Innovations in Weather Forecasting for a Changing Climate

The Environmental and Energy Study Institute (EESI) held a briefing about the role of weather intelligence technology in helping predict and prepare for extreme weather events. As greenhouse gas emissions rise, climate change is triggering and intensifying extreme weather events like heat waves, droughts, and hurricanes – as well as making them harder to predict. This briefing highlighted how advanced forecasting technology can face the challenges of climate change and enable improved warning systems for weather events. 

Nonprofit, government, and private sector panelists discussed the current state of weather intelligence technology and federal policy opportunities for improving weather intelligence, including the promotion of public-private partnerships.

Highlights

KEY TAKEAWAYS

Meteorologists are trusted by the public and are at the frontline of climate communication. The National Weather Service works with emergency managers and decision-makers across the country to prepare for weather events and improve response in communities that need the most help. Weather intelligence uses forecast data to provide a customized set of recommendations on how users should respond to the forecasted weather. Artificial intelligence (AI) is transforming the weather and climate landscape. Information derived from AI can be used to improve early-warning systems, know when and where to relocate people, refine insurance planning, and allocate disaster resources. Investment in weather forecasting saves lives. Due to the collaborative nature of the industry, investment in the public sector supports the entire industry, while underfunding in the public sector limits the entire industry.

 

Rep. Eric Sorensen, U.S. Representative (D-Ill.)

• Sorensen served as a local TV meteorologist in central and northwestern Illinois for over two decades before joining Congress.
• Meteorologists are trusted as a source of weather and climate information, which helps people make educated decisions like when to stay off the road because of winter weather and when to be aware of heat waves.
• Extreme droughts and floods make farming more difficult today than in the past, even with new technology. Communicating weather and climate information to farmers and supporting them in this work is critical.
• Sorensen was the first meteorologist to be involved in reauthorizing the National Oceanic and Atmospheric Administration (NOAA) when he helped draft the Weather Act Reauthorization of 2023 (H.R. 6093).
• Climate change is leading to unprecedented extreme weather, so it is critical to provide adequate funding to weather forecasting services that inform the public.

 

Q: In your two decades as a meteorologist, did new tools become available to make informing the public easier?

Sorensen

• The program Climate Matters, run by the nonprofit Climate Central, gives broadcast meteorologists the tools to localize the impact of a changing climate to their specific television market. Audiences did not connect with conventional climate imagery like ice melt in the Arctic, so Climate Matters provided meteorologists with the tools to communicate how climate change would affect their audiences directly, which increased engagement.

 

Michelle Mainelli, Deputy Assistant Administrator, National Weather Service

• Climate change is creating weather extremes. In 2023, there were 28 weather and climate disasters in the United States that each cost one billion dollars or more. These disasters are becoming more frequent.
• The National Weather Service (NWS) works with emergency managers and decision-makers across the country, but the agency has less of a presence in many vulnerable communities, which is an issue the NWS is attempting to address.
• To ensure successful outreach, NWS customizes warnings to specific locations and reports through a variety of media.
• Weather reporting is an active role. It requires meteorologists to travel to where events are happening and work with decision-makers in creating action plans for weather events.
• The NWS is following its 10-year Transformation Roadmap to further prepare for weather events and improve response in communities that need the most help.

 

