Safe City: A Study of Channels for Public Warnings for Emergency Communication in Finland, Germany, and Greece

Abstract

Public warning systems are an essential element of safe cities. However, the functionality of neither traditional nor digital emergency warnings is understood well enough from the perspective of citizens. This study examines smart city development from the perspective of safety by exploring citizens’ viewpoints. It investigates people’s perceptions of the ways in which they obtain warnings and information about emergencies involving health risks. Data were collected in the form of focus group interviews and semi-structured interviews in Finland, Germany, and Greece. The results suggest that people place a lot of trust in their social network, receiving text messages, and their ability to use web-based search engines in order to obtain public warnings. The study discusses the challenges identified by citizens in the use of conventional radio and television transmissions and sirens for public warnings. Based on the results, citizens demonstrate informed ignorance about existing mobile emergency applications. Our results imply that it is not sufficient to build emergency communication infrastructure: the development of smart, safe cities requires continuous work and the integration of both hard and soft infrastructure-oriented strategies, i.e., technological infrastructure development including digitalisation and education, advancement of knowledge, and participation of people. Both strategic aspects are essential to enable people to take advantage of novel digital applications in emergency situations.

1. Introduction

Cities are complex environments; many challenges are presented by the need for people to be able to live in safe and healthy environments. Modern societies promote both hard and soft infrastructure-oriented strategies in their attempts to tackle a range of urban problems, and to make cities more efficient and liveable places for both residents and visitors. These development efforts are called smart city initiatives. Hard strategies involve the technological development of infrastructure, including digitalisation, while soft strategies refer to the advancement of knowledge and participation of people [1]. In brief, various smart city activities can be characterised as intelligent and efficient responses to a range of urban needs, such as enhanced public safety, environmental protection, or city services [2].

A smart city is also a safe city; one that considers the protection of both people and the environment [3]. In their seminal work, Giffinger et al. [4] identified six key characteristics of a smart city: competitiveness (smart economy), social and human capital (smart people), participation (smart governance), transport and ICT (smart mobility), natural resources (smart environment), and quality of life (smart living). Safety is a necessary and integral aspect of each of these characteristics, but so far it has received less attention than other fields of smart initiatives in smart development strategies [3,5,6].

To protect people and the environment, safe cities possess clear response plans to various major crisis situations, such as fire, natural disasters, chemical spillage, crime, violence, and terrorism, among others [7]. Public warning systems are an essential element of effective emergency management and response plans. Here, safe cities may apply technologies that are not solely reliant on digital innovation, but which are also able to use older solutions that do not necessarily depend on recent, intelligent development efforts [5]. However, the functionality of traditional [8] and digital emergency warnings [9,10] is still not understood well enough from the perspective of citizens.

This study explores citizens’ perspectives of public warning systems. More specifically, the study investigates the following research question: What are people’s perceptions of the ways in which they gain information about emergencies that involve health risks? The current understanding of people’s views and preferences, as well as their overall involvement in crisis management, is still limited [11,12]. However, people’s knowledge and views of existing warning systems are important for the further development of citizen-centric smart initiatives in the area of safety.

The empirical data for this study is based on 15 focus group interviews conducted in Finland, four conducted in Germany, and 23 individual interviews conducted in Greece during the first half of 2021. Our study discusses the roles of both conventional and digital emergency communication and identifies challenges in the use of these technologies. Our study then contributes to the literature on smart cities by bringing the safety aspect to the forefront.

2. Smart City

Smart cities have been studied extensively from a range of perspectives in recent decades. Even though there is no universally agreed definition of what constitutes a smart city [13], there are three recurring characteristics or dimensions shared by studies of smart cities. These are a technology dimension, a human dimension, and an institutional dimension [14,15]. The main aim of all smart city initiatives is to increase the quality of life of its citizens [16].

