How much solar power can my location generate?
The most useful number for comparing places is kWh/kWp/year: how many kilowatt-hours of electricity one kilowatt-peak (kWp) of solar panels produces in a typical year. A “5 kW” rooftop system is 5 kWp, so multiply the map value by 5. A place scoring 1,000 kWh/kWp/year would give a 5 kW system about 5,000 kWh a year; a place scoring 1,600 would give about 8,000 kWh.
SolarMapped colours the whole map by this value for fixed panels tilted at their best angle. Click anywhere on land for the local figure, the month-by-month pattern and the best tilt.
| City | kWh per kWp per year | 5 kW system, per year |
|---|---|---|
| Helsinki, Finland | ≈ 970 | ≈ 4,900 kWh |
| Berlin, Germany | ≈ 1,050 | ≈ 5,300 kWh |
| New York, United States | ≈ 1,400 | ≈ 7,000 kWh |
| Madrid, Spain | ≈ 1,620 | ≈ 8,100 kWh |
| Sydney, Australia | ≈ 1,580 | ≈ 7,900 kWh |
| Cairo, Egypt | ≈ 1,820 | ≈ 9,100 kWh |
Example PVGIS values, rounded, for crystalline-silicon panels at their optimal fixed tilt facing the equator, with 14% system losses. Click any city on the map for the exact current figure.
What is solar potential?
People use “solar potential” for two different things. Solar irradiation (or radiation) is the sunlight energy that reaches a surface, measured in kWh per square metre. PV output is the electricity a solar system actually produces from that light, in kWh per kWp.
PV output is the more practical measure. Panels lose efficiency when they get hot, reflect some light at low angles and pass energy through cables and inverters, so output isn’t simply proportional to irradiation. SolarMapped therefore leads with PV output and shows irradiation as a secondary figure.
How does location affect solar panel production?
- Latitude. Closer to the equator the sun is higher and the year more even. Far from it, winters are dark and production is concentrated in summer.
- Cloud cover and climate. Cloudy, humid regions get less direct sun than dry ones at the same latitude, which is why deserts lead the map.
- Temperature. Solar panels are slightly less efficient when hot, so a very hot site loses some of its advantage.
- Tilt and orientation. Fixed panels do best facing the equator (south in the northern hemisphere, north in the southern) at an angle close to the latitude, flatter near the equator.
- Shading. Hills on the horizon are modelled where data exists; nearby buildings and trees are not, and they matter a lot on real roofs.
How SolarMapped calculates solar production
The map layer and the per-location report both come from PVGIS, the free Photovoltaic Geographical Information System from the European Commission’s Joint Research Centre. The map was pre-computed on a 2° global grid and smoothed; the report on the right runs a fresh PVGIS calculation for the exact spot you pick (rounded to about one kilometre) and caches it.
Assumptions: crystalline-silicon panels, 14% system losses, fixed free-standing mounting facing the equator, tilt optimised for the location, and terrain horizon shading where available. Changing system size scales the result directly. Real systems vary with panel quality, roof shape, shading, dirt and snow, and year-to-year weather. See Sources and method for the details and data licences.
Does SolarMapped account for clouds and weather?
Yes, and it is the single biggest reason the map is not just a pattern of latitude bands. The sunlight totals are measured from weather satellites, not calculated from a clear-sky formula, so the real cloud cover, haze and dust over each place is already inside the number. Air temperature and wind come from a separate reanalysis dataset and are used to model how much efficiency the panels lose when they run hot.
That is why Madrid reaches about 1,620 kWh/kWp/year while Berlin manages about 1,050, even though they sit only twelve degrees of latitude apart. Cloud, not sun angle, is most of the difference. It is also why the monsoon shows up in the timeline: India and West Africa dip in the middle of the year despite the sun being at its highest.
What you get is a long-term average, not a forecast. The figures blend many years of weather, so they describe a typical year for the location. A single real year will come in sunnier or cloudier, and short-term conditions on any given day are not part of the model. The report lists the exact dataset and year range used for your location.
Does it account for snow on the panels?
No. Snow is the clearest known gap in these estimates, and it works in two directions that the model leaves out entirely:
- Snow covering the panels blocks output until it slides or melts off. This is not modelled. The 14% system-loss figure covers cables, inverter, mismatch and dirt — it is a flat, year-round allowance and contains nothing for snow.
- Snow on the ground is bright, and a tilted panel picks up light reflected off it. That can lift winter output. The model assumes a fixed, bare-ground reflectance all year, so this bonus is missing too.
These two partly cancel out, and the gap matters least where snow is most common. In the far north the deep-winter weeks contribute only a few per cent of the annual total, so even losing them completely barely moves the yearly figure. Late winter and early spring are where it really bites — by March the sun is strong again while snow can still be sitting on the roof, and that is when the model is most likely to be optimistic.
If you are planning a system in a snowy climate, treat the winter months here as an upper bound and ask a local installer about panel tilt, which affects how readily snow slides off.
How does solar radiation change through the year?
Drag the timeline under the map, or press play, to step through 52 weeks of a typical year. The map then shows solar radiation — the sunlight energy reaching one square metre of panel that week, in kWh/m² — instead of the annual average. It is built from the same PVGIS hourly data as the location report, averaged over 2021–2023.
The weekly layer averages a shorter run of years than the annual figure does, so it is a noisier read on the weather: one unusually cloudy or clear year has more influence on an individual week than it does on the annual total. Treat the shape of the curve as reliable and any single week as approximate.
Watching a year run through is the clearest way to see why latitude matters. Northern Europe and Canada almost switch off between November and February, while the tropics barely change. Places with a wet season, such as India or West Africa, dip in the middle of the year even though the sun is high, because cloud blocks it. Pick a location first and the report lists that spot's value for every week.
Solar potential by location
Country and city pages with location-specific solar data are coming. Until then, use the search box above to look up any city.