������Ƶ

Local news media serves an important role in resilience, and it’s not one that social media is replacing

The local press—broadcast media, newspapers, and radio—can be a vital part of resilience, particularly during disasters. Local reporters know the local context and appear to be a familiar and trusted voice in a community, in contrast to what opinion polls suggest about public attitudes toward media in general.

Nearly all of the emergency managers, relief organizations, and National Weather Service experts we talked to championed the role of the local media. Routinely during a crisis, emergency experts send real-time information to reporters and editors, which they can then spread to their viewers and readers. During a crisis, it’s exactly what people need: accurate, reliable, and localized information. A question for further exploration is whether local media could do more to report on resilience-building beyond just disaster response.

Social media, on the other hand, was seen as a double-edged sword, one that can complement local media, especially in reaching a broad audience, but is not always a trusted (or fact-checked) source. In worst-case scenarios, social media can take on a life of its own, a vehicle for quickly spreading false information, which can hinder rather than help emergency responders and the public.

“One of the reasons that we have had a lower death rate with tornadoes is because we have such great media presence and push telling people what they need to do. It has become almost normal for people to take cover and know what to do once the sirens go off. And to take it seriously.” —Nicole Hawkins, Director of Regional Preparedness at the American Red Cross, St. Louis, MO

“If Roger [Tulsa’s Emergency Manager] and the TV stations are supported, then the community is supported. We try and understand our role. TV stations reach the public, and we can’t compete with that.” —Steve Piltz, Meteorologist in Charge at the National Weather Service, Tulsa, OK

“We noticed that all our local TV stations were posting the same warning information on their Facebook pages, so we asked, ‘Why aren’t we putting it on our Facebook page?’ We attempt to learn from how broadcast media does this, so that we can not only echo, but reinforce it.” —Gregory Gust, Warning Coordination Meteorologist at the National Weather Service

“I think [social media] is the best and worst thing…that’s happened to emergency management in my entire career. Best because I think it’s great that we can get information out directly to people without a filter, and we can actually communicate… The downside of it is once information gets out, whoever it comes from, if it is not true it’s very, very difficult to correct that.” —Mark Diedrich, Director of the St. Louis County Office of Emergency Management, St. Louis, MO

Strong social and professional networks are key to preparing for, responding to, and recovering from weather disasters

It may seem counter-intuitive, but in some ways chronic severe weather has been helpful, at least in terms of building resilience in the locations we visited. Of course, adversity is nothing to wish for, but when that’s the hand a community is dealt, there tend to be battle-tested routines and relationships. What’s more, the emergency responders often have an enthusiastic appreciation for these networks because they have seen many times over that such relationships allow them to respond better and more quickly.

As a result, each of the counties we looked at has multiple ways for partners to communicate and come together. The National Weather Service (NWS), for example, has a chat page where NWS employees, local media, and emergency managers can exchange facts in the lead up to a storm to ensure accurate and timely information. Emergency managers engage in a range of exercises and simulations routinely to practice response but also to bring all the key governmental and community players together. For relief organizations, there are Multi-Agency Resource Centers that collect the assistance-providers in one place as a one-stop shop for those affected by a disaster. And finally, the Voluntary Organizations Active in Disaster (VOADs) and Community Organizations Active in Disaster (COADs) convene everyone who has a piece of the emergency puzzle to coordinate actions and allow for communication in “blue sky time” (i.e., before and after disaster situations, when the sky is figuratively, if not literally, blue).

“I think Red Cross, MEMA [Massachusetts Emergency Management Agency], and our partner organizations realized that we all needed to work together instead of all working separately to the same goal.” —Debra Duxbury, Disaster Program Manager at the American Red Cross, Essex County, MA

“The time to build those relationships is on a sunny August afternoon, not during a disaster. And then you need to work to maintain those relationships.” —Rick Smith, Warning Coordination Meteorologist at the National Weather Service Norman Office, Caddo County, OK

“If you have trust and understanding going in, then you weather it going out. That’s why relationships are so important and you need to build them beforehand.” —Roger Jolliff, Director of the Tulsa Area Emergency Management Agency, Tulsa, OK

“Really, relationships is what it comes down to. It’s that we know each other, we talk in blue skies, and then when something does happen we are able to work through it. It is much easier than trying to figure it out at 2 in the morning after a tornado just touched down.” —Mark Diedrich, Director of the St. Louis County Office of Emergency Management, St. Louis, MO

“It was really amazing how we, over time, have learned to work together and to not just say this is our thing, that’s your thing…because it really was like that years and years ago. It was really hard and it was reflected in the recovery of people. So now we have this ability to move faster.” — Nicole Hawkins, Director of Regional Preparedness for the American Red Cross, St. Louis, MO

Public complacency and disaster fatigue can be major barriers to resilience

Public and stakeholder opinion and outreach can be one of the most important—and challenging—aspects of resilience.

First, it can be hard to convince a community to invest in risk mitigation for something that might or might not happen in any given year. In Tulsa, for example, we heard that many area residents seem to believe long-held myths that tornadoes cannot hit the city. Local experts were concerned that this causes a lack of urgency when it comes to preparedness, and said it could be one reason why there seem to be fewer storm shelters in Tulsa than in other areas of Oklahoma. These myths—that tornadoes can’t hit cities or won’t cross water—are not only generally untrue, an unseasonal F2 class tornado actually hit the city of Tulsa the day we arrived in Oklahoma in August 2017. In St. Louis, relief experts we spoke to suggested that communities more recently affected by flooding generally prepare more for future floods, even if other communities are just as much at risk for future floods. Sometimes, it seems like human nature to play disaster roulette, hoping that your number will never come up, but it inevitably will in areas that are at persistent risk for severe weather.

Then there’s resilience fatigue, or the resilience trap, a public perception that resilience and preparedness measures actually remove risks altogether. Again in Tulsa, deadly flooding in 1984 and 1986 led to a series of successful flood control investments, but local experts worry that city residents think that means the danger of flooding is gone. Indeed, we heard that the city is once again allowing construction in vulnerable areas and failing to upgrade aging infrastructure (the American Society for Civil Engineers gave Oklahoma failing grades on dams, levees, and water infrastructure in a , for example). The unfortunate truth we saw in the locations we visited is that it can take another disaster to renew public support for resilience.

So, a key question is: How can cities build physical and social resilience to disasters without becoming victims of their own success? How can communities create an enduring commitment to resilience without generating self-defeating complacency?

