Can AI Predict Natural Disasters?

Artificial intelligence (AI) has rapidly emerged as a game-changer across various fields, transforming the way we work, live, and interact with technology. Its impact is already significant and continues to expand into key areas like automation and robotics, healthcare, finance, education, and now, even weather and natural disaster forecasting.

Numerous US authorities and organizations are actively exploring the use of AI to predict natural disasters. Among these entities are NOAA, FEMA, the United States Geological Survey (USGS), NASA, and IBM’s The Weather Company. The growing interest in AI from such influential organizations reflects the recognition of its vital role in improving natural disaster forecasting. And the rationale is evident. AI excels at analyzing vast amounts of data from diverse sources, identifying patterns and trends that may indicate the likelihood of specific events, and significantly accelerating data processing.

Here are some ways AI can be used in predicting and responding to natural disasters:

  1. Data analysis: AI can process data from satellites, weather stations, sensors, and other sources to monitor environmental changes and detect anomalies that may indicate the potential for a natural disaster.
  2. Machine learning models and predictive analytics: AI can use machine learning algorithms to learn from historical data about past disasters and their contributing factors and make predictions about future events based on similar patterns. This includes:
    • processing data from sensor networks deployed in disaster-prone areas to detect changes in environmental conditions and predict potential disasters;
    • analyzing satellite images and detecting changes in landscapes that may indicate impending disasters, such as wildfires, floods, or landslides;
    • analyzing seismic data and detecting patterns that may indicate an impending earthquake;
    • analyzing data from buoys, ocean sensors, and weather monitoring stations to predict events like hurricanes, storm surges, and tsunamis.
  3. Early warning systems: AI-powered early warning systems can send alerts to communities at risk, allowing them to take preventive measures and evacuate if necessary, minimizing the impact of disasters.
  4. Social media analysis: AI can monitor social media platforms for real-time information and user-generated content about potential disasters, helping authorities respond more effectively.
  5. Simulation and modeling: AI can be used to create complex simulations and models that simulate different disaster scenarios, helping researchers and authorities understand potential outcomes and plan accordingly.

Many AI systems have demonstrated a high degree of accuracy. For instance, Google’s DeepMind, developed a machine learning model that outperformed traditional methods in predicting rainfall up to six hours in advance. In another example, researchers at Harvard developed an AI that can accurately predict earthquake aftershocks, potentially saving lives by providing early warnings.  AI can help predict the likelihood of wildfires by using data on weather conditions, vegetation, and historical fire patterns, and can help fire departments allocate resources more effectively and even predict the path a wildfire might take once it has started. 

According to the Washington Post, NOAA estimates the cost of the most extreme weather and climate disasters impacting the United States during the past seven years at more than $1 trillion. and resource consumption, and preserving ecosystem services and biodiversity. A study by the National Academies of Sciences, Engineering and Medicine, estimated that the potential annual net benefits of improved forecasts and early warning systems for four types of hazards (tornadoes, hurricanes, inland flooding, and coastal flooding) could range from $511 million to $3.3 billion.

Despite the advantages of AI, there are challenges and limitations to what AI can do:

  1. Data Availability: In some regions, data collection infrastructure may be limited, reducing the effectiveness of AI models, which require large and diverse datasets to make accurate predictions. 
  2. Model Complexity: Some AI models used for predicting natural disasters are computationally intensive and require substantial computing resources. Deploying and maintaining such models can be a challenge, especially in resource-constrained environments.
  3. Interpretability: Some AI models can be difficult to interpret, making it challenging to understand the reasoning behind specific predictions. This lack of transparency can raise concerns, especially in critical decision-making processes.

While AI can augment decision-making, human expertise and domain knowledge remain crucial in understanding the context and nuances of natural disaster predictions.

Predicting natural disasters remains a challenging task, and no method, including AI, can offer perfect predictions. Nevertheless, AI can enhance prediction accuracy and speed in specific cases. When combined with traditional forecasting tools and the expertise of meteorologists, geologists, and disaster response professionals, AI has the potential to significantly improve disaster preparedness and response efforts.

Moreover, AI can’t predict sudden, unexpected changes in conditions, such as a sudden volcanic eruption or an earthquake triggered by unknown factors. These uncertainties underscore the need for human oversight and the integration of AI predictions with other forecasting methods.

AI technology will continue to advance and more data will become available, so we can anticipate further progress in predicting natural disasters. Many people believe that AI will bring about several transformative changes in natural disaster prediction and public notifications. 

