by C. Weiner* (4/14)
- As of 2013, 61% of Colorado’s electricity was generated from coal, 26% from natural gas, 10% from non-hydro renewables, and 3% from hydropower.
- Power plants can be characterized by how they are used to meet the constantly changing demand for electricity.
- Plants that are least expensive to run are usually operated first, followed in order by plants that are more expensive to operate.
- The costs of running power plants typically do not include ‘externalities’, such as greenhouse gas emissions, air quality impacts, land impacts, and water consumption.
It is easy to take for granted light that comes on at the flip of a switch because in general Colorado and the rest of the United States have extremely reliable electricity systems. But electricity generation at the scale demanded by homes, businesses, and industries requires vast resources and highly sophisticated infrastructure. This fact sheet will identify fuel sources for electricity generation, operating characteristics of different fuel sources, costs and benefits of these fuel sources, and other considerations for electricity generation in Colorado and in general.
Data source: International Energy Agency
Data source: U.S. Energy Information Administration
Data source: U.S. Energy Information Administration
What fuels the current electric system?
The world consumed over 18,000 terawatt-hours of electricity in 2010, with the U.S. alone consuming over 20% of that total (U.S. Energy Information Administration). The electricity consumed worldwide is enough to power over 2.2 billion average Colorado homes. As shown below, a mix of coal, natural gas, renewable, and nuclear energy make up the dominant share of fuel sources used to generate this vast amount of electricity.
Despite the tremendous size of the world’s current electricity system, global electricity generation is expected to almost double from 2010 levels by 2040, driven largely by increasing standards of living in developing countries and by the fact that almost 20% of the world’s population in 2010 did not have access to electricity (International Energy Agency).
In the United States, similar percentages f fuel sources are used to generate electricity, with coal at 37%, natural gas at 30%, nuclear at 19%, renewables at 12% (including hydropower at 7%), and other sources at 2%. Colorado is much more coal-dependent than the country as a whole, although that reliance is decreasing in favor of increasing supplies of natural gas and non-hydro renewables.
Despite significant growth in rooftop and other distributed solar electricity systems, growth in wind energy systems has been unparalleled among renewables because of the relatively low cost of electricity from utility-scale wind farms on Colorado’s eastern plains. This shadows a national trend in which wind energy was the single largest source of new electric generating capacity added in 2012, with natural gas following and solar electricity gaining. To understand why our current electric system is fueled the way it is and how decisions about using different fuel sources are made, it is important to first understand how power plants operate.
How are electric generating stations operated?
Coal can be used to create electricity when it is pulverized and burned to boil water into steam. The steam is used to spin a turbine and the turbine shaft spins a generator to generate electricity. Steam generators are also known as ‘thermal plants’. Natural gas and nuclear plants can also make steam to create electricity. However, in many natural gas plants the combustion gases are utilized directly to spin a turbine and generator. This is referred to as a simple cycle or ‘combustion’ gas turbine. If after passing through the combustion turbine the exhaust gases are used in a second stage to produce steam, this is called a combined cycle. The average efficiency of a steam generator (thermal plant) is about 33%, the average efficiency of a natural gas combustion turbine (simple cycle) is about 30%, and the average efficiency of a natural gas combined cycle plant is about 45%.
Plant efficiency alone is not widely valued by utility managers. Generation must ‘follow’ the load (or demand): as load increases or decreases across the grid, generation for that grid must be increased or decreased to match load at any given time (as well as to provide the appropriate amount of reserves). Power plant units can be characterized by how they are used to meet this ever-changing demand.
A ‘baseload’ generating unit is run fairly steadily throughout a day and throughout the year. They are dispatchable, which means that they can be controlled to run at specified levels over specified time periods. Together, baseload units can provide a significant portion of overall generating resources on a grid at a given time. They are often steam generators that cannot ramp up or down quickly and that operate most efficiently when run near capacity. Coal, nuclear, geothermal, natural gas, and hydro are all potential baseload resources.
‘Cycling’ or load following units are dispatchable but are ramped up or down throughout a day to meet changing demand. Hydropower and combined cycle gas plants are commonly utilized as load following resources. ‘Peaking’ units are a type of load following unit that are utilized primarily to meet peak demand on a given day. Although they are less efficient than thermal plants, gas combustion turbines can be started or stopped relatively quickly, can be ramped between load levels quickly, and can operate at a wider range of power outputs than thermal plants. Combustion turbines are therefore most often used as peaking units. To some degree, solar can be used a summer peaking units when the peak is due to increased demand for air conditioning on sunny days.
