If you’re interested in energy and climate issues, then you should definitely check out the writing of David Roberts at grist.org, a news/blogging site that focuses on environmental issues. I don’t always agree with him, but I find that his analysis is usually insightful and his twitter feed is full of links to great articles, both his own and others’.
That’s why I was so surprised to see two successive tweets last week that (in 140 characters, no less) raised some major red flags about renewable energy in the U.S. The first claimed that the intermittency of wind resources is “NOT a big deal”:
The second seemed to downplay concerns that renewable energy requirements in states around the nation would lead to increases in energy costs:
Now, Mr. Roberts is not the author of either of the articles to which he linked, but by sharing them to his 11,000+ followers, he’s certainly giving them his endorsement. Which is why I was disappointed to find, upon reading them, that neither of them is a very good analysis, and both downplay some very real issues that will have to be dealt with as renewables become a larger part of our nation’s energy mix.
Wind Intermittency: A Bigger Deal than a Former Vestas Employee Might Admit
The first red flag when I opened this article was the author: “Chris Varrone, Founder and President of Riverview Consulting and Former Chief Strategist, Technology R&D at Vestas Wind Systems.” Vestas is a fairly well known wind developer, and when I see an article written by a former employee that says “Wind is way better than people are saying!”, I approach it with a bit of skepticism. In this case, that skepticism was warranted.
Mr. Varrone’s assertion that we need not be worried about the intermittency of wind is based on three claims: (1) the electricity market already has a category for these type of costs (called ancillary services, distinct from energy and capacity costs) so they’re already being included, (2) even when you calculate the impact of wind on the capacity and ancillary services markets, it’s very small, and (3) experiences in Europe show that we can increase wind penetration without affecting these integration costs.
The first claim, that “intermittency is no big deal” because the electricity markets “have already accounted for this” by separating payments to generators into energy, capacity, and ancillary services, is almost laughable. Yes, electricity markets do indeed separate their payments into these categories, and yes, wind only receives payments for energy. But the issue is the effect of increasing wind penetration on the capacity and ancillary services payments to other generators, not whether wind is receiving these types of payments. As we add more wind to the electricity mix, we increase the need for ancillary services, and as basic economic theory tells us, increased demand leads to an increase in the price of the good demanded. Increased payments in the ancillary services market are an external cost associated with wind generation: they do not show up in the price of purchased wind, but the consumer has to pay them all the same.
Wind has a similar effect on the capacity market. As Varrone points out, “the value of wind capacity is low, about 10 to 20% of nameplate capacity” (i.e. 100 MW of wind can be expected to provide 10 to 20 MW of generation during peak demand hours). This low capacity value reflects the fact that wind turbines generate less energy on those hot summer days when air conditioners are running and loads are high. So while wind energy can displace energy from dispatchable resources like coal and gas, utilities will still need to develop these resources to ensure that they can meet capacity demands during those peak hours. This, too, is a cost that arises directly from wind penetration but is not included in the price that utilities pay to the wind generators.
Varrone concedes that there might be an effect on the ancillary services market, but says (without providing any references) that this effect is almost negligible: “Many studies have found the cost of wind integration to be in the $3 to $5 per MWh range.” Compared to the overall cost of wind (the California Energy Commission Cost of Generation Model estimates the levelized cost of wind generation at ~$65/MWh for the highest quality wind in California), that’s not so much. But arriving at the $3-$5 number requires a fairly selective reading of studies, as far as I can tell. This graph (produced by consulting firm E3* for their work analyzing AB32 for the California Public Utilities Commission, source document here) shows the results of a variety of wind integration studies, many of which are well outside of Varrone’s range, especially at higher levels of wind penetration.
Varrone’s final argument is that experiences in Europe, specifically in Northern Germany and Denmark where “wind accounts for 50% or even 100% of electricity demand for certain periods”, has shown that these problems will not increase as we add more wind to the grid. But this leaves out an important detail about the electrical grid in those areas, hinted at in his quote from Mr. Peter Jørgensen, Vice President of International Relations at Energinet, Denmark’s grid operator: “We are able to balance the present system with strong interconnectors, market-based trade with the neighboring countries, and good wind forecasts.” While all of these factors are important, the key is the “strong interconnectors.” As Varrone points out, quoting Jay Apt from Carnegie Mellon, “[t]he United States ranks among the bottom among developed nations in terms of the reliability of its electricity service.” This means that we don’t have those strong interconnectors between areas that Denmark and Northern Germany do.
Furthermore, even if we did have those interconnectors, we don’t have access to the extensive Hydro resources in Norway and Sweden that Denmark and Northern Germany and Denmark do. According to the Energy Information Agency, Norway had 30.788 GW of installed capacity in 2008, 28.246 GW of which (92%) was Hydro. In contrast, Washington (highest in the nation in terms of Hydro capacity as percent of total installed) had 68% Hydro in 2009 (Excel file), with 20.815 GW out of 30.629 GW total. Sweden also has a significant amount of Hydro, with 16.352 GW (48%) of 33.943 GW in 2008. While less than Washington as a percent of the total, it still far exceeds the 7% that Hydro makes of the U.S. total installed capacity (from the same spreadsheet as the Washington statistics above).
Why is this important? Because as the Bonneville Power Administration notes, “[h]ydropower is flexible” and “can be increased or decreased very quickly to match variations in demand for electricity.” While some conventional resources can also be ramped to accommodate these variations, hydro has the advantage of low operating costs, as it is powered by water rather than costly fossil fuels.
So really, the comparison that Varrone draws to Europe is inappropriate due to the differences between the two systems. He also points to Iowa as having high wind penetration and no operational problems, which is a more promising data point. However, Iowa is part of the Midwest ISO, and Iowa’s total generation capacity in 2009 (15.8 GW, from Excel file linked above) represents just over 11% of the total generating capacity in MISO as of 2006. So while wind may provide a substantial portion of Iowa’s load, it still remains a small portion of the overall load in the area.
Meanwhile, there is a data point for what happens when wind becomes a large share of the generation in an electrically isolated area. Texas, electrically distinct from the surrounding states as the Electricity Reliability Council of Texas (ERCOT), has been adding wind generation rapidly over the past number of years. Unfortunately, due to the geographic remoteness of the resource, it has been leading to headlines like this: “Loss of Wind Causes Texas Power Grid Emergency.” ERCOT had to shed 1,100 MW of interruptible load to prevent rolling blackouts in the state.
I think that wind power has a significant part to play in our energy future, but it is not a panacea. There are very real issues with drastically increasing the amount of wind in our electricity mix (which will be necessary to decrease emissions in the most cost-effective manner), and dismissing these concerns with a wave of the hand only does a disservice to wind and will result in confusion amongst the public later when the claims of the wind industry turn out to be no more than hot air. (Yeah, it’s a terrible pun. I couldn’t resist).
Since this post went way longer than I expected it to, I’ll hold off on addressing the second article posted by Roberts until a later post. If you made it this far, I must say, I’m impressed. More to come soon.
* Disclosure – I used to work full-time for E3, and now do some work for them on an as-needed basis. However, I did not have anything to do with the study in question.