“Complacency is our greatest enemy in emergency preparedness.” —Roger Jolliff, Director of the Tulsa Area Emergency Management Agency, Tulsa, OK

“I think resilience is cyclic. The energy that was here in the 80s and the 90s, I don’t think we have the same level of energy anymore. People will tell you now that Tulsa doesn’t flood.” —Roger Jolliff, Director of the Tulsa Area Emergency Management Agency, Tulsa, OK

“When it comes to weather, most people have short memories, and unless they have had it affect them, a lot of people don’t take it that seriously, until they see it. And then they tend to change their tune. But then, if it doesn’t happen again for however many years, then it’s like hmm…” —Jim Kramper, Warning Coordination Meteorologist at the National Weather Service, St. Louis, MO

“Unfortunately, sometimes it takes an event before they realize they need to do better preparing. Still, a lot of people, even in this area, don’t think that anything is going to happen to them.” — Brent Nelson, Walsh County Emergency Manager, Walsh County, ND

Resilience to severe weather can prepare a community for other types of disasters, and vice versa

When a disaster–any type of disaster–does occur in the United States, there is typically a groundswell of support for the victims in the community, state, and even across the nation. Amid the devastation, that community camaraderie can be a beacon lighting the way to recovery.

Disasters can also leave behind new relationships, practices, and expectations that make communities better prepared for adversity, from terrorism to tornadoes. “Boston Strong” became the local motto after the 2013 Boston Marathon bombing, for example. Regional emergency managers believe that terrible experience tested their capabilities and taught them, and the community, how to better respond to any disaster. One expert even said the bombing enhanced public trust in the governor, who was seen as making the right emergency decisions. As a result, that expert now has faith that when the next disaster comes, the public will heed the government’s warning (to evacuate from the path of a storm, for example).

Even when a government has lost public trust, a disaster can teach useful lessons. In St. Louis, nearly every local expert or official we met talked about the 2014 protests of the police shooting of a young black man in Ferguson (a municipality in St. Louis County), even when we asked about weather. One emergency manager, for instance, regretted not viewing the civil unrest as a disaster from the very beginning, in the sense that some of the people of Ferguson needed the kinds of relief and support a community needs after a natural disaster. The scramble to help people who were trapped behind the protests helped emergency managers and volunteer groups think differently about preparedness for other kinds of disasters.

Many emergency professionals and volunteers in the St. Louis region also pointed to the risk of a catastrophic earthquake on the New Madrid fault as a helpful driver for area resilience, even though the likelihood of such an earthquake in any given year is relatively low.

“It’s interesting because we have had disasters that are obviously weather, but we also considered Ferguson, so civil unrest, a disaster. But you almost look at it in the same way. It’s like alright, what do we need to do to get the resources and get in there.” —Cathy Vaisvil, Community Partnerships Manager at United Way and St. Louis COAD Member, St. Louis, MO

“We have been warning of this impending doom of the earthquake ever since I’ve been here for twenty years. We figure if we can plan to respond to that, we can pretty much plan for anything else.” —Mark Diedrich, Director of the St. Louis County Office of Emergency Management, St. Louis, MO

“I also think that we generally have, based on recent experiences, a higher level of compliance [with resilience measures] amongst our citizens. It is simply because we are fortunate that our decision makers make good calls at certain times.” —David Woodbury, Hazard Mitigation Grants Coordinator at the Massachusetts Emergency Management Agency, Essex County, MA

Resilience efforts are not one-size-fits-all, particularly when it comes to rural areas

What makes a community resilient may be unique to that community for any number of reasons—anything from demographics to topography to social vulnerability to continental weather patterns. We did observe a significant difference between sparsely populated rural areas and cities, including in risk perception. Tornadoes, for example, have a lower chance of causing immense damage in a rural county than in a city due to lower population density and fewer structures. Indeed, the Fujita scale of tornado severity largely measures the damage a tornado wreaks, rather than just its physical characteristics, such as wind speed. So, if a tornado happens in an open field and causes no damage, it may not even get a “score” or be recorded, no matter how ferocious the funnel.

Rural counties sometimes have little choice but to be self-reliant, if they are remote and dispersed. In Walsh County, North Dakota, for example, the Red Cross decided to focus on preparedness, rather than response, given that flooding and other disasters may compromise their access to affected individuals. Some rural areas, such as Walsh County, also have the advantage of strong social networks, given that their populations can be less transient than many city populations. At the same time, even as these individuals may have different expectations, when it comes to self reliance and assistance, we also heard that in some cases, they may be more resistant to help or instruction from outside experts, organizations, or agencies.

“In a rural area people learn to live on their own, because you have to. You know that if a storm happened you need to handle it. Here we are used to it, we are ready for it.” —Michael Attocknie, Tribal Administrator of Caddo Nation, Caddo County, OK

“There are a couple roads that can get you to Walsh County, but it’s not like the inner city where you can take 15 routes to get to that one destination. So when a certain road is flooded out, it impacts a lot of different services that can be provided. That’s why we do a lot of pre-station things.” —Jessica Kulzer, Disaster Program Manager for the American Red Cross, Walsh County, ND

“We’re for the most part very self-sufficient… A lot of what happens is that a lot of the damage is taken care of by the neighbors, they don’t even bother to report it.” —Brent Nelson, Walsh County Emergency Manager, Walsh County, ND

“[Talking about a sheriff in a rural town.] And he said okay thanks a lot. Then his parting words were, ‘well, you know pretty much the bottom line is most people up here, we tend to take care of ourselves. We tend to take care of our own.’ And that’s about the way they do it, too. You just aren’t going to hear much from them.” —Jim Kramper, Warning Coordination Meteorologist at the National Weather Service, St. Louis, MO

Social vulnerability is often tightly linked to disaster resilience

A few experts we talked to felt that focusing on bad weather was missing the point: underlying social vulnerabilities often shape the vulnerability of a community, not the severity of the weather. Indeed, in many of the communities we visited, lower-income, minority, and other at-risk populations were living in areas particularly prone to flooding precisely because the cost of living was lower in those locations. In other words, in these communities, natural disasters are more likely to hit the people least able to afford both resilience and recovery.

Oklahoma’s Caddo County, for example, is a bad weather superstar: Tornadoes, ice storms, floods, and droughts are all part of the landscape. The emergency manager and the tribal administrator of Caddo Nation, an American Indian First Nation, told us, however, that compared to systemic challenges, such as land rights, health issues, and widespread poverty, weather is something they can handle. Indeed, to them, building resilience to the weather meant addressing the underlying systemic social issues. In Tulsa, the chief resilience officer emphasized the importance of looking at resilience holistically, instead of focusing on one kind of threat or hazard or one sector of the local economy, and incorporating social vulnerability.

At the same time, there are often gaps between organizations offering help and the most vulnerable populations. Many of the experts we talked to recognized that there was a gap and expressed a desire to close it. This is an area that may be ripe for better data and analysis and shared lessons learned.

