The Fifth Risk

Kathy Sullivan was the top official within the National Oceanic and Atmospheric Administration (NOAA), which oversees the National Weather Service (NWS). She was also one of three officials that DJ Patil wanted to meet with when he became the Chief Data Scientist of the United States. While completing his doctorate, Patil immersed himself in data from the NWS when he suggested a new statistic for looking at weather: how predictable the weather is at any given moment.

The Department of Commerce collects a lot of data as part of its mission, and the majority of it comes from the NWS. Much of that is critical to daily life within the country, and Lewis writes that “[w]ithout that data [from the NOAA], and the Weather Service that made sense of it, no plane would fly, no bridge would be built, and no war would be fought—at least not well” (163).

When Wilbur Ross was nominated to run the department at the end of November 2016, he met with a former Bush administration staffer who had worked within the Commerce department. That official explained that it was essentially geared toward science and technology, to which Ross replied, “Yeah, I don’t think I want to be focusing on that” (164). Then, in October of 2017, Barry Myers arrived as the new head of the NOAA. He was the CEO of AccuWeather, a for-profit weather company. AccuWeather advertised itself as being more accurate than the NWS. However, all private companies depend on NWS data for their predictions. Myers had long been involved with the NOAA, wanting to keep NWS data from being available to the public because private companies could profit off of individuals’ desire for a forecast. The Bush official who worked with Ross noted, “The more people have access to the weather data, the better it is for the country” (176). Lewis also references how David Friedberg used NWS data to start WeatherBill (later the Climate Corporation) to offer insurance to farmers affected by weather events.

In 2013, President Obama signed an executive order in 2013 to make all unclassified government data publicly available. Patil arrived in Washington a year later with the intention to get as much out of those data as possible. Among the benefits, journalists were able to look through the Department of Health and Human Services data on Medicaid and Medicare. From there, they found an unusual number of opioid prescriptions. Patil tells Lewis that “[w]e would never have figured out that there was an opioid crisis without that data” (177). When Trump took office, Lewis notes, Patil watched much of this data disappear; in particular, climate change data was removed from the websites of the Environmental Protection Agency and the Department of the Interior.

When Sullivan became the head of the NOAA, she became interested in the relationship between people and their government, wanting to create a weather-ready nation after walking through the aftermath of a tornado in Joplin, Missouri, in 2011. That tornado had killed 158 people with thousands more injured. Despite the tornado warning being issued 17 minutes before (four minutes ahead of the average), people didn’t seek shelter. She wanted to understand why this was and started to bring behavior economists and psychologists into the NOAA to help collect data on how people respond to risk.

Kim Klockow was among the first scientists Sullivan brought in to achieve her goal. Klockow worked at the National Weather Storm Prediction Center in the NWS building in Norman, Oklahoma. She wanted to investigate how one could influence people to respond to their benefit when confronting storms. She interviewed people who lived in areas that had been hit by tornadoes and found that not only did they not have enough time to escape, but they also did not think that it would hit them in particular. The NWS also implemented new “impact based” warnings, which described what effect a tornado could have. Klockow thought that these were “intellectually dishonest” because “[h]ow could you warn about the impact of a storm whose force you would only be able to discern after the fact?” (208).

The part concludes with Lonnie Risenhoover, the emergency manager in Beckham County, Oklahoma, on the day a tornado was hitting Elk City. By coincidence, a team of researchers at the Storm Prediction Center was trying out a new model, and their data encouraged Lonnie to activate the tornado sirens. Only one fatality occurred as a result. Lewis finishes the book recounting a dinner he had with Hank Jenkins-Smith and Carol Silva, the co-directors of the University of Oklahoma’s Center for Risk Management, asking questions about who is more likely to survive a tornado. The one question he doesn’t ask but considers is whether someone who has experienced a tornado is more likely to survive than someone who hasn’t.