Regional Variability and Uncertainty of Electric Vehicle Life Cycle CO2 Emissions across the United States
详细信息 查看全文
- 作者：Mili-Ann M. Tamayao ; Jeremy J. Michalek ; Chris Hendrickson ; In锚s M. L. Azevedo
- 刊名：Environmental Science & Technology
- 出版年：2015
- 出版时间：July 21, 2015
- 年：2015
- 卷：49
- 期：14
- 页码：8844-8855
- 全文大小：720K
- ISSN：1520-5851
文摘
We characterize regionally specific life cycle CO2 emissions per mile traveled for plug-in hybrid electric vehicles (PHEVs) and battery electric vehicles (BEVs) across the United States under alternative assumptions for regional electricity emission factors, regional boundaries, and charging schemes. We find that estimates based on marginal vs average grid emission factors differ by as much as 50% (using National Electricity Reliability Commission (NERC) regional boundaries). Use of state boundaries versus NERC region boundaries results in estimates that differ by as much as 120% for the same location (using average emission factors). We argue that consumption-based marginal emission factors are conceptually appropriate for evaluating the emissions implications of policies that increase electric vehicle sales or use in a region. We also examine generation-based marginal emission factors to assess robustness. Using these two estimates of NERC region marginal emission factors, we find the following: (1) delayed charging (i.e., starting at midnight) leads to higher emissions in most cases due largely to increased coal in the marginal generation mix at night; (2) the Chevrolet Volt has higher expected life cycle emissions than the Toyota Prius hybrid electric vehicle (the most efficient U.S. gasoline vehicle) across the U.S. in nearly all scenarios; (3) the Nissan Leaf BEV has lower life cycle emissions than the Prius in the western U.S. and in Texas, but the Prius has lower emissions in the northern Midwest regardless of assumed charging scheme and marginal emissions estimation method; (4) in other regions the lowest emitting vehicle depends on charge timing and emission factor estimation assumptions.