Because solar and wind are not dispatchable–they cannot be controlled to provide a specified output at any given time – and because their fuel is free, these resources are used to generate electricity whenever they are available. Load following and sometimes even baseload resources must therefore not only be managed to be responsive to changing demand, but also to changing supplies of wind and solar.
In addition to the responsiveness of different types of power plants to changing supply and demand, cost is also a foremost consideration for those operating the electric grid. Plants that are least expensive to run are typically operated first to meet the demand for electricity at any given time, followed in order by plants that are more expensive to operate. Historically, this has been the reason behind widespread use of coal, but decreasing natural gas prices have resulted in greater use of gas and decreased use of coal to meet demand.
Hypothetical Electricity Load Profile on Western Grid (NREL)
The graph from the National Renewable Energy Laboratory’s (NREL) Western Wind and Solar Integration Phase 2 Study shows a simulated mix of fuel sources used to meet demand during a week in March in the American West. As you can see, baseload resources like nuclear and coal fluctuated very little, hydropower was also fairly steady but was somewhat responsive to changing demand and the availability of wind/solar, gas combined cycle (CC) plants swung wildly according to changes in demand and wind/solar availability, gas combustion turbines (CT) were reserved primarily for peak demand, and wind and solar resources were dispatched whenever they were available.
How much do different power plants cost to build and operate?
Price is a foremost consideration for power providers both when running power plants and when considering what new electric generating plants to build in the future. One simplified way of comparing the costs of different fuel sources is to use a concept called the ‘levelized cost of energy’ (LCOE). LCOE accounts for both the capital expenses to build a new plant and provide transmission as well as the lifetime operating expenses of a plant. It is represented as ‘cost per kilowatt hour’. U.S. Energy Information Administration estimates of LCOE in 2018 (due to the time it takes to build new plants) for selected technologies are shown below.
It must be noted that these estimates are nationwide, include a 3% increase in the cost of capital for carbon-intensive technologies to reflect the possibility of a carbon tax or penalty, are based on utility-scale developments, do not account for applicable tax credits or other financial incentives, and assume various capacity factors. For example, it is assumed that solar photovoltaic plants will generate maximum output 25% of the time, combined cycle natural gas plants will run at maximum output 87% of the time, and peaking ‘advanced combustion’ natural gas plants would only run 30% of the time. In reality, these capacity factors will vary according to the weather and climate of the region in which the plant is built, site and time specific costs, and the cost of dispatching other existing fuel sources to generate electricity.
This last point is particularly important since the actual cost of building and running a new power plant must be weighed against not only the cost of building and operating another type of plant but also against the cost to operate existing plants. While adding solar may result in ramping down expensive natural gas peaking plants, adding a new baseload natural gas plant may result in ramping down less expensive coal. In another example, decreasing demand for electricity through energy efficiency incentives is often seen as a low cost way of meeting electricity demand – cheaper than adding new power plants. Power providers base resource decisions on both the LCOE of new plants as well as the avoided cost associated with using less fuel from existing plants.
What are costs of different fuel sources not captured by LCOE?
Although LCOE is a simplified indicator of the price consumers will pay for generating electricity from different fuel sources at a given point in time, it does not account for ‘externalities’. An externality is a consequence of an economic activity that is not reflected in the price of that activity. Negative externalities related to electricity generation include air pollution and land use.
It is important to look at the entire life cycle of a fuel source or generation plant – from extraction to operation to plant retirement – when considering externalities. The National Renewable Energy Laboratory has taken results from thousands of life cycle assessments (LCAs) of various energy technologies to come up with reliable greenhouse gas (GHG) emission estimates for these technologies over their entire useful lives.
|Life Cycle Greenhouse Gas Emissions Estimates for Various Energy Technologies (NREL)|
|Technology||Grams of CO2e
|Concentrating Solar Power||13-46|
It is clear from these estimates that ‘clean energy technologies’ are in fact much cleaner than coal when considering GHGs even over the entire lifecycle of the technologies. For wind and solar, most of the GHG emissions are associated with raw material extraction, materials processing, and manufacturing. For coal plants the majority of GHG emissions are associated with operations (i.e. burning coal to generate electricity). Other technologies like natural gas, hydropower, geothermal, and bioenergy require further scrutiny before more definitive estimates of lifecycle GHGs can be made.
