Whether in the form of a carbon tax or cap-and-trade regime, climate-change policy is coming and will have a profound effect on electric suppliers and consumers. EPRI studied the effects of high...
Wind Integration and the Cost of Carbon
Renewables are greenest when displacing coal, not gas.
show a striking difference in the impact on gas and coal-fired generation (see Figure 2) .
In the Eastern Interconnection study, nearly all the wind generation displaces coal-fired steam generation. By contrast, in the New England footprint, wind displaces gas-fired combined cycles without affecting coal-fired generation much even up to a 24 percent wind addition. So wind integration in the ISO-NE footprint provides half the carbon mitigation per MWh as compared to the Eastern Interconnect footprint. In the NYISO and WECC studies, wind also displaces gas only, while in SPP, wind displaces an even mix of gas and coal-fired resources. Combustion turbines provided little generation in most of the base case scenarios, while both hydro and nuclear units are unaffected due to their zero marginal costs.
The results for ISO-NE, NYISO and WECC are expected because model assumptions for gas prices exceed comparable coal costs. Adding zero marginal cost renewables displaces units higher in the generation stack: combined cycles over coal units. By contrast, the impact of wind on coal in SPP and in the Eastern Interconnection as a whole is surprising.
NREL explains that more spinning reserve is needed in the respective study footprints due to wind intermittency. Only gas-fired units can provide these reserves, with their quick-start, flexible characteristics. These reserves must be in the same balancing area as the wind additions to comply with contingency conditions. So in the Midwest and Great Plains states, where there is little gas to displace, significant gas-fired resources have to be added, forcing coal units to turn down.
Figure 3 shows where active gas-fired generation might be sufficient for reserve requirements. To avoid considering inactive units, generation by MWh rather than MW of capacity is depicted. Gas predominates along the coasts (East, West and Gulf), except for Washington, where hydro provides ample reserves. So along the coasts wind displaces gas, with commensurately less carbon reduction than in coal-dominated areas.
The wind integration studies quantify this pattern (see Figure 4) . Notably, the WECC study includes a low gas price scenario ($3.50/MMBtu) along with the base case ($9.50/MMBtu). In the low gas price scenario, coal is the marginal resource instead of gas, so adding wind has approximately twice as much impact on carbon emissions as compared to the base case.
Because wind and other renewables incur above-market costs, building generation where it displaces the most carbon is the most cost effective option. A high quality, unsubsidized wind resource near load can compete with gas-fired generation, when gas is selling for pre-shale discovery prices. Offshore wind, currently double the cost of on-shore wind, needs both a very high capacity factor and a major cost reduction to compete. This is illustrated by quantifying the levelized cost difference between wind and gas-fired combined cycle generation in a financial model showing the dependence on the most sensitive parameters: gas prices and wind net capacity factor (NCF). The model assumes the gas unit is non-urban and the wind turbine capital and construction costs are moderate with siting in accessible terrain (see Figure 5) .
Putting all the economics together,