Amory Lovins vs. Stewart Brand - Part One (The “Land Footprint Myth”)

Three weeks ago Mr. Amory Lovins released a very pointed critique of Stewart Brand’s chapter on nuclear in Brand’s new book, Whole Earth Discipline. After reading both Brand’s and Lovins’ pieces, I understood why Lovins was so critical of Brand. It was because Brand was quite critical of Lovins in his book (p. 99):

In early 2009, in Ambio magazine, Amory Lovins declared: “Nuclear power is continuing its decades-long collapse in the global marketplace because it’s grossly uncompetitive, unneeded, and obsolete.”

How can someone [Lovins] so smart be so wrong about a subject he knows so well? [Emphasis added]

Ouch. It’s now clear to us why Mr. Lovins came out with his critique of Brand when he did.

Over the past couple of weeks, we’ve been able to digest Mr. Lovins’ latest claims in his new study (pdf) and have generated quite a few thoughts to share. In Lovins’ response to Brand’s chapter on nuclear, Lovins takes Brand to task on four issues he believes are myths about nuclear: baseload energy, land footprint, the need for all options, and the role of government. Because there is a lot to discuss about each topic, we’re going to present blogposts addressing each of the myths to show how Lovins’ latest critique is nothing more than the usual cherry-picked junk that we’ve always seen from Lovins.

Lovins’ supposed “footprint myth”
One issue the Lovins clan has with Brand is the claim that wind and solar generating facilities need a tremendous amount of land to produce the same amount of electricity as nuclear plants. Here’s the quote from Brand (p. 81):

As for footprint, Gwyneth Cravens points out that “A nuclear plant producing 1,000 megawatts takes up a third of a square mile. A wind farm would have to cover over 200 square miles to obtain the same result, and a solar array over 50 square miles.”

Here’s what the Lovins study says in response after making their own calculations (p. 16):

windpower is far less land-intensive than nuclear power; photovoltaics spread across land comparable to nuclear if mounted on the ground in average U.S. sites, but much or most of that land (shown in the table) can be shared with lifestock or wildlife, and PVs use no land if mounted on structures, as ~90% now are. Brand’s “footprint” is thus the opposite of what he claims.

When comparing land footprints among the three technologies, the Lovins study used a total nuclear lifecycle footprint of 119 square meters/GWh from a study written by two national lab scientists (Fthenakis and Kim). As usual, the new Lovins study cherry-picked only one chart from F&K’s study and that was a chart showing the amount of land nuclear plants need during the entire life cycle of a nuclear energy facility (mining, power plant, etc.). F&K’s study, however, didn’t just look at nuclear, they also showed the amount of land needed for the life cycle of all other technologies. Below is the chart:
As can be seen from the chart highlighted in red, nuclear’s life cycle land use requirements come in several orders of magnitude lower than wind’s and solar’s. Yet this chart and the study’s conclusions are ignored in Lovins’ paper and only the number for nuclear is used.

As well, if we continue to use the 119 square meters/GWh land use for nuclear, other studies cited in the Lovins paper also show nuclear uses much less land than wind and solar. Below is a chart from the source the Lovins study uses for its solar number (pdf). Even using the lowest range of land needed per year from the last column shows that solar needs at least 42 times and wind needs more than 1,100 times the amount of land as a nuclear plant.
For wind’s footprint, the Lovins study cited “the Bush Administration’s 20% Wind Energy by 2030” study but again cherry-picked the data to support its claims. Here’s the full paragraph from the wind study (pdf) of which only the last half was mentioned in the Lovins study (p. 110-111):

Wind development also requires large areas of land, but the land is used very differently. The 20% Wind Scenario (305 GW) estimates that in the United States, about 50,000 square kilometers (km2) would be required for land-based projects and more than 11,000 km2 would be needed for offshore projects. However, the footprint of land that will actually be disturbed for wind development projects under the 20% Wind Scenario ranges from 2% to 5% of the total amount (representing land needed for the turbines and related infrastructure). Thus the amount of land to be disturbed by wind development under the 20% Wind Scenario is only 1,000 to 2,500 km2 (100,000 to 250,000 hectares)…

So, for 305 GW of wind, the required area of land is estimated to be 50,000 square kilometers or 19,300 square miles. Dividing 19,300 by 305 GW, we find that a wind farm requires 63 square miles of space per GW. If we multiply that area by three to account for wind’s 30% capacity factor compared to nuclear’s 90%, we find that a wind farm requires nearly 200 square miles of land “to obtain the same result as a [1,000 MW] nuclear plant.” Close to what Brand and Cravens said.

Yes, the actual land “disturbed” by a wind turbine is only 2-5% of that, however, a wind turbine needs a huge amount of open area to produce meaningful quantities of electricity. This requirement can’t be ignored, even though Lovins calls it “erroneous,” else wind turbines would be stacked right next to each other. It would be disingenuous to tell the Iowa farmers that a wind farm doesn’t take up much land when all they need to do is walk outside their homes and see their entire horizon blanketed by turbines.

Further, on closer look at Lovins’ sources, cherry-picking again appears. On pages 13 and 14, Lovins cites a study written by Spitzley & Keoleian (pdf) which Lovins picks a few convenient nuclear numbers and ignores the rest. Yet those authors also wrote a study analyzing all technologies, not just nuclear. Below is a picture of S&K’s page 31, which shows the amount of land needed for all technologies. Highlighted in red are the numbers that show nuclear uses much less land than wind and solar.
Three sources cited in the Lovins study concluded that nuclear uses much less land than solar and wind. Clearly, the authors of those studies consider the open areas between wind turbines and the large arrays for solar plants a requirement to function. Yet the Lovins study clearly manipulated the numbers from those sources to fit its own beliefs. Thus, it’s not Brand and Cravens who believe in a land footprint “myth”, it’s Mr. Lovins.

Stay tuned as we’ll get into what qualifies as baseload energy.