In this edition of Author Talks, McKinsey Global Publishing’s Ramya D’Rozario chats with Thomas E. Weber, journalist and educator, about Cloud Warriors: Deadly Storms, Climate Chaos—and the Pioneers Creating a Revolution in Weather Forecasting (Macmillan Publishers/St. Martin’s Press, Summer 2025). Weber explores the role of improved forecasting and enhanced communications in building a weather-ready economy. By viewing weather literacy as a leadership imperative, leaders can leverage the power of technology and data to adopt smart forecasting tools, inform decision-making, and forge a more resilient future. An edited version of the conversation follows. You can also watch the full video at the end of this page.
What is weather literacy, and why is it important?
The more I researched forecasting, the more I began to think about weather literacy. I want to encourage everyone to learn more not only about different weather dangers but also about how to be a smart consumer of the forecasts.
Being a smart consumer of forecasts means different things for different people. For example, it’s important to know the difference between a tornado watch and a tornado warning. I grew up in Michigan, where tornado warnings were our biggest worry. Now I live in New York City. Hurricanes are a concern there, yet wildfires also occur at the city limits during dry weather. People in the Pacific Northwest are accustomed to rain, but in 2021, they experienced a deadly heatwave. Being weather literate in our era really means learning about and preparing for all kinds of weather, and developing plans for how you will react if you’re in danger.
What does weather literacy mean for businesses or organizations?
A long-range forecast is one example of weather literacy in business. Long range refers to seasonal forecasting. For a business, the weather could affect demand for products or services, or it could affect supply chains. Unusually warm or cold weather translates into different energy usage as people turn to heating or air-conditioning. Long-range forecasting can be very useful for business. Short-range forecasting involves realizing the potential impact of dangerous weather on your employees and your operations. You may have workers who work outdoors, or you may have a facility that’s in the path of severe weather.
As I speak with company representatives, I’ve learned that one good way to encourage weather literacy within the enterprise is to have someone on the team designated as a “weather guru.” More often, companies bring a trained meteorologist onto the team. The goal isn’t just to have someone who does weather forecasting. For a business, the objective is to have someone who can translate that forecast information into impact and into decision-based information that managers need and can use.
Being weather literate in our era really means learning about and preparing for all kinds of weather, and developing plans for how you will react if you’re in danger.
How has weather tracking evolved over the ages?
The evolution of weather forecasting is amazing, and it goes back centuries. It starts with people who tried to predict conditions for agriculture. If you consider modern weather forecasting, where we have decent [forecasting] skill, there are three “key waves” for skill enhancement.
The first wave in modern forecasting occurred in the 19th century when the telegraph emerged. The telegraph enabled people in the western part of the US to send a message to the East about the weather that was on the way, since big weather patterns predominantly move from the West to the East.
The second wave of modern forecasting arrived with the space age. In the ’50s and ’60s, very important technologies suddenly emerged: satellites, radars, and computers.
Now we’re entering the third wave of modern forecasting. There are several advances underway that will make forecasting even more accurate. Artificial intelligence is one technological advance. Another technology is the Internet of Things, which refers to devices that have sensors and are connected to networks. For example, your phone can indicate barometric pressure. Scientists are working on using that information from millions of telephones to make the weather forecasts more accurate.
An important component of this third wave is that weather scientists increasingly look to the social sciences to determine how to communicate warnings more effectively. A very important challenge is to make sure those forecasts translate to good outcomes.
For a business, the objective is to have someone who can translate that forecast information into impact and into decision-based information that managers need and can use.
How have start-ups and AI affected the field of weather forecasting?
There’s no question that AI is quickly becoming an important tool for forecasters. There is so much excitement about AI in weather forecasting. But it’s important to remember that there’s a public-safety element in weather prediction. It will take time for forecasters to gain experience with these systems to have confidence in them.
AI also helps interpret data from the Internet of Things. These new types of data sources help forecasts become hyperlocal. In the book, I reference a start-up company that uses simple SMS text messages to send basic forecast information to farmers in Ghana. These messages help guide their planting and fertilizer use. Basic information about when it will rain—so that people don’t apply and waste fertilizer that will wash away—is making a huge difference for farmers.
