This wraps up the serialized edition of our residential Rooftop Revolution report. Read the whole report in short segments below, or download the report and other resources.
Part 1 – The Coming Solar Electricity Transformation
Part 2 – Solar Costs and Grid Prices On a Collision Course
Part 3 – The Solar Opportunity
Part 4 – Positive Policy Changes for Distributed Renewable Energy
Part 5 – Planning for Phasing Out Incentives
Top 42 Metropolitan Areas of the U.S., by Population
Although in the text we refer to the “Top 40 metropolitan areas,” in this analysis we actually used the top 42, because their cumulative population is just over half the national population. The list is at Wikipedia (see endnote).36
Year of Grid Parity by Three Measures
The following tables provide detail for the charts showing the cumulative population at grid parity in the major metropolitan areas, by our three measures (average grid prices, time-of-use prices, and economic grid parity). Metropolitan areas are listed next to the year they reach residential solar grid parity (through 2021), along with the cumulative U.S. population that represents.
The Cost of Solar
In general, the parameters of this grid parity analysis err on the side of conservative. For one, we largely used the average retail electricity price for comparison, despite the fact that many utilities could implement time-of-use prices that would better reflect the value of solar during period of peak electricity demand. In a few states, most prominently California, TOU pricing can increase the value of solar electricity from 5 to 200% or more.37 Furthermore, it’s possible to make a re- turn installing solar prior to grid parity, because the lifetime savings will still exceed the cost of the system.
Secondly, the estimated lifetime of a solar array for this analysis is 25 years, despite ample evidence that solar panels can still produce over 80% of their original annual output after that time period. If calculated over 30 or more years, the cost of solar is significantly lower.
The only issue hanging over our cost estimates is the potential demand for backup or energy storage to support a significant distributed solar market. While research suggests that dispersion of solar power production over a wide geographic area can smooth output variations from weather, solar power doesn’t work when the sun goes down. Until solar reaches the 15% threshold, however, few places in the electricity system will require additional backup or storage. As more re- newable energy has come onto the grid in recent years, utilities have shown an interest in backing down output from coal and natural gas power plants. As the latter have additional capacity, they can be used to firm up output from many distrib- uted solar arrays.
Other Studies of Rooftop Solar PV Potential
We aren’t the first to estimate the potential for rooftop solar in major U.S. cities. To validate our estimates of residential rooftop, we found five other studies of rooftop solar potential. The results of these studies are listed below along with our estimate. Where possible, we included the figure for residential rooftops (if so, the study result is in italics). See below for more detail.
San Diego County
Anders, Scott, et al. Potential for Renewable Energy in the San Diego Region. (San Diego Regional Renewable Energy Group, August 2005). Accessed 12/7/11 at http://tinyurl.com/7k9qby6.
- Population of 3 million
- Residential rooftop solar potential of 2772 MW (AC), 6310 GWh per year
- Commercial rooftop solar potential of 1624 MW (AC), 3263 GWh per year
- Total: 4400 MW (>100% of peak demand) for 9,600 GWh per year (50% of energy sales)
- SDG&E peak demand in 2004 was 4,065 MW with energy sales of 19,000 GWh
New York City
Navarro, Mireya. Mapping Sun’s Potential to Power New York. (New York Times, 6/16/11). Accessed 12/12/11 at http://tinyurl.com/7j2llmk.
- City population of 8 million
- Two-thirds of city rooftops could sport solar PV, generating 5850 MW. Half of peak demand, and 14% of year-round consumption.
San Francisco
Garbesi, Karina, PhD, and Emily Bartholomew. The Potential for Solar Electricity Generation in San Francisco. (Environmental Law and Justice Clinic of Golden Gate University Law School, 6/1/01). Accessed 12/12/11 at http://tinyurl.com/6svn2qs.38
- Population of 805,000
- Low-end estimate of 547 GWh per year (378 MW based on PVWatts estimate of 1446 kWh per kW of DC capacity)
- horizontal, not tilted panels
- Rooftop space found by estimating 35% of land covered by buildings (average city-wide), a total rooftop area of 36 million sq. meters, with 5-30% availability. Low end estimate was ~15%.
- City electricity consumption of ~6000 GWh per year, peak demand estimated at 1077 MW for 2012.
Seattle
Liddell, Ryan. Estimating Rooftop Solar Electricity Potential in Seattle from LiDAR Data. (Presentation to Northwest GIS Users Group, October 17 – 21, 2011). Accessed 12/12/11 at http://tinyurl.com/6uar3fl.
- 563,000 population
- 26 TWh per year potential
- Study assumes 30% of rooftop space is unusable. Change assumption to 80% unusable, and it’s 7.4 TWh potential (79% of production).
- City uses 9.4 million MWh per year (city utility, EIA), peak demand of 1845 MW
Austin
Wiese, Steve, et al. A Solar Rooftop Assessment for Austin. (American Solar Energy Society, 2010). Accessed 12/12/11 at http://tinyurl.com/6wg6tnw.
- 1 million people in service territory
- 536 million s.f. of rooftops
- 204 million s.f. suitable for PV
- 142 million s.f. estimated
- 2,446 MW (84% of existing power plant capacity) for 3.3 TWh per year (28% of existing energy use)
Solar Installation Capacity
While the converging cost of solar and electricity suggest a forthcoming explosion of solar power, competitive cost is just one of several hurdles that solar power must overcome. Sufficient labor and manufacturing must keep pace with exponentially increasing demand for solar power installations. The electricity system may have to adapt to over- come technical limitations. And various interest groups, including incumbent utilities, present political opposition to widespread democratization of the electricity system.
By 2018, 1 in 6 Americans will live in a major metropolitan area at solar grid parity, with a potential universe of 30,000 MW of rooftop solar on residences (sufficient capacity to supply over 2.7% of residential electricity needs).39 The following chart shows the growth in U.S. solar capacity from 2007- 10 and a best fit line through 2020. The trend line is based on the installation trend with the federal (state and local) solar subsidies in place.
Could installations in the U.S. accelerate beyond their current, torrid pace to meet this potential? Evidence from Germany suggests that they could.
From 2000-2010, the German market grew enormously, from annual installations in the tens of megawatts to over 7,000 megawatts per year. On a per capita basis, Germany installs as much solar per year (35,000 MW) as the U.S. would need in total (30,000 MW) to reach its residential solar grid parity potential in 2018.
References
- 36 – Wikipedia contributors. Table of United States Metropolitan Statistical Areas. (Wikipedia, The Free Encyclopedia. Wikipedia, The Free Encyclopedia, 2/10/12). Accessed 2/21/12 at http://tinyurl.com/7h3zheg. ↑
- 37 –Farrell, John. How Electricity Pricing Can Boost Distributed Solar – Part 1. (Institute for Local Self-Reliance Energy Self-Reliant States blog, 1/12/12). Accessed 1/13/12 at http://tinyurl.com/7tywc7c. ↑
- 38 –San Francisco Energy Information. (sfenergywatch.org, 2011). Accessed 12/12/11 at http://tinyurl.com/7z2mdwz. ↑
- 39 – We calculate that the average solar installation in the U.S. produces 1320 kWh AC per year per kW of DC capacity. Thus, 30,000 MW would produce ~39.6 billion kWh annually, 2.74% of residential electricity sales in 2010 (1.45 trillion kWh, as reported by the Energy Information Administration). ↑