An LCA of natural gas is of particular interest at this time since prices are low and its market share is growing. Although it is accepted that burning natural gas emits approximately half of the GHGs as burning coal, so-called ‘fugitive emissions’ – leaks – of raw natural gas (mostly methane) during extraction and/or distribution may result in lifecycle GHGs even greater than coal-based electricity. Research is currently underway in this area.
Beyond GHGs, air pollution from electric generation plants is an externality that can be harmful to human health. Burning coal and natural gas for electricity generation emits sulfur dioxide, nitrogen oxides, and particulates. Sulfur dioxide contributes to acid rain and respiratory illness, nitrogen oxides contribute to smog and respiratory illness, and particulates contribute to smog, haze, respiratory illness, and lung disease. Burning coal also emits mercury and other heavy metals. Although airborne mercury concentrations are low and of little concern, mercury entering water – either directly or through the air – can accumulate in fish and the animals and humans that eat fish.
|2018 Levilized Costs of Energy (U.S. EIA)|
|Technology||Levilized Capital Costs
|Operation and Maintenance Costs
|2018 Projected Total LCOE
|Natural Gas – conventional combined cyle||$0.017||$0.050||$0.067|
|Coal – conventional combustion||$0.067||$0.033||$0.100|
|Natural Gas – advanced||$0.034||$0.071||$0.105|
|Coal – advanced||$0.086||$0.038||$0.123|
The average emission rates in the United States from coal-fired generation are: 2,249 lbs/MWh of carbon dioxide, 13 lbs/MWh of sulfur dioxide, and 6 lbs/MWh of nitrogen oxides. Compared to the average air emissions from coal-fired generation, natural gas produces half as much carbon dioxide, less than a third as much nitrogen oxides, and one percent as much sulfur oxides at the power plant (U.S. EPA). Overall, emissions of SO2 and NOx from the electric power sector have decreased dramatically from almost 16 MST (million short tons) of SO2 and over 6 MST of NOx in 1990 to under 4 MST of SO2 and under 2 MST of NOx by 2012. This has been the result of 1990 amendments to the Clean Air Act which established a cap-and-trade program for SO2 and controls on NOx, federal rules, state regulations, and a recent shift toward natural gas from coal generation (U.S. Energy Information Administration).
Every utility-scale electric generating technology alters the land. Of course all technologies require materials for their construction – whether plants are made of concrete, steel, or other. In addition, wind farms require land for the turbines, for access when developing the site, and for mining of essential turbine components. Coal and nuclear require land for mining, land and often a water source for power plants, and often land for disposal of waste. Large-scale hydropower alters the land around the waterway when dams are constructed. Solar arrays take up land, and mining is required to extract minerals needed for the solar panels. Natural gas electricity generation requires land for drilling wells, for pipelines, and for the power plants. The chart below from Kammen (2011) shows median land use (in square meters) per gigawatt-hour of electricity generation for different energy technologies.
Although the cost for power providers to acquire and use water for plant operations is built into the LCOE, the fact remains that water use for electricity generation in the arid West is significant. Hydropower is generally the largest water consumer among electricity generating technologies at 5,300 L/MWh (based on California data), followed by coal and nuclear plants that recirculate water at over 3,000 L/MWh. ‘Once-through’ coal and nuclear plants consume closer to 1,100 L/MWh, while combined cycle natural gas plants vary from 380 to 680 L/MWh depending on the cooling technology. Wind, solar photovoltaics, and dry-cooled combined cycle natural gas plants consume 15 or fewer L/MWh (Kammen, 2011). For perspective, this means that a 500 MW coal plant that recirculates water would consume about 13.2 million liters (3.5 million gallons) of water per day whereas the same size dry-cooled combined cycle gas plant would consume 180,000 liters (47,550 gallons).
Note that recirculating cooling systems consume more water than once-through systems since the water loop in recirculating systems is exposed to the air (in a cooling tower) before completing its cycle and some evaporation takes place when exposed. Although once-through systems initially withdraw more water than recirculating systems, they end up consuming less. Most thermal plants in the West utilize recirculating cooling systems. Also note that dry-cooled systems that use air as the cooling medium do not operate as efficiently as wet-cooled systems, so although they use less water they require more fuel per level of electricity output.
Taken altogether, many of the negative environmental and human health impacts over the lifecycle of various utility-scale energy technologies can be summed up in the following table. (Note that descriptions of ‘low’, ‘moderate’, and ‘high’ for each technology are relative to the other technologies in the list.)
What are benefits of different fuel sources not captured by LCOE?