Here are some ways AI is likely to revolutionize natural disaster prediction:

  1. Enhanced Data Processing: AI will continue to excel in processing vast amounts of data from diverse sources, including satellites, sensors, weather stations, and social media. This capability will lead to more comprehensive and up-to-date information, allowing for more accurate predictions.
  2. Improved Early Warning Systems: AI-powered early warning systems will become even more effective in providing timely alerts to communities at risk. By analyzing real-time data, AI can rapidly detect the onset of natural disasters and issue warnings, giving people more time to prepare and evacuate safely.
  3. Higher Prediction Accuracy: As AI algorithms and machine learning models evolve, they will become better at identifying patterns and trends in historical data. This will lead to improved prediction accuracy for all kinds of disasters.
  4. Integration of Emerging Technologies: AI will likely be integrated with other technologies, such as Internet of Things (IoT) devices and drones, to gather real-time data and provide more comprehensive situational awareness during disaster events.
  5. AI-based Simulation and Modeling: Advanced AI-driven simulation and modeling will help researchers and emergency planners simulate various disaster scenarios. This will enable better understanding of potential outcomes and aid in devising more effective disaster response strategies.
  6. Dynamic Risk Mapping: AI will enable dynamic risk mapping that adapts in real-time as disasters unfold and new data becomes available. This will let authorities adjust response plans as the situation changes, improving overall effectiveness.
  7. Automated Damage Assessment: AI will play a significant role in post-disaster damage assessment. By analyzing satellite imagery and drone data, AI can rapidly assess the extent of damage to infrastructure, helping prioritize response efforts.
  8. Personalized Alerts and Communication: AI-driven personalized alerts can be tailored to individual needs, so the right information reaches the right people at the right time. This will mitigate information overload and improve emergency response coordination.
  9. More Reliable Predictions for Rare Events: With improved algorithms and data analysis, AI will be better equipped to predict rare and extreme events, such as tsunamis and volcanic eruptions, which are notoriously difficult to forecast accurately.
  10. Global Collaboration and Data Sharing: AI will facilitate global collaboration among experts and response teams. It will enable cross-border data sharing and analysis, leading to more comprehensive disaster prediction and preparedness.

It’s essential to acknowledge that while AI offers tremendous potential for advancing natural disaster prediction, there are challenges that must be addressed. These challenges include data privacy, the interpretability of AI models, and ensuring equitable access to AI-driven technologies. Responsible and ethical deployment of AI will be crucial in maximizing its benefits to mitigate the impact of natural disasters and protect communities worldwide.

And it’s important to note that AI predictions are not and will probably never be infallible. They depend on the availability and quality of data, and even with the best models, natural disasters often involve elements of randomness and uncertainty that can’t be completely accounted for. Therefore, while AI tools can greatly assist in predicting natural disasters, they should be used as part of a comprehensive disaster management strategy, not as a standalone solution.

The Most Important Issues in Emergency Preparedness, According to AI

We asked the AI engines what the most important issue was in emergency preparedness.  While none of them gave us the same answer, most of them focused on communication as especially important. We agree. Here are their answers by source:

OpenAI.  

“Emergency preparedness is a multi-faceted discipline that addresses a range of potential crises, from natural disasters to man-made events. The “most important” issue can vary based on regional vulnerabilities, current events, and individual perspectives. However, several fundamental issues are paramount in emergency preparedness:

  • Risk Assessment and Understanding: Which events are most likely to occur in a given area. 
  • Communication: Ensuring that timely and accurate information reaches all affected populations.
  • Public Education and Training: Educating the general public on basic preparedness steps, like creating emergency kits or making evacuation plans, as well as specific groups on their particular needs.
  • Resource Allocation: Agencies must have the necessary resources—human, financial, and material—to respond to emergencies. This includes medical supplies, food, water, and shelter resources.
  • Interagency Cooperation and Coordination: Emergencies often require a coordinated response from multiple agencies, including local, state, and federal entities. 
  • Infrastructure Resilience: Ensuring that critical infrastructure, like power grids, transportation networks, and healthcare facilities, can withstand or quickly recover from disasters is crucial.
  • Evacuation Plans: Well-communicated routes and destinations are vital when evacuation is a requirement.
  • Continuity of Operations Plans (COOP): Businesses, government agencies, and other entities need plans to ensure they can continue essential functions during and after a disaster.
  • Regular Drills and Simulations: Regularly practicing emergency plans through drills and simulated exercises ensures that gaps and weaknesses are identified and addressed.
  • Recovery and Rehabilitation: Planning for the post-disaster phase is vital. This involves rebuilding, providing long-term medical care, and addressing psychological trauma.

While all of these are critically important, if forced to pick the “most important,” many experts might lean towards communication. Effective communication underpins all other elements of emergency preparedness and response. Without it, risk assessments can’t be acted upon, resources may not be allocated properly, and the public won’t know how to protect themselves.

BingAI.

There are many issues in emergency preparedness that are important and challenging, but based on the web search results, one of the most pressing ones right now is addressing the gaps in the nation’s emergency management capabilities across all levels of government. According to a report by the U.S. Government Accountability Office (GAO), FEMA has not determined how to close these gaps, which include areas such as disaster housing, cybersecurity, and climate resilience. The report also found that FEMA has not assessed the federal government’s own capabilities to respond to and recover from disasters, and has not prioritized the use of preparedness grants to address the known gaps. The GAO recommended that FEMA take several actions to improve the national preparedness system and ensure that the nation is ready for any potential hazards.

Another issue that is related to emergency preparedness is ensuring that vulnerable populations, such as older adults, are adequately protected and supported during disasters. FEMA has announced that its 2023 National Preparedness Month campaign will focus on preparing older adults for disaster, especially those from communities that are disproportionately impacted by disasters. Older adults face unique challenges and risks during emergencies, such as mobility issues, chronic health conditions, and social isolation. 

Claude.

Here are a few of the most important issues in emergency preparedness:

  • Climate change – More extreme weather events like hurricanes, floods, and wildfires are requiring better preparedness plans and infrastructure. Planning for disasters that may be worse than historical events is crucial.
  • Aging infrastructure – Much of the infrastructure for emergency response like dams, levees, and power grids are aging. Upgrading and strengthening these systems is an urgent need to be able to withstand disasters.
  • Cybersecurity threats – Essential systems like power grids and communications networks are vulnerable to cyber attacks. Hardening these systems and having backup plans is critical.
  • Supply chain weaknesses – Recent disasters have revealed problems with supply chains for essential goods like food, water, fuel, and medical supplies. Diversifying sources and storage of key items can improve resilience.
  • Pandemic risks – COVID-19 demonstrated gaps in pandemic preparedness on a global scale. Planning and infrastructure for test kits, protective equipment, medical facilities, vaccines, public communication, etc. needs improvement.
  • Inequality in impacts – Disadvantaged communities often are disproportionately impacted by disasters. More focus is needed on assisting vulnerable populations before, during and after emergencies.
  • Public awareness/engagement – Educating and involving the public in preparedness, such as having emergency kits and plans, improves community response capacity. More work is needed to motivate and enable families to be ready.

Bard.

The most important issue in emergency preparedness right now is effective communication. In the event of a disaster, it is critical that emergency responders and the public be able to communicate effectively with each other. This is especially important in large-scale disasters, when communication systems may be overwhelmed or damaged.

There are a number of challenges to effective communication in emergencies, including:

  • Disruption of communication infrastructure. Natural disasters such as earthquakes and floods can damage communication infrastructure, such as cell towers and power lines, making it difficult for people to communicate.
  • Language barriers. It’s important to have communication plans in place that can reach people who speak different languages. This is especially important in areas with large immigrant populations.
  • Information overload. In the event of a disaster, people may be bombarded with information from many sources, including news media, social media, and government agencies. This can make it difficult to find accurate and reliable information.

To address these challenges, emergency planners need to develop communication plans that are:

  • Redundant. There should be multiple ways for people to communicate, in case one communication channel is disrupted.
  • Multilingual. Communication plans should be translated into multiple languages.
  • Fact-checked. Emergency planners should work with news media and social media platforms to ensure that accurate and reliable information is being shared.

By addressing these challenges, emergency planners can help to ensure that effective communication is possible in the event of a disaster. This will help to save lives and reduce the impact of disasters on communities.

Hyper-Reach Increases Texting Capacity

We’ve been growing a lot lately, which is a great thing. We now have customers in more than 30 states from coast to coast and the communities we serve have a combined population of more than 10 million people. 

All that growth means we need to keep growing our calling, texting, emailing and messaging capacity. And we do. Because we’re leveraging the power of the Amazon cloud – as well as other cloud services (for redundancy and reliability) – we have practically unlimited scalability. 

But we also want to speed up our messaging throughput, so that we don’t just keep up, but actually improve our service to you. 

