Land Requirements

The objective of this section is to discuss the profound limitations that are imposed by the diffuse nature of all renewable energy sources and to further discuss what land management policies should be considered to provide the world with renewable energy that is also sustainable. We assert that the biggest, generally unappreciated, limitation of all current renewable energy systems is the huge amount of land that is required to displace even a modest amount of fossil fuels. Moreover, making matters worse, many of the current renewable technologies are extremely wasteful of land (or sea) area, using much more than is theoretically needed and thus ultimately making these renewable technologies economically, ecologically, and politically unviable if they are used for anything more than small-scale projects.

The world currently uses about 15 TWa (terawatts average) of power from all fuel sources combined and this usage is expected to increase to more than 30 TWa by 2050. Moreover, many of the fuel sources that are currently providing 15 TWa today are non-renewable. Therefore, we expect that more than 15 TWa of new power must come online in the next 40 years. A very conservative estimate is 18 TWa of new baseload is needed by 2050. This is based on looking at the depletion rates just from crude oil alone. The 18TWa of new capacity translates into the equivalent of 450 new 1 GWa (one Gigga Watt average) needing to be built every year for the next 40 years to meet demand. The 18,000 new 1 GWa power plants is so large a number that we consider any technology that is incapable of scaling to provide at least 1 TWa (i.e. 1,000 power plants producing 1 GWa each) over the next 40 years to be a failure from the point of view of providing planetary scale power---even if it is an effective technology in certain niche markets.

So, which renewable technologies can scale to at least 1 TWa? In what follows we consider that question just from the point of view of the land usage. Please see the Resource Limitations section of this web site for an extension of this discussion from the point of view of rare materials like the technology metals tellurium used in thin film solar cells.

To better appreciate the scale of the land needed consider Fig. 1. This is a bar chart of the average power density of different renewable technologies. Average power density is what a baseload power plant produces---it is measured in average Watts per square meter [Wa/m^2]. Note that the peak power density, which is often quoted for some renewable sources like solar, is not utilized here because it is not a good figure of merit for baseload power. Baseload (or average power) is also the desired quantity when discussing land resources requirements.

 

Fig. 1. Average Power Density & Area Used for 1 GW baseload Renewable Power Plant. This critical figure shows just how incredibly difficult it is to capture diffuse renewable energy. Many different renewable power technologies are compared and the corresponding best-in-class power density, measured in W/m^2, is shown in the bar chart. Additionally, four other scales are presented: (1) the area needed to make a 1 GW baseload power plant, (2) the length of the edge of a square needed to fully power 500,000,000 people in the USA in the year 2050 at the current per capita energy consumption of 250kWh/day/person, (3) the fraction of the continental USA the hypothetical square requires, (4) the approximate land cost when sold at $10M per square km ($40k per acre). The best-in-class renewable resources are represented in this figure and it covers most types of renewable energy technology where public data is currently available. As can be seen even the most power-dense flows listed are very diffuse and require a lot of land. The inset of the USA shows what 1% and 10% looks like at several separate locations---see corresponding Frac. US scale. NOTES: Those items listed with a TRIANGLE symbol are some of the largest solar power plants that are functional or under construction/planning in the USA. The conversion efficiency of bio-fuel and geothermal heat sources to electricity using a heat engine was assumed to be 50%. The solar insolation was assumed to be 225 Watts per sq. meter average, which is the annual day-and-night average insolation found in sunny Los Angeles California and Phoenix Arizona. Geothermal is assumed to be run in renewable-mode so that non-renewable heat-mining is avoided. The XE technology, presented in the last two bars, is currently under development. Source of data: various government and academic reports provide in-depth analysis, however, a good place to start is the easy to read and number-rich, text from Cambridge University physics Professor David MacKay, "Sustainable Energy--without the hot air", ISBN 978-0-9544529-3-3 and it is also available free at www.withouthotair.com .

 

The bar chart in Fig. 1, is plotted in increasing order of average power density for many different renewable power sources. Additionally, on the right hand side there are a number of conversions that have been performed to address area and costs required to completely power the US in the year 2050. The first of these conversions is the number of square kilometers needed for a 1 GWa power plant, the next is the edge length of a square enclosing that area, the third scale is the percentage of the United States and the fourth scale is a rough estimate of the land cost (in trillions of 2010 dollars).

Let’s take a quick tour of the numbers. Lowest of the performers is renewable-mode geothermal, at less than 0.001 Wa/m^2 of average electricity generated. This occurs when geothermal energy is harvested from the earth slowly enough (renewable-mode) to allow the depths of the earth to remain at a hot constant temperature even though cold water is being passed though the earth to make steam to run a turbine. While there are a few locations on the planet where there are significant amounts of geothermal energy most places are not good locations for much geothermal power. For example, it would take almost 3,000 times the area of the USA to provide sufficient power from this technology in the year 2050 to supply all Americans’ needs. Although, geothermal has some excellent niche applications it will never scale to provide the USA or the world a significant fraction of its total power needs.

The next batch of renewable technologies in Fig. 1 are many of the bio-fuel alternatives, starting with corn ethanol. Of particular, interest is the much celebrated cellulosic based ethanol derived from switch-grass. This has a capacity of about 0.01 Wa/m^2—more than 10 times better than corn ethanol. Yet, it would still take more than 250% of the area of the country for powering the USA in 2050—it is literally off the fraction graph of Fig. 1. To better appreciate the amount of land cellulosic ethanol a new figure was developed and the result is the largest red square of Fig. 2 (below).

