Studying the data from flooding events like this recent event in Michigan, in addition to the ... [+]
It’s typically major storms like Hurricane Katrina and Superstorm Sandy that come to mind when thinking about devastating storm damage. Tied with Hurricane Katrina, Hurricane Harvey was the costliest tropical cyclone on record in the U.S., inflicting $125 billion in damage, largely from the unprecedented rainfall-triggered flooding across southeast Texas. But there are many other extreme storms in the U.S. that while generally unnoticed by the public, still have major economic impact.
Just last year, flooding on the Mississippi River in Illinois, Iowa, Missouri, Nebraska, and nine other states caused an estimated $6.2 billion in damage, according to the National Oceanic and Atmospheric Administration (NOAA). River flooding is among the top 100 costliest weather disasters since NOAA began tracking the cost of severe weather damage back in 1980.
A more recent devastating flooding event hit central Michigan earlier this month and is one of those events that will likely have a major economic impact. Midland, Saginaw and the surrounding Michigan counties saw up to 7 inches of rain in about 36 hours, forcing more than 10,000 people to evacuate and contributing to the failure of the Edenville and Sanford dams. In addition to the devastating impact on area residents, the flooding impacted major businesses in the area, including the Dow Chemical company, closing its headquarters as well as the manufacturing complex. .
Studying the data from flooding events like this recent event in Michigan, in addition to the learnings behind other catastrophic storms, can ensure infrastructure is safely designed and operates safely and reliability under extreme weather.
The U.S. Army Corps of Engineers (USACE) established the Extreme Storm Event Database, which includes the largest precipitation events, to help hydrologic engineers make informed decisions for dam and levee construction, modification and operations across the United States.
Support for this database comes from private weather organizations too, collecting data from major weather events, and not just the events that make national headlines. For example, a thunderstorm event on August 23, 2018, dropped copious amounts of rain across northern parts of Phoenix, Arizona prompting an evaluation of the storm rainfall for making infrastructure changes.
A recent project conducted by a private weather company, in conjunction with the USACE, contributed to storm data in the Extreme Storm Event Database. The team added information for over 600 storms, including a detailed summary of each storm’s precipitation characteristics such as location, duration, maximum precipitation, and a depth-area-duration table.
The meteorological statistics and precipitation analytics company, MetStat*, is building its own repository that includes everything in the USACE database, plus numerous other events that have been assembled over the past 25 years. With nearly 4,000 storms across North America documented, the company manages the largest known extreme storm database in the United States.
Both the government and private databases support new research and technology such as machine learning, where storm precipitation and flooding consequences can be correlated. Researchers at Rice University are using an advanced form of deep learning to create computer systems that learn how to accurately predict extreme weather events, using minimal information about current weather conditions. A recently published study by a team at the university shows that with further development, artificial intelligence systems could help with early warnings for extreme weather,
The work with these extreme weather databases offers more than just a historical record of storms, but also an opportunity to create actionable insights for water resource managers, flood control districts, dam owners, emergency managers, insurance companies and others.
*MetStat was acquired by DTN in 2019. The author works for DTN.