“Honestly, weather is one of the least of our worries. We are still worrying about white encroachment, and social issues like obesity, heart diseases, and food deserts.” —Michael Attocknie, Tribal Administrator of Caddo Nation, Caddo County, OK

“[Tulsa] is not only divided geographically, but also through the lexicon. When you mention resilience to African Americans they think of racial equity. When you talk to people who are white they don’t think of racial equity first. They think of weather and climate.” —DeVon Douglass, Chief Resilience Officer for the City of Tulsa, Tulsa, OK

“It’s not that weather isn’t a big deal. I was in a tornado when I was a child and our house was one of the worst hit. My mom was hurt, my sister was picked up and blown 100 yards, I was thrown down the stairs with glass in my feet – it was really traumatic. But people rallied behind us. It shook us, it shook us to the core, but people helped us get through it. It’s just not the same as systemic issues that are a constant daily issue.” —DeVon Douglass, Chief Resilience Officer for the City of Tulsa, Tulsa, OK

“Many people don’t realize that they need to call the Red Cross within 72 hours after their house burns down if they want assistance…When someone is watching their life burn down, they just want to go and heal with their family. But if they do that they can miss the Red Cross’ window and not realize.” —Polly Edwards, Environmental Director and Emergency Manager of Caddo Nation, Caddo County, OK

“The reason a lot of those homes have basement backups and flood issues is because of where the homes were placed. They should have never been built there in the first place. Impoverished people… have been pushed into undesirable housing stock, actively and passively, for generations.” —Lance LeComb, Manager of Public Information and Spokesperson for the Metropolitan St. Louis Sewer District, St. Louis, MO

Caddo is a rural county located in the southwest region of Oklahoma. Small towns such as Anadarko, with a population of 7,000 people, are scattered around the many farms and energy operators (including oil, natural gas, and wind) that dominate the land area. Caddo County is also home to multiple American Indian First Nations, many of whom have lived there for generations, but only after being forced off their ancestral lands. 

One legacy of 19th century federal policy is that the county is still checkerboarded in places, so that First Nations, such as Caddo Nation, do not have continuous property rights, but rather a patchwork of city, county, and private jurisdictions. To make matters even more complicated, Caddo County has some of the worst weather in the United States. Since 1982, Caddo County, Okla., has had 30 FEMA disaster declarations.* And that’s only scratching the surface.

* For our research purposes we looked at Major Disaster and Emergency declarations, but did not include fire management declarations. We also filtered out any disaster declarations that were not weather-related, such as oil spills or terrorist attacks.

Ice storms are one of the most damaging storm types for Caddo. During ice storms, one of the biggest threats is the accumulation of ice on power lines—the weight of which can pull down lines, even the poles themselves, and cause widespread power outages. Two ice storms of this nature, for example, occurred in Caddo County during November and December 2015. The local power companies estimated that the November ice storm for 14,397 customers and cost $2.1 million—a significant number considering the county only has a population of 29,300.

Blizzards, too, are dangerous in Caddo. One especially damaging storm on Christmas Eve in 2009 was both powerful and ill-timed, causing one emergency manager to call it “.” In Caddo County, five to seven inches of snow fell, accompanied by winds of up to 60 mph. In nearby Oklahoma City, all interstates around and inside the city were shut down, which caused thousands of people to be stuck in their cars for hours. More cars than usual were on the road due to the holiday and some drivers made the rash decision to abandon their vehicles and walk. This dangerous decision also caused the cleanup to slow down due to the difficulty of clearing the snow with stranded cars on the roads.

Floods were the second most common weather disaster in Caddo County, but the most bizarre and powerful flood to hit Caddo County actually began as a tropical storm. In August 2007, Tropical Storm Erin struck the Gulf of Mexico before moving southwest into the United States. Once on land, the storm was downgraded to a tropical depression, but days later the storm regained its power as it moved over southwest and central Oklahoma (some called it a “second eye of the storm”). Without much warning, the region received a momentous amount of rain. In Caddo County, 9.3 inches of rain fell within a 24 hour period. Creeks and rivers were overflowing and highways became impassable—some roads reported rushing water up to a foot in depth—creating fatal conditions for drivers. Unfortunately, four people died in Caddo from the floods. There were also successful water rescues, including of a mother and her two children, who were swept off the road while driving. The flooding became so severe that 150 bridges, in addition to roads, had to be closed, and at least 100 homes sustained water damage—a significant impact for a small, rural county.

Tornadoes can also be a force to be reckoned with in Caddo, but potentially less so than the other weather types. Since 1982, Caddo County has had five tornado-related FEMA declarations. However, this is a fairly small number of declarations relative to the large number of tornadoes that continuously occur in Caddo County. In fact, according to the Oklahoma Climatological Survey, Caddo County had the highest number of recorded tornadoes in Oklahoma since 1950—109 tornadoes in total. This may not even be the complete total as some tornadoes go unnoticed, particularly in rural areas.

So why are FEMA declarations for tornadoes so few in Caddo County? One rationale can be illustrated by the tornado event that occurred on May 3, 1999. On that day, . The tornadoes ranged in intensity, including seven F0 tornadoes, two F1, one F2, and one F3. A few houses were damaged, three people were injured, and barns, powerlines, and trees were knocked down. Even though Caddo had nearly one fifth of the tornadoes during the storm event—which impacted over 20 counties—the county accounted for only .4 percent of the 675 injuries and none of the 40 fatalities. This could mean that Caddo County is well-prepared for tornadoes, but perhaps also that tornadoes have less of an impact on rural areas, compared to urban areas with more structures and higher population density.

Finally, agriculturally-dependent Caddo is also heavily affected by droughts. The above graph from the U.S. Drought Monitor shows the frequency, intensity, and extent of droughts in Caddo County since 2000. From this graph, it is easy to see that Caddo County is no stranger to water stress, and in fact lived through a prolonged and intense drought just recently. From the end of 2010 to mid-2015, Caddo County had a nearly non-stop drought ranging from abnormally dry to exceptional drought. The only break came in the middle of 2012, but unfortunately, the drought came back with extreme intensity a few months later. In fact, from 2006 to 2016 Caddo County recorded . This means that the county had a drought classification (of varying severity) for more than half of that 10-year time period. To make matters worse, when the rain did come back in mid-2015, that was problematic, too. In May and June 2015, heavy rains plagued the county and caused multiple flash floods. During the May floods, one highway was reported to have up to four feet of flood water on it, and multiple passengers had to be water rescued from their vehicles. Two FEMA declarations were issued in two months due to the floods—an extreme change from the years-long drought Caddo was experiencing so shortly before.

“Honestly, weather is one of the least of our worries,” stated Michael Attocknie, the tribal administrator for Caddo Nation, an American Indian First Nation headquartered in Caddo County, Oklahoma. 

This frank admission halfway through our conversation with Attocknie came as a surprise, given that Caddo County, by the numbers, appears to have some of the worst weather in the United States. There are tornadoes—the highest frequency recorded in Oklahoma, a notable feat in the state. There are ice storms that weigh down power lines and knock out electricity to a scattered, rural population. There are floods. There are droughts. A hurricane even traveled up from the Gulf of Mexico not so long ago, bringing a heavy downpour when the eye reformed right over Caddo.

No matter what type of weather system we asked about, however, Attocknie and Polly Edwards, Caddo Nation’s emergency manager and environmental director, just listened politely and reassured us it was nothing to lose sleep over. “In a rural area, people learn to live on their own, because you have to. You know that if a storm happens, you need to handle it. Here we are used to it, we are ready for it,” soothed Attocknie.