The first dimension, technology, is essential in smart city initiatives. This means that building and developing smart cities begins with networked infrastructure where different, often state-of-the-art information and communication technologies (ICT) (e.g., mobile phones, computer networks, fibre optic networks, internet services, sensors and connected devices, and the internet of things) play a central role [17,18]. As Siokas et al. [15] note, digital tools have a dual role in smart city initiatives. They digitise data on the spot and provide high-speed services. Siokas et al. [15] emphasise that from the smart city development perspective, it is essential to be clear what the needs and goals of a smart city project are, and to select and use appropriate technologies effectively.

A critical issue or backbone of current network infrastructure is the issue of connectivity as it helps people to access data and services from everywhere [16]. People are continuously connected through smartphones and other digital devices, bringing up the issue of the secure use of ICT. From the smart city perspective, it is important to note that growing connectivity needs efficient and secure communication networks [19].

The second key dimension in smart city discussion is the human perspective. As Hollands [20] notes, progressive smart cities begin with the human approach instead of believing that technologies themselves automatically improve cities. In other words, although technology alone is not enough, smart cities need to put technology at the service of their citizens and not vice versa [1]. Smart city projects first need to be centred on the needs of the residents. This means that the development of smart cities should not be restricted only to the most up-to-date technologies [21,22].

The human perspective also means that smart cities foster the development of human and social capital, in other words, skills, competencies, communities, and networks of relationships. This means that smart people are an essential component of smart cities. The smart people concept comprises elements such as qualification levels, affinity to life-long learning, creativity, open-mindedness, and participation in public life [4].

The third dimension of smart city initiatives highlights the role of institutions in the development of smart cities. This refers to governance structures, policies, regulations, and directives, in other words, the entire administrative environment that is needed to support the design and implementation of smart city initiatives [14]. This perspective has gained more importance in recent smart city research [15]. At best, smart city initiatives foster citizen participation and the development of collaborative and inclusive smart communities [15,23].

3. Safe City

Safety and security are essential building blocks for an attractive urban environment and quality of life in cities [24]. Put differently, quality of life is an important feature of smart cities, comprising not only factors such as culture, health, and housing, but also safety [4]. This means that cities increasingly need to be not only socially, economically, and environmentally smart and sustainable, but also to continuously pursue strategies for safe city development [25].

A safe city can be considered to be a subsystem of a smart city, covering all safety aspects within a city [3]. This means that safe cities aim to reduce and respond to threats or emergencies to allow people to live and move in healthy environments. A safe city thus “ensures protection of society, property, and environment” [5] (p. 838). To this end, it has technologies and services both to prevent and to respond to various crisis phenomena [7].

Safety refers to the “condition of being safe from risk or danger” [26] (p. 22). It is important to note that safety and security are different concepts, even though the terms are often used interchangeably. Kourtit et al. [27] emphasise that safety refers to protection from an unintended threat, harm, or catastrophe, while security refers more to a state in which people feel protected against deliberate threats or harmful actions. Yuen [28] refers to Lang’s notions [29] (p. 234), who highlighted two dimensions of safety and security needs for people in urban environments. On one hand, people have physiological needs that refer to freedom from bodily harm, e.g., the need to feel safe from various types of accidents. On the other hand, people also have psychological needs that refer to having a sense of place, i.e., the need to know where one is and be able to orient oneself in the area. Brands et al. [30] present findings revealing that people experience safety in urban environments when there is either an absence of negative features (e.g., dangers, threats, or feelings of fear) or the presence of positive aspects (e.g., feeling at ease, feeling free, feelings of familiarity). They also found that people feel safe in cities when they do not have to be alert or cautious; in other words, when they do not have to think about safety at all.

Safe cities utilise a range of technologies and services in their efforts to produce urban safety. These consist of systems such as emergency communication networks, public address and general alarm systems, environmental monitoring, fire and chemical control systems, and communications and mass media [7]. Importantly, in smart city development efforts, the safety aspects require consideration of not only the functioning, maintenance, or repair of technologies, but also of organisational and social matters, such as people’s different levels of knowledge and competence in using technological solutions [31]. Not everyone is aware of the technologies available or proficient in their use. People may also have subjective interpretations and understandings of technology usage that may involve a lot of invisible work—even peculiarities—and sometimes these could lead to unintended consequences, such as focusing on attempts to maintain a technology instead of utilising its useful properties [32]. Additionally, in a crisis, people may encounter a range of stress factors such as fear, anxiety, or panic, which may hamper their information processing and use of technology [33]. This means that in safe city development efforts, people may need a lot of technological and social support to implement smart solutions in their everyday lives.

