How Companies Like IBM Are Helping Predict Weather Better With AI

Stormy sea
Credit: Shutterstock photo

Each year, natural catastrophes result in severe economic costs and impact the lives of millions globally. While the forces of nature cannot be controlled, accurate and reliable weather forecasts will enable governments and administrations to take appropriate measures and contain the magnitude of the resultant damage. Likewise, precise weather predictions would benefit businesses by allowing them to take more appropriate decisions by factoring in weather related impacts.

Over the years, weather predictions have greatly improved, however, there is still room for growth. Here’s a look at how advanced technologies such as Artificial Intelligence (AI) are being leveraged to transform weather forecasting.

Weather isn’t always pleasant

On an average, approximately 60,000 people globally die from natural disasters each year, over the past decade. It is estimated that since 1980, natural disasters caused overall losses of almost $4.8 trillion out of which approximately 28% was insured according to Munich Re. In 2018, the insured losses due to natural disasters in the U.S. totaled to $52 billion. The data from National Centers for Environmental Information suggests that the U.S. has sustained 241 weather and climate disasters where the overall damage costs reached or exceeded $1 billion since 1980. The cumulative cost for these 241 events exceeds $1.6 trillion.

The World Bank estimates that the hydrological and meteorological hazards are responsible for 90% of total disaster losses worldwide and thus “improving the prediction of hydromet hazards—by getting accurate, timely predictions into the hands of decision-makers and the public—can save lives and money.”

The role of AI

Today, thousands of weather-focused satellites, terrestrial stations and IoT devices are constantly gathering real-time data. However, the extraction of insightful, accurate and timely information from the constantly increasing silos of ‘big data’ remains a challenge. This is where AI can play a crucial role.

A research paper titled “Using Artificial Intelligence to Improve Real-Time Decision-Making for High-Impact Weather” demonstrates how the “use of artificial intelligence techniques along with a physical understanding of the environment can significantly improve the prediction skill for multiple types of high-impact weather.”

IBM (IBM) has been working on projects related to weather forecasts for a long time, and is a leading player in meteorology today: “In 1996, IBM began exploring the 'business of weather,' hyper-local, short-term forecasting and customized weather modeling for clients.” In 2016, IBM acquired The Weather Company’s B2B, mobile and cloud-based web properties—, Weather Underground, and The Weather Company. 

In May 2019, IBM announced an AI-based tool dubbed IBM Weather Signals, designed to help companies predict how fluctuations in weather can impact business performance, even months in advance. Such information can be vital for industries (such as consumer packaged goods, aviation, energy, services) that are sensitive to changes in daily or seasonal weather. Better temperature and weather forecasts can help the energy sector anticipate peak demand and adjust production accordingly. Information on rainfall, drought, or floods is important for farming related decisions. The aviation sector obviously also benefits from better weather insights.

This month, IBM launched the new IBM Global High-Resolution Atmospheric Forecasting (GRAF) system—a high precision, rapidly updating global weather model that updates hourly (instead of every 6-12 hours) and at a 3 km resolution (down from 10 km-12 km) to provide a clearer picture of weather activity around the globe by assimilating weather data more rapidly into actionable forecasts.

IBM brings the technology behind some of the world’s most powerful supercomputers for weather forecasting by running the new system on an IBM POWER9-based supercomputer flexing NVIDIA V100 Tensor Core Graphics Processing Units (GPUs).

An NVIDIA (NVDA) blog post points out that, “weather models are data heavy.” There is humongous amounts of data from weather radar systems, satellites, aircraft, weather balloons, barometers and thermometers on the ground. Crunching such data into usable information is the key. The IBM GRAF system monitors about 1.5 billion points in split seconds using NVIDIA GPUs.

“When heavy rainfall is expected, including monsoon-driven storms, GRAF has the potential to provide detailed, potentially more accurate weather forecasts for how much rain is expected in the next 15 hours,” according to Todd Hutchinson, director for numerical weather prediction at The Weather Company. Forecasts for 26 million locations across the world will be produced by The Weather Company.

In the future, adoption of AI-based forecasting services could translate to better business, but more importantly result in a larger good for all, both in tangible and intangible terms.

“Every one dollar invested has the potential of generating at least three dollars’ worth of benefits in weather and climate services—a win-win,” reads a World Bank report.

The author has no position in any stocks mentioned. Investors should consider the above information not as a de facto recommendation, but as an idea for further consideration. 

The Weather Company is a unit of IBM and the names have been interchangeably used in the information related to the project.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

In This Story


Other Topics

Artificial Intelligence

Prableen Bajpai

Prableen Bajpai is the founder of FinFix Research and Analytics which is an all women financial research and wealth management firm. She holds a bachelor (honours) and master’s degree in economics with a major in econometrics and macroeconomics. Prableen is a Chartered Financial Analyst (CFA, ICFAI) and a CFP®.

Read Prableen's Bio