US rooftops could physically host 1,118 gigawatts of solar panels and generate 1,432 terawatt-hours of electricity a year, equal to 39 percent of national electricity sales at the time of measurement. That figure comes from the most detailed federal assessment of rooftop solar ever produced, and it treats every suitable roof in the country as one large, distributed power plant. This report compiles the verified numbers, ranks states by how much of their own electricity their roofs could supply, and sets the physical potential against what the US has actually built. Every statistic below is sourced. Figures the Roofing Brief derived from published data are labeled as calculations.
Executive Summary
- US rooftops have a technical potential of 1,118 GW of solar capacity and 1,432 TWh of annual generation, equal to 39 percent of national electricity sales in the study year. Source: NREL, 2016.
- Small buildings alone account for 731 GW and 926 TWh, about 65 percent of the national rooftop potential. Source: NREL, 2016.
- California has the highest offset potential: its rooftops could generate 74 percent of the electricity its utilities sold in 2013. Source: NREL, 2016.
- Wyoming has the lowest, at 14.2 percent, because it has the highest per-capita electricity use of any state at 30.3 MWh per year. Source: NREL, 2016.
- Cloudy, low-consumption states such as Maine and New York can offset more of their total electricity use than sunny, high-consumption states such as Arizona. Source: NREL, 2016.
- The US had installed 279 GWdc of solar of all types by year-end 2025, equal to about 25 percent of what rooftops alone could hold. Roofing Brief calculation from SEIA and NREL data.
- The federal estimate is a physical ceiling, not a forecast. It excludes cost, ownership, grid limits, and policy. Source: NREL, 2016.
Key Findings
- Rooftop solar technical potential in the US is 1,118 GW of capacity. Source: NREL, Rooftop Solar Photovoltaic Technical Potential in the United States, 2016.
- That capacity could generate 1,432 TWh of electricity per year. Source: NREL, 2016.
- 1,432 TWh equals 39 percent of US electricity sales in 2013, the study baseline year. Source: NREL, 2016.
- Small buildings of 5,000 square feet or less hold 731 GW of the potential and 926 TWh per year, about 65 percent of the total. Source: NREL, 2016.
- Medium and large buildings hold the remaining 386 GW and 506 TWh per year, about 35 percent of the total. Source: NREL, 2016.
- Only 26 percent of small-building rooftop area was found suitable for panels after removing roofs that were too shaded, too small, or wrongly oriented. Source: NREL, 2016.
- The analysis used airborne LiDAR and geographic information system methods across 128 US cities, representing roughly 23 percent of US buildings, then modeled the rest of the country statistically. Source: NREL, 2016.
- The 2016 estimate nearly doubled NREL’s earlier 2008 figure of 664 GW and 800 TWh, driven by better roof data and improved panel efficiency assumptions. Source: NREL, 2016.
- California could offset 74 percent of its 2013 utility electricity sales with rooftop solar, the highest of any state. Source: NREL, 2016.
- A cluster of New England states could each offset more than 45 percent of their electricity sales despite below-average sunlight, because per-capita consumption is low. Source: NREL, 2016.
- The three lowest-offset states are North Dakota at 24.6 percent, West Virginia at 22.9 percent, and Wyoming at 14.2 percent. Source: NREL, 2016.
- Wyoming ranks last mainly because its per-capita electricity use of 30.3 MWh per year is about 250 percent of the national average, driven by industrial demand. Source: NREL, 2016.
- Washington State, which has the lowest population-weighted solar resource in the continental US, could still offset about 27 percent of its electricity use from rooftops. Source: NREL, 2016.
- Roughly 3.3 million US homes are built or need roof replacement each year, representing about 30 GW of new rooftop solar opportunity annually. Source: US Department of Energy, citing NREL.
- The US installed 43.1 GWdc of solar of all types in 2025 and reached 279 GWdc of cumulative capacity, with solar making up 54 percent of new generating capacity that year. Source: SEIA and Wood Mackenzie, 2025 Year in Review.
What “Technical Potential” Means
Technical potential measures how much electricity rooftop solar could produce if every suitable roof were fully used, given current panel technology and the physical characteristics of the roofs. It is a ceiling, not a prediction. It does not account for installation cost, who owns the building, whether the grid could absorb the power, financing, permitting, or homeowner interest. NREL states this directly: the estimate is the upper bound of what is physically achievable, not what is economical or likely. Reading these numbers as a target rather than a limit is the most common error journalists and advocates make.
