In less than a generation, the way we receive weather information has been revolutionised, thanks to those handy mobile soothsayers called apps! John provides some words of scientific insight into the weather icons on your phone. Where do they come from and how much trust should we really place in them? As we all know, when they go wrong, it REALLY affects our lives.
Shining a bright light on a dark science
How’s your weather app doing today? Good news! Mine promises mostly sunny skies today. Actually when I click on this cheerful symbol, it breaks the day down into a helpful icon for each hour. It’s not actually indicating clear blue skies and sunshine throughout. There will be some puffy cloud and, for a couple of hours, showers are expected. Even more usefully, each icon has an accompanying probability of rainfall, precise to the nearest 1%. Amazingly, I can get this hourly precision out to two weeks ahead. And because I spent a bit of money for this app, it gives me a few other ‘bells and whistles’ too.
A friend shows me a different app that she uses. It was free. It also gives an hourly breakdown of the expected weather. Impressive. But hang on a minute. This app expects the day to be chillier, quite grey and indicates light rain for five hours of the day. There’s no clue as to the probability of that rain – so presumably it’s a certainty?
A shower of confusion
Wait a minute! Which app do I believe? How can I plan my day? The confusion is compounded just a few hours later, when both apps have changed their forecast; but they’re still not the same as each other. They both seem quite precise in their own way. But inevitably, at least one of them is bound to be precisely wrong.
Sounds familiar? How can we trust these forecasts when they keep changing? Surely in this day and age, we should expect better?
Well – yes and no. There are scores of weather apps to choose from. But let’s look at how app forecasts, often very good, are generated and why they sometimes go very wrong.
It’s not easy – (Pinning the tail on the donkey!)
Meteorology is a fiendishly complicated science. The weather we experience at any given hour of the day is the result of mind-blowingly intricate interactions between air molecules all around the world.
Imagine a molecule of air as a (tailless) donkey. We know roughly where the donkey is as it is released into a huge field. But, based on that information alone, good luck trying, blindfold, to pin the tail on the same donkey after it has randomly wandered around the field in the intervening few days.
From the meadow back to meteorology – the same principle applies; but the ‘meadow’ is the size of the planet! Even if we know exactly what each molecule of air is doing right now, predicting where that molecule will go in the future depends on knowing the precise laws of physics which govern its behaviour, and having a powerful enough super-computer to do the calculations required.
How powerful? As much as 14,000 trillion calculations per second drive the Met Office’s supercomputer. Yet, even then, the atmosphere is too complex for one of the most powerful machines in the world to be accurate all the time. There are inevitable errors. With each new run of the model – several times a day – the forecast will change as it adjusts to the latest errors observed by the eyes in the sky – from satellite, aeroplane and radar.
This illusion of certainty that many weather apps portray is a confidence trick.
All I want to know is: “Will it rain?”
The tension between simplicity and accuracy is a seemingly unsolvable issue with weather app providers. They know that most of us just want a “Yes”-“No” answer to whether it will rain. The reality is less clear-cut.
It’s a game of probabilities
Predicting the weather, hard enough for the next few hours, gets increasingly difficult the further ahead we look because the errors get larger. Sure, the computers will still make a prediction. Indeed, they’ll come up with a neat icon for a Saturday afternoon in 6 months’ time if they’re asked to. But the reliability of such predictions becomes vanishingly small beyond the range of a few days.
… apps in-built to your iPhone or Android devices often use inferior computer models
To muddy the waters further, depending on your app, the forecast will come from one of a number of meteorological centres, each of which has a supercomputer that models the atmosphere in a subtly different way. Each model has its own unique flaws. And some are better than others. The Met Office and BBC Weather apps use some of the best-performing and detailed models in the world. However apps in-built to your iPhone or Android devices often use inferior computer models. As a result their predictions will be, on average, poorer.
To try to take account of model flaws, some centres analyse how consistent their predictions for a particular outcome are. If very diverse, then the weather is deemed to be inherently unpredictable, and vice versa. Using this method, some apps display a probability of a certain weather outcome: for example, a 32% chance of rainfall.
Such probability forecasts tend to give a much more realistic likelihood than the bold icons that often accompany them on the same app.
But probabilities to the nearest percentage point? Weeks ahead? Really? Be wary of those precise claims at longer ranges.
Precision v Accuracy – The illusion of certainty
“OK”, I hear you say. “Surely the forecast for my postcode should be OK for just the next few hours, right?” Well it depends.
There’s a difference between precision and accuracy. Weather models provide precise forecasts for specific points. But what if there’s a hill or a lake or a coastline between that point and your postcode? Even the most powerful models cannot resolve every crease and fold in the landscape. Very local effects which the models don’t ‘know about’ can make a big difference, even in the next few hours. From clear to cloudy, from -3C to +3C, over just a couple of miles.
- Most weather apps cannot be corrected by a human forecaster when the weather goes wrong
- Met Office and BBC Weather Apps use some of the best high-resolution models in the world
- The inbuilt weather apps that often come with your phone and tablet use less accurate computer models
- Meteorologists admit that weather apps are often misleadingly precise in their forecasts
- The prediction of showers, fog and low cloud are hardest for weather apps to successfully achieve
- ‘Impactful’ weather types (such as rain), are given more weighting than others, when the app decides which icon to display
- Icons are sometimes at odds with the rainfall probabilities next to them because they are derived from different models
So, for all these reasons, (and a few more besides!), weather apps are, like all machines, prone to going wrong. To the question: “Will it be sunny tomorrow”, your app may boldly answer “Yes”. But the reality is often more of a “Maybe” – and that’s a fact.
“Ok John, but you’re a grumpy ‘glass half empty’ kinda guy”, I hear you say. “You’re picking holes in a technological marvel of the modern age! Let’s hear it for weather apps!”
Yes. They can be brilliant, are getting better, and have become the primary source of weather information for millions of us. But sometimes they are dreadfully bad.
That’s why here at weathertrending, we’re unique in giving you a daily heads-up in our QuickCasts on whether your apps will be more or less reliable. You won’t hear that level of honesty anywhere else.
John and Sara give some hints on when (and when not) to trust your apps here
Please do sign up to our email list to gain exclusive content from John and Sara straight to your inbox: