Wind and Nuclear Power: As Different as Chalk and Cheese

April 2011

• The March 11 earthquake in Japan and the subsequent nuclear disaster have catalyzed opposition to nuclear power generation in many jurisdictions, including Ontario.
• Some commentators have suggested nuclear power should be replaced by a more benign source, such as wind. But the two forms of power have very different characteristics, and are not practically interchangeable.
• Given the operating characteristics of wind generation, 34,000 MW of wind capacity and 10,000 MW of natural gas-fired generation capacity would be required to replace Ontario's nuclear.
• That amount of wind capacity would require 14,200 km² of land - the equivalent of a strip 14 km in width around the shorelines of southwestern Ontario.
• For the average residential consumer, the annual cost increase would be $632 (including HST).

Background The March 11 Japanese earthquake and tsunami and the subsequent nuclear disaster at Fukushima have sparked questions on the merits and risks of nuclear power generation.

In Germany, state elections have put the nuclear-opposing Green Party firmly back in the policy driver's seat and so the future of nuclear generation in Germany is increasingly bleak.

The most extreme voices in the conversation advocate - at least at some point - phasing out nuclear power completely. One option theorized, particularly by supporters of renewable energy, is to replace nuclear with wind power. However, the two forms of generation are as different as chalk and cheese. Too often, opinions expressed about what forms of generation we should use simply gloss over such differences or downplay their significance. So, to illustrate the differences in this case, we allowed ourselves to think, "what if?"

Attributes of Nuclear Generation in Ontario In 2010, nuclear generation in Ontario provided 82.8 TWh (82.8 million MWh or 82.8 billion kWh) of energy. Nuclear power operates at a generally constant or "flat" rate of output, so this is equivalent to a continuous, around-the-clock output of 9,452 MW. This average output represents an average "capacity factor" of about 90% from an installed nuclear capacity of about 10,500 MW. All nuclear output in Ontario is paid a contract price; early-2011 prices for Bruce 'A', Bruce 'B' and Ontario Power Generation output are, respectively, about $72/MWh, $51/MWh (a floor price that currently is the effective price paid) and $56/MWh. Based on 2010 output quantities from the plants, the average weighted price paid is about $56.50/MWh.

Attributes of Wind Generation in Ontario The capacity factor for wind is much lower, since wind energy is only available to the degree the wind blows. For our analysis, we evaluated the Ontario Independent Electricity System Operator (IESO) hourly wind data for the calendar years 2009 and 2010. Over that period, the installed capacity of wind farms included in this data rose from 903 MW to 1,328 MW. On a normalized (producing a constant, year-round quantity of installed capacity) basis, the average capacity factor over that period was 27.8%. This means that for every 1,000 MW of installed wind capacity, the average output over the course of a year would be 278 MW. More importantly, the output can vary greatly, with 1,000 MW producing as little as zero when there is no wind, and as much as 949 MW.

Cost-wise, the vast majority of output from Ontario wind plants is paid contract rates. These rates vary but in the not-too-distant future the dominant price paid will be the Feed-in-Tariff rates of $135/MWh for onshore projects and $190/MWh for offshore projects.

Notional Quantity of Wind Required To replace 82.8 TWh of annual energy now produced by nuclear power with wind at an assumed capacity factor of 27.8%, an installed capacity of 34,000 MW of wind would be required - almost 26 times as much wind capacity as Ontario had at the end of 2010.

Wind Land Use In our analysis we assumed that all turbines would be onshore. The next consideration is the spacing of the wind turbines. A recent study from Johns Hopkins University in the U.S. and Katholieke Universiteit Leuven in Belgium suggests that placing the wind turbines 15 rotor diameters apart, more than twice as far apart as is currently common, results in more cost-efficient power generation.

For our analysis we assumed spacing of 10 rotor diameters. We also assumed a per-turbine rated output of 3 MW and a rotor diameter of 112 meters. Each turbine would then have a footprint of 1.25 km². This means 11,333 turbines would be required for a nominal output of 34,000 MW and these turbines would require 14,200 km² of land.

The best onshore sites for wind turbines are commonly found along shorelines, often in southwestern Ontario. To illustrate the scale of wind installation required for this task, 14,200 km² is equivalent to a band about 14 km wide and about 1,000 km long, starting at Collingwood, running clockwise around Georgian Bay and Lake Huron (excluding the Bruce Peninsula), down to Windsor, east along the Lake Erie shore to Niagara and along Lake Ontario, back up to Toronto.

Output Variability of Wind Wind generation data from Ontario shows that installed wind capacity of 1,000 MW would produce an hourly output that could range from zero to 949 MW. Ontario would need 34,000 MW of wind capacity to produce 9,452 MW on average, but with this much installed capacity, the actual hourly output would range from zero (or close to zero) to 32,200 MW. What do we do when the wind output is more than we need?

A system peak hour in Ontario might represent about 24,000 MW of energy demand. If 9,452 MW of output is anticipated but over 32,000 MW materializes, close to 23,000 MW of unanticipated energy is attempting to come onto the grid. This quantity would overwhelm the 14,500 MW of other generation online at that moment.

