In This Issue
Authors
The Bridge: 50th Anniversary Issue
January 7, 2021 Volume 50 Issue S
This special issue celebrates the 50th year of publication of the NAE’s flagship quarterly with 50 essays looking forward to the next 50 years of innovation in engineering. How will engineering contribute in areas as diverse as space travel, fashion, lasers, solar energy, peace, vaccine development, and equity? The diverse authors and topics give readers much to think about! We are posting selected articles each week to give readers time to savor the array of thoughtful and thought-provoking essays in this very special issue. Check the website every Monday!

Future of Weather Forecasting

Monday, February 22, 2021

Author: Joel N. Myers

In The Signal and the Noise, the noted statistical analyst Nate Silver (2012) examined forecasts in many categories and found that most demonstrate little or no skill and have made little or no progress in accuracy over the decades. The lone exception he found was weather forecasting. Before attempting to predict what is ahead for weather forecasting, it is important to understand how weather prediction became such a success story.

Background

Benjamin Franklin was the first to observe that northeastern storms (nor’easters) begin in the Southwest; he proposed models to describe the progression of storm systems and published some of the first weather forecasts.

During World War I English mathematician Lewis Fry Richardson postulated that fluid dynamics equations could be used to forecast the state of the atmosphere. It took him months to calculate a 6-hour forecast by hand, only to get impossible results due to errors in his calculations. But his mathematical approach was vindicated in the 1940s with the invention of computers. Since the 1970s, weather forecast models have been run on the world’s fastest, most powerful supercomputers, spurring advances in both computer technology and weather modeling.

Government Forecasts

Through World War II, weather forecasts were mainly issued by the government. After the war a few meteorologists who had left the military began services that offered customized forecasts to weather-sensitive businesses, but until the 1960s forecasts available to the public came almost entirely from the US Weather Bureau, now the National Weather Service (NWS). They were limited in scope and accuracy, typically covering only today, tonight, and tomorrow, with few details, and they often missed major events. In those days, weather forecasters were often the butt of cartoons and jokes (“The only job where you can get paid for being wrong!”). Many meteorologists were passionate and driven to improve the accuracy of forecasts and build respect for their trade.

Since then, NWS has greatly improved the available data—with Doppler radar, satellite observations, and more detailed and accurate computer-generated forecast model output—and thus the quality of its forecasts and warnings.

Commercial Forecasting Innovations

Much of the increased value and utility of weather forecasts for the public and for business has been spurred by commercial weather companies, such as AccuWeather, which I founded as a Penn State graduate student in 1962.

These companies have introduced hundreds of innovations—forecasting techniques and products, patented technological advances, improvements in accuracy and hyperlocalization, better television displays, mobile apps, colorful and more meaningful newspaper maps, and clear communications. They also consider the potential impact of weather forecasts to help people and businesses make the best decisions, thereby saving lives, protecting property, and increasing business efficiency and profitability.

Advances in communications technology and cell phones have expanded the timeliness, perceived accuracy, and value of forecasts and warnings through better displays and enhanced wording and communications.

Partnerships

The American weather enterprise, consisting of meteorologists working in government, universities, and commercial weather companies, has advanced the science of weather forecasting from general predictions for the next day or two to timely life-saving warnings, specific hyperlocal minute-by-minute forecasts for the next several hours, highly detailed forecasts for the next several days, and even daily forecasts 90 days out.

Until recent decades, all observational data came from government agencies. Today, commercial companies also provide unique datasets from lightning detectors, microsatellites, drones, crowdsourcing, mesonets, and in-vehicle sensors, stored and distributed via the internet and cloud technology.

The hyperbolic rise in data and how to best access, store, process, analyze, display, and utilize them will be a challenge for the weather enterprise.

As data from all sources grow exponentially and computer models improve, forecasts and warnings will become even more accurate, timely, and hyperlocal and extend further ahead. Yet the hyperbolic rise in data and how to best access, store, process, analyze, display, and utilize them will be a challenge for all parts of the enterprise.

The Forecast for Forecasting

With growth in the scale of data and the need for fast access to information, new ways of curating data will be necessary. Many off-the-shelf technologies will be challenged in balancing speed and cost. It will take creativity and innovation in data engineering to meet the expectations of consumers and businesses.

Perhaps the greatest change in weather forecasting has been from meteorologists’ reliance on their experience, education, and knowledge of atmospheric -processes to their increasing dependence on computer models and parameters derived from those models. In the future this trend will accelerate, with algorithm weighting of hundreds of computerized forecast models, using artificial intelligence to choose the best outputs from each forecast source. Combinations of algorithm-based input will vary with the weather, parameters, location, seasonality, model performance, and other factors, to achieve the greatest accuracy and best impact depending on the purpose of the forecast.

Over the next several years, the greatest benefits from continually improving weather forecasts are likely to come from the incorporation of forecasts in new technologies and systems, such as wearables, energy management, predictive analytics for business, intelligent homes, smart cars, agriculture, supply chains, and transportation.

Weather and climate events have an annual impact of trillions of dollars on the global economy and remain a top concern of leaders worldwide (WEF 2020). Companies are increasing their proactive use of weather information to analyze their operational data and key performance indicators to identify relationships and turn them into predictive analytics at a localized level to reduce weather impacts and drive improved results.

Each sector of the weather enterprise will continue to play an important role. The NWS provides the basic data and models; universities conduct research and train future meteorologists; and the commercial sector, the primary interface to end users and driven by competition among companies, leads innovation in weather forecasts and warnings, hyperlocalization, detail, personalization, presentation, communication methods, usefulness, and the impact of weather on people’s health and activities.

Conclusion

In 1900 a hurricane struck Galveston, Texas, and killed 12,000 people. Since then loss of life from flooding, lightning, hurricanes, tornadoes, blizzards, and other severe weather has decreased in this country by an order of magnitude because of the success chronicled by Nate Silver. That trend will continue because of the increasing accuracy and precision of weather forecasts and warnings effectively communicated and better tailored to the specific needs of people and businesses. This is already resulting in rapidly falling fatality and injury tolls from severe weather events because of better decisions by people and companies and more profitable outcomes.

The American weather forecasting success story is one of progress and an exemplary public-academic--private partnership. With further advances in the science of meteorology and the unique contributions of each sector of the weather enterprise, progress in weather forecasting will increasingly benefit the nation’s economy, public safety, and quality of life for everyone.

References

Silver N. 2012. The Signal and the Noise: Why So Many Predictions Fail—but Some Don’t. New York: Penguin Group.

WEF [World Economic Forum]. 2020. Global Risks Report. Geneva.

About the Author:Joel Myers is founder, CEO, and chair of AccuWeather.