8 Sep 2014

A hurricane-force storm in September?

As I looked at the longer term trend for our weather this morning, one weather computer model (the US Global Forecast System) threw up a particularly interesting theme for the fourth week in September – a hurricane-force storm.

Now before you all start getting overly excited and reaching for the panic button, for now, it simply remains, in my opinion, an unlikely outcome.

But I hear you say, how can I say that? This has been predicted by a powerful weather computer model, using a supercomputer that has the same level of computing power as many tens of thousands of PCs combined. Surely it’s going to happen?

hurricane_GFS_WZ_wpGFS forecast for 22nd September: Wetterzentrale

Well, this is where the value of human forecasters comes into play, to look at something that a computer model has generated and question whether or not it is likely to come to fruition.

The things that go against this exact scenario happening at the moment are;

– it is still two weeks away – a very long time in weather forecasting

– it is based on an ex-hurricane that hasn’t even formed yet

Why are the chances of this exact outcome happening low?

The seeds of the hurricane-force storm that the weather computer model has generated in two weeks are only just moving off the west coast of Africa and have to overcome many hurdles in order to develop.

Clusters of thunderstorms, known as tropical waves, roll off the west coast of Africa during the summer and early autumn into the Atlantic Ocean.

rb-animated

Tropical wave off west coast of Africa (red/orange): NOAA

However, in order for them to develop into a hurricane, they need to avoid to things that stop them from growing stronger – wind shear, cool waters and dry air.

Hurdles to overcome

Wind shear refers to the wind at different heights in the atmosphere blowing at varying speeds and directions. If wind shear is strong, it has a tug-of-war effect on potential storms, ripping them apart.

In order for hurricanes to form, sea surface temperatures need to be above 26C, to provide enough tropical energy to be transferred upwards. Sea surface temperatures have been below average in some parts of the Atlantic, meaning that this has lessened the potential so far this season.

Another big player is the presence of dry, dusty air that has been blown across the Atlantic from the Sahara desert. This dry air fights the moist tropical air that hurricanes need to thrive in, inhibiting storm development.

dust_atlantic_NASA_wpDust being blown over Atlantic from Africa: Nasa

Even if this potential storm manages to overcome some of these hurdles, which it looks as though it might, there is still one more significant hoop to jump through – interaction with the jet stream.

The remnants of whatever storm happens to form needs to interact with the jet stream at precisely the right time and position over the mid-Atlantic Ocean in order to deepen explosively.

It’s like a surfer trying to ride a wave. If he catches the wave at the right moment, then game-on. However, if he misses the moment, then nothing happens.

Unlikely outcome, but not to be ignored

Even though I’ve demonstrated that in order for this hurricane-force storm to happen, a lot of things need to fall into place, the possibility should never be ignored.

What this weather computer model has shown us meteorological folk is that in two weeks’ time, the ingredients will be present to deliver the potential for some wet and very windy weather.

During the coming 7-10 days, weather computer models from all around the world will be closely monitored to hone in on the detail as things become clearer closer to the time.

Don’t forget, you can get the latest forecast on the Channel 4 Weather website. I’ll also be posting updates on Twitter – @liamdutton

Tweets by @liamdutton

One reader comment

  1. anon says:

    re the weather, may I ask a question please? some time ago it was difficult to predict the weather then as computing power increased I think I noticed the forecasts were pretty accurate. Perhaps its my perception of things but for the last 10 years or so the forecasts have become less accurate but I presume computers have become even more powerful and other methods of predicting weather have also improved so does this indicate anything?

    Is there any statistical evidence that records actual weather with that forecast and would any pattern in this show something happening more recently re the weather?

Comments are closed.