Different electric generating technologies also have a number of benefits not reflected in the LCOE. These include grid reliability, job creation, and long-term price stability.
Land Use Associated with Energy Technologies. ‘PV Multi Si’ stands for multicrystalline silicone solar photovoltaics. ‘CSP’ stands for concentrating solar power. ‘CPV’ stands for concentrating photovoltaics. Although onshore wind is shown to take up a relatively large amount of land per unit of energy generated, the land around the turbines is still useable by agricultural producers, residents, and many species of wildlife.
Fossil fuel, hydropower, and nuclear plants are much less variable than wind and solar resources. Integrating large amounts of wind and solar into the grid while delivering uninterruptable service to electric customers is a challenge. In a recent study of integrating wind and solar into the Western U.S. electricity grid, the National Renewable Energy Laboratory (2010) found that integrating a 35% penetration of these renewables could be done without significant changes to infrastructure by: increasing the size of the area from which solar and wind could be drawn when balancing electricity supply with demand; decreasing grid scheduling time intervals; incorporating better forecasts of wind and solar into scheduling; and other actions. Above a certain level of penetration by these more variable resources, however, our grid would simply not function as reliably unless there was large-scale storage for wind and solar electricity. And LCOE does not account for any power required to meet electricity demand when intermittent resources are not available.
|Negative Externalities of Energy Technologiesaccount|
|Technology||GHG Emissions||Air Quality Impacts||Land Impacts
(based soley on land area)
|Consumptive Water Use|
|Natural Gas||Moderate – High
|Moderate||Moderate||Moderate – Low|
|Nuclear||Low||Low||Low – Moderate||High|
|Solar PV||Low||Low||Low – Moderate||Low|
Another benefit associated with electricity generation is job creation. A comprehensive study conducted by Wei et. al (2010) summarized the range of jobs generated by different energy technologies from previous studies. As shown, the average of the range for job-years per gigawatt-hour from solar PV, energy efficiency, and solar thermal were highest, with coal and natural gas the lowest. It should be noted that these ranges do not Technologiesaccount for job losses between sectors, as could be the case when replacing large amounts of coal with natural gas or renewable energy for example. The ranges also include both utility-scale projects and small-scale distributed generation projects. It is also possible that jobs per gigawatt-hour will decrease as installers gain more experience with a given technology.
The graph below shows the growth in job-years over a business-as-usual (BAU) scenario accounting for both improvements in energy efficiency and various levels of federal Renewable Portfolio Standards (RPSs), in which targets are set and achieved for renewable electricity as percentages of overall electricity generation by 2030. This graph accounts not only for job gains due to increasing use of energy efficiency and renewable energy, but also accounts for job losses from fossil fuel sectors as a result. It does not account, however, for the location and concentration of those jobs. For example, the manufacture of solar PV panels may take place in China, whereas construction of a new natural gas plant may utilize mostly local American workers.
Jobs Growth Associated with Renewable Portfolio Standards (Wei et. al)
A third significant benefit associated with certain fuel sources is long-term price stability. It is important to electric utilities and their consumers that prices for electricity don’t increase substantially from one year to the next. Although LCOE does account for estimates of changing fuel prices over time, it does not capture this inherent value of price stability. Free fuel in the form of wind, water, and solar energy provide a more stable price than fuel such as coal, natural gas, or nuclear which is subject to swings in commodity prices due to numerous variables. The threat of a potential carbon tax also makes clean energy technologies appear much more stable in terms of price than carbon-emitting fuels. And although government subsidies provided to renewables like wind and solar have tended to be more volatile than subsidies related to traditional energy technologies, this uncertainty affects a power provider’s decision to invest in certain fuel sources more than it affects utilities and consumers already buying power from a constructed plant. All power plants are subject to maintenance costs which are fairly predictable, especially for mature technologies.
|Average Jobs Associated with Energy Technologies|
per Gigawatt- Hour
Taken altogether, many of the positive grid reliability, job, and price stability impacts of various energy technologies can be summed up in the following table. (Note that descriptions of ‘low’, ‘moderate’, and ‘high’ for each technology are relative to the other technologies in the list.)
It should also be noted that energy development can have significant macroeconomic impact aside from job creation in forms such as severance taxes, property taxes, and lease and royalty payments. While the oil and gas industry is by far the largest source of energy-related public revenues in Colorado, an apples-to-apples comparison of energy sources according to their macroeconomic impacts by unit of energy produced or delivered is not available at this time.
