UK long range forecast: the next six weeks

Weeks three to six (from 16 Nov) lean a little warmer than normal across the UK on average. Each week is compared with what is normal for the time of year. The bigger the difference, the stronger the signal. It tells you how a week leans; it cannot tell you which day will be wet.

Week by week, UK average

Averaged across our eleven cities. Bars to the right are warmer or wetter than normal; to the left, colder or drier.

This week
2 Nov to 8 Nov
A little colder than normal
−0.5°C
Drier than normal
−18 mm
A little colder than normal, drier than normal.
Next week
9 Nov to 15 Nov
Close to normal
−0.2°C
Close to normal rainfall
−2 mm
No clear signal either way: expect typical weather for the time of year.
Week 3
16 Nov to 22 Nov
Warmer than normal
+1.5°C
Wetter than normal
+5 mm
Warmer than normal, wetter than normal.
Week 4
23 Nov to 29 Nov
A little warmer than normal
+0.9°C
Wetter than normal
+5 mm
A little warmer than normal, wetter than normal.
Week 5
30 Nov to 6 Dec
A little warmer than normal
+0.6°C
A little wetter than normal
+2 mm
A little warmer than normal, a little wetter than normal.
Week 6
7 Dec to 13 Dec
A little milder than normal
+0.5°C
A little wetter than normal
+2 mm
A little milder than normal, a little wetter than normal.

Source: ECMWF and Open-Meteo. Updated 4 Nov, 13:18. ECMWF EC46, 51 runs, updated daily; rain normals shown are Birmingham's, the most central city

By nation

The leaning for each capital, week by week. Small differences between nations are within the noise; large ones usually mean a north-south split in the pattern.

Week of	England (London)	Scotland (Edinburgh)	Wales (Cardiff)	Northern Ireland (Belfast)
2 Nov	A little colder than normal (−1.1°)Drier than normal	Close to normal (−0.1°)Drier than normal	A little colder than normal (−1.4°)Drier than normal	Close to normal (−0.4°)Drier than normal
9 Nov	Close to normal (+0.3°)Wetter than normal	Close to normal (−0.4°)A little drier than normal	Close to normal (−0.2°)Close to normal rainfall	Close to normal (−0.1°)A little drier than normal
16 Nov	Warmer than normal (+1.6°)A little wetter than normal	A little warmer than normal (+1.4°)A little wetter than normal	Warmer than normal (+1.6°)Wetter than normal	A little warmer than normal (+1.1°)A little wetter than normal
23 Nov	A little warmer than normal (+1.2°)Wetter than normal	A little warmer than normal (+0.6°)A little wetter than normal	A little warmer than normal (+1.2°)Wetter than normal	A little warmer than normal (+0.6°)A little wetter than normal
30 Nov	A little warmer than normal (+0.8°)Wetter than normal	A little warmer than normal (+0.5°)A little wetter than normal	A little warmer than normal (+0.6°)A little wetter than normal	Close to normal (+0.3°)A little wetter than normal
7 Dec	A little milder than normal (+0.6°)A little wetter than normal	Close to normal (+0.4°)Close to normal rainfall	A little milder than normal (+0.5°)Close to normal rainfall	Close to normal (+0.3°)A little wetter than normal

City by city: every city page carries its own six week outlook.

How to read a long range forecast

Read it like betting odds. “Week 4 leans milder than normal” means the 51 runs of the model, taken together, put more weight on mild outcomes than usual for that week. It does not mean every day will be mild, and it says nothing about any single day.

Signal strength matters more than direction. We call anything within half a degree “close to normal”, because the models have little skill at that size. A leaning of 1.5 degrees or more across a whole week is a clear signal at this range.

Watch how it changes. The extended-range model updates every day. A leaning that holds steady for several updates is more trustworthy than one that flips. For the days you are actually planning, use the 14 day forecast, which shows the spread of the runs day by day.

Long range forecast: questions

Can anyone forecast the weather six weeks ahead?

Not day by day. What the extended-range models can do, with some skill, is say whether a whole week is more likely to be warmer or colder, wetter or drier than normal. That is what this page shows, and nothing more precise.

What is 'normal' here?

The ECMWF extended-range model is compared with its own long-term climate for the same time of year (its re-forecasts of past years). That removes the model's own biases, which is why we use it rather than comparing raw model output with station averages.

Why do the weeks further out sit close to normal?

Because the 51 runs disagree more and more, and their average drifts back towards the climate. A week three to six leaning of more than about one degree is a real signal; smaller than half a degree is no signal at all.