So we’ve recently increased our text messaging throughput by 50%. Which means that when you send a message, it’s going to be delivered faster than ever before.  

It’s our great privilege to serve our customers. And because we appreciate that opportunity we keep looking for ways to serve you better. Please let us know your ideas for how we can do better. And thanks for being a Hyper-Reach customer.

New User Credential Process for Hyper-Reach

As of March 20, Hyper-Reach implemented an enhanced username/password scheme for new and changed PINs, also known as passwords. If you’re an existing Hyper-Reach customer, this won’t affect you or your staff immediately, but you’ll still want to know about it for three reasons: 

  1. This new username/(PIN)password approach provides for more protection for your user accounts. 
  2. The new approach is closer to what most cloud-based systems use today, so it will be more familiar to users who are new to Hyper-Reach. 
  3. Because the new approach is required for any new and changed passwords, your organization will need to use it for any new users or if anyone on staff needs to change their password. 

The new requirements for a username are pretty standard: 

-It must be unique. If you try a username that’s taken, the system will give you an error message. Just pick a different name.

-You can use any combination of printable ASCII characters, except the “%” sign.  That’s almost everything on a standard US keyboard.

-It must be at least 8 characters long. You can use easily remembered usernames such as your email address, first and last name, etc. 

The requirements for a PIN or password are a little different from those for a username:

-Your PIN/password can’t be the same as your username.

-It must be at least 8 characters long. 

-The PIN/password must contain at least 1 digit.

-It must contain at least 1 “special character” – mostly, those are the characters on the top of your keyboard, such as ~, `, !, @, #, etc.

-You must use at least 1 uppercase AND 1 lowercase alpha character.

Examples of valid PIN/passwords are “A2NcF$rTxx”, “1@3$aBcD” and “135&(_xYz”. 

Because these password practices are common for many systems, most people shouldn’t have any problem with them. 

If you’re an existing Hyper-Reach customer and you want to keep your existing usernames and PINs, you are welcome to do that. The system only requires these changes for new or changed PINs/passwords.

And if new users need access for launching campaigns using the IVR/telephone, we can help with that. Just contact customer service for help.

Another Pandemic Just Around the Corner?

COVID-19 wasn’t the world’s first global pandemic and it won’t be the last. A study by the National Academy of Sciences put the probability of a pandemic with similar impact to COVID-19 at 2% in any given year. And that’s based on historical data. The same study estimates that the chance of novel disease outbreaks will grow by 3X in the next few decades. So what if a new pandemic happened this year?

The study’s authors did the math and concluded that it’s highly likely a similar pandemic will happen in the next 59 years. They also note that the probability is the same in any one of those 59 years. So what if it happened this year?

One analyst we follow – Zeynep Tufekci – recently wrote “[the world] is facing the possibility of a pandemic of a far more deadly pathogen [than COVID-19.]”  She was talking about avian influenza – the disease behind the recent huge increase in the price of eggs.  And she’s not alone in warning of the threat.

For birds, avian flu is already a worldwide disaster. It’s been found in 47 US states, as well as 37 European countries, Latin America, the Caribbean, and Asia. Almost 60 million domestic poultry and more than 100 million in the western hemisphere have died or been culled because of the flu.

Fortunately the particular strain of flu we’re talking about – H5N1 – doesn’t infect mammals very frequently. But that appears to be changing. Last October more than 50,000 mink were destroyed when H5N1 was found at a mink farm in northern Spain. Seals, sea lions, bears, foxes, skunks and humans have all been found with the virus. 

And among humans, it’s pretty deadly. Of the known cases, 56% of H5N1 infections have resulted in death. Of course, the mortality rate can be much lower and still create great havoc, as COVID-19 showed us. 

Because of the increasing numbers of interactions between humans, animals and diseases, the risk of diseases such as H5N1 mutating into something that’s highly transmissible among people goes up. 

The mink incident is especially troubling. As Tufecki writes: “When the coronavirus infected Danish mink farms in 2020 and the minks generated new variants that then infected humans…the outbreaks were uncontrollable. If different strains of flu have infected the same person simultaneously, the strains can swap gene segments and give rise to new, more transmissible ones. If a mink farmworker with the flu also gets infected by H5N1, that may be all it takes to ignite a pandemic.” Others agree: “This outbreak signals the very real potential for the emergence of mammal-to-mammal transmission,” according to Michelle Wille, a University of Sydney researcher. Adds Isaac Bogoch, an infectious disease specialist, “This is an infection that has epidemic and pandemic potential. I don’t know if people recognize how big a deal this is.”