The National Numbers
The federal assessment found that after excluding unsuitable roofs, the US has enough usable roof area to host 1,118 GW of solar capacity generating 1,432 TWh per year. To put the capacity in context, that is more than the entire installed generating capacity of the US power system across all fuels at the time of the study. The generation figure, 1,432 TWh, equals 39 percent of the 2013 electricity that utilities sold nationwide.
| Metric | Value | Source |
|---|---|---|
| Rooftop PV capacity potential | 1,118 GW | NREL, 2016 |
| Annual generation potential | 1,432 TWh | NREL, 2016 |
| Share of 2013 national electricity sales | 39 percent | NREL, 2016 |
| Small-building share | 731 GW / 926 TWh (about 65 percent) | NREL, 2016 |
| Medium and large building share | 386 GW / 506 TWh (about 35 percent) | NREL, 2016 |
| Cities analyzed with LiDAR | 128 | NREL, 2016 |
| Prior 2008 estimate | 664 GW / 800 TWh | NREL, 2016 |
The building-size split matters for roofing. Nearly two-thirds of the national resource sits on homes and small structures, the same roofs that get torn off and replaced on a 15 to 30 year cycle. Any policy or product strategy that focuses only on large commercial and warehouse roofs addresses barely a third of the physical potential.
State Rankings: Offset Potential
The most useful way to rank states is not by raw generation but by how much of their own electricity use rooftops could cover. This offset percentage depends on two things: the quality of the solar resource, and how much electricity the state consumes. A state with modest sunlight but low consumption can rank higher than a sunny state with heavy industrial demand. That is why the ranking surprises people.
| State | Rooftop offset of state electricity sales | Note | Source |
|---|---|---|---|
| California | 74 percent | Highest offset of any state | NREL, 2016 |
| New England cluster | more than 45 percent | Low per-capita use offsets weaker sunlight | NREL, 2016 |
| Washington | about 27 percent | Lowest solar resource in continental US | NREL, 2016 |
| North Dakota | 24.6 percent | Among the three lowest | NREL, 2016 |
| West Virginia | 22.9 percent | Among the three lowest | NREL, 2016 |
| Wyoming | 14.2 percent | Lowest; highest per-capita use at 30.3 MWh/year | NREL, 2016 |
NREL notes explicitly that several below-average-sunlight states, including Minnesota, Maine, New York, and South Dakota, have similar or greater offset potential than higher-resource states such as Arizona, because consumption, not sunshine alone, drives the percentage. The full 50-state table is published in NREL’s data catalog for the study; this report cites the verified anchor points rather than reproducing unverified state figures.
Why Sunlight Is Not Destiny
The counterintuitive rankings come from simple arithmetic. Offset potential is roof generation divided by state consumption. Wyoming has good sun but the highest per-capita electricity use in the country, 30.3 MWh per year, roughly 250 percent of the national average, because of energy-intensive industry. That large denominator pushes its offset to 14.2 percent, the lowest in the nation. California combines a large stock of suitable residential roofs, strong sun, and moderate per-capita household use, producing the highest offset at 74 percent. The lesson for anyone modeling rooftop solar by region is that local electricity demand matters as much as the solar resource.
Potential Versus Reality
The physical ceiling is large. The built reality is far smaller. The US installed 43.1 GWdc of solar of all types in 2025, a 14 percent decline from 2024, and reached 279 GWdc of cumulative installed capacity across utility-scale, commercial, and residential segments combined. Residential rooftop solar was close to flat year over year, falling about 2 percent.
| Metric | Value | Source |
|---|---|---|
| US solar installed in 2025 (all types) | 43.1 GWdc | SEIA and Wood Mackenzie, 2025 |
| Cumulative US solar, year-end 2025 (all types) | 279 GWdc | SEIA and Wood Mackenzie, 2025 |
| Solar share of new US generating capacity, 2025 | 54 percent | SEIA and Wood Mackenzie, 2025 |
| Change in residential solar, 2025 | down about 2 percent | SEIA and Wood Mackenzie, 2025 |
| US retail electricity sales, 2024 | about 4,000 TWh | US EIA |
Original Synthesis: Roofing Brief Calculations
The following figures are derived by the Roofing Brief by combining public datasets. They are calculations, not published statistics. Inputs and limitations are stated for each.
1. All US solar equals about a quarter of rooftop capacity alone
Roofing Brief calculation: 279 GWdc / 1,118 GW = about 25 percent. Every solar panel the US has ever installed, across utility-scale fields, commercial roofs, and homes, adds up to roughly one quarter of what American rooftops alone could physically hold. Inputs: cumulative US solar capacity of 279 GWdc at year-end 2025 (SEIA and Wood Mackenzie); rooftop technical potential of 1,118 GW (NREL, 2016). Limitation: the 279 GWdc figure includes ground-mounted utility-scale solar, so the rooftop-only share of realized capacity is well below 25 percent. The two figures also use different rating conventions, so treat this as an upper bound on rooftop realization.