There’s more to it than just rural hardiness, however. “We live with our families, we live with our grandparents,” Attocknie noted. “Because we live like this, stories can be passed down, and knowledge is shared.” Edwards added, “For Caddo people, resiliency is bred into them.”

Edwards certainly has a point: Caddo peoples have lived with the weather in this region for a long time (archaeological evidence dates back at least 2,500 years).

For generations, the Caddo were a large network of people, whose homeland originally spanned modern day northeast Texas, northwest Louisiana, southwest Arkansas, and southeast Oklahoma. With the arrival of European and then American settlers, the Caddo went from being a populous society of many groups and dialects to only a few hundred refugees. In 1868, after many years of displacement, they were resettled in Oklahoma, two hundred miles west of the heart of their homeland. A 19th century law (the Dawes Act), intended to force native integration into U.S. culture, checkerboarded the area into plots as small as 40 acres, mandating ownership of some plots to non-native settlers. (No brief paragraph from an outsider could do justice to the history of the Caddos. We encourage you to read more here and here.)

Today, the Caddo Nation can claim 5,000 members. “We are still worrying about white encroachment,” acknowledged Attocknie, but his biggest concern is “social issues, like obesity, heart disease, and food deserts.” Poverty and unemployment rates in Caddo County are well above the national average, as well. In other words, it’s not the weather that’s the problem, per se, but rather the underlying, systemic social challenges.

Take ice storms, for example. In Caddo County, ice storms are not frequent, but they are one of the most significant weather events because they affect such a wide area, reaching even the most remote corners of the county. For the Caddo people who live in Caddo County (some have spread out to Kansas or other areas, but are still affiliated with the Nation), ice storms are also serious because they can weigh down power lines and cause power outages. Given the hazardous conditions, dispersed population, and large area, power restoration can be a challenge.

A power outage of days or even weeks is not necessarily a catastrophe, though. Many residents have strong social networks, so vulnerable individuals (elderly residents, for example) often have neighbors looking after them. Plus, nearly everyone knows to prepare when a storm is coming. Again, Attocknie asserted, “It is a natural part of our life that we need to prepare for winter. We are used to it.” Even so, Edwards is working to establish a Caddo Nation mass communication system for such emergency situations.

Ice storms do lay bare, however, an underlying social vulnerability.  Many Caddo people depend on well water, so no power means no water, which can be dangerous, especially for individuals who already have health issues. When that happens, Polly Edwards is in charge of finding ways to get water to people who need it, a difficult task in hazardous conditions. So, what starts out as bad weather can become a health crisis. Moreover, this health crisis extends at a slower burn under blue skies, given that the well water is untreated. Edwards, who is also the environmental manager, noted that there are concerns that the water may be contaminated.

Fortunately, there is a under Caddo Nation, which the tribal leaders plan to tap. Unfortunately, such large infrastructure projects are costly, a challenge for many sparsely populated rural areas. Caddo Nation also has to contend with laws such as the Dawes Act and the , which remain in force, as well as competing claims from other Indian nations, cities, and the state. That means that if the Caddo Nation wants to pipe potable water to their members and other nearby tribes in Southwest Oklahoma, Michael Attocknie has to negotiate with every jurisdiction on the checkerboard—and some not even on the board.

So, while an ice storm is a severe weather event no one looks forward to, true resilience to it is complicated in a place like Caddo County because the nature of the vulnerability is complicated. On the other hand, if the tribal administration is addressing the chronic challenges of obesity, poverty, and vulnerable drinking water supplies in their community, they’re also de facto building resilience to severe weather.

Since 1972, Essex County, Massachusetts has declared 27 weather disasters. This large number of declarations can be attributed to three main geographic factors. First, Essex County has a long coastline along the Atlantic Ocean, making it susceptible to remnants of tropical storms and strong nor’easters. Second, the county lies on the lower Merrimack River valley where many rivers and tributaries are susceptible to flooding. And third, it’s in the Northeast United States, and unsurprisingly is prone to cold temperatures and heavy snow.

This dangerous combination of coastline, rivers, and snowmelt, subsequently results in frequent flooding across the county. Between 2001 and 2010, Essex was hit by six major floods. In 2001, occurred throughout March as snow and then rain deluged the rivers. In 2005, the remnants of caused floods that came up through western Massachusetts. In other years, the floods came from the coasts, bringing debris onto roads and causing severe beach erosion. The most infamous recent flood though was the , when up to 12 inches of rain fell in a short time period, causing rivers and tributaries across the county to swell to record levels. As a result, many main roads were closed, including a major commuter route to Boston, and schools were shut down. The rain also brought additional fears that multiple dams could fail. Fortunately, only one dam failed, and the nearby residents were evacuated safely beforehand.

Snowstorms also come in multiple forms in Essex, including inundation by heavy snow, ice that brings down power lines, and the most damaging: blizzards and nor’easters. During the , Essex County received up to 29 inches of snow and had hurricane-level winds. In nearby Boston (only 10 miles away) winds reached 76 mph during the storm. One town in Essex was suspected to have similar extreme wind levels, but conditions were so severe that at the height of the storm the wind sensors were lost and the exact wind speeds remain unverified. A storm even more severe than the blizzard of 2013, however, occurred in 1978, and continues to hold many of the state’s records for snow totals and coastal flooding. At the peak of the storm, visibility was near zero as 79 mph winds whipped snow through Boston and its surrounding areas. Essex, being a commuter area, was highly affected when the storm hit during evening rush hour, stranding 3,500 cars on the highway as commuters headed home from Boston. Essex also recorded immense snowfall totals. The coastal town of Rockport recorded 32.5 inches of snow. To make matters worse, the storm caused major coastal flooding all along the Essex coastline. By the end of the storm, 2,000 homes on the Massachusetts coast were destroyed or severely damaged by the floods.

Essex County is also affected by tropical storms and hurricanes. in 2011 and Hurricane Sandy in 2012 have shown that hurricanes can cause incredible damage in the northeast region. In fact, Hurricane Irene caused the state of Massachusetts $34.7 million in damage from wind alone and another $24.13 million from inland flooding. Essex’s earliest FEMA declaration for a tropical storm dates back to 1985 when Hurricane Gloria made landfall in Milford, Connecticut and traveled northeast through Massachusetts and into Maine. And even when a tropical storm doesn’t make landfall, remnants can intersect with other weather patterns to bring additional rain and wind, exacerbating other severe weather in Essex.

“We did an exercise several years ago where we had a firm go out and interview random citizens on the street and ask: At what point would you evacuate?” recalled Mike Russas, the chief of response and field services at the Massachusetts Emergency Management Agency (MEMA). “One guy said, ‘At a hurricane Cat-9.’”

To put this response into perspective, the categories of hurricanes are based on the , which ranges from category 1 (74-95 mph winds) to category 5 (157 mph winds and above). The scale indicates that if a category 5 hurricane occurs, “catastrophic damage will occur,” and “most of the area will be uninhabitable for weeks or months.” So, the Massachusetts gentleman was telling the emergency management team that he would only evacuate if the hurricane were apocalyptic and literally off the scale.