4. Public Warnings: Channels of Emergency Communication

Public warning systems are important in the production of urban safety. A public warning can be defined as “safety communication to inform a population about a threat from an imminent or ongoing emergency” [34] (pp. 168–169). The purpose of warnings is to provide information about imminent or ongoing emergencies and to influence or control people’s behaviour by instructing them on how to respond to such emergencies, in order to either avoid or minimise detrimental consequences [34].

Laughery [35] points out that effective warnings accomplish two objectives. First, they get noticed and encourage people to encode the content of the warning. Second, effective warnings provide understandable information for people to be able to make informed decisions. Therefore, warnings need to contain information about the hazard and potential consequences, as well as instructions regarding safe and unsafe behaviour. Laughery [35] also emphasises that people’s compliance decisions may depend on several factors, such as the design of the warning system, various familiarity issues (e.g., previous experiences in the environment), and situational factors (e.g., the situation in which the warning is presented).

Several channels can be used to transmit emergency warnings, such as radio, television, newspapers, social media, and mobile phones [36]. One of the more widespread and traditional tools to alert the public to severe danger is sirens. Sirens may theoretically provide many advantages in emergency communication, but existing research has found that people do not pay much attention to them, nor do they understand the meanings of different sounding signals [37]. Yet for example, a recent study in France has shown that the public identifies sirens as the most effective alert system over alternatives such as mobile phone-based alerting tools, even though they are rarely used by the authorities [8]. This result is even more confusing, as the same study shows that in a mock emergency, most people did not know how to respond to the siren, or even if they did know, they did not react appropriately upon hearing it [8]. Another problem with siren-driven alert systems is related to their audibility. Optimising the warning siren system placement is a challenging task, particularly in complex urban environments [38].

Digital technologies such as warning apps are increasingly being developed to distribute public warnings. These are software applications that run on mobile devices, such as mobile phones or tablets, to deliver warning messages from authorities to the public during emergencies. Such apps can be effective channels for emergency communication, provided that members of the public install the apps and use them. Fischer-Preβler et al. [34] studied non-users’ intentions to start using and users’ intentions to keep using a warning app. They found that these intentions were positively affected by people’s perception of the trustworthiness of the information delivered, their motivation to comply with the expectations of key referents such as family and friends, and their belief that the use of the warning app will be effective in protecting them. Fischer-Preβler et al. [34] also found that the use intention was negatively affected for both non-users and users if the app took up too much system capacity, i.e., battery charge or storage space, on a smartphone.

It is essential that digital technologies provide the correct amount of information in order for people to react appropriately to warnings. Short disaster warnings (90 to 140 characters) have been found to be uninformative, causing fear and confusion as well as milling, i.e., seeking additional and confirming information before taking any protective action [10]. People may seek to confirm information by other means, such as surveying the environment or communicating with other people [39]. An important challenge is therefore to develop warning messages that are brief enough for people to be able to take in the information, but long enough to reduce milling, thus shortening warning response delays [40]. Zhou et al. [41] found evidence of the advantage of using information graphics, i.e., different formats of data visualisation in crisis communication. For example, combinations of time- and action-related icons with hazard messages have been found to increase people’s intentions to act in hazardous situations [42].

Another challenge is the audio output for alerts. Users want the audio to be distinguishable when heard and of an appropriate volume to make them aware of the situation, without causing any annoyance that may lead to them ignoring alerts or uninstalling the app [43]. An additional challenge lies in formulating and designing warnings that will also reach people with diverse disabilities [44]. Overall, it is important to consider inclusiveness policies when new emergency communication tools and protocols are being developed [45].