H5N1 is just one of many pathogens that could pose a serious problem. Nipah virus, Zika, Mers, swine flu, yellow fever, Marburg virus and Ebola are just a few of the diseases that scientists have identified as capable of evolving and spreading around the world. 

Given the probabilities and the damage such diseases can cause, are we ready?  A lot of people don’t think so. As the Atlantic wrote just six months ago: “.. many public-health experts, historians, and legal scholars worry that the U.S. is lapsing into neglect, that the temporary wave of investments isn’t being channeled into the right areas, and that COVID-19 might actually leave the U.S. weaker against whatever emerges next.” The article goes on to describe various criticisms of the investments that the US is making in disease monitoring and prevention. 

Socially, we also seem to be unprepared for the next pandemic. More than 2 years after the first vaccines were available, less than 70% of the US is fully vaccinated. Less than half of Americans wear masks even occasionally while in public, including me, by the way. And opposition to both vaccination and mask wearing mandates is as high as 30% in some of the polls we’ve seen. It’s probably actually higher than that. 

As a public safety or emergency management professional, your job is to prepare for events that can harm the public. The point of this article is just to remind you that one of those events – a massive disease outbreak – will be a constant and potentially imminent threat for many, many years to come. 

One step in preparing is having a mass emergency notification system.  If your community doesn’t have one or isn’t happy with the one it has, it’s time to check out Hyper-Reach. You can book an online demo here. 

Request a Demo

911, Smartphones and False Alerts

Although Spring is here, we’re still interested in some recent articles we saw about skiing, Apple devices and 911.  

Apple has a feature that automatically calls 911 when some devices detect that a user is in trouble, for example, in a fall or a crash. The device will attempt to alert the user, ask if they are OK, and then call emergency services (aka 911) if the user doesn’t respond. 

While this feature seems potentially useful, it was causing havoc in some ski areas this winter. Apple 14 phones were calling the 911 centers in places like Summit County, Grand County and Vail, CO from the ski slopes in those areas. 

When people ski, they fall, even if the falls aren’t bad. And ski slopes have professional ski patrols on the scene. So even though there was no need for 911 to respond, the calls were coming in at a rate of 20 or more per day and taking up valuable resources. 

While this problem will undoubtedly get fixed, or at least greatly minimized, this news got us thinking about automated devices generally and their connection to 911 and emergency responders. 

There are tens – maybe hundreds – of millions of devices that are, or could be, programmed to automatically call 911 or other emergency responders. These include smartphones, watches, smart speakers, security monitoring systems, fire detection, automobiles, and wearable health monitors. The GPS device on my bike has a crash detection feature that can notify my wife in the case of a bad fall. And there are emerging categories of “smart” clothing that can monitor your heart rate and other vital signs, as well as smart glasses, rings, and more. 

At an institutional level, there are now devices meant to automatically detect the sound of gun fire, explosive vapors, flooding and other hazards. This is in addition to traditional fire and burglary alarms. 

And beyond completely automated systems, there are an increasing number of devices that make it easy to report a hazard, including so-called “panic button” apps on smartphones, other wearable panic buttons, smart speakers, etc. Most Android and iOS phones come with a built-in SOS feature to call 911 and send messages to emergency contacts, in addition to any number of specialized apps available for the purpose.

While we think the increased availability of monitoring and emergency calling systems has great potential, it also comes with a lot of issues that must be dealt with, including: 

– False alarms. This is already a problem across the US. An Arizona State University paper said 20 years ago that “phantom” wireless calls to 911 were between 25-70% of all 911 calls in some jurisdictions and a more recent estimate we found claimed that 90% of all calls to 911 were false alarms. 

– Privacy. For these devices to be effective – at least the mobile ones – they need to track the location of the user. Which raises all kinds of new privacy and security concerns, since calling emergency services means letting the government know where you are.

– Hacking. Because there are many different companies that are offering emergency calling capabilities, there are a variety of security standards that are, or aren’t being used. That could open the door to potential hackers to figure out a way to send a crippling number of calls to emergency responders. 

It may be that none of these problems are insurmountable. But the technology is evolving rapidly, so there are bound to be mistakes and possibly great disruptions along the way to figuring it all out. 

In the meantime, we think the best posture is to be aware, adopt those technologies that seem promising, and offer useful guidance to your residents, who are going to adopt some, many or all of these technologies and might very well appreciate your helpful advice. With that in mind, we’ve drafted a guide that you can freely take and adapt to your own uses with your residents.