Russas has seen in other studies, exercises, and his own personal experiences that this “Cat-9” guy is not alone in his sentiments, though he certainly is on the extreme side of the evacuation skepticism spectrum. Just last year, for example, MEMA worked with the Federal Emergency Management Agency (FEMA) on a behavioral study, contacting hundreds of people across the state. The results were similar: People said they would be hesitant to evacuate. Russas explained one of his takeaways from the study: “I think there is an absolute difference in results from a behavioral perspective from states like Louisiana, who have been slammed, who have personally suffered through an event, versus a New Englander saying, ‘I’ve been through Hurricane Bob and I can suffer through it.’”

In short, the study showed a certain overconfidence in Massachusetts about severe weather. Part of this, we were told, has to do with local color in Massachusetts. Mikael Main, a Boston area native and regional manager at MEMA with responsibility for Essex County (and many other counties), explained the hardy culture: “It is just that mentality of the New Englander, the mind frame of being able to take care of yourself. We don’t need a lot of assistance from other people, we can do this ourselves, we are used to this stuff.”

Sometimes, this mentality can be beneficial to emergency preparedness, such as in a snow storm. Frequently, MEMA and their partners will open a shelter during a snowstorm, but few people come. Instead, affected people tend to rely on their social networks to get through the storm safely. Main explained, “People don’t want to stay in a shelter; they want to be in their own home. They want to be with their friends and family and take care of each other. With the warming centers, what we see a lot more of now, is people will come in, charge their phones, warm up, and then go home and sleep for the night.” In some ways, this mentality could be a sign of resilience—that snowstorms are something people in Massachusetts have dealt with and can bounce back from with little harm.

But the culture cuts both ways: The same stoicism that makes Massachusetts ready for snow storms may be making them more vulnerable to hurricanes.

Essex County, in particular, lies in a precarious location along the coast. It’s already prone to coastal and inland flooding (the main reason it made our cut as a severe weather hotspot), so a hurricane—and more importantly, its storm surge—could cause significant damage. Russas is well aware of this. “We have done emergency exercises where we took the Sandy track and had it travel literally 50 miles north and had it barrel into Boston with a major surge,” he explained. “And [Essex] gets totally inundated with water.”

The findings from that study are located in their state’s hazard mitigation plan, and shows detailed possibilities of what would be inundated by water if a category 1, 2, 3, or 4 hurricane hit Massachusetts. For example, in Essex County, a category 1 hurricane would affect 9 percent of the county’s total area, threaten 21,409 people, and wind damage would create 56,631 tons of brick and tree debris (an adult elephant weighs between 2 and 7 tons). A category 3 would affect 12.5 percent of the county, threaten 48,321 people, and create 143,560 tons of brick and tree debris. The situation is even worse in nearby Suffolk County, where Boston—with its large population and expensive infrastructure—sits on the edge of the water. Based on MEMA’s estimates, if Boston were hit by a category 4 hurricane, 269,737 people would be affected.

Now take this one step further: What if some of the population decides to not evacuate, or is reluctant to evacuate? Again, Russas is on top of this. “That absolutely plays a part in our preparedness and messaging,” he told us. “We know ‘okay they are really not going to want to leave their home because they think they can weather the storm.’ If we look at storm surge and we know that it will impact 97,000 people, that has an impact if we need to time those evacuations. If we know there is going to be some resistance or reluctance to evacuate; we need to back that in. If we know that they are going to wait until the last minute to get into their cars and move, we need to back that off into our planning factors.”

MEMA is also attempting to better inform the public about evacuations through a campaign called Know Your Zone. The campaign publishes maps and preparedness information, which explains where the location of evacuation zones and advises residents in those zones to be prepared to evacuate in the event of a hurricane. This way, the public will already know if they live in a vulnerable area and hopefully this will mitigate against the natural reluctance. But, of course, a change in mentality takes time.

In the meantime, there is one factor that plays in Massachusetts favor: a good track record in recent emergency decisions made at the state level. “We are fortunate that our decision makers make good calls at certain times,” said David Woodbury, a MEMA coordinator in the mitigation unit. “The two that stick out in my head,” Woodbury continued, “are when the city shutdown following the Boston Marathon bombing, which was a real tough call, I remember being there when it was made, and then shortly after that we had a very intense blizzard and Governor Patrick shut down all the roadways. He took a lot of heat for it, but in the same blizzard, New York did not shut down and I-495 in Long Island was a disaster zone and got a lot of media coverage. So the people here thought, ‘Okay so maybe that wasn’t so bad of an idea.’”

The hope is that if another emergency were to happen, say a hurricane barreling towards Boston, and the state government had to make another tough call for mandatory or voluntary evacuations, the people of Massachusetts would listen, based on the trust the state has built with its citizens in recent years. And maybe, that trust will overpower the recalcitrant Massachusetts mentality and help people evacuate to safety. But just in case, the folks at MEMA are ready with a “stoic New Englander” discount factor.

St. Louis sits in a precarious place, nestled between two big rivers and crisscrossed with tributaries. The French trappers who founded the city in 1763 chose a spot high on a limestone bluff, close enough to the Mississippi River to get their goods to market but far enough away to avoid floods. But the city has long outgrown that origin story, if not the romance of the riverine location. Today, flooding is part of life in St. Louis, with six major floods since 2008 and two record-setting floods in the last three years alone. What’s more, with no surrounding mountains to run interference, the city is wide open to cold weather sweeping in from the north and hot, humid weather rising up from the south, and is prone to tornadoes. And when all else fails, there still can be extreme wind or monster hail. For all these reasons, St. Louis County and City landed on our top ten list for disaster-prone areas.

occurred during the holiday season in December 2015. Heavy rains poured seven to ten inches across St. Louis County on December 26th and 27th, causing multiple creeks and rivers to overflow. The rain fell so quickly that Deer Creek rose 11 feet in three hours, and the Meramec River crested at record levels. As a result, flooding was widespread and damaged nearly 900 buildings, including two wastewater treatment plants, the latter alone estimated to cost $16 million. The rains eventually subsided on December 28th, but the Meramec, Missouri, and Mississippi rivers all remained at high flood levels into January. Eighteen months later, the flood records set in 2015 were broken by . The floods hit many of the same areas, especially towns along the Meramec River, and shut down two major highways for multiple days.

Floods were not exactly unknown in St. Louis before 2015, and in fact the area had a similar incident as recently as 2008, when remnants of Hurricane Ike dumped rain over the region. There have also been near misses, such as the snowmelt in the summer of 2011 from record snowfall in the Rocky Mountains. States all along the Missouri River faced severe flooding as major dams had to release an unprecedented amount of water downriver. Fortunately, by the time the flood waters reached the confluence of the Mississippi River, most had subsided. Still St. Louis had a series of flash floods from saturated soils, but compared to their upstream neighbors, this was a lucky break.