2. Rooftop potential still covers about a third of today’s larger grid
Roofing Brief calculation: 1,432 TWh / about 4,000 TWh = about 36 percent. NREL’s 1,432 TWh potential equaled 39 percent of 2013 electricity sales. Held constant against 2024 retail sales of about 4,000 TWh, the same potential now equals roughly 36 percent, because national electricity use has grown since the study year. Inputs: NREL rooftop generation potential of 1,432 TWh (2016); US retail electricity sales of about 4,000 TWh in 2024 (EIA). Limitation: the roof stock and panel efficiency have also grown since 2013, so real current potential is likely higher than 1,432 TWh; this calculation holds the 2016 estimate fixed and only updates the denominator.
3. Reroofing offers about 30 GW a year, and the US captures a fraction of it
Roofing Brief calculation: residential solar captured on the order of 15 percent of the annual reroof and new-build solar opportunity in 2025. The federal estimate puts the new-build and roof-replacement cycle at about 30 GW of rooftop solar opportunity per year across 3.3 million homes. SEIA reported residential quarterly installations near 1.1 GWdc, for example 1,106 MWdc in the first quarter of 2025 and 1,088 MWdc in the third quarter, implying roughly 4.4 GWdc for the full year. That is about 15 percent of the 30 GW annual opportunity. Inputs: 30 GW per year reroof and new-build opportunity (US Department of Energy, citing NREL); residential quarterly installs near 1.1 GWdc (SEIA and Wood Mackenzie). Limitation: the full-year residential figure is a Roofing Brief estimate extrapolated from quarterly data, and the 30 GW opportunity spans both residential and small-commercial roofs, so the true capture rate for all reroof-driven segments is lower than 15 percent.
Chart Recommendations
- Offset potential by state, ranked bar chart. Data: state rooftop offset percentages. Source: NREL, 2016. Insight: shows that consumption, not sunshine, sets the ranking. Citation-worthy because it corrects the assumption that sunny states lead.
- Potential versus installed, single stacked bar. Data: 1,118 GW potential against 279 GWdc installed. Sources: NREL and SEIA. Insight: visualizes the roughly 25 percent realized share. Citation-worthy as a one-glance gap.
- Where the potential sits, donut chart. Data: 65 percent small buildings, 35 percent medium and large. Source: NREL, 2016. Insight: nearly two-thirds of the resource is on homes and small structures.
- Offset share against a growing grid, two-bar comparison. Data: 39 percent of 2013 sales versus about 36 percent of 2024 sales. Sources: NREL and EIA. Insight: rising demand slowly erodes the offset share of a fixed potential.
Methodology
Source selection prioritized primary federal data. The core dataset is NREL’s Rooftop Solar Photovoltaic Technical Potential in the United States, a peer-reviewed national assessment that used airborne LiDAR across 128 cities, geographic information system processing to identify suitable roof planes, and hourly PV generation modeling. National installed and market figures come from the SEIA and Wood Mackenzie Solar Market Insight 2025 Year in Review. Electricity sales figures come from the US Energy Information Administration. The Department of Energy page on rooftop potential is used only where it restates NREL findings. Inclusion required a traceable primary or official source with a stated year and geography. Numbers that could not be verified against these sources were excluded rather than estimated. Where this report presents a derived figure, it is labeled a Roofing Brief calculation with its inputs and limitations. Conflicting numbers were resolved in favor of the original NREL report over secondary summaries. The NREL potential reflects a 2013 electricity baseline and technology assumptions of its publication year, and it should be read as a physical ceiling, not a forecast. Last updated July 2026.
Source Quality Ranking
- Tier 1, primary and official. NREL, Rooftop Solar Photovoltaic Technical Potential in the United States, 2016. US Energy Information Administration electricity sales data. US Department of Energy rooftop potential page restating NREL.
- Tier 2, credible trade and market bodies. SEIA and Wood Mackenzie Solar Market Insight 2025 Year in Review for installed capacity and market segments.
- Tier 3, reputable trade journalism. PV-Tech reporting used only to cross-check the building-size breakdown against the NREL figures.
- Excluded. Unsourced roundup pages, AI-generated summaries, and any figure that could not be traced to a primary or official dataset.
Citation Format
Source: National Renewable Energy Laboratory, Rooftop Solar Photovoltaic Technical Potential in the United States, 2016.
Source: SEIA and Wood Mackenzie, U.S. Solar Market Insight 2025 Year in Review, 2026.
Source: U.S. Energy Information Administration, Electricity retail sales, 2024.
Source: U.S. Department of Energy, Solar Rooftop Potential, citing NREL.
Most Quotable Statistics
- US rooftops could generate 1,432 TWh a year, 39 percent of national electricity sales. Source: NREL, 2016.
- California’s roofs could cover 74 percent of the electricity its utilities sell. Source: NREL, 2016.
- Wyoming’s roofs could cover only 14.2 percent, the lowest in the nation. Source: NREL, 2016.
- Nearly two-thirds of rooftop solar potential sits on homes and small buildings. Source: NREL, 2016.
