A team that includes Derek Lemoine of the University of Arizona finds that improving short-term temperature forecasts could reduce U.S. heat mortality by 18% to 25% in the year 2100. The study says timely and accurate warnings give people time to change plans and protect their health when extreme heat is likely.
To reach these results, the researchers used day-ahead National Weather Service forecasts for the contiguous United States from the summer of 2004. They combined those forecasts with observed climate data from Oregon State University’s PRISM Climate Group and county-level mortality records from the Centers for Disease Control and Prevention. After removing deaths caused by non-weather factors, they found that forecast accuracy is key. The greatest risk came when forecasts underestimated hot conditions.
The team surveyed professional meteorologists in early 2025 about future forecasting technology and used those responses to build optimistic, pessimistic and perfect-accuracy scenarios. Depending on technological progress and warming, better forecasts could largely offset projected increases in heat-related deaths, while declining investment could worsen outcomes.
Difficult words
- forecast — a prediction of future weather or conditionsforecasts
- mortality — the number of deaths in a population
- timely — happening at the right time
- underestimate — to think something is smaller or less severeunderestimated
- contiguous — connected together in one continuous area
- offset — to reduce or balance the effect of something
- scenario — a description of a possible future situationscenarios
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- How could better short-term forecasts change what people do during very hot weather?
- Should governments increase investment in forecasting technology to reduce heat deaths? Why or why not?
- What actions can people take when they receive a heat warning in their area?
Related articles
Latin American groups build AI to study gender violence
Groups in Latin America create open, local AI tools to study gender inequalities and violence. Projects like AymurAI search court documents, protect sensitive data on local servers and help governments and civil society with evidence.
How civil society adapts to AI and surveillance
In April 2026 Global Voices and IRIS published ten case studies from the Global Majority showing how civil society groups respond to AI and algorithmic platforms. Responses include co‑opting, countering and innovating, with local and cross‑border strategies.