Snow has also been a significant problem for St. Louis. In fact, the metropolitan area has suffered through a whole variety of winter weather in recent history, from heavy snowfall in 2014, to a blizzard in 2011, to a major ice storm in 2006, all of which caused significant damage. The was estimated to cost the city at least $1 million just for snow removal, for example. The National Guard had to deploy to rescue people across the county in 2011 and delivered emergency services door-to-door to the hundreds of thousands who lost power in 2006.

A history of major snowstorms in St. Louis has not precluded extreme heat days in the summer. Two of the  struck consecutively in 2006 and 2007. First in 2006, heat indexes ranged from 100 to 115 from mid-July to mid-August. To make matters worse, severe thunderstorms caused the worst power outage in St. Louis history, leaving 700,000 buildings without power—some for over a week—in the middle of the heatwave. In total, 794 people had heat-related injuries and eight died across the St. Louis area. Then in August 2007, another heatwave caused the third warmest August on record in St. Louis. In just over a week, 519 people in St. Louis County and 422 people in the city had heat-related injuries, and eight people died. Nearly 500 of the heat injuries were due to attendance at outdoor concerts during the heatwave.

Tornadoes also strike a constant fear in the hearts of the people of St. Louis, especially lately: There have been four significant tornadoes within a three year span in recent history. The first of the tornado events happened in April 2011 when an and traveled 21.3 miles. Homes were leveled, the international airport had windows blown out and roof peeled back on the main terminal, but fortunately only five people were injured, and there were no deaths. Nearly 2,500 buildings sustained damage. Two years later in April 2013, two and damaged almost 400 homes. This event was followed one month later in May 2013 by that damaged an additional 600 homes and 100 businesses. Another tornado appeared in April 2014, damaging another 100 buildings.

The final weather threat facing St. Louis is the most common, and perhaps the most bizarre. Severe weather in this area comes with extreme winds and monster hail, and though it is typically not fatal, it can be costly.  reached 4.5 inches in diameter in May 2011. Imagine grapefruit size blocks of ice being blown down by severe thunderstorm winds onto your roof, car, and windows, breaking or harming everything it its path. In April 2012, two to three inch hail resulted in 150,000 hail damage claims—estimated to cost $995 million—across the state, including St. Louis. In May 2004, 1 inch hail piled up a foot deep and clogged storm drains causing floods on the interstate highways. It also rained hail on a dealership lot full of brand new BMWs and Mercedes Benz. And in 2001, , estimated to cost several hundred million dollars, after hail 2.75 inches in diameter hit ten thousands of homes and vehicles, damaged every single vehicle at the Ford Motor assembly plant, and harmed 22 jetliners at the international airport and 10 fighter aircrafts at the Missouri National Guard.

St. Louis has frequent and severe floods. For anyone who lives in or visits the area (or looks at a map), this will come as no surprise. After all, the city was built below the confluence of two of the largest rivers in the country, the Mississippi and the Missouri, and is criss-crossed with tributaries. Indeed, St. Louis just had had two record breaking floods within 18 months, in December 2015 and April 2017.

Given that history, we were surprised, however, that the emergency experts we spoke to in St. Louis seemed less concerned about floods than they were about other disasters. Mark Diedrich, the director of emergency management for St. Louis County, for example, told us that “earthquakes are our highest [concern], then tornadoes are probably right under that. They are a little less frequent, but they’re damaging and like earthquakes there is very little warning. And then we probably get into severe storms—ice or heat—and then floods.” Other experts concurred: earthquakes at the top of the list and flooding at the bottom. In fact, Nicole Hawkins, the director of regional preparedness for the St. Louis branch of the Red Cross, politely pushed back on our assumptions about her priorities. “You know, I worry about floods the least,” she corrected.

Based on our reading of the area’s history of severe weather, that was not what we expected. Why does flooding, a chronic severe weather event, rank so low? And why haven’t recent record-breaking floods raised the sense of urgency?

According to the people we met, it’s because flooding is as familiar as an old shoe for St. Louis, and they feel ready for it.

St. Louisans certainly have a great deal of practice in flood preparedness and response. The rains pour, the rivers swell, and people know what they have to do because the same thing may have happened last year and the year before that, too. “One thing I like about floods,” Diedrich explained, “with the exception of flash floods, is that you know they are coming, you have time to prepare, and you can put things in place. And usually, the people who are affected have been affected before, so they probably have more preparedness.”

This individual preparedness is certainly helpful, but may not be sufficient for every severe storm. The December 2015 flood, for example, was particularly damaging because it caught area residents off guard. December is an uncommon time of year for floods and the flooding occurred between Christmas and New Year’s Day, when many people were not really paying attention to the fact that a record amount of rainfall was falling. The unusual volume of water meant some areas flooded that had not flooded for years, and in other areas, flooding was worse than usual. Individual residents were not prepared for that, but fortunately, the emergency management community was.

In fact, one of the main reasons why St. Louis is so ready for floods is because they have a strong network of emergency experts who communicate amicably and frequently. Together, they are ready to respond to any situation and to help their community recover quickly. These are connections that are tough to make for the first time in the heat of a crisis.

“Really, relationships is what it comes down to,” Deidrich said. “It’s that we know each other, we talk in blue skies, and then when something does happen we are able to work through it.” Years ago, St. Louis did suffer from a lack of strong relationships, so the experts know to appreciate what they have now. Hawkins commented, “It was really amazing to see how we have learned over time to work together and to not just say this is our thing, that’s your thing, this is your thing, because it really was like that years and years ago. It was really hard and it was reflected in the recovery of people. Now we have this ability to move faster.”

One of the most impactful results of the collaboration among disaster relief organizations is the (MARC)—a one-stop shop for disaster recovery. Before the creation of the MARC, disaster victims would have to call each organization (the Red Cross, the Salvation Army, etc.) and recount their story over and over again to ask for various kinds of help. Hawkins explained, “You can imagine if it were you and you are going through the most traumatic thing to ever happen to you, how frustrating that could be.” MARCs were the solution to this dilemma. Now, the relief organizations set up booths in a single location—such as a local high school or community library—accessible to individuals affected by the disaster, where they come and recount their story once to all the various partner organizations. “They can get a lot of help in one day versus lots and lots of days,” Nicole emphasized. This was particularly necessary for the recent floods, which involved 14 days of rolling MARCs around the region. As the floods crept from west to east across the state, the MARCs were picked up and moved to best serve the affected people. Without strong relationships, this proactive response would have been unlikely.

So, how did St. Louis become so well connected?
At the end of the day, it’s hard to say. Some experts pointed to a specific disaster as the catalyst. Others thought it was a result of strong leaders, such as Mark Diedrich, who understand the merit of relationships. Whatever the impetus, the local and regional Community Organizations Active in Disaster (COAD), a volunteer-based association that convenes local organizations and government agencies, is critical. COADs are not unique to St. Louis, or Missouri for that matter, but in many areas the COADs are much less active or are strapped for funding, sometimes causing them to be temporarily or permanently suspended.  

