Scientific Article Review: ‘Land-use intensity of electricity production and tomorrow’s energy landscape’

 

     This article by Jessica Lovering, Marian Swain, Linus Blomqvist, and Rebecca R. Hernandez, was published in June 2022 in the journal of the Public Library of Science (PLOS). In a lot of ways this paper just confirms what we already know, that some energy sources like wind, solar, and biomass are very high in land-use per unit of energy produced. The metric used in the paper is land-use intensity of energy (LUIE) – “(measured as hectares occupied per terawatt-hour of electricity generated in a given year [ha/TWh/y]) for real-world electricity generation–not hypothetical or modeled electricity generation–across all major sources of electricity and a broad geographic distribution.” This study aimed for more accuracy than previous studies, citing methodological weaknesses in those previous studies.

     Global electricity use is set to continue rising as developing countries grow their economies and modernize. The authors note that the current land-use for energy around the world amounts to 0.4% of ice-free land, compared to agriculture at 30-38% of ice-free land. Agricultural land-use has been dropping in recent years as agricultural intensification grows more food with less land. Energy land use is set to rise drastically, especially as more wind, solar, and biomass are added to the grid.

     The authors distinguish between direct land use intensity and indirect land use intensity. This varies considerably between resources. In addition to this, dual use capabilities vary between resources. While it is true that wind has a much lower energy density and takes up much more space than, say, natural gas, it is also true that both of those resources can accommodate agriculture for dual land use. With agrovoltaics, growing certain crops under solar panels, solar can accommodate some, but much less agricultural “co-generation.” The main issue with wind is the land required to accommodate spaced wind. In order for the wind resource to be optimally available the turbines must be spaced a certain distance apart, typically thousands of feet apart, or about 5 per square mile. Thus, the spacing footprint for wind is vastly larger than the actual land footprint for wind.

     The authors of the paper rightly distinguish between land footprint and spacing footprint. However, the much larger spacing footprint for wind presents other problems: people don’t want to live within a wind farm and often don’t want to live on the border of one either. There have been many complaints about the flashing lights, noises, and vibrations. Much more land must be leased per unit of energy produced for wind projects than for drilling or pipeline projects. Wind can only produce about 10 MW per square mile. Of course, that resource can last for thousands or even millions of years while the natural gas will deplete in decades to a century. That means it takes about 100 square miles of wind turbines to equal the amount of energy produced at a 1000 MW natural gas plant and that energy is intermittent and unpredictable while the gas energy is predictable and always available.

 

Note: This is on logarithmic scale. The same data is plotted with a few notes below on a linear scale.

     For both wind and natural gas, the authors used two different metrics: footprint and spacing. Footprint is the actual footprint of the facilities. Spacing includes all the land between facilities. In the case of wind that means the land between turbines and substations but for natural gas that means the land between wells, access roads, and pipelines. Both wind and natural gas require spacing. Natural gas requires it to optimally produce the resources without interference between wells. Wind requires it for a similar reason since harnessing wind changes local wind characteristics. One might even say as I have noted before that harnessing wind is a form of geoengineering. That spacing requirement is a hard limit for wind. An issue likely not accounted for is that in the subsurface there may be natural gas reservoirs at different depths from different reservoirs, additional resources that may be tapped from the same or near locations in the future. In terms of energy density, natural gas has a much higher energy density than wind. In the analysis given here it is about six times more energy dense than wind. They also note that natural gas LUIE with spacing is nearly that of ground-mounted solar PV. This is misleading since ground-mounted solar PV takes up all of that land while spaced natural gas just takes up a fraction of it. One could say the same for wind, which also takes up a fraction of actual land than does solar.  

     Since most decarbonization scenarios rely on huge increases of onshore wind and solar which both have very high LUIE it stands to reason that land use is on the rise for these facilities and calls for acceleration mean corresponding land use acceleration. A 2018 study find that 35% of onshore wind developments face opposition of some sort. Energy writer Robert Bryce has kept a database of wind and solar opposition that shows it is very strong and not likely to slow. These developments will no doubt decrease and fragment natural habitats much more than fossil fuel facilities. Offshore wind and floating solar do not require land except for transmission and rooftop solar does not require land.

     Compared to other land use studies, the conclusions here show natural gas as having a much higher LUIE than other studies have shown. Their analysis shows natural gas LUIE as 12-58 times that of Vaclav Smil’s 2010 study. That concerns me regarding overall accuracy. They explain that away poorly by simply noting that Smil probably based his estimates on land-efficient natural gas operations. I believe most current natural gas ops aim to be as land efficient as possible as land footprint is a key ESG metric. LUIE values for natural gas have no doubt been decreasing drastically in recent years due to longer well laterals, more wells per pad, higher production per well and pad, and stacked plays which allow closer well spacing. Very significant production improvements have also led to a much-increased amount of energy produced per unit of land. They do not really address this rather huge discrepancy between their results for natural gas and Smil’s.

However, even with that questionable discrepancy, they note:

“Our results suggest that production of electricity to meet decarbonization goals could become a significant new driver of land-use and land-cover change with implications for habitat and biodiversity loss, food security, and other environmental and social priorities. An expanding footprint is not inevitable: the LUIE for integrated PV, nuclear, the footprint of wind, and geothermal are each less than coal or natural gas, which together, currently generate more than 60% of the world’s electricity.”

     I think that statement is misleading since it suggests, or rather says outright that the LUIE of “the footprint of wind” and “integrated PV” is less than coal or natural gas, which is technically true but incomplete. It requires the leasing of much more land per unit of energy produced. It takes a certain amount of space for wind, much more than for natural gas, regardless of the actual land footprint of facilities. It is the “space” required between turbines that is at issue. That is a hard limit. Natural gas has some spacing limits but not the hard limit that wind has. The spacing requirement for wind means that according to their data it is neatly 100 times that of the turbine footprint but for natural gas it is less than 5 times from footprint to spacing. It doesn’t take a rocket scientist to determine that for replacing natural gas with wind an expanding footprint is indeed inevitable, directly contradicting the conclusion in the statement, since the energy density of coal and natural gas is much higher than that of wind and solar. It is already coming to pass as large wind farms and solar farms are built. Other future footprints not considered include the additional mining and land use for the needed expansion of the transmission system. According to some estimates, the transmission system would be needed to be expanded by 2 or 3 times what it is today to accommodate high levels of wind and solar as intermittent resources.  

 

References:

Land-use intensity of electricity production and tomorrow’s energy landscape. Jessica Lovering, Marian Swain, Linus Blomqvist, Rebecca R. Hernandez. Public Library of Science (PLOS One).  Land-use intensity of electricity production and tomorrow’s energy landscape | PLOS ONE