Here in the US, start-ups play a big role in providing specialized forecasts, some of which are aimed at very niche applications. We see companies that provide wind forecasts for drone operations. We also see companies that provide technology to help self-driving cars “understand” road conditions ahead, based on a local map of the weather.
What future tech trends do you see in weather forecasting?
When I look at the trends in tracking the weather and improving predictions, I see not only steady progress but also reinforcement, as we incorporate more artificial intelligence into the system.
Technology gives additional support to our advanced warnings of danger, whether it’s a tornado or a hurricane, to improve our forecasting capability. I would like to see better use of personal technology for public alerts about the weather.
Our phones know so much about us. Wearable devices increasingly add to that knowledge base. Yet there’s real potential for personal technology to help individuals make better use of weather information.
How can the new generation of weather forecasters and scientists be better prepared in an AI-driven future?
I have spent time with undergraduate students who want to pursue meteorology. Some of them would like to become forecasters for television or for the National Weather Service. Other students want to conduct research, and still others are looking at the growing number of corporate meteorology-related jobs. For example, it would be pretty surprising for insurance companies not to have a good meteorology staff.
No matter the pursuits of the next generation, students are very aware that they need to understand computers, coding, and AI, and that these fields will be critical for their future. One challenge that some young people have voiced is learning how to talk about climate and weather in a way that keeps some people from automatically tuning out.
What are some ways to bridge the gap between forecasting and emergency action on the ground?
Bridging the gap between forecasting and action on the ground is one of the biggest questions in today’s weather world. There has been talk over the last couple of decades about how the forecasts keep getting better, but the outcomes don’t always match the quality of the forecasts.
One way to bridge that gap is better communication. That means closing the gap between the technical language of meteorology and translating forecast information into impact and decisions. This has actually become a priority at the National Weather Service. Local forecast offices work more closely with local officials, not just to share the forecast but to really provide decision-based information.
I can share an example of how we can bridge that gap. There’s a new initiative created by the New York State government in Albany, known as the State Weather Risk Communication Center. It’s a small staff of trained weather experts who look at the forecasts and then translate that information into what public officials need. They have a color-coded system of green, yellow, and red. Their “customers” include every entity, from the Department of Transportation, which gauges road conditions and the need to apply salt, to school superintendents who make decisions about school closures, to first responders who want to know about the possibility of dangerous conditions that might result in rescues. It’s very innovative to see a group dedicated to turning forecast information into decision-based information.
Bridging the gap between forecasting and action on the ground is one of the biggest questions in today’s weather world.
Why did you write this book?
Initially, I wanted to better understand how forecasting works and how good it is. We all look at the weather every day.
At its core, my interest as a journalist has always been the intersection of technology, science, and society. As I delved more deeply into forecasting, I realized that there was an urgent story here. That story is about the work needed, not only to improve forecasts themselves, particularly for extreme weather, but also to bridge that gap between better forecasts and better outcomes.
I wrote this book to help everyone feel more weather literate—and to help leaders, whether they’re in business, government, or any organization, think more deeply about forecasting and communications.
I realized that there was an urgent story here. That story is about the work needed, not only to improve forecasts themselves, particularly for extreme weather, but also to bridge that gap between better forecasts and better outcomes.
Was there anything that surprised you in the research, writing, or response?
One thing that surprised me repeatedly is that forecasts are much more accurate than most people generally believe. Hurricane forecasts have become amazingly accurate in forecasting the track of a storm days out.
Now, public officials make decisions about evacuations with lead times that would’ve seemed unbelievable 20 years ago. And we have confidence in those forecasts. Ultimately, the average person notices when a forecast is off but not when it’s accurate.
I hope readers come away with a better understanding of just how good our forecasts are and how much they’re improving. Everything involves probability. When you hear that there is a 40 percent chance of rain, you have to accept that there’s a 60 percent chance that it won’t rain. People face difficulty in working with probabilities, but that acceptance is very important in understanding weather forecasts.
What constitutes the ‘weather enterprise’?
