Wellington averages 2,094 sunshine hours a year in NIWA’s 1991 to 2020 climate normals. That is 28 hours behind Auckland, a difference of 1.3 percent, or about five minutes a week. The city’s reputation and its climate record do not match.
Wellington against the other main centres
| Location | Hours per year |
|---|---|
| Nelson | 2,497 |
| Christchurch | 2,128 |
| Auckland | 2,122 |
| Wellington | 2,094 |
| Hamilton | 2,034 |
| Palmerston North | 1,772 |
Nelson is genuinely sunnier and Palmerston North is genuinely duller. The Auckland gap that most people assume is wide is not. Those Wellington figures come from the Kelburn station at 125 metres, and a shaded gully and an open ridge in the same suburb can differ from each other by more than Wellington and Auckland differ on paper.
What those hours turn into on a roof
Sunshine hours are a weather statistic, not an output figure. For output, the European Commission’s PVGIS model gives roughly 1,500 kWh per kW installed per year for Wellington, assuming a north-facing array at 30 degrees with 14 percent system losses. On those assumptions a 6kW array models at about 9,000 kWh a year.
Treat that as a planning number rather than a promise. It is modelled from reanalysis data on a grid tens of kilometres across, so it describes the region and not your roof. We do not use it to rank one city against another either: the same model puts Nelson below Wellington despite Nelson having the most measured sunshine in the country, which is a fair sign that it handles New Zealand’s terrain loosely.
How uneven the year is
This is the number that actually matters, and it is the one regional averages hide:
| Month | Hours | Month | Hours |
|---|---|---|---|
| January | 231.8 | July | 121.0 |
| February | 211.4 | August | 147.8 |
| March | 204.0 | September | 164.4 |
| April | 156.4 | October | 193.3 |
| May | 133.1 | November | 211.7 |
| June | 101.2 | December | 218.0 |
June averages 101 hours against January’s 232, a little over two to one. The PVGIS output model lands on almost the same seasonal shape from entirely different data, which is a good sign the ratio is real: it puts June generation at roughly 40 percent of January’s.
So a system sized to cover your summer daytime use will cover a much smaller share of it in June. That is the argument for sizing against twelve months of your own bills rather than an annual average, and it is the main reason people add storage.
Cloudy days, and why panels prefer the cold
Panels generate from diffuse light as well as direct sun, so an overcast day produces less rather than nothing. Worth knowing: the 2,094 hour figure counts bright sunshine only, so what the system makes on grey days sits on top of that number rather than inside it.
Panels also run on light rather than heat, and they lose efficiency as they warm up. A cool, bright Wellington day is good generating weather, which is part of why the output modelling does not punish the region the way its reputation suggests it should.
Does the Kāpiti Coast differ?
For sunshine, not meaningfully. NIWA does not publish a separate Kāpiti normal in the main table, and modelled output for Paraparaumu comes out close to Wellington’s. What does vary by address is exposure, shade and wind zone, and the last of those has consequences beyond generation. See building consent for solar panels in Wellington.
What decides it on your roof
Regional figures settle whether solar is viable here. They do not settle whether it is worth it for you. Three things do:
- Which faces get the sun, and how much usable area you have. The solar roof check reads this from satellite data in a few seconds.
- How a system gets sized, and what it costs. That comes off your own twelve months of usage. The home solar page covers sizing, batteries and payback.
- What the building is for. Daytime load changes the maths, which is why commercial sites get sized differently to houses.
For worked local examples, our recent installs show how the same sized system ends up looking different from one roof to the next.
Common questions
How many sunshine hours does Wellington get?
2,094 a year in NIWA’s 1991 to 2020 normals, measured at Kelburn. Auckland is 2,122 and Christchurch 2,128, so the three are within about 1.6 percent of each other.
Do solar panels work on cloudy days?
Yes, at reduced output, because panels generate from diffuse light too. The annual sunshine figure counts bright sunshine only, so cloudy-day generation is additional to it.
How much power will solar generate in Wellington?
Modelled at roughly 1,500 kWh per kW installed per year for a north-facing array at 30 degrees. A 6kW system models at about 9,000 kWh annually, before any allowance for shade on your particular roof.
Does solar still work through a Wellington winter?
It works, at roughly 40 percent of midsummer output in June. A system sized only against summer use will feel short in winter, which is why sizing starts from a full year of bills.
Is Wellington too windy for solar?
Wind does not reduce generation. It does change how the array has to be fixed down, and in most of the city that has consent implications, covered in the consent guide.
We size from twelve months of your own power use, and the quote shows the equipment, the expected output and the payback before you commit. Site visits and quotes are free.
Sources: sunshine hours from NIWA mean monthly sunshine, 1991 to 2020 normals, Wellington measured at Kelburn Aws. Modelled output from the European Commission’s PVGIS tool, PVGIS-ERA5, north facing at 30 degrees with 14 percent losses.