In St. Louis, that has not been an issue. The COAD has been in steady operation since 2003. Warren Robinson, the chair of the regional coalition of COADs, credits chronic severe weather as the glue that holds the COADs together. “You have to have some way to keep them engaged,” he said in regard to the many partner organizations affiliated with the COAD, “and by golly, we have kept them engaged the past few years.”

For Robinson, who is double-hatted as the director of emergency management of Jefferson County, the COAD is an opportunity to help communities in a way that complements his day job. As a county official, Robinson humbly acknowledged the limitations of leading a government agency: “Our hands are tied in a lot of ways. We need to make sure [victims] were impacted by certain [kinds of] disasters and certain criteria are met.” The COAD, on the other hand, has a long-term recovery fund that is not bound by the same exclusions and deadlines and can reach people government assistance doesn’t always reach. A weather event has to hit certain damage thresholds, for example, in order to qualify as a Federal disaster and unlock Federal assistance dollars. Even when a disaster doesn’t meet those broader thresholds, it can be just as devastating for the affected individuals.  In other cases, people may not know much about the government services available to them or how to get those services.

In particular, the experts we spoke to expressed concerns about low-income, at-risk, and minority communities in the region. According to them, people in these communities don’t communicate well with government and volunteer aid organizations, and the organizations don’t communicate well with vulnerable populations, either. “It’s great to call 211 [a fast track helpline for community services], but if [people] don’t know they are supposed to call 211 or even that anybody is there to help them, it’s a problem,” commented Ben Perrin, a disaster services manager and St. Louis COAD chair. A number of the people we interviewed noted that there’s a shortage of educational outreach programs in these areas and little representation in the COADs, which is doubly problematic. First, individuals and families in these communities often lack the personal resources to prepare for or bounce back from a disaster—to repair a damaged home, for example. In that sense, these populations are least likely to be resilient and most likely to need help. Second, low-income populations tend to have a higher vulnerability to weather disasters because of where they are located—land and property costs are often lower in a floodplain, for example. Lance Lecomb, manager of public information for the Metropolitan St. Louis Sewer District (MSD), observed that “the reason a lot of [these] homes have basement backups and flood issues is because of where [these] homes were placed. They should have never been placed there in the first place.” Renters and transient populations may not even be aware that they are in vulnerable locations, which points back to the need for better, more targeted education and outreach. Everyone we spoke to in St. Louis acknowledged this gap.

For over a decade, the St. Louis COAD has been a way to connect the many different pieces of disaster preparedness and response to make the process fit together better. The cross-agency communication that this model fosters then allows the emergency actors to expand their reach to help a greater number of people, and hopefully to improve the collective ability to find those who are falling through the cracks. For emergency actors, they’ve seen the proof that this model works, and expressed that efficacy justifies their investment of time and effort.

The COAD model that works so well in St. Louis could be applied anywhere. The same goes for the MARCs and other powerful emergency networks that St. Louis has sustained. However, to do so in a sustained way requires leaders who understand the value of relationships and who are willing to put in the time to maintain them in “blue sky” times (i.e., when there’s no active disaster situation). In St. Louis, recent record-breaking floods strained the system and highlighted resilience gaps in underserved communities, but also proved St. Louis is ready, thanks to the network of relationships.

Since 1996, more than 30 of the U.S.’s have hit Tulsa. Though Tulsa was not always the eye of the storm, the unusually high number of damaging weather events indicates that the county lies on an unrelenting and varied storm path.  Over 20 years, Tulsa has seen the likes of tornadoes, heatwaves, wildfires, hail and severe wind events, droughts, floods, and blizzards.

Often times, these storm types can intersect in disastrous ways going from one extreme to the next. For example, from July 2012 to May 2015 Tulsa experienced 24 months of exceptional (D4), extreme (D3), or severe (D2) drought. Once the , the weather rebounded powerfully with widespread heavy rains, which caused major flooding throughout the remainder of the month. The May 2015 floods shut down highways, inundated farms, and closed down numerous businesses. Then in December, the rains continued, and resulted in nine flood incidents across Tulsa County. The heavy winter rains broke a daily rainfall record and —only a few months after the county was plagued by prolonged drought.

However, the occurred Memorial Day weekend 1984. A severe storm brought 24 hours of nonstop rain that averaged 9.35 inches, with some areas measuring up to 14 inches. Intense floods spread across the city, damaging thousands of homes and businesses, killing 14 people, and ultimately cost between $150-180 million. After the storm subsided, the public outrage grew, and resulted in a $140 million flood prevention project to ensure this flood severity never occurred again.

In terms of tornadoes, Tulsa typically sees lower level classifications (EF0 and EF1), which can usually be managed without fatalities or much damage. Yet even these weaker tornadoes can cause significant destruction when they hit in a populated area. In March 2016 an for 10.5 miles and caused damage to over 300 homes, injured seven people, and hit multiple churches, businesses, and barns. On another occasion, an EF1 tornado just 80 yards wide touched down for only .4 miles—next to the Tulsa international airport. The small tornado and its strong winds damaged airplanes, cars, and an airport hotel in a matter of minutes. The , however, was an EF4 in 1993, which touched down in northeast Tulsa and traveled for 5.5 miles. The EF4 also was accompanied by an EF3 tornado and together they injured 130 people and caused $100 million in damage.

Heatwaves can also be a fatal and frequent occurrence in Tulsa. Beginning as early as May and peaking in frequency in August, heatwaves can cause hundreds of heat-related illnesses within just a few days. In June and July 1998, in Oklahoma. In Tulsa, at least five people died, including two 40-year-men and a 39-year-old woman. Another heatwave hit Tulsa in August 2007 while the city hosted the PGA Championship. Reportedly, 1,000 fans went to the first aid stand for heat-related illnesses, a fourth of which were deemed serious. In the city of Tulsa even more people were treated for heat illnesses and two elderly men died.

On the opposite side of the weather spectrum, severe snowstorms also have a devastating impact on Tulsa. During the in February 2011, a half-inch of sleet fell in Tulsa before twelve to sixteen inches of snow quickly blew on top. Across Tulsa County there was near zero visibility as 35 mph winds whipped the snow in all directions and created large snowdrifts. Businesses, schools, highways, and the airport shut down, multiple schools and a casino had roofs collapse from the weight of the snow, boat docks and boats were damaged or destroyed on Grand Lake, and hundreds of traffic accidents occurred.

How can city leaders convince the public that it’s a good idea to spend money on resilience before an emergency occurs? This is the multi-billion dollar question all Americans should be asking in the wake of the devastating 2017 hurricane season. It is also a question we heard in Tulsa from emergency managers who worry that the local public is growing complacent.