Excess Generation Again, Ontario data shows that the annual amount of wind energy output generated above the desired output level of 9,452 MW would amount to 27.1 TWh or 32.7% of total potential wind output. This unanticipated excess output would exacerbate or create a number of undesirable outcomes, including:

• surplus base load generation
• dispatched-off situations
• subsidized exports

Surplus base load generation is a well-known dynamic. It is not uncommon in Ontario and the problem will grow, especially once two additional Bruce 'A' units come online in 2012 and as Green Energy Act-related wind and solar generation comes into service.

"Dispatching-off" is the process of paying a generator not to produce. If we assume a competitively-procured wind price of $110/MWh (18.5% less than the current onshore FIT price of $135/MWh), the potential annual liability from paying the wind generators not to produce when wind supply exceeds demand would be about $3 billion.

Subsidized exports occur when surplus power is allowed onto the grid but then exported since it exceeds Ontario demand. The power is bought at a high contract price, and sold in a low-price export market. Since at least some revenue is returned on the export sale, the economics of subsidized exports may be more attractive than dispatching-off. However, the strategy is limited by the ability of the grid to accept and transport the surplus power to the export market.

For our analysis, we relaxed the requirement to have exactly 9,452 MW of wind production and instead allowed wind output to exceed it by some buffer. Allowing a buffer quantity of up to an additional 1,000 MW, an annual 3.4 TWh of the 27.1 TWh of potential surplus wind output might be exported instead of being dispatched off. We then assumed that this additional wind output does not displace other Ontario generation but instead contributes to incremental exports; also, that energy is exported at an average price of $40/MWh. About $374 million of dispatched-off payments would be avoided but in return for paying for the output, the province would only receive $136 million for the exported power. The result then is that this buffer-quantity of exported wind power would still be subsidized to the tune of $238 million.

Back-Up Generation Required If we install 34,000 MW of wind capacity to get an anticipated base load output of 9,452 MW, we have to anticipate that the hourly output could be as low as zero when there is no wind. It's not unreasonable that 9,452 MW of back-up generation would be required in these periods. This generation must be firm and dispatchable and with coal soon to be out of the Ontario supply mix, natural gas-fired generation would be the back-up of choice. Assuming an availability of 95% of installed capacity, the required installed capacity would be about 10,000 MW - equivalent to about eleven of the cancelled 900 MW Oakville generating plants.

If we assume a 50/50 split between simple-cycle and combined-cycle gas turbine plants and a resulting annual fixed cost payment of $135,000/MW, the annual fixed cost for 10,000 MW of natural gas-fired generation would be $1.35 billion.

Ontario data indicates that about 27.1 TWh of energy output would be required annually from gas-fired generation to fill the hours when wind output is below 9,452 MW.

Assuming an average heat rate of 8.25 MMBtu/MWh, average marginal maintenance costs of $3.50/MWh and a plant-gate natural gas cost of $7.00/MMBtu, the marginal cost of gas-fired generation would be $61/MWh. Assuming spot market revenue of $40/MWh, the total marginal cost net of spot market revenue would be $21/MWh. The total net annual marginal cost for the 27.1 TWh of required natural gas-fired generation would then be $569 million.

Combining the capital and marginal costs, the total annual cost of natural gas-fired generation would be $1.92 billion.

Additional Wires Investment Required In our analysis we assumed that each MW of new wind capacity would require $200,000 of new wires (transmission and/or distribution). Assuming an annual cost metric of $0.18/year per wires dollar invested, the 34,000 MW of wind would require an additional wires investment of $6.8 billion and have a resulting annual cost of $1.22 billion.

Additional Ancillary Services The operating range noted earlier clearly points to the potential for, at best, operating challenges and, at worst, operating mayhem.

If it were even possible to integrate an additional 34,000 MW of wind into the Ontario grid, the IESO would require significant, incremental expenditures in the areas of planning, forecasting and most significantly, ancillary services.

In our analysis we assume an (admittedly, arbitrary) additional, related annual expenditure of $150 million.

Cost Summary At a unit price paid of about $56.50/MWh, the 82.8 million MWh of current annual nuclear output has an annual cost of about $4.7 billion.

For the theorized 34,000 MW of wind generation and 10,000 MW of natural gas-fired generation, the total annual cost is about $12.4 billion, resulting in a combined unit cost of about $150/MWh.

The total additional cost to replace current nuclear output with wind and natural gas would then be $7.7 billion or about $93/MWh on each unit of nuclear output. When this 165% increase is spread across all Ontario consumption, the tax-exclusive increase for most Ontario consumers would be about $56/MWh or 5.6 cents/kWh. On provincial-average residential consumption of 800 kWh/month, the annual, HST-inclusive increase would be $632.

Conclusion Aside from generating electricity from a different source of energy, wind energy and nuclear energy have a number of fundamentally different characteristics. No one could seriously propose replacing all of Ontario's current nuclear capacity with wind, but considering such a strategy helps to illustrate the differences in these types of generation, and illustrates how the design and development of Ontario's generation mix must balance a number of complex constraints.

Are Ontarians Subsidizing Exported Power? Read more »