Video recording, resources and tips from the CASW Connector Chat

Climate change is creating conditions for extreme weather as well as disasters like fires and floods. At the same time, the federal databases – and even data archives – that inform understanding and predictions about such events are disappearing.

This CASW Connector Chat on July 16, 2026, co-organized with the American Statistical Association, walked through the pipeline from data analysis to media publication with a statistician, a climate scientist and a journalist. They addressed how the changing data landscape is affecting the questions they can ask and answer and offered tips for reporting on this dynamic and challenging beat.

Speakers:

Christopher K. Wikle, the Curators’ Distinguished Professor of Statistics at the University of Missouri. He specializes in spatio-temporal statistics, with primary applications to geophysical processes, weather, climate, complex biological processes and the environment.

Jason Samenow, the founder and chief meteorologist for Capital Weather, a company dedicated to providing trustworthy D.C. area weather coverage. He was previously The Washington Post‘s weather editor and also worked as a climate change analyst at the U.S. Environmental Protection Agency.

Sarah Kaplan, a climate reporter at The Washington Post covering humanity’s response to a warming world. Her reporting focuses on extreme weather as well as climate change impacts on food, water and biodiversity. She was part of the team of Post journalists recognized as a finalist for the 2025 Pulitzer Prize in National Reporting for coverage of Hurricane Helene’s human and environmental toll. She previously reported on Earth science and the universe at the Post.

The event was moderated by Regina Nuzzo, a professor of mathematics and data science at Gallaudet University, co-host of the Normal Curves podcast, and a science journalist specializing in statistics and the scientific process.

Below you’ll find a recording of the Chat, takeaways and tips from the presenters.

This tip sheet is a living document; to share additional resources, contact the CASW Connector team at connector@casw.org. And sign up for our newsletter to receive updates about future Connector Chats and resources for science writing.

Tip Sheet: Reporting on Climate Disasters

Editor’s Note: On the same day as this Connector Chat (July 16, 2026), the National Academies of Sciences, Engineering and Medicine released a new report, Extreme Weather Events and Climate Change Attribution; here’s the press release.

By Sarah Kaplan, The Washington Post

Any time a major weather disaster strikes, your audience (and often editors) will ask, “is this because of climate change?”

If you’ll forgive me stepping on my soapbox for a moment, my take is that journalists should avoid saying that climate change “caused” a disaster — weather is the product of a lot of factors, and disasters can and did occur even before humans started pumping our atmosphere full of greenhouse gases. Plus, how disastrous an event becomes is often determined by other human factors — building codes, evacuation planning, emergency response, etc.

But there’s no denying the extremely robust research linking human-caused warming to many of the worst kinds of weather extremes: heat waves, hurricanes, wildfires, flash floods, etc. Scientists agree these events will occur more often on a hotter planet, and that they will worsen as the world continues to warm. Every single heat wave, for example, is starting from a higher baseline and has the potential to reach unprecedented temperatures because of climate change. For this reason, I don’t shy away from referring to these events as “climate disasters” or “climate extremes” — even if scientists have not yet quantified exactly how much a given event was influenced by climate change, we know enough to know that influence was there.

Other phrasing I use often:

  • “Climate change makes these kinds of events more frequent and intense.”
  • “This event bears the characteristic fingerprints of climate change.”
  • “Rising global temperatures set the stage for this event.”
  • “Experts say human-caused climate change contributed to this disaster.”

The most authoritative resource for research on how extreme events are influenced by warming is the United Nations’ Intergovernmental Panel on Climate Change. The most recent assessment report includes a chapter on weather and extreme events that examines how climate change will affect heat, heavy precipitation, river floods, droughts, extreme storms, fire weather, etc.

For less technical explanations, Climate Central’s extreme weather toolkits include science explainers and other resources to understand climate connections to extreme heat, wildfire, tropical cyclones, coastal flooding and other types of weather.

One type of extreme weather where the research is less settled is tornadoes. I find this explainer from an Ohio State tornado researcher helpful for understanding why tornadoes are so complicated and what the science is beginning to tell us. My former colleague Scott Dance also does a good job in this piece for the Post (may require subscription).

