October 29, 2010
Doing the Math: Comparing Germany’s Solar Industry to Japan’s Fukushima Reactors
By Sara Mansur and Devon Swezey
Updated Below (3/24/2011)
Grist environment writer Christopher Mims has written a widely read post comparing Japan's Fukushima nuclear reactor complex to solar photovoltaic energy in Germany. The post, "Germany's Solar Panels Produce More Power Than Japan's Entire Fukushima Complex," implies that solar PV may be an adequate substitute for aging nuclear reactors in both Germany and Japan.
But an analysis of the electricity generated by Germany's solar PV industry and Japan's Fukushima Daiichi reactors finds that Germany's entire solar PV capacity, installed at a cost of at least $86 billion, generated only half the amount of electricity generated by the Fukushima plants in 2010.
"It's worth noting that just today, total power output of Germany's installed solar PV panels hit 12.1 GW -- greater than the total power output (10 GW) of Japan's entire 6-reactor nuclear power plant."
There are two problems with this.
First of all, the total installed capacity of Japan's Fukushima six-reactor Daiishi plant is actually 4.5 GW. The total power output of Japan's entire Fukushima complex, which consists of ten reactors--six at Daiichi and an additional four at Daini--is 8.8 GW. So Germany's peak solar PV output of 12.1 GW is nearly three times greater than Japan's Daiichi reactor complex.
Does that mean that solar in Germany is somehow equivalent to three of Japan's nuclear complexes? The answer is no, and this leads to the second problem with Mims' post.
The 12.1 GW that Mims cites is the total power generated at one peak time of day. But Mims' numbers don't tell us anything about what we really care about, which is electricity generation.
As Mims himself notes, solar power production varies with weather and the time of day--it doesn't supply 12.1 GW of power continuously. Rather, looking at total electricity generated over a year gives us a much more accurate, apples-to-apples comparison of each technology's contribution to a country's energy needs.
According to Mims:
"To find out how much energy those panels generated today in total, you'd have to calculate the area under that curve in the lower right hand corner."
Fortunately, we've run those calculations, and they present a much different picture than the one implied in Mims' post.
In 2010, Germany's cumulative installed solar PV stood at 17.3 GW. In 2009, Germany's PV solar capacity factor--the ratio of actual energy output over the year and the energy the plant would have produced at full capacity--was 9.5%. This is quite low for solar PV, which typically has capacity factors around 15%, and is likely due to the fact that Germany doesn't actually get that much sun. If we assume the same 9.5% capacity factor for 2010, then Germany's 17.3 GW translates into about 14,397 GWh of actual annual electricity generation from solar cells.
By comparison, in 2010, Fukushima's six Daiichi reactors--which have a nameplate capacity of 4.5 GW--produced 29,221 GWh of power generation.
That is, one nuclear power plant complex produces more than twice the power generation of Germany's entire installed solar industry.
Furthermore, Germany's entire solar PV output is equal to a little more than one percent of Japan's total electricity generation.
So could we feasibly replace the power generated from nuclear in Japan with electricity from solar?
The German solar industry was built over 20 years with expansive government support. Using an estimate of $5 per watt of installed solar PV capacity, we estimate the country's 17.3 GW in installed solar capacity to have cost at least $86.5 billion dollars. The actual costs are likely higher, since this estimate assumes 2010 module prices, while costs have substantially declined in the past decade.
As Breakthrough's Jesse Jenkins, Ted Nordhaus, and Michael Shellenberger make clear in today's Atlantic:
Present day renewables remain too expensive and undependable for any economy in the world to rely upon at significant scale. So Germany, despite its vaunted solar feed in tariffs, will rely more heavily upon coal, which it has in abundance, as it retires its aging nuclear fleet. The US, already in the midst of a natural gas boom, will use more gas.
In the comments to this post, Christopher Mims deflects attention from the factual errors and misleading comparison in his original post. The facts of the matter are that Mims provided not only incorrect information about the capacity of the Daiichi nuclear reactors, but also made completely irrelevant comparison between total installed, intermittent solar capacity in a country with a solar capacity factor of 9.5% to a nuclear power complex that produces vastly more electricity.
Particularly since this post has been widely viewed, now listed as the most viewed post on Grist's site, Mims owes it to his readers to correct his original post and provide not only factual information but also an accurate comparison that will allow Grist's readers to draw more reasonable implications.
Mims has asked us to compare the cost of building, operating, and maintaining the Daiichi reactors to the equivalent in Germany's solar PV installations. The answer to Mims' question is that the equivalent in power generation from solar PV would cost at least $100 billion more than the generation from the Daiichi reactors.
This was calculated using across-the-board conservative estimates for the cost of solar and high estimates for the cost of nuclear.
Here are those calculations:
The cost of constructing a new nuclear plant today is estimated to be between $3,000-$7,000/kw. If we assumed cost of $6,000/kw capacity, we'd find that Fukushima's Daiichi reactors cost $27 billion. This estimate is likely on the high end, since the cost of nuclear technologies has risen in the past two decades.
As we calculated in the post, it cost more than $86 billion for Germany to produce less than half as much generation as Daiichi from its total installed solar PV. So if we scaled to the equivalent amount of energy, we would find that it would cost at least $145 billion in solar technologies to produce the same amount of energy as Fukushima's Daiishi.
Now, neither of those estimates includes operation and maintenance costs. According to the EIA, fixed and variable operation and maintenance costs for advanced nuclear technologies cost about 3.3 cents per kWh. That's $26.5 billion in operation and maintenance costs for the Daiichi reactors' lifetime generation until 2009. The equivalent maintenance and operation costs in solar production (using the EIA's estimate of 1.1 cents/kWh) is $9.6 billion.
In sum, it costs $155 billion dollars to construct, operate and maintain solar technologies that will produce less than the equivalent amount of power as $53.5 billion in new nuclear technologies.