Sending Alerts Through Alexa-Enabled Smart Speakers

We’re always looking for new ways to deliver alerts to the public, which is why we developed an interface with Amazon’s Alexa smart speaker. 

With the Hyper-Reach Alexa interface, you can send messages to your residents who enable Hyper-Reach on their Alexa-enabled devices. Those include Amazon’s Echo devices, but also much more, including speakers from Sonos, Bose and Bang & Olufsen. All a user has to do is say “Alexa, enable Hyper-Reach” and follow Amazon’s instructions. There’s no cost to your residents, and this feature is included in all of our public safety packages, so there’s no extra cost to you. 

Smart speakers are an important – and growing – way for people to communicate. There are more Alexa-enabled devices than home telephones and that keeps increasing every year as people disconnect their landlines and smart speakers continue to grow. 

And the value of smart speakers goes beyond just the number of people you can reach. As we point out in this month’s issue, Alexa is a useful way to reach people with access issues, such as visually impaired and physically disabled folks.  So Hyper-Reach’s Alexa capability lets you reach people in your community who are especially difficult to reach. 

We’ll keep working on new ways to help you reach your residents. To find out more about Alexa and how you can use it, book a demo or ask your Hyper-Reach customer service representative.

Extreme weather forecasts can save lives — but that doesn’t mean the public always listens

Extreme weather forecasts can save lives — but only if people listen and respond. The best forecast is useless if people don’t act on the information, a reality that applies as well to emergency alerts. So we did a roundup on what experts are saying about why people ignore or respond inappropriately to weather warnings. 

Normalcy bias or optimism bias

Normalcy bias, or normality bias, is a cognitive bias which leads people to disbelieve or minimize threat warnings. Consequently, individuals underestimate the likelihood of a disaster, when it might affect them, as well as its potential adverse effects

Normalcy bias makes it difficult for us to engage in “worst-case” thinking and plan for a serious failure or disaster. This kind of bias causes people to assume that, although a catastrophic event has happened to others, it will not happen to them. People often base their decisions on previous experiences, such as other storms they’ve lived through. 

Part of the challenge is that forecasts are uncertain, so the area covered by a warning is necessarily larger than the area that’s actually affected. Most people who receive warnings don’t experience the actual event, which can cause them to discount future warnings.  For tornadoes, for example, meteorologist Dr. Kim Klockow-McClain puts it this way, “Even within a given event, less than 1 per cent of the spatial extent of a tornado-warned area will actually experience a tornado, and about 70 per cent of all tornado warnings will result in false alarms.”

The consequences can be deadly. In 2011, one of the deadliest tornadoes in US history in Joplin, Missouri, killed 158 people and injured more than 1,000 others. NOAA’s assessment of the relevant NWS warnings and forecasts found that some residents had become desensitized, and that “initial siren activation has lost a degree of credibility for many residents.”

About 70% of people reportedly display normalcy bias during a disaster. The normalcy bias can manifest in response to warnings about disasters and actual catastrophes.

The challenge of “probability” and how people process information

Translating weather risks and emergency alerts into terms the public can understand is important, but difficult. 

Dr. Marshall Shepherd – director of the University of Georgia’s atmospheric sciences program – talked about this challenge on an NPR podcast “The Science of Extreme Weather” (The Pulse : NPR). 

For example, he talked about the “cone of uncertainty” – the projected path and intensity of a hurricane or tropical storm – a concept that can be hard for people to wrap their heads around and is often misconstrued. The cone suggests a 2 out of 3 chance that anywhere within that cone will be the center of the storm. For many people, this is not an easy idea to grasp.

In one example, a TV reporter went to grocery stores in Sarasota, Florida, 3 days prior to Hurricane Ian making landfall to see how people were preparing. The folks she interviewed were “just doing their regular shopping”. Nobody seemed especially worried. Sarasota was in the “cone of uncertainty”. 

Some experts believe that most people do not ignore the information they get in warnings.  Dr. Jen Henderson, with the Cooperative Institute for Research in Environmental Sciences says, “…from the interviews and focus groups we’ve done, people are not complacent. They’re all taking action, it’s just not the actions we’d expect or we can see.”  Julie Demuth at the National Center for Atmospheric Research agrees: “For the most part, people don’t disregard weather warnings. But that doesn’t mean they’re always going to do what we want them to do.”

