Can big data predict the weather?

As a Brit I always talk about the weather no matter if it’s sunny or rainy, I plan my weekends depending on what the forecast is and I get disappointed when it’s wrong. Data is being used to help with these predictions we see on TV, but big data is also helping companies to understand the trends, predict potential natural disasters and understand climate change.

We have been trying to predict the weather for thousands of years and today it is a lot more accurate than it was even 10 years ago. Weather companies are now able to look back at the masses of data from previous years to find trends and patterns in weather. But it’s not just the old data that is helping any more, and it’s not just the weather companies that are effectively using this data.

One motor insurance company is using weather data to aid its scoring points system for drivers that use telematics boxes. By looking at the weather they compare the driver’s acceleration and braking speed along with cornering to score them. By using this data the insurance company is able to give more accurate policies to their customers, as they can now take in account of the driving condition as well as the driving style.

The data is also being accessed by airlines, by looking at the data they are able to predict wind speeds and direction and can set the airplane course accordingly, helping to reduce emissions. The Met Office has predicted that by using this data for flights they have helped to save ’20 million tonnes of CO2 emissions annually’. The planes themselves also have sensors on them to help collect data on temperature, direction and speed of the wind.

The data available can also help with research, IBM has set up a research and analytics project named Deep Thunder to help cities and countries around the world predict severe weather conditions. One city they are currently working with is Rio de Janeiro, due to the climate the city suffers from many floods and landslides. IBM is using data that describes the atmosphere to create algorithms of how storms are likely to develop. IBM has said that with this model they can ‘predict up to 40 hours ahead of time how much rain will fall in a particular location – with 90% accuracy’

And of course, the data is being used by the weather companies themselves. By looking at the data available and how storms or high winds, like Doris, Ewan and even Stella are travelling across the world companies are able to predict the direction it will take, whether it will hit the North or South and how powerful the damage will be.

While the weatherman may not always get it right, big data has massively improved the reliability of predictions that are being made. Not only is the data helping with day-to-day forecasts it is also potentially saving lives. With more of the data being collected and more research being conducted the predictions are only going to get more accurate, and maybe one day natural disasters might be predicted weeks, or even months before they happen.

Do you think big data will continue to help predictions? How accurate do you think it can get?

KDR Recruitment is the home of the UK’s best Information Management and Analytics jobs.

This blog was originally published on LinkedIn. To read the original article click here.

You might also like: