|Fig. 1: Insolation map of the U.S., after . Yellower color means sunnier.|
This paper will look at how the United States Of America could provide itself with the raw energy that it consumes, were supplies of Coal, Natural Gas, Oil, and Nuclear Fissile Material to disappear entirely. The reason for this discussion is because the supplies of these materials are finite on the earth due to the fact that they are either not created on earth, as in the case of radioactive material, or take a very long time to be created from decomposed organic material, as in the case of coal, natural gas, and oil. The problem with using these fuels is that when they are used to produce energy, the raw material that goes into the process (ie the coal, oil, or Uranium) becomes largely useless after it has been used to produce energy, and becomes an unusable waste product. However, were sources of energy such as the sun or wind to be used, there would be no worry of source depletion since that energy continues to be incident upon the earth at a constant rate, whether or not humans capture it for use as electricity.
I will not concern myself here with how long it will take for these sources of fuel to run out, but they will run out eventually, given any nonzero rate of consumption. I also will not concern myself with expected rates of consumption on future dates, or expected improvements in current technologies. Lastly, I will not discuss the environmental impact of any of these technologies. What I will look at is the current energy budget of the United States as of 2009, meaning the total quantity of energy, in Joules, used by the United States over the course of the year 2009. I will then look at some of the most promising current technologies for electricity production (as of 2010), that do not rely on the consumption of non-renewable fissile material. From there, I will look into what scale of these technologies would be needed in order to provide the United States with the amount of energy that it currently uses, without the use of nuclear fissile material or hydrocarbons. Lastly, I will discuss some of the difficulties in incorporating those technologies with our current infrastructure and usage patterns.
|Table 1: 2009 US Energy Consumption|
The United States energy consumption for 2009 is shown in Table 1. The numbers for the large, traditional energy sources in the top half of the table are taken from the 2010 BP Statistical Review Of World Energy.  They are quoted in units of million tons of oil equivalent (Mtoe). The survey states that 1 tonne of oil equivalent is roughly equal to 42 gigaJoules of energy, so one Mtoe is just 4.2 × 1016 Joules. The number for the newer "renewable" sources are from the US Department Of Energy.  These numbers have larger error bars on them (especially geothermal and biomass) since it is not always made clear what technologies are defined in each category. From this table we can see that hydro-electricity, solar, and wind only make up a small percentage (about 8%) of the total energy usage of the United States. The other 92% of energy comes from fuels which will at some point run out. Already we can see that going without these energy sources will be a tall order!
On of the main reasons the United States (and the rest of the world) relies on these sources so predominantly now is that people are good at using them; we've been using these sources for a long time, so the coal, oil, and natural gas we use are all relatively cheap to get out of the ground, and the engines that burn them are fairly efficient. But there is another very important reason the world uses these materials, one that isn't as obvious- that we can (relatively) easily control the rate at which we produce electricity with these fuels, with the exception of nuclear energy.
The way the United States' electricity needs are largely met is by providing a certain baseline amount of energy with fairly constant supplies of nuclear energy (nuclear energy is hard to turn on and off quickly) and by providing most other electricity by burning a combination of coal, oil, and natural gas. This is largely because energy usage fluctuates wildly over the course of a day, and it is fairly easy to turn turbines on and off at relatively short notice. That way, when more energy is needed, turbines can be turned on, and when the usage demands drop back down, those same turbines can be easily turned off. As I will soon discuss, that's not so easy to do with energy sources like solar and wind power, which come and go as they please, and often have peak production patterns which run opposite to usage patterns.
Because, for the sake of this paper, I am not not allowing the United States to rely on coal, oil, natural gas, or nuclear fissile material, we must instead take stock of available alternative methods of electricity generation. The technologies I will discuss here are technologies that can be used for the remaining lifetime of the earth because of where they draw their energy. I will specifically look at wind, solar, geothermal, and the burning of hydrocarbons such as wood and dung. First though, I would like to dispel a common myth regarding these technologies: that they are completely renewable, and therefore unlimited. These technologies all operate by converting energy that is currently abundant on the earth and which will never stop being abundant. However, this requires a physical object be made to affect this conversion. Accordingly, if the material that goes into the production of each of these units is not 100% recycled and re-used upon their being decommissioned, these technologies are also not truly "renewable." This is because they would still be using up finite resources (in the form of metal, concrete, or silicon) and discarding them as waste at the end of their energy-producing lifetimes. While the energy that these technologies convert into electricity is abundant, free, and essentially infinite, the physical objects required to do these operations use up finite resources as hungrily as does the burning of coal, so these technologies are only as "renewable" as their production and recycling processes.