This question was more than academic when we arrived in Tulsa in August 2017, given that an unusual F2 tornado had caught the city off guard earlier the same day. Tornado season generally runs from March through June; the last tornado to hit the Tulsa area in August was in 1958. Most of the city was also asleep at 1:19 am when the tornado touched down, so few people would have been watching the news. In addition, tornadoes can form quickly, and in this case, emergency managers had little to no advance warning.

Over the course of our visit, we saw in local papers and television that many Tulsans were shocked—and even outraged—by the tornado and the lack of warning. In explaining the backlash, a number of the experts we spoke with pointed to local beliefs that a tornado could not actually hit the city and also that warning technology has improved to the point of near infallibility.

The first belief, we were told, was grounded in folklore, which took firm root over time as multiple tornadoes seemed to inexplicably bypass the city. Roger Jolliff, the emergency manager of Tulsa, explained: “Lore, attributed to Native Americans and the initial settling of the Tulsa community, says a tornado will not pass the Arkansas River. The tornado of ’99 made people believe it. Then we had so many times where tornados did not touch down on the city. One time we had a tornado warning for 45 minutes in Tulsa and two local TV stations followed the storm recording it, but still, it never touched down.” Jolliff noted that people also generally think tornadoes can’t hit cities with tall buildings.

Despite what looks like circumstantial evidence, there is no weight to these myths. Steven Piltz, the meteorologist in charge at the Tulsa National Weather Service, told us that he is certain a major tornado could hit Tulsa, and is concerned that these stories could leave the public unprepared. “Draw a straight line between Moore, Oklahoma [devastated by an F5 tornado in 2013] and Joplin, Missouri [another destructive F5 tornado in 2011],” Piltz pointed out. “What’s in the middle? Tulsa.”

Unfortunately, such myths can create a false sense of security, which then can lead to complacency. And as Jolliff warned us, complacency can be immensely detrimental to preparedness as individuals hedge their bets against the risks. “Safe rooms in the Moore area are in the thousands,” Jolliff observed. “We have hundreds.”

The second harmful narrative—that tornadoes can be predicted with perfect accuracy—can also stoke a false sense of security. There is no question that the technology and science behind weather prediction has improved drastically in recent years, but the hard truth is that there is still much scientists don’t know about complex storm systems. The weather can still be unpredictable.

This was exactly the case that day in August. From the very beginning of the storm, Jolliff kept his eye on the radar and remained in frequent contact with Tulsa’s network of emergency managers, national weather service experts, and meteorologists. Early in the evening, all information pointed to an extreme thunderstorm, not unusual for the area and time of year. When the winds escalated above 58 mph, Jolliff issued an extreme thunderstorm warning to Tulsa, and then began to wait and watch.

Unfortunately, the tornado developed out of the thunderstorm before anyone caught it on the radar. Once it touched the ground, the tornado ran fast, traveling three miles through Tulsa in six minutes before dissipating in the next county. The funnel did significant damage, including to a 20 story building, but fortunately there were no fatalities.

Steven Piltz, a longtime internationally recognized expert on tornado prediction, noted that the conditions that formed this particular tornado hurt his team’s ability to predict. “Yes, it will happen again,” Piltz warned. “We will miss the beginning minutes of some tornadoes. And we will miss some altogether.  That will happen, but it comes with the territory. Our standard is still that we want to be able to warn for every tornado, but the public needs to be aware that it doesn’t always happen.”

Why is the public’s demand for perfection so detrimental? For one, unrealistic expectation can degrade trust in experts even when they do everything in their power to predict accurately. And we saw across all the locations we visited that trust in institutions and experts is vital for major disasters, so that people will heed warnings to evacuate or seek shelter, for example.

Roger Jolliff stood by his decision that night to not sound the sirens in Tulsa, given that he only knew about the tornado after it had already touched down. “It’s a tough call, but it goes with the territory,” Jolliff explained. “As an emergency manager, you need to make that split second decision. It’s healthy, and it’s how we get better. You need to learn and grow from it.”

But even though he stood by his decision, Jolliff acknowledged the concerns of the public and answered questions about the night’s events at a . Jolliff told us that one Tulsan who called him earlier in the day to express her anger and resentment for the lack of warning called him back after seeing the briefing to tell him she now understood why he made his decision.

Steven Piltz also agreed that even though it appeared that Tulsans demanded perfection when it came to the sirens, they also accept the truth when experts are upfront with them. “Everyone seems to trust everyone else enough that they may…not like what you did, but [they] think you are telling…the truth.” Piltz added, “In that way, Tulsa stays under control.”

Emily Gallagher and Sharon Burke conducted all of the research, interviews, and writing for this report, with research, interview, and writing assistance from Braxton Bridgers for the St. Louis and Walsh stories. Maria Elkin edited the text, selected the photographs, advised on local media, and lead the overall conceptualization of the website. Ellie Budzinski designed the site and selected photographs. Kirk Jackson helped conceptualize the site and contributed all the data visualizations. Ross van der Linde supervised all of the communications work on this project.

Over the course of this project we conducted interviews with 41 experts from 26 organizations in 6 counties. We chose the final 5 counties in this report based on a variety of criteria, including severe weather type, geographic region, population size, and demographics. We also drew from the research in our earlier report, which included federal data on U.S. counties with the most chronic severe weather. Our goal was to create a narrative report, one that gave a more personal view of the data from a diversity of voices.

In each location, we sought out key people with some role in the broader emergency management community and a deep knowledge of the local context. In all locations, we drew from five main categories of organizations:

1. Emergency Management Agencies
2. The National Weather Service, local branches
3. Relief Organizations
4. Local Media Stations
5. The National Guard or the U.S. Army Corps of Engineers

We did not always draw from all five categories in each location, and sometimes we talked to multiple people in one category. We aimed for consistency across all the counties we visited, but we also wanted to allow the experts in each area to tell their own stories, including by following their recommendations for whom to interview.

Below is a list of all the organizations that were gracious enough to take the time to meet with us, and we could not have written this report without them:

• American Red Cross, Eastern North Dakota Chapter
• American Red Cross, Massachusetts Chapter
• American Red Cross, St. Louis Area Chapter
• Caddo Nation of Oklahoma
• Casco Bay Estuary Partnership
• City of Portland, Maine, Department of the City Manager
• City of Tulsa, Resilient Tulsa Office
• Disaster Resilience Network
• Maine Voluntary Organizations Active in Disaster
• Massachusetts Emergency Management Agency
• Massachusetts National Guard
• Metropolitan St. Louis Sewer District
• National Weather Service’s Grand Forks’ Forecast Office
• National Weather Service’s Norman Forecast Office
• National Weather Service’s St. Louis Forecast Office
• National Weather Service’s Tulsa Forecast Office
• News on 6 WARN
• Oklahoma National Guard
• South Central Climate Science Center
• Southern Climate Impacts Planning Program
• St. Louis Community Organizations Active in Disaster
• St. Louis County Police, Office of Emergency Management
• Tulsa Area Emergency Management Agency
• U.S. Army Corps of Engineers, St. Louis District
• Walsh County Office of Emergency Management
• WDAY-TV Weather