Attribution science

The emerging science of climate attribution, which uses statistical analysis to determine what role warming played in a given disaster, is getting stronger every year. Attribution studies can often quantify whether climate change made an event more likely or whether it intensified its impacts — data that can bolster your reporting and strengthen your audience’s understanding of the connection between greenhouse gas pollution and its devastating effects.

Some of my go-to sources for attribution research are:

  • World Weather Attribution network: This group, based out of Imperial College London in the U.K., uses peer-reviewed methods based on climate modeling to perform rapid attribution studies of high-profile weather events. Their analyses are useful because they publish while events are still in the news (unlike studies that are often published in peer-reviewed journals two years after the fact).
  • ClimaMeter: Another rapid attribution research group, based out of the Institute Pierre-Simon Laplace in France. They use a different methodology than WWA that compares modern events to historical analogs to assess the influence of climate change.
  • Climate Central: A U.S. based nonprofit that tries to improve climate communication, this group has developed several tools that can help journalists make the connection between climate and extreme heat, humidity, tropical cyclones and ocean temperatures. Their Climate Shift Index includes a map that visualizes how warming has influenced the likelihood of that day’s temperatures around the world — making it helpful for real-time reporting on heat waves and other climate-influenced phenomena.
  • Climate Attribution: This project based out of Columbia University maintains several databases of published research connecting climate change to changes in the Earth system, specific extreme events and the impacts of warming on humans. They also have a “source attribution” database of studies identifying how much various emitters have contributed to warming.
  • Carbon Brief’s map of climate attribution studies: This dataset visualizes the results of nearly 1,000 studies on extreme weather events and trends. Note that just because there are more from North America, Europe and Asia doesn’t mean these regions are seeing greater climate impacts — it just means more research is being done on those places. But if you are covering a particular part of the world and want to know what kinds of attribution research have been done there, this can be a helpful tool.

Impact attribution studies are especially useful for helping your audience understand how climate change is already hurting people. For example, when covering the health effects of an ongoing heat wave, you could cite this study finding that climate change was responsible for more than one third of heat-related deaths around the world between 1991 and 2018 — illustrating that more people are dying from heat exposure because of human-caused warming. Or if you’re writing about wildfire smoke, you could link to this study projecting that climate-fueled increases in fires could lead to a 73% increase in the number of people who die from smoke pollution.

Other sources of useful climate and weather data

  • Climate at a Glance: This site from the National Centers for Environmental Information has a lot of useful interactive tools for understanding how climate has changed over time at the local, national and global levels. I often use this as a data source for stories I’m already working on and as inspiration for future reporting. For example, if I was writing about springtime drought in Arizona and wanted to show how it had gotten worse since the pre-industrial era, I could use the state time series tool to plot the Palmer Drought Severity Index between 1895 and 2026, revealing a dramatic downward trend. On the other hand, if I knew I wanted to write about the effects of drought and was trying to determine the best place in Arizona to report from, I might use the “departure” and “county mapping” options to see which part of the state was experiencing the most extreme drought anomalies.
  • NOAA’s Daily Weather Records Summary: This tracker shows how many new weather records have been set in the past week, month and year. It can be especially helpful if you’re covering temperature and rainfall extremes. For example, you could say, “32 U.S. weather stations set all-time temperature records over the past seven days” (a real stat as of July 8!) to underscore the severity of a recent heat wave.
  • Climate Pulse: This tool from Copernicus, the European Union’s top climate agency, provides visualizations of sea and air temperatures going back to 1940 — very helpful for reporting on global temperature records and trends.
  • Berkeley Earth: This nonprofit produces a ton of great climate research and visualizations, but one of the most helpful for reporters is a dataset of cities, states and countries that tells you how much average temperatures have increased at each location. In addition, they have a new synthesis tool that also includes climate projections, which can be useful if you’re reporting on impacts in a specific place or if you’re a local reporter trying to help your audience understand what climate change will bring.
  • NOAA/Climate Central Billion-Dollar Disaster database: This database, formerly maintained by the National Oceanic and Atmospheric Administration, tracks the rising toll of extreme weather and climate disasters since 1980. NOAA stopped updating the dataset in 2025, but Climate Central has taken up the mantle. Just remember that disaster costs are determined not only by the severity of the weather event, but the degree to which people are exposed and vulnerable. A big part of why hurricanes and wildfires have become so costly in the U.S. is the fact that millions of people have moved into the most hazard-prone areas (may require subscription).
  • First Street’s Risk Factor tool: Formerly the First Street Foundation, now a for-profit company, this group produces sophisticated climate models that allow you to look up flood, fire, wind, heat and air pollution risk for any property in the U.S. They also produce a lot of interesting reports on shifting climate hazards. Though most of their more sophisticated analyses are behind a paywall, they often partner with journalists, so it’s worth reaching out to them if you have a story you think their data could be useful for.
  • NOAA’s Sea Level Rise Viewer: This tool, which allows you to visualize U.S. flood impacts at different degrees of sea level rise, is useful for reporting on coastal flooding and storm surge.
  • Flood Inundation Mapping Services: This is a new, experimental tool developed by NOAA’s Office of Water Prediction that allows you to see what areas will likely be inundated from a forecasted river flood — kind of similar to the sea level rise viewer. NOAA has a PDF explaining how the tool can be used (note that it’s still under development and not available for the whole country yet). But it could be quite powerful, especially for local reporters who want to be able to tell their audiences about specific roads or neighborhoods that are expected to flood.
  • Climate.us: After the Trump administration cancelled the U.S. Global Change Research Program and shuttered its Climate.gov website, a nonprofit staffed by several former federal employees and contractors has tried to resurrect some of the program’s work. Note that many of the datasets and tools are only current as of 2025, when the federal government stopped updating them. But the nonprofit says it aims to restart and expand data collection with the help of independent researchers.