Thomas Cavett, Vice President of Government Affairs and Strategy, Tomorrow.io

• Tomorrow.io is a weather intelligence software company focused on translating weather information into individualized suggestions. Tomorrow.io’s nonprofit partner, TomorrowNow, has demonstrated that access to weather intelligence information can dramatically increase productivity and crop yields for small farmers in the Global South.
• Most current weather forecasts inform users about temperature, precipitation, and other conditions, but they are less clear about what that means for a given individual or business. Weather intelligence uses forecast data to provide a customized set of recommendations on how users should respond to the forecasted weather.
• The use of logic-based artificial intelligence (AI) in combination with human meteorologists makes it possible to customize information without requiring meteorologists to be available to answer questions for every employee of a company.
• A lack of conventional, ground-based radar equipment in developing countries and in the open ocean limits forecasting data in these areas. This leads to significantly more infrastructure damage from extreme weather events in developing countries and can limit the effectiveness of U.S. forecasting since many major storms form over other areas of the world. U.S. military operations abroad can also be impacted by a lack of forecasting infrastructure.
• Tomorrow.io is launching 30 satellites that will use microwave sounding alongside conventional radar to collect high-quality weather data covering the entire globe within 30 minutes, four to six times faster than current state-of-the-art technologies.
• Three main types of satellites will make up the Tomorrow.io constellation. The first type is the Time-Resolved Observations of Precipitation structure and storm Intensity with a Constellation of Smallsats (TROPICS), which is technology developed by the National Aeronautics and Space Administration (NASA) and the Massachusetts Institute of Technology (MIT) Lincoln Lab. The second type is the Pathfinder radar satellite, two of which are in orbit today. The third style of satellite is the operational radar, which is larger and has a scanning system that covers a wider swath of Earth at any given time than the Pathfinders.

 

Pierre Gentine, Maurice Ewing and J. Lamar Worzel Professor, Columbia University; Director, Learning the Earth with AI and Physics (LEAP), National Science Foundation Science and Technology Center

• AI is transforming the weather and climate landscape. Columbia University, New York University, the University of California Irvine, and the University of Minnesota are partnering through Learning the Earth with AI and Physics (LEAP) to use AI to improve climate models.
• Funding for climate adaptation, such as for relocating vulnerable populations or reorganizing cities facing flooding, is inadequate compared to funding for reducing greenhouse gas emissions.
• The cost of damage from climate events has been rising. However, populations and industries in the most impacted areas have not relocated, calling for new forms of climate adaptation.
• AI can enhance weather forecasting. For example, AI is used to quickly predict flooding, which is useful in warning public and private stakeholders. AI can also use satellite data to monitor disasters such as wildfires and hurricanes to predict their trajectory. This information can be used for early-warning systems, relocating populations, insurance planning, and allocating disaster resources.
• AI weather forecasting is operational, but it is not widely used right now in the United States. Europe is leading on implementation.
• AI forecasting can predict weather from one day to two weeks. It cannot predict events further in the future, especially extreme events.
• The LEAP partners are also looking at climate projections and trying to use AI to improve the current physical models like those produced by the Intergovernmental Panel on Climate Change. They have found that AI can build up the physical models to address known deficiencies in the projections and decrease uncertainty.

 

Dan Stillman, Co-Founder and Meteorologist, Capital Weather Gang, Washington Post

• Unpredictable weather patterns lead to inaccurate forecasts surrounding extreme weather events, which puts lives at risk.
• The world is seeing an increase in extremely warm days due to climate change, and current models are frequently unable to accurately predict these temperatures. This can be dangerous, particularly in the summer since people relying on forecasts to guide their behavior may be unprepared for the real, significantly more extreme, temperatures.
• In October 2023, Hurricane Otis was predicted to make landfall in Mexico as a tropical storm but instead rapidly intensified to a category five hurricane. Existing models are less able to predict this rapid intensification, which causes storms to dramatically increase in strength in a short period of time and is tied to climate change.
• Forecast accuracy has improved over the years, but progress is slowing down. New models and methods of gathering data are needed to reverse this trend.
• AI trained on archived historical data can provide faster and cheaper modeling, which can be valuable in combination with traditional models.
• It is crucial for data to be communicated effectively to the public, meaning social science needs to be involved in improving meteorology.
• Investment in weather forecasting saves lives. Due to the collaborative nature of the industry, investment in the public sector supports the entire industry, while underfunding in the public sector limits the entire industry.

