Need-- They need to keep pushing, and that's been historically quite difficult to forecast.

So there's something that we call the spring predictability barrier, which happens, um, around the time that we see a transition between the different phases of the El Niño cycle.

So last winter, we had what we call a La Niña event.

That was when the surface waters were colder than normal.

And in the springtime, that previous event has died away.

It's decayed.

We're back to near normal conditions, but we're ahead of starting to see the buildup of the next event, which is coming this year.

Uh, how does this interact? Does it, does it-- Does global warming make the forecasting and the impacts of all this harder to understand?

(laughing)

On the subject of sort of beautiful optical effects, there's, we're talking mainly about clouds forming within this troposphere, the lowest layer of the atmosphere, but actually, you can get clouds even higher than that.

So there are types of clouds called noctilucent clouds that form very high up, about 80 kilometers, uh, above the, the surface.

It's very, very dry up in, uh, the mesosphere, um, but it's very, very cold.

It's about a -120 degrees Celsius there.

So e- even though it's very dry, it gets cold enough that you can still form these very, very thin ice clouds, and you can see those sometimes, if you're lucky, um, particularly at sort of dusk or in the morning when you get that, as, as Owain was describing, you get this sort of horizontal path of the light through the atmosphere, um, which gives you that more sort of red-y colors.

You can sort of see them very like ghostlike clouds almost.

Of the uncertainties, the major uncertainties in climate models, our understanding of cloud, is that one of the bigger problems that we have in the models?