Statistical Methodology for Climate Data Analysis

By Chris Wikle, University of Missouri

Statisticians have worked alongside climate scientists for years to develop methods that can provide inference on weather events and climate, explicitly considering uncertainties associated with data, model and forcing (e.g., Berliner et al., 2000; Tebaldi et al., 2005; Tebaldi and Knutti, 2007). The development of new methods for climate science continues to be an area of active research in the statistics community (see Ensor and LaLonde, 2026, for a recent comprehensive overview). In addition to general considerations of uncertainty quantification (Sain et al., 2025), there are methodological challenges with climate data and processes related to the inherent spatio-temporal dependence and complex multiscale process interaction that is present in these systems (e.g., see Cressie and Wikle, 2011). Specifically, there have been significant methodological challenges dealing with sampling networks that are distributed preferentially (e.g., sampling locations depend on the value of what is being measured; Diggle et al., 2010; Amaral et al., 2024) and accommodating the fact that extreme events and nearby locations are dependent (e.g., see Naveau et al. 2020; Huser and Wadsworth, 2022). Indeed, accounting for dependence in extreme events can make an important difference in event attribution (e.g., Zhang et al. 2022; Risser et al. 2025, 2026) and increasingly, neural inference methods and data-driven causal inference methods are being used to scale these methods to practical scales (e.g., Zhang et al., 2026; Risser et al., 2025b).

Examples of Sarah’s Reporting Using Climate Attribution

These are gift links.

Coverage of heat events

  • July 2, 2026, news story about the July 4 heat wave that references WWA study calling it “virtually impossible” without climate change.
  • June 25, 2026, story on deaths in Spain from June extreme heat event.
  • Here’s another story I wrote after the 2021 Pacific Northwest heat dome, which incorporated attribution science as well as a variety of other data sources to strengthen the reporting — an analysis of mortality data, looking at maps of the urban heat island to find the worst-affected zip code — but also focused on a real person affected by a disaster. I think it’s an example of how you can use data and human storytelling together to have the strongest possible impact.

Examples of attribution science being used in attempts to hold polluters liable for climate impacts

Hurricane Helene and Floods Above coverage

  • This October 1, 2024, story incorporated multiple kinds of attribution research to strengthen the message — that climate change is increasing flood danger in places that didn’t previously perceive themselves to be vulnerable.
  • The Washington Post used a huge variety of data sources for our 2024 investigation of how rising global temperatures are allowing record amounts of water vapor to flow through the atmosphere, contributing to an intensification of floods.
  • July 20, 2025, story on the gap between the very accurate forecasts about Hurricane Helene and people’s preparedness to evacuate.