- Every solar panel the US has ever installed equals about a quarter of what rooftops alone could hold. Roofing Brief calculation from SEIA and NREL data.
Data Limitations
The NREL potential uses a 2013 electricity baseline and the panel efficiency of its publication year, so it understates today’s physical potential from more efficient modules and a larger roof stock. It is a technical ceiling and ignores cost, ownership, grid capacity, and policy. The state rankings anchor on verified NREL figures; this report does not reproduce unverified per-state values. Installed-capacity figures from SEIA are direct-current ratings across all solar segments and are not directly comparable, unit for unit, with NREL’s modeled rooftop capacity. The reroof capture calculation extrapolates a full-year residential figure from quarterly data and should be read as an estimate.
Downloadable Dataset: Recommended Fields
A citable dataset built from these sources should include: state name; rooftop capacity potential in GW; rooftop generation potential in GWh; offset share of state electricity sales; population-weighted solar resource; per-capita electricity consumption in MWh; small-building share of potential; installed solar capacity to date; source citation; and study year.
Press Summary
American rooftops are a power plant the country has barely switched on. The most detailed federal study of the question found that suitable US roofs could hold 1,118 gigawatts of solar panels and generate 1,432 terawatt-hours a year, equal to 39 percent of national electricity sales. Nearly two-thirds of that potential sits on homes and small buildings. The rankings defy expectations: California’s roofs could cover 74 percent of its electricity use, while sunny Wyoming could cover just 14.2 percent, because it consumes so much power per person. Low-sun states such as Maine and New York outrank Arizona on offset share. Against this ceiling, the US had installed 279 gigawatts of solar of all types by the end of 2025, roughly a quarter of what rooftops alone could hold, and residential installations were flat. The gap between what roofs could do and what they do is the story. Source: NREL, SEIA, and EIA data.
Suggested Headlines
- US Roofs Could Power 39 Percent of the Country, Federal Data Shows
- California’s Roofs Could Cover 74 Percent of Its Electricity. Wyoming’s, Just 14 Percent
- Why Sunny Arizona Ranks Below Cloudy Maine for Rooftop Solar
- Nearly Two-Thirds of America’s Rooftop Solar Potential Sits on Homes
- Every Solar Panel Ever Installed Equals a Quarter of What US Roofs Could Hold
Frequently Asked Questions
How much electricity could US rooftops generate?
US rooftops could generate 1,432 terawatt-hours per year, equal to 39 percent of national electricity sales in the 2013 study baseline. Source: NREL, 2016.
How much solar capacity could fit on US rooftops?
Suitable US rooftops could host 1,118 gigawatts of solar capacity. Source: NREL, 2016.
Which state could offset the most of its electricity with rooftop solar?
California, whose rooftops could generate 74 percent of the electricity its utilities sold in 2013, the highest offset of any state. Source: NREL, 2016.
Which state has the lowest rooftop solar offset potential?
Wyoming, at 14.2 percent, mainly because it has the highest per-capita electricity use of any state at 30.3 MWh per year. Source: NREL, 2016.
What share of rooftop solar potential is on small buildings?
About 65 percent, or 731 gigawatts and 926 terawatt-hours per year, sits on buildings of 5,000 square feet or less. Source: NREL, 2016.
How was the rooftop potential measured?
NREL used airborne LiDAR and geographic information system methods across 128 US cities, representing about 23 percent of US buildings, then modeled the rest statistically. Source: NREL, 2016.
How much solar has the US actually installed?
The US reached 279 GWdc of cumulative solar capacity of all types by year-end 2025, about 25 percent of the rooftop technical potential. Sources: SEIA and NREL; Roofing Brief calculation.
Does rooftop solar potential still meet today’s electricity demand?
The 1,432 TWh potential equals about 36 percent of 2024 retail electricity sales of roughly 4,000 TWh, down from 39 percent of 2013 sales because demand has grown. Sources: NREL and EIA; Roofing Brief calculation.
Why do sunny states like Arizona rank below cloudy states?
Offset potential is generation divided by consumption, so low-use states such as Maine and New York can offset more of their total electricity than high-use sunny states such as Arizona. Source: NREL, 2016.
How much rooftop solar opportunity comes from reroofing each year?
About 30 gigawatts per year, from roughly 3.3 million homes built or reroofed annually; US residential solar installed on the order of 4.4 GWdc in 2025. Sources: US Department of Energy citing NREL, and SEIA; Roofing Brief estimate.
Related Roofing Brief Research
- Solar Roofing Adoption Statistics: US Home Solar Data 2026
- Solar Shingle Cost and Adoption Report (2026)
- Impervious Roof Area and Stormwater Report (2026)
- Cool Roof Energy Savings Report (2026)
- Green Roof Mandates and Incentives by City (2026)
Reviewed by The Roofing Brief Team. Last reviewed July 2026.