Bigger, wetter thunderstorms are one of the manifestations of climate change, and are altering risk calculations for millions of American homes and businesses.
The good news: Satellite imagery and AI can improve alerts of impending floods, as well as insurance options for at-risk properties.
In 2023 we‘ve seen rainstorms cause major flooding in New York City, Vermont, Florida, Texas and California. The impact of flooding rains has been even broader than the headlines suggest, and is out of whack with historic patterns.
Thunderstorms were the single costliest type of natural catastrophe for insurers in the first half of 2023, according to reinsurer Swiss Re. They now account for 68 percent of catastrophe losses, about twice the usual level. That includes a record $34 billion in insured losses in the U.S. in 1H 2023, with 10 different flooding events causing losses in excess of $1 billion.
“With severe thunderstorms as the main driver for above-average insured losses in the first half of 2023, this secondary peril becomes one of the dominant global drivers of insured losses,” said Martin Bertogg, Head of Catastrophe Perils at Swiss Re, one of the largest reinsurance companies.
The emergence of stronger storms is a byproduct of warmer air and seas. For every 1°C increase in temperature, 7% more water vapor is carried by the same air volume. Increases in water vapor provide more energy for severe weather – prompting descriptors like “rain bombs” and “atmospheric rivers.”
The Risk
Thunderstorm severity is important because it is changing the risk profile of regions not previously known for flooding, expanding the geography and impact of heavy rain events.
“Climate change is intensifying rainfall and floods,” according to the National Climate Assessment released last week, which noted that ”changes in precipitation amount and duration … have combined with land-cover change and increasing property values to increase overall economic damages from floods.”
A new model from First Street Foundation projects that heavy rainfall events will become more common, and about 20% of the country can now expect their “1-in-100-ear” severe flood will instead happen every 25 years. In 20 counties at high risk, it’s now every 8 years.
“The magnitude of the changes in expected rainfall intensity are startling for many areas in the United States, and it is important that Americans are fully aware of this consequence of climate change that can impact their lives and homes,” said Dr. Jungho Kim, First Street’s senior hydrologist and lead author of the peer-reviewed study.
“The data is startling, and it should be a wake-up call,” Chad Berginnis, the executive director of the Association of State Floodplain Managers, told The New York Times.
A key question is whether the current wave of infrastructure spending is keeping pace with the science on heavy rainfall events. As cities seek to update infrastructure to be more resilient, some are longing for better data from the government.
The First Street data indicates that 12.6 million properties nationwide face significant flood risks despite not being required by the federal government to buy flood insurance. The prospect of floods in new places will also test emergency managers’ ability to warn residents about impending events.
The Opportunity
Technology is creating better tools to meet these challenges. For instance, the combination of satellite mapping and artificial intelligence can enable earlier alerts of flooding rains.
Google says its free Flood Hub tool now extends local flood forecasts up to 7 days in advance, up from 48 hours last year. The longer forecast is enabled by AI refinement of weather data and real-time satellite imagery, which Google then runs through two AI models:
- A Hydrologic Model, which forecasts the amount of water flowing in a river,
- An Inundation Model, which predicts what areas are going to be affected and how deep the water will be.
“Governments, aid organizations, and individuals can use Flood Hub to take timely action and prepare for riverine floods,” Google said in a blog post about Flood Hub.
Earlier this year, the Department of Homeland Security began working with climate adaptation specialist Floodbase, using near real-time data on flood risk to help strengthen FEMA’s flood analytics for the National Flood Insurance Program (NFIP). Floodbase uses satellite imagery and hydrology-based deep learning models to model flood events.
Floodbase specializes in parametric insurance, which allows policyholders to buy protection based on a specific measure of a disaster – a parameter like total rainfall, wind speed, or magnitude of an earthquake. Traditional indemnity insurance compensates the insured based on an estimate of financial losses, which takes time. Since parametric insurance payouts are triggered by a specific condition, it has the potential to provide faster relief to storm victims.
One example of this concept is a community insurance pilot program in New York, where the Center for NYC Neighborhoods purchased parametric insurance for rainfall-related flooding, with payouts to be used for emergency grants to low- and middle-income residents facing recovery expenses.
“This pilot is designed to make financial resources immediately available and prevent households from spiraling into worse financial hardship,” said Carolyn Kousky, Associate Vice-President for Economics and Policy for the Environmental Defense Fund.
Subscribed
Resources (Nov. 21, 2023)
Thanksgiving and Climate Conversations
- What do you say when climate comes up over holiday dinner? One of our favorite newsletters, Talking Climate with Katharine Hayhoe, offers tips for better conversations while explaining and “depolarizing” climate change. Hayhoe has a section in this week’s newsletter about having constructive climate conversations at the holiday table.
Links
- I’m a Climate Scientist. I’m not Screaming Into the Void Anymore: This essay by Dr. Kate Marvel was widely shared across social media this week. Marvel, a climate scientist at Project Drawdown, shares the sense of optimism she gained from last week’s National Climate Assessment (she was a lead author). A worthwhile and hopeful read.
- Polar Ice Researchers Issue New Warning on Sea Level Rise: The report warns that global average temperatures settle at 2 degrees Celsius above the preindustrial baseline could translate into more than 40 feet of sea-level rise — “a melt that would take centuries and reshape societies across the globe.”
- World Briefly Smashes 2-Degree Warning Limit for First Time: We keep seeing new warming records, and this is not a happy one (particularly given our previous link).
Research and Data
- Climate Central looks at the science of thunderstorm potential, and charts long-term change using a severity indicator known as Convective Available Potential Energy (CAVE). The biggest historic changes have been in the Eastern half of the U.S. in spring and summer.