A funny phrase that people in the US weather community use to describe three legs of a tripod represents how forecasting works in the United States.
- The federal government and the National Oceanic and Atmospheric Administration (NOAA). The government provides significant infrastructure and public-safety forecasting, and it sponsors the weather satellites and radars we need.
- Work at research universities. Research helps advance our understanding of weather. At the University of Oklahoma, for example, I was able to venture out with storm chasers into tornadoes and gather data on what causes them to form. That work is important for improving our predictions.
- Private enterprise. This involves companies that engineer better radars and build satellites. They also provide custom forecast solutions for special applications.
These three legs of the weather enterprise naturally work together, but they also compete in a relatively healthy way. Cooperation between these three components has been an explicit strategy in the US for several decades. It has produced advances that make US forecasting enviable.
Describe your experience while conducting research for this book.
One of the great pleasures of writing a book is being able to take the time to do in-person reporting.
The most exciting experience was storm chasing through Oklahoma with researchers from the National Severe Storms Laboratory [NSSL]. Twisters, a movie that came out fairly recently, is about getting instrumentation close to severe thunderstorms and tornadoes as they form and gathering related data. I found myself in the back of a pickup truck that contained mounted instruments, witnessing hail pounding down on us and the darkest sky I’ve ever seen.
I’m not an unusually brave person. Yet the researchers are so skilled, and they monitor the conditions so closely, that I was not afraid at any point. That was an amazing experience.
Another experience—a visit to San Diego’s public utility company—showed me the weather enterprise concept, the three legs of the tripod, working together.
They have established a very robust in-house weather forecasting capability. In Southern California, wildfires are such a danger. I had the chance to visit the weather experts at San Diego Gas & Electric to observe their own network of weather stations and their own special weather model. Then I visited the California Department of Forestry and Fire Protection [CAL FIRE]. CAL FIRE has a different type of weather forecasting and software that it uses to help guide firefighters once a fire is underway. This is another specialized tool that has been developed by a start-up in the weather world.
I went to the San Diego office of the National Weather Service, where the forecasters have a life-and-death responsibility of issuing red flag warnings. These warnings inform people when conditions are most dangerous, when to take care of ignition sources, and when to be prepared to evacuate.
I had the chance to see an entire ecosystem of forecasting and electric utility—the people who have to fight the fires and the public safety officials who have to keep the public abreast of the warnings. They all speak constantly with each other. This is something the public doesn’t see. Yet it’s crucial to the work of keeping people safe and keeping our economy sound as well.
What would be one takeaway from your book for readers?
If there’s one takeaway from this book, I hope it’s a desire to become more weather literate. I thought I was pretty savvy about the weather until I began conducting research. I pay a lot more attention to the weather now.
Here in New York City, if I hear about a flash flood possibility, I think twice about when I might want to ride the subway. If I’m outside the city, I think about where I would go [for safety] if the waters began rising.
One can apply that perspective across the board. So I hope readers will appreciate not only the skill and accuracy of modern forecasting but also the importance of being aware of how to act on and use that information in their daily lives.
Twister versus Twisters, which one is more accurate?
Both the movie Twister and the more recent movie Twisters are pretty beloved in the weather community, even if cinema overdramatizes some aspects of storm chasing.
Twister, the movie with actress Helen Hunt, was accurate in some ways because of the concept of trying to get an instrument inside the middle of a storm. The characters have a big oil drum filled with probes that they’re trying to maneuver into the path of a tornado.
There’s a real-life analogue of that project at the National Severe Storms Laboratory. The lobby of the laboratory in Norman, Oklahoma, proudly displays the movie prop and the real-life version of the tool that NSSL worked on. There was a lot of science in the original Twister. The newer movie, Twisters, accurately reflects the idea of having more sophisticated radars to help gather information about tornadoes. In the real world, that process probably doesn’t look like people running around trying to get close to the storm and setting up tripods.
A long-range plan for the National Weather Service involves replacing the weather radars that we currently use, which date back to 1988, with more modern types of radar, such as what you would find on a US Navy ship.
The need to better image the storm is an example of real science at the heart of the process.