As a practitioner, it’s worth noting that the academics don’t all agree. Dr. Shepherd thinks the public and policymakers need to be trained on how to consume information better. But Susan Joslyn, associate professor of psychology at the University of Washington disagrees. She studies the way people make choices when given weather information. “People can’t absorb and use information unless it’s tailored to how they’re thinking about it and their decisions. To evacuate or not.” Professor Joslyn believes that people can handle more complexity than they are given credit for. 

Despite the lack of consensus, we can hope that continued research will help make information more relevant and impactful.  As an article on the Weather Network puts it: “Bridging that gap — between basic weather information and peoples’ response to it — is a key area of research for the future of severe weather communication…”

There’s no easy solution to the challenges here. Data from the 2020 NHS Data Digest, suggests that the rate at which the adult population becomes prepared or maintains preparedness for emergencies has stalled over the years, despite the fact that extreme weather events are becoming more common and more extreme. So there’s a critical need to encourage, guide, and assist individuals and communities to move from thinking about a potential emergency and actually doing something about it. 

One benefit of emergency alert systems such as Hyper-Reach is that you can deliver information to residents in a highly localized and specific way. And you can use the alert system to tell residents exactly what actions you want them to take. 

Greater precision might give you the ability to communicate more effectively with the public. Dr. Joslyn’s research shows that most people can understand numeric likelihoods, so that if they’re told there’s a 20 or 30 percent chance of something happening, they make better decisions than they might have made without this information.

Joslyn’s work implies that it’s better for people to have the numeric information because they’re going to make their own estimates anyway, based on past experiences. If they’ve experienced false alarms in the past, they may end up underestimating the risk, but more accurate information about risks and uncertainty can reduce the misestimating – and that could save lives.

Taking the message to where people are – on social media

Traditional media such as radio and cable television, is on the decline, particularly among younger residents. More and more people are consuming content on social media. And that’s both a challenge and an opportunity for emergency managers trying to get their message out. 

We live in a world where 53% of Americans get their news from social media.  And that number is growing.

In 2022, there were an estimated 270 million active social media users in the United States. That’s about 81% of the total US population, a number that grew 12.5% compared to the previous year.

Facebook is the most popular social network in the US, with 228 million active users. There are 186 million active Instagram users, and TikTok is popular and growing among adults under 30.

The average Facebook user spends approximately 20 hours on the app per month and overall social media usage is estimated at more than 60 hours per month. 

With the development of modern society social media has evolved into a viable communication tool in all aspects of life including emergency notifications. And that’s why many Emergency Managers already use social media when sending important messages to their community. 

We agree that social media is a great tool to share your information if it’s used as an integral piece of an overall communication strategy. Here’s why:

  1. Milling. People automatically turn to social media and the Internet to gather more information to both learn new things and to confirm information they’ve received.  Emergency managers use the term “milling” to describe when people try to confirm an alert or other emergency communication. And social media is an important way for people to find confirming information. 
  1. Snowballs.  Social media gets repeated among its recipients, a phenomenon called the social media snowball effect.  Even if a post initially reaches only 10% of its intended audience, it has the potential to reach many times that number. People tend to share information they consider valuable with their friends and family and each of those recipients can share that information in turn.  And social media lets them do it instantly with only one click across different social tools
  1. Growth. There are more social media channels than there were 10 years ago and we’d bet good money there will be more in another 10 years. Two examples are TikTok and  Nextdoor. We are especially excited about Nextdoor because it’s built around neighborhoods, which makes it perfect for geo-targeted alert messages. And Nextdoor has a service for public agencies that lets you send messages to everyone in your jurisdiction without needing them to follow you. As one head of emergency services put it: “DO NOT ignore this platform.  It is growing and agencies who work with NextDoor have direct access to subscribers”  In fact, we’re so excited about Nextdoor, we built an integration, which you can read about here.

Some emergency notification providers think that social media should not be considered as an emergency notification tool because of some of its limitations. For example, one emergency notification provider argues that social media should not be used for sending out emergency alerts and community notifications because it does not reach people reliably. For example, a Facebook post will only reach people if they are connected to Facebook and had liked your page previously. Your posts might or might not appear in their news feed and you might reach only about 10% of your population. 

What this argument misses is that social media is one tool in a tool kit.  And you don’t use a hammer when you need to drill a hole. 

Social media can be a great tool for sharing information. But, you should not rely on it as your only means of sending out emergency alerts. Consider it as a part of your communication plan and use it in a bundle with other proven emergency communication channels such as:  text messages, calls, IPAWS WEA alerts etc. That way you will fill in possible gaps and cover as many people as possible. 

Hyper-Reach emergency notification system integrates with the most popular social media channels for sharing emergency & community alerts such as Facebook, Twitter and – now – Nextdoor.  We also give you access to every other tool, so you have the most powerful of toolkits. To discover more, book our demo.

More accessible emergency alerts for people with disabilities

To be as effective as possible, emergency alert and notification systems need to reach as many people as they can. This includes folks with access and functional needs, such as disabilities that make it difficult to respond to emergencies quickly. 

The Federal Emergency Management Agency (FEMA) defines access and functional needs as including older individuals and people with physical, sensory, behavioral, mental health, intellectual, developmental and cognitive disabilities. Also included are people with limited English language skills, access to transportation, and/or financial resources to prepare for, respond to and recover from an emergency. And that’s a lot of people. 

According to the Centers for Disease and Control Prevention (CDC), the United States counts around 61 million adults with disabilities. That’s about 1 in 4 adults, or roughly the population of Italy.

47.7 million of Americans have some hearing disabilities and 56.4 million people have some difficulty seeing.  Emergency alerts often overlook these vulnerable populations, which means that people with disabilities often encounter communications that are inadequate or effectively unavailable. This study by the National Council of Disabilities looked at several major disasters, including Hurricane Sandy and the Boston Marathon bombings, and outlined these common barriers to effective communication during emergencies:

● Televised emergency announcements by officials that do not include American Sign Language interpreters;

● Inaccessible emergency notification systems;

● Inaccessible evacuation maps;

● Websites with emergency information that is not accessible to screen readers used by people who are blind or have low vision;

● Shelters at which no one is able to communicate with people who are deaf or hard of hearing;

● Emergency communication using language that is inaccessible to people with intellectual or developmental disabilities and people with limited English proficiency. 

According to this fact sheet on aging in the United States created by the Population Reference Bureau, the number of Americans ages 65 and older is projected to nearly double from 52 million in 2018 to 95 million by 2060, and the 65-and-older age group’s share of the total population will rise from 16 percent to 23 percent. And while many older people stay sharp and capable, an older population means more people with diminished eyesight, hearing, mobility and thinking skills. 

This Emergency communication survey from 2014 shows how people with disabilities received, verified, and shared emergency alert information during their most recent public alert:

An external file that holds a picture, illustration, etc.
Object name is i1936-900X-15-5-567-t04.jpg

Because the survey is a few years old, we believe that the use of smartphone apps, IM and other web-based services has increased. But the general point remains: the more notification channels you use the better your chance to cover those with access and functional needs.  As the National Association for the Deaf puts it “There is no “one” system that is best for alerting citizens in an emergency. …the message should be sent out to as many people and in as many formats as possible (by television, radio, phone/TTY, computer, cell phone, text messaging, pager, and other means).”

As an emergency service provider, we’ve been working hard to make our Hyper-Reach notification system as accessible as possible:

  • We use as many notification channels as possible to send out a message: text, phone calls, social media, IPAWS, browser push notifications, and Alexa.
  • Our telephone calling capability fully supports TTY/TDD messaging.
  • Alexa is especially helpful for people with vision and mobility problems, because of its touch-free audio interface. One commentator called Alexa “the most successful product on the planet” for accessibility for the blind and others have noted how valuable it is for folks with physical disabilities. 
  • Our IPAWS interface gives you access to the Emergency Alert System for broadcast channels, such as TV and radio. 
  • And with Google Translate we let you send your message in almost any language that people use. 
  • Our new interface with Nextdoor adds even more, by making it easy to deliver messages that can then be spread from neighbor to neighbor. 
  • We also make it as easy as possible to enroll in emergency alerts.  For example, we give you a call-in number that lets people without computers or computer skills enroll with just a phone call. 
  • And Hyper-Reach has an automated hotline, which lets you record a message with information you want the public to know. People call in to hear your message, which can be easily updated as the situation changes. Your residents don’t need a computer or even be able to read to access the service. So it’s perfect for the elderly, those with vision impairment, and the illiterate. You can even record the message in multiple languages. 

We’re focused on making Hyper-Reach the best system for reaching the most people – with and without disabilities, access or functional needs. And we’re always looking for new ways to help.  So if you’ve got a suggestion, drop us a note at jveilleux@ashergroup.com.  Or for a demo, click here. 

And if you want to do even more, consider Deaf Link, whose Accessible Hazard Alert System can deliver messages as videos in American Sign Language.