Weather Forecasts and the Internet of Things

Weather forecasting. Some days, like maybe this past week, it seems like a bit of a joke. When it rains at your picnic or you brought your heavy jacket only to carry it around with you, it doesn’t seem like much of a science. Actually, weather forecasting has actually been improving, using better technology and data science to help make better predictions. And weather is now being seen through the lens of the Internet of Things (IoT).

Why is having accurate weather reports so important?

Both short and long range forecasts have tremendous impacts in many different ways.  Extreme weather forecasts are needed to make proper evacuation calls and to help with emergency response. Weather has a direct financial impact on many businesses, such as transportation, retail, and construction. Traders factor forecasts into market pricing for energy and agricultural products. Long term patterns allow scientists to see how climates change. Accurate weather predictions have large financial and societal consequences; it’s much more than deciding on bringing that sweater along or not.

Studies do show that weather forecasting has improved for a number of decades. Average daily temperatures forecasts, three, five, and seven day predictions, and extreme weather modeling have all gotten better. Meteorologists – and of course the public – know there is room for improvement. The data modeling of weather is continuously being enhanced, with multiple complex models taken into consideration for analysis. Large supercomputers process vast data sets, comparing that against models, past weather patterns and, historical data.

And in what many people saw as an unusual business move, IBM bought the digital assets of The Weather Company in 2016. Why would IBM do this? It also goes to the IoT. The Weather Company was collecting weather input from over 100,000 Internet connected devices, such as weather stations, aircraft, buildings, and more. They had also built a robust platform for the intake of all of this diverse data. IBM saw a big opportunity. With the combination of this data ingestion engine, massive weather data, and Watson, IBM’s cognitive computing platform, IBM saw a natural convergence. Watson could enhance it skills and capabilities and IBM could deliver timely and highly accurate weather forecasts to businesses. The first product launched as a result of this acquisition was an extremely localized weather forecast – between .2 and 1.2 miles – to give businesses detailed, very confined, weather information. Clearly there is much more to come of this.

Like so many things, weather forecasts are changing as we re-envision Internet data collection, big data analytics, and advanced computing. Weather is truly one of the greatest big data problems and opportunities.


Follow this blog

Get every new post delivered right to your inbox.