The Physics Police

The Physics Police

Friday, September 12, 2014

Solar Power and Birds

The Ivanpah Solar Power Facility is a "thermal tower" solar plat in the Mojave Desert, California. Rather than converting sunlight directly into electricity as do photovoltaic solar panels, thousands of mirrors reflect sunlight at a central tower. Water in the tower is heated to 1,000 degrees Fahrenheit. The resulting steam runs an electric turbine. Storing solar energy in the form of heat allows the plant to generate electricity 24 hours a day. Latent heat used overnight is replenishing the next day.

This feature allows the plant to avoid the major drawback of other renewable energy production systems, like wind farms and photovoltaic arrays, which only output energy when the wind is blowing, and the Sun is shining, respectively.

Ivanpah Solar Power Facility (No, it's not Burning Man.)
Criticisms of this pilot plant include the cost, and environmental impact. As for the cost, pilot plants are always expensive. Huge savings are typically seen when a technology is scaled up. The facility cost about $5,561 per kW, only one and a half times the cost of building a new coal facility which runs about $3,500 per kW. Anyway, cost is a concern for the economists. I'm interested in the environmental impact, because I'm an environmentalist.

Unforeseen in the Environmental Impact Statement and prior to construction, some birds were found to have been burnt by when they flew into the area of focused sunlight. US Fish and Wildlife Service (FWS) began an investigation in July, 2012.

Shortly after the plant formally opened in February, 2014 the Wall Street Journal ran an attack piece by Cassandra Sweet with accompanying correspondent interview.
A giant solar-power project officially opening this week in the California desert is the first of its kind, and may be among the last, in part because of growing evidence that the technology it uses is killing birds.
This WSJ article leaves out the fact that all buildings with windows kill some birds, and that all methods of generating power harm the environment. The question is, how many birds are killed by this facility due to heat damage from fling into the focal range? And, how does the environmental impact of this facility compare to other types of facilities?

By the time the FWS published a preliminary report of their investigation in April, 2014, the news story was long forgotten. The report got little attention until August, when Ellen Knickmeyer and John Locher published an egregiously misleading article in the Associated Press.
Federal wildlife investigators who visited the BrightSource Energy plant last year and watched as birds burned and fell, reporting an average of one "streamer" every two minutes, are urging California officials to halt the operator's application to build a still-bigger version.
Here's what report actually says about the "streamers":
... these events represent the combustion of loose debris, or insects.
In reality, the report investigation found only 47 birds injured by solar flux in the 1-year study period. So, at least 47 of the "streamers" might have, in fact, been birds. But not all of them. The article attempts to fool its readers by knowingly conflating the streamers with bird deaths. I trust my readers know better than to fall for such an obvious syllogistic fallacy.

Credulous readers of the AP article would walk away thinking that the report showed 5,591 times more bird fatalities than were actually reported. That's misleading by 3.7 orders of magnitude!

How does this compare to some other anthropogenic sources of avian mortality?
  • An estimated 1.4-3.7 billion birds are killed each year by cats.
  • As many as 980 million birds crash into buildings annually.
  • 174 million birds die from power lines every year.
  • Up to 340 million birds perish from vehicles/roads.
  • Approximately 6.8 million birds die flying into communications towers.

And how does this compare to other types of power generation?

Avian Mortality from different types of power generation.

One study compares estimates of the number of avian fatalities per GWh by energy sector.
  • 0.27 for Wind
  • 0.60 for Nuclear
  • 9.40 for Fossil Fuel
Whereas I calculate the total (solar flux + other causes) fatalities per Gigawatt-hour:
  • (141 birds per year) / (1000 Gigawatt-hours per year) = 0.141 for Ivanpah

That's right. The Ivanpah plant costs the least number of bird's lives per Gigawatt-hour. It's terrible to measure electric power in terms of dead birds. But if you really care about protecting birds, you must question whether the benefits outweigh the risks, not to your own political ideology or world-view, but to the birds.

Moral and environmental consideration of a technology must include the cost of the alternatives!

Solar thermal towers help protect birds by replacing more hazardous means to generate power.


  1. I fully agree with your wider point - that fossil fuels almost certainly kill more wildlife per GWh than solar or wind. However, I think you need to be cautious about how you use these particular data from the FWS study. They collected those 141 dead birds over the course of a year and five months, which included seven months of construction when the towers weren't fluxing - so only ten months of active generation, not twelve. Also, they do NOT claim that these 141 birds (including 47 killed by the heat) represent ALL of the birds killed in that period - they're just the birds they found on the dates they visited during that period. More dead birds may have been carried off by scavengers etc.

    Of course, even if that 47 represented just 10% of bird deaths, the total would still be far, far lower than the "one every two minutes" suggested by the dodgy media coverage, and much lower than bird deaths from fossil fuels. But still, worth being careful with the phrasing - it's really important that we don't end up accidentally being misleading ourselves!

  2. Thanks for noticing these two points. 47 in 10 months is makes a rate of 56.4 per year. The referenced literature indicates carcass recovery can be as low as 30% due to scavengers. That justifies a conservative 56.4 / (0.3) = 188 birds per year instead of the 141 figure I used. That means 0.188 fatalities per GWh, still smaller than Wind power.