Innovation Blog

Here at Innovation Age we know that innovation can be exciting and frustrating in equal measure so as professionals working in the field of innovation we've distilled our experiences here to hopefully inform and inspire your own innovation journey.

Innovation in Data

This week’s storm over the south of England provided a great example of how 25 years of innovation in data collection, processing and communication have benefitted us all.

The story begins in 1987 when a storm with a similar impact was brewing in the Atlantic and heading for Southern England, but nobody knew it. While forecasters were aware of the storm, its track was expected to be further south with strong winds over France with only rain predicted for England. As a result, in the days before 24-hour news coverage, we all went to bed on the night of October 15th with no idea of the carnage that would ensue during the night. As is now well documented 18 people lost their lives during those chaotic hours, in part because of the lack of warning that was given.

In comparison, the storm this week, which similarly hit overnight after a relatively quiet day, came as a surprise to no one. Warnings were issued during the preceding days that proved to be 100% accurate. While a few people still tragically died, the accurate forecast undoubtedly saved many lives and showed the benefit of data innovation since the previous storm.

There are three key aspects to how data is used and each is highlighted in this example:

Collection – in 1987 readings were taken from mainly a few ground stations and weather vessels moored in the Atlantic, which hampered forecasters in two ways. Firstly it provided low resolution readings with hundreds of miles between data points and secondly it meant that what was happening in the upper atmosphere had to be worked out from observing what happened at ground level. Since 1987, however huge amounts more data is available from a variety of satellites providing the visible, thermal, radar and many other views from space to add to the terrestrial measurements.
Processing – in 1987 even predicting the weather a day ahead was a considerable challenge for the super-computers available. When coupled with the comparatively few models available this meant that the degree of accuracy possible led to the track of the storm being missed by over a hundred miles. Nowadays with more computing power and, crucially, more models available, forecasters are able to run multiple scenarios and average the results to provide greater confidence in the predictions. As a result, today’s 5-day forecast is as accurate as 1987’s 1-day forecast.
Communication – even had all the information been available in 1987 it is questionable whether the mechanisms existed to provide the public with adequate warnings. The latest TV weather forecast provided on the evening before the storm was at 9:30pm and there was no recognised method of alerting people to severe weather. But in the wake of that storm the Met Office instigated the now familiar coloured severe weather warnings and they have become well recognised by the public and worked so well in preparing people this week.

So, significant advances have been made in all three aspects between the two storms, which resulted in a dramatically improved outcome the second time around. While each innovation on its own would’ve provided some benefit it took all three to work in harmony to save lives on Monday morning – you could say that as far as data was concerned, it was a perfect storm.

By Brian Carrington, October 2013

Click here to see previous blogs.