The first technology I will look at is solar energy as manifested by electricity derived mostly from photovoltaic cells and thermal concentrators, one of the most heavily touted sources of "renewable" energy. I would first like to make a note that this method solar energy production is "renewable" in the sense that the sun itself will never keep assailing the earth with energy in the form of photons for many more years, and that when it does stop, we will have bigger problems than fulfilling our energy budget. Looking at Figure 1 below, we see that the United States receives an average of about 4KWh/m2/day over its entire surface; from this we can calculate the total amount of solar energy incident upon the United States every year.  The United Nations lists the United State's surface area as 9.6 × 106 Km2 , so doing some math we see that the total amount of energy coming from the sun and striking the United States in 1 year is roughly 5.0 × 1022 Joules.  As we can easily see, this is about 500 times as much as the United State's total energy budget for 2009, which was 9.5 × 1019 J; so if just 1% of the entire United States could be covered in solar panels, there'd theoretically be more than enough electricity to fulfill demand.
Despite this rosy picture of nearly boundless energy, one of the main downsides to using solar energy is that the amount of energy produced is proportional to the intensity of the sunlight, so if it's a cloudy day (or night time) solar arrays won't be producing as much energy as at peak sunlight hours. What this really means though, is that regions that rely on air conditioning in the hot summer months will be at peak usage during peak production times (during the sunniest parts of hot days), which is good. But colder regions (like the Midwest and East Coast) have opposite usage patterns since their biggest draw comes from heating during the cold winter months, when there isn't much sun.
Secondly, I will look at electric energy produced by wind turbines, another favorite form of "renewable" energy because the winds will presumably never stop blowing as long as there is life on the earth. One of the main cited downsides of wind energy is that it too comes on and off when it pleases, though that issue is softened by the fact that the wind generally dies slowly, making it relatively easy for grid operators to shift resources around. According the the US's National Renewable Energy Lab (NREL), 3.7 × 107 GWh (1.3 × 1020 J) of annual potential exists in the United States.  This number is from a report that only takes into account regions with a certain minimum wind speed and assumes 5MW/Km2 of installed space. While this number most likely has large error bars, and was made by a group that supports wind energy, the total production capacity cited by them is on the order of 1020 J, which is roughly equal to the US energy budget for 2009. This suggests that wind energy likely won't be a feasible sole means of energy production, but it definitely has room for growth beyond its current 0.1% share of production, and it is slated to grow significantly, if current the trends outlined in Fig. 2 are any indication.
|Fig. 2: US Yearly Installed Wind Power Capacity. Data from eia.doe.gov .|
Taking a look at geothermal energy, we see that 3.5 × 1017 J were produced in 2008  from a slated mean capacity of roughly 9,057 MW.  Looking at untapped resources, we see an estimated 540,000 MW of additional potential capacity.  Extrapolating from that, we see that geothermal has a potential energy production of roughly 2.0 × 1019 Joules per year, about a fifth of current usage. However, that doesn't take into account small-scale thermal heating used to warm buildings, which is regrettably poorly documented, but which would further increase geothermal's potential impact. Geothermal also has the nice property that it's very constant, which means that it could provide a nice baseline energy that runs year round.
Lastly, looking at biomass we see the form of energy production most similar to current combustible fuels. The key difference though is that biomass can be completely renewable, and have no impact on foodsources if it comes from:
The key here again is that the biomass must be harvested at sustainable rates, meaning that the harvest rate and production rate must be equal, or it is again not renewable. One issue with biomass however, is that it is also used heavily to produce fertilizers and paper, so it is difficult to greatly expand its use without risk of unsustainable consumption, or causing ripples in the food production industry. There is however potential for methane harvesting from kitchen compost and organic scraps that otherwise go to landfills, as well as human fecal matter. Estimates say that 5 to 7 million tons of sludge (human fecal matter) were created in the United States in 2002 , and that the dry biomass has an energy density of 10,000BTU/lb.  Conservatively guessing that dry biomass is about half the density of wet biomass, we get about
While this number is by no means accurate, even if I was off by a factor of 10, this value is still nowhere near making a significant contribution to the United States' energy budget of ~1020 Joules.
Now that I have shown the energy-producing potential of solar, wind, geothermal, and biomass energy, we can start to look into how they could fulfill the raw amount of energy required by the United States. However, we also need to figure out how to integrate that power so that it is always readily available at all hours of the day, every day of the year. This issue is compounded by the problem that large parts of the country are at peak energy usage when solar is at its minimum production. One advantage of the current grid is that it relies heavily on the use of combustibles (ie coal and natural gas) that can be switched on with short notice. Fortunately, there are a two methods of storing energy that have been proven effective: hydropump storage (pushing dam-loads of water up and down hills), and compressing and storing air underground.