 

Fig. 2. Renewable Energy Area Requirements in the USA needed to support all energy needs, using 100% electrical power, for 500 million people, i.e. the population expected circa 2050, each with a per capita power use of 250 kWh/day. Each equivalent area square is based on using the best of kind technologies as seen in Fig. 1. The largest red square corresponds to cellulosic ethanol in Fig. 1 and is not even the lowest land efficiency technology! The smallest green square corresponds to the use of XE's eTracking system and current state-of-the-art triple-junction solar cells---this corresponds to the "3-Junction" bar in Fig. 1. Finally, this figure begins to shows why central power stations will become the predominant power suppliers and not solar roofs. Calculations show there is nowhere near enough usable roof area to even use efficient solar to supply the needed power requirements of the USA, which is a common misconception of many in the solar industry---although roofs do comprise a market and can generate cash flows. NOTES: (1) Same notes as in Fig 1. (2) Please understand that XE is not implying through this figure that we expect only one energy technology, such as ours, for the entire USA. We expect that there will be many technologies used, however, this image suggests the daunting scale of the problem and why land is to become one of the most important energy generation resources.

 

Yet another favorite renewable power source that has been much discussed in recent years is genetically optimized algae for the production of diesel fuel. Of particular interest are those algae that are allowed to consume carbon dioxide directly from the exhaust of power plants to reduce green house gas emissions and boost the growth of the algae---this reduces the amount of work the algae has to do to concentrate the CO2. The graph in Fig. 1 shows the optimized algae at about 10% concentration of CO2 in the growth water. Some of the best experiments done to date have occurred in 2009 at the cost of $600M by Exxon Mobil and Synthethic Genomics [Discover Mag. Jan/Feb 2010]. In these experiments 2,000 gallons of algae-based fuel were obtained from each acre of sun-drenched water area per year. Let’s derive the average power density in order to appreciate how distorting claims made in non-standard units like gallons (or tons etc..) actually are, even if they are factually correct:

where the 50% factor was introduced to account approximately for the conversion of a fossil fuel into electricity. From this we can see from the “Frac. US” scale in Fig. 1 that it takes the equivalent of just over 50% of the area of the USA (obviously in a water environment) dedicated to algae growth to power the country in 2050 using genetically optimized algae. Even if we were not interested in generating electricity and only in liquid-fuels for traditional automobiles we would still need an impractical amount of ocean area to be allocated for algae farming. NOTE: Bio-fuels do make sense for certain niche applications where a portable high energy density storage is the only option, such as for military and some aviation uses, but surely not as a general power source.

Next in Fig. 1 we finally arrive at ground-based wind turbines and solar. These systems come in a number of configurations using different technologies. Nonetheless, the range of average power for state-of-the-art power plants is no more than about 2-9 W/m^2. Of particular interest to us are the state-of-the-art solar power plants that are currently scheduled for construction in the USA, these power plants are indicated with a TRIANGLE symbol. Even at the upper end range of 9 W/m^2 for solar it would take an area of over 100 km^2 to make just one 1 GWa power plant! Moreover, to power the USA in the year 2050 using the current per capita energy of a typical American (250 kWh/day/person) would take an area of 580,000 km^2. This is equivalent to an area having extent of 760 km × 760 km. On average these systems would take about 10% of the land area of the USA—see the inset of Fig. 1.

Additionally, the sunniest deserts in the USA provide about 7.5 kWh/m^2/day for a general 2-xais tracker and about 5 kWh/m^2/day for the direct component, which is the component used in 2-axis concentrator tracking technologies—see for sn example table of data see Insolation at Bakersfield California. Therefore, for a 25% efficient solar panel (e.g. a state-of-the-art HCPV panel!) having a 50% area efficiency one American would need to have about 400 m^2 of desert or roof area completely devoted to his or her immediate living energy needs. Unfortunately, the data in Fig. 1 (see TRIANGLE symbol tagged labels) shows that today's solar plants are being designed to only cover about 10% of the land so that one American actually would need to use 2,000 m^2 of land area to meet their total energy needs---not just the current electrical component of energy usage.

At this point the reader might be thinking that the whole matter could be more easily settled if instead of using utility-scale power plants we were to use all of the roof space on homes and businesses. After all it is well known that a good bit of current electricity usage can be offset by the installation of solar panels on a roof. Unfortunately, the matter is not so simple. The required area that needs to be devoted to each person in the US is 10-100 times more area than the usable roof area in the USA can provide! The usable roof area is approximately 10 m^2/person (about 80% of people live in cities in multi-story dwellings). So, we expect that some form of utility-scale solar is the only long-term viable option to solve a significant fraction of the energy generation problem. Non-utility-scale solutions will still play a part and will have a market but utility-scale power is the only option that provides a path to a solution---this is simply the consequence of physics and the scale of the problem we face.

So, we have outlined here why land area is more critical than is generally appreciated, especially if it is the intent of humanity to derive a significant fraction of our energy from renewable sources. The point made herein is not that we expect solar, or any other technology, to entirely power the US by 2050 using any particular renewable energy technology, but rather that even trying to power a 30%-50% fraction of the US by renewables is going to be an exercise in futility unless policy makers start to grasp the incredible scale of the energy problem that lies before us and start to focus on only those technologies that can deliver the highest energy density at the lowest cost. Currently only the most advanced (and expensive) forms of solar energy, at around 40W/m^2 in Fig. 1, appear to be just barely capable of the job of producing 1 TWa over the next 40 years from the point of view of land resources required.

We would ask policy makers to: (1) start focusing much more funding on renewable energy systems that can scale to no less than the terawatt level when land and other material resources are considered as limiting factors; (2) to begin the process of allocating 1%-2% percent of strategically located land area in the continental USA for the exclusive use of utility-scale renewable power plants; and (3) to have an honest debate, to include the general population, on what we are willing to give up in land area in order to become an energy independent nation---that number better be at least 1% of the continental USA or renewable energy will remain forever out of reach.

For more information feel free to contact us here.