Between the operating characteristics, LCOE, and positive and negative externalities described above, one can get a sense of the complex set of factors that go into decision-making in the electric utility sector. Additional considerations include the long-term reality that supplies of non-renewable resources like coal and natural gas will diminish and prices will increase as a result. Modern estimates of U.S. coal supply using current rates of consumption are around 200 years. Modern estimates of U.S. natural gas are perhaps harder to pin down because of a recent boom in domestic shale gas production but are likely similar. The long-term supply of certain rare earth minerals used in wind and solar electricity generation must also be considered, but research is underway to find alternatives.
The attractiveness of different fuel sources is also deeply affected by laws and regulations governing the energy sector. Both federal and state RPSs and other policies can be significant drivers of a shift toward renewable electricity. State and federal regulatory bodies like state Public Utilities Commissions and the Federal Energy Regulatory Commission write regulations and rules based on legislation that can also drive shifts toward one fuel source or another. In addition, the ownership structure of an electric utility can determine how much that utility is subject to both federal and state regulation. Typically, investor-owned, private utilities are regulated monopolies subject to state supervision while state regulation over rural electric cooperatives and municipal utilities is much more limited. Utility ownership types can also affect fuel source decisions in that investor-owned utilities must provide a profitable return to shareholders while cooperatives and municipal utilities are primarily beholden to their members or citizens.
|Positive Attributes of Energy Technologies|
|Technology||Grid Reliability||Average Job-Years/GWh||Price Stability|
|Hydropower||High||Moderate (small hydro)||High|
The world currently consumes an enormous amount of electricity and that amount is expected to grow further as demand increases. Meeting demand for electricity in a responsible and cost-effective manner requires highly sophisticated analyses of different variables including a utility’s current fuel source mix, current and anticipated electric load profiles, the cost of adding new capacity, consideration of externalities, current and anticipated law and regulation, and meeting the expectations of utility ownership. (Taken together, accounting for these variables is referred to as ‘integrated resource planning’.)
Different fuel sources offer different advantages and disadvantages, and no one (or even two) fuel sources are ‘silver bullets’ for the electric power industry. Any deep transformation of the electric power sector in Colorado and elsewhere will require a tremendous amount of creativity, foresight, focus, and partnerships as well as continued research and development.
Fthenakis, V. and Kim H. C. (2009). ‘Land Use and Electricity Generation: A life-cycle analysis. Renewable and Sustainable Energy Reviews 13 (6-7), 649-668.
International Energy Agency (2012). Key World Energy Statistics 2012. Retrieved from www.iea.org/publications/freepublications/publication/kwes.pdf on January 16 , 2014.
Kammen, D. (2011) ‘An Assessment of the Environmental Impacts of Concentrator Photovoltaics and Modeling of Concentrator Photovoltaic Deployment Using the SWITCH Model’. Retrieved from https://rael.berkeley.edu/sites/default/files/Kammen%20et%20al%202011-%20EIA%20of%20CPV.pdf on April 9, 2014.
National Renewable Energy Laboratory (2010). The Western Wind and Solar Integration Study Phase 2. Retrieved from www.nrel.gov/docs/fy13osti/55588.pdf on January 16, 2014.
National Renewable Energy Laboratory (2013). Life Cycle Assessment Harmonization. Retrieved from www.nrel.gov/analysis/sustain_lcah.html on January 16, 2014.
U.S. Energy Information Administration (2013). Annual Energy Outlook 2013. Retrieved from www.eia.gov/forecasts/aeo/pdf/0383%282013%29.pdf on January 16, 2014.
U.S. Energy Information Administration (2014). Power Plant Emissions of Sulfur Dioxide and Nitrogen Oxides Continue to Decline in 2012. Retrieved from www.eia.gov/todayinenergy/detail.cfm?id=10151 on February 17, 2014.
U.S. Environmental Protection Agency (2013). Air Emissions. Retrieved from www.epa.gov/cleanenergy/energy-and-you/affect/air-emissions.html on February 17, 2014.
Wei, M., Patadia, S., and Kammen, D. (2010). ‘Putting Renewables and Energy Efficiency to Work: How many jobs can the clean energy industry generate in the U.S?’. Energy Policy 38, pp. 919-931.
*C. Weiner, Colorado State University, clean energy Extension specialist. (4/14)
Colorado State University, U.S. Department of Agriculture and Colorado counties cooperating. CSU Extension programs are available to all without discrimination. No endorsement of products mentioned is intended nor is criticism implied of products not mentioned.
Go to top of this page.