First looking at compressed air, I will take a look at the United States' only compressed air energy storage facility, in McIntosh, Alabama. This 110MW facility's "19 million cubic feet [of air] is stored at pressures up to 1080 psi in a salt cavern up to 2500 feet deep and can provide full power output for 26 hours."  That translates to
of total energy storage capacity. However, there is only one of these plants in the United States, largely because it is more profitable to store compressed natural gas in them instead. However, if the United States' 4364 billion cubic feet of natural gas storage (in 2010) were all converted to compressed air, that would yield us with 2.3 × 1018 Joules of potential storage. At 2.4% of yearly usage (about 9 days' worth), that's not a small amount of storage capacity to be able to contain at any given time, especially since there is then the option of re-charging these cites when usage once more drops below production. While this could easily be used to provide electricity when consumption increases during peak hours on an hourly timescale (assuming it was recharged during low-usage night hours), this would still not be enough energy to keep the coldest parts of the country lit and warm during the winter months, in the absence of any other power being produced. Unfortunately, potential for expansion of these compressed air tanks is fairly limited since they are generally made from depleted aquifers and hydrocarbon wells, and those aren't things that can be easily produced on demand.
Next I will look to hydropump storage systems, which push water up a large hill when there is excess electricity, and release it back down hill when it is needed. In 2009 the US had 21.5 GW of hydropump storage capacity,  but unfortunately there is no time scale listed for that 21.5 GW, so there is no way to know how much energy is represented. However, we can see how this stacks up as a percentage of average power use. The total energy use for 2009 was 9.5 × 1019 J, which over the year averages to 3.0 × 1012 Watts, or 3000 GW. From this we see that our 21.5 GW only comprises 0.7% of the average power consumed, so we can't expect stored hydropower to foot the energy bill when all else fails. The US also seems to have no plans for expanding its hydropump storage capacity anytime in the foreseeable future, suggesting that it's probably very expensive to install, and that there probably aren't very many viable cites to expand capacity to. 
It would seem that it is possible for the United States to provide itself with power without the use of any non-renewable combustible fuels, largely though the use of geothermal, solar, and wind power for baseline needs, and using compressed air and hydropump storage to take into account fluctuations on a daily basis. For the issue of what to do when solar production gets weak in colder regions during winter, it is likely that solar energy would have to be transported via transmission lines from sunnier, more temperate regions, which experience lower than average energy usage during those times. However, winter usually brings strong winds, and so wind farms should be able to support much of the baseline energy needs, if managed properly. This suggests that one solution isn't going to suit the entire United States, and that distinct approaches will have to be taken on the regional level to best match usage with production. It also may turn out that the entire United States may have to integrate to move electric power over longer distances than it currently does.
There are some specific issues that I did not discuss - transportation fuel for vehicles, cost, mass production feasibility, fluctuation numbers, etc - that factor heavily into feasability. However, the goal of this paper was to ascertain whether or not it would be physically possible to produce the energy consumed by the United States currently in the absence of all non-renewable combustible fuels. As shown, that quantity of energy can be realistically produced through wind power, can theoretically be produced many times over by solar power, can be chipped away at by large-scale geothermal (~20%), can be contributed to by household geothermal for heating purposes (contribution unknown), and will get roughly 6% from current (difficultly changed) hydropower and biomass usage. None of that takes into account new technologies or improvements in current technology (with the exception of geothermal). Secondly I have shown that daily power fluctuations can feasibly be buffered through the use of compressed air storage with some additional capacity coming from hydropump storage.
These new energy sources cannot be plugged directly into the current energy grid without some additional storage infrastructure and clever source choices, but it seems that it would be possible to do, both from a net production standpoint and a usage fluctuation standpoint.
 "Statistical Review of World Energy 2010," British Petroleum.
 "U.S. Installed Wind Capacity and Wind Project Locations," U.S. Department of Energy, 5 Mar 10.
 "US Consumption By Energy Source," U.S. Energy Information Agency, August 2010.
 "Population by Sex, Rate of Population Increase, Surface Area and Density," Table 3 in Demographic Yearbook 2005, United Nations Statistics Division.
 "Concentrated Solar Power Resource Potential," U.S. Energy Information Administration, 29 Apr 03.
 "Estimates of Windy Land Area and Wind Energy Potential by State," U.S. National Renewable Energy Laboratory, 4 Feb 10.
 C. F. Williams et al., "Assessment of Moderate- and High-Temperature Geothermal Resources of the United States," U.S. Geological Survey Fact sheet 2008-3082, 2008
 D. Tilman et al., "Beneficial Biofuels - the Food, Energy and Environment Trilemma," Science 235, 270 (2009).
 J. A. Oleszkiewicz and D. S. Mavinic, "Wastewater Biosolids: An Overview of Processing, Treatment and management," J. Env. Eng. Sci. 1, 75 (2002).
 "Biosolids Success Stories," Water Environment Federation.
 J. Gardner and T. Haynes, "Overview of Compressed Air Energy Storage," ER-07-001, Boise State Office of Energy Research.
 "An Updated Annual Energy Outlook 2009," U.S. Energy Information Administration," SR/OIAF/2009-03, April 2009.