carbon emissions reductions (up to 20% of the base case) can be achieved most efficiently with geothermal power and demand reductions via energy efficiency or conservation. Deep emissions reductions (up to 89% for this closed system), however, may require the build-out of very large fleets of intermittent renewables and improved power system flexibility, communications, and controls. At very high penetrations, combining wind and solar power improved renewable portfolio performance over individual build-out scenarios by reducing curtailment, suggesting that further reductions may be met by importing uncorrelated out-of-state renewable power. The results also suggest that 90–100% carbon emission reductions will rely on the development of demand response and energy storage facilities with power capacities of at least 65% of peak demand and energy capacities large enough to accommodate seasonal energy storage. 1 Introduction In response to a growing concern over global warming, the last decade has seen a surge in proposals for reducing the carbon dioxide emissions associated with electric power generation, many of which include large build-outs of renewable technolo- gies including wind, photovoltaics (PVs), concentrating solar power (CSP), geothermal, wave, and tidal power. This paper seeks to determine how the temporal characteristics of electric power demand, the variability of renewable resources, and the controls employed by renewable technologies influence the potential for a renewable portfolio to displace carbon-based generation and to reduce carbon dioxide emissions at very high penetrations. Furthermore, we seek to understand which of these factors has the strongest influence on the carbon abatement potential of a given technology, and in the case that a limit to the carbon abatement potential of intermittent renewables exists, what technologies are needed to achieve complete decarbon- ization of the electricity grid. In the past, economic analyses of the carbon abatement potential of renewables have tended to assume that renewable Department of Civil and Environmental Engineering, Stanford University, Stanford, CA, 94305. E-mail: ehart@stanford.edu; Fax: +1 650 7237058; Tel: +1 650 7212650 † Electronic supplementary information (ESI) available. See DOI: 10.1039/c2ee03490e Broader context The reliable integration of renewable resources on to the electricity grid represents an important step toward decarbonizing the electric power sector and mitigating global climate change. This step is complicated by both the variability and the uncertainty associated with power output from renewable resources, like wind and solar power. Analyses that seek to quantify system reliability, reserve requirements, and the carbon dioxide emissions associated with operating these reserves have historically relied on simu- lations with high temporal resolution (typically an hour or less) and with stochastic treatments, both of which increase the computational complexity significantly. However, energy-economic models capable of analyzing the costs and economic impacts of different decarbonization strategies or policies typically use time scales of one year and cannot accurately resolve the phenomena associated with intermittent renewables. In this paper, we develop a parameterization of the results from higher temporal resolution simulations that can be implemented in large-scale energy-economic models. This effort contributes to the improved economic treatment of renewable power sources in analyses used by policymakers and may provide additional insight regarding technological cost targets for innovators. 6592 | Energy Environ. Sci., 2012, 5, 6592–6601 This journal is ª The Royal Society of Chemistry 2012 Dynamic Article Links C < Energy & Environmental Science Cite this: Energy Environ. Sci., 2012, 5, 6592 www.rsc.org/ees ANALYSIS