 

Kaitlyn McGrath, Meteorologist, WUSA9

• Meteorologists are trusted by the public and are at the frontline of climate communication. Studies show many people appreciate local television that reports climate impacts on local weather and extreme weather events. Social and digital media have expanded the roles of meteorologists beyond local television.
• Many people recognize that climate change has impacted their life when communicators use personalized examples, including naming specific events which are attributed to climate change such as wildfire smoke, flooding, and the early blooming of certain flowers like cherry blossoms.
• Media meteorologists work with the NWS, NOAA, NASA, and local universities to inform the public on climate change.
• Climate Central releases free weekly graphics that quantify the impacts of climate change. These are used by solution journalists, who connect climate issues to climate action.
• For example, Climate Central released a graphic that detailed the winter temperatures required for fruit trees to produce fruit (cold temperatures help fruit trees rest, preparing them for spring growth). These cold temperatures are not consistently being met because of climate change. WUSA9 connected with Virginia Tech scientists developing solutions to this problem through genetic modification and developing topical ointment for the trees. WUSA9 aired this story paired with these possible solutions, while also explaining how this issue could impact grocery store prices.
• Climate Central produced another graphic which shows that Washington, D.C., trees prevent 276 million gallons of stormwater runoff, absorb one million pounds of air pollution, and remove 40,000 tons of carbon pollution annually. WUSA9 connected this information to a local nonprofit, Casey Trees, which plants trees in Washington, D.C., with the support of volunteers. The goal of the segment was to educate viewers and give people an avenue to be part of a climate solution.
• Not all climate data needs to be presented to people as a full report; sometimes climate data is most effectively presented when integrated with daily weather reports and life.
• The Climate Shift Index, a new product by Climate Central, compares how often a given temperature will occur in the current climate with the frequency of that same temperature in a climate without global warming. Products and graphics like these help the public understand the impacts of climate change on their local weather patterns.

 

Q&A

 

Q: How can we effectively communicate novel weather events that people have not experienced before, like extreme heat or derechos, to the public?

Mainelli

• Localizing the big picture to individual communities and visualizing what a weather event will look like in the context of their lives is important.
• Mapping efforts, such as the heat risk map by the Centers for Disease Control and Prevention that visualizes where high heat will occur in individual communities, make weather information more accessible.

Cavett

• Commercial businesses often receive broad, vague, or hyper-scientific weather information, which causes these businesses to make binary decisions on whether or not they should continue operations. Meteorologists must issue information in a more direct, action-focused format, such as suggest windows where operations can continue or precautions that must be taken.

Gentine

• Informing the public about how much a specific weather event can be attributed to climate change is important.

Stillman

• Communicating weather events through accessible avenues such as on local news or on news apps is a primary focus.
• Visualizing what damage caused by weather events will look like is essential to informative weather forecasting.
• Communicating the confidence of an extreme event occurring and when a high confidence is reached increases the efficacy of weather forecasting and event preparedness.

McGrath

• Telling people how to address and prepare for events in a direct manner and using visualization catered to specific community conditions will more effectively communicate weather information.

 

Q: Are you seeing the audience for your forecasts and modeling growing to include people like civil engineers, architects, and municipal officials, who are dealing with the increase in extreme weather events in their planning?

Gentine

• The U.S. Army Corps of Engineers is using climate and weather data more than in the past, but that work is limited in effectiveness by people choosing to remain in the path of extreme weather events. It is important to create incentives to convince people to leave dangerous areas.

Cavett

• As AI models become more mainstream, the weather industry will need to determine what infrastructure is needed to support these new models.
• There is a risk of unqualified people using AI models to create forecasts with limited accountability.

Mainelli

• Observation equipment is the foundation of weather forecasting. As observation tools evolve, it is essential to get the most out of new technologies.

 

Q: What is one thing that would change the game in terms of your ability to do what you do and the ecosystem of weather forecasting?

Cavett

• Contextualizing weather and specific events is the focus of work to restructure weather forecasting.

Gentine

• Centralized data infrastructure, especially on the cloud, will improve the ability to accurately forecast weather.

Stillman

• AI will allow weather forecasting models to run more times with slight variable changes at a lower cost, which will more accurately relay the likelihood of a weather event occurring and what the most extreme weather results could look like.

McGrath

• Climate attribution products such as the Climate Shift Index by Climate Central will allow the public to understand how much a specific weather event or condition can be attributed to climate change.

Mainelli

• Embracing emerging technologies will reshape both weather forecasting and community preparedness for weather events.
• Weather forecasts do not need to be perfectly accurate, but they do need to be communicated effectively.

 

Compiled by Emily Phillips and Meghan Tinnea and edited for clarity and length. This is not a transcript.

 

Photos