5. Research Setting and Methods

As part of the ‘GIDPROvis’ research project [46], this study investigates people’s views of the existing information and communication channels for potential emergency situations involving health risks. Due to the nature of our research project, there is a strong focus on emergencies resulting from chemical hazards and accidents. To investigate people’s perceptions of public warning systems, we interviewed citizens in three countries: Finland, Germany, and Greece. These countries represent the home countries of our EU-funded research project partners responsible for data collection. The focus on interviews was twofold. The first part of the interview guide centred on people’s understanding of chemical hazards and risks, while the second part examined their views on public warning systems. This study focuses on the latter part of our interviews.

The original plan was to implement focus group interviews in all three countries. Due to the global COVID-19 pandemic, all interviews had to be held online. In Finland and Germany, we conducted focus group interviews in winter and spring 2021. Because of unforeseen technological and scheduling difficulties, focus groups could not be implemented in Greece. Therefore, in Greece, the data were collected in the form of online semi-structured interviews in the spring and summer of 2021. For the same reason, some of the planned focus groups changed into semi-structured interviews with individual interviewees in Finland.

The key feature of focus group interviews is group interaction and comments made by participants to each other. However, one of the key challenges of focus groups is managing either non-talkative or dominant participants. Here, we followed the recommendation by Lobe et al. [47] to keep the online focus group size small and the interviews well-structured to allow better interaction. Most of the focus groups had between two to four participants and one group had five participants. Small group size and well-structured focus group interviews helped us to ensure that as with the individual interviews, all focus group participants were able to express their perspectives on all questions.

In Finland, we conducted 11 focus groups with 36 participants and four semi-structured interviews with individual interviewees. These participants were recruited in three ways. Firstly, convenience sampling (referring to ready availability) was used. This is the most common way to recruit participants for focus groups [48]. Secondly, focus group participation was advertised to the members of two associations: Enter and SIGCHI Finland. These associations are interesting as they operate in the area of digitalisation in society. The members of the Enter association are volunteer senior citizens who offer free in-person support for the elderly in the use of computers and other everyday digital technologies. SIGCHI Finland promotes the furthering of knowledge and encourages interest in the field of human-computer interaction in society. Thirdly, five of the 11 focus group interviews were conducted with 21 second- and third-year university students who were undertaking a bachelor’s degree programme to become high school science teachers (including chemistry). All the interviewees lived in southern Finland.

In Germany, we conducted four focus group interviews with 11 participants, and in Greece, 23 semi-structured interviews were held with individual participants. Convenience sampling was used in participant recruitment in both Germany and Greece. In Germany, the interviewees lived in the northern or north-western part of the country (Hessen, Lower Saxony, Nordrhein-Westfalen, and Schleswig-Holstein). In Greece, most of the interviewees lived in or around Athens, but some participants lived in other parts of the country.

The interviewees in our study consisted of students, employees, and pensioners. Apart from the students in the five focus groups in Finland, who had chemistry-related backgrounds, the interviewees represented a range of different professional fields. This was relevant for the purposes of our study, as it meant they were not educated or trained to deal with chemicals, chemical risks and hazards, or chemical communication in their everyday occupational activities. In total, 74 people between the ages of 18 and 84 participated in interviews in the three countries. Table 1 summarises the information on interviewees in three countries.

Before the interviews, all interviewees received a data protection notice containing information about the processing of personal data. All research participants also provided fully informed consent to participation in advance. All documents were provided to the research participants in their native language. The interviews were conducted in Finnish, German, and Greek in their respective countries. All interviews lasted between 45 and 90 min, offering space for an intellectually intensive discussion. All interviews were recorded and later transcribed verbatim.

In the interviews, we discussed the ways in which people currently obtain information on emergencies involving health risks. We also asked about interviewees’ experiences of the existing warning apps used for emergency communication (112 in Finland [49], NINA in Germany [50], and the forest fire risk warning application of the General Secretariat for Civil Protection in Greece [51]).

In the analysis phase, we first compiled data from each country separately. We then read the transcriptions several times to familiarise ourselves with the empirical data. Next, we conducted a content analysis by systematically examining the whole empirical data set and focusing on themes and patterns emerging from the data [52,53]. This resulted in the identification of three categories of technologies for delivering public warnings. These were communication media, sirens, and mobile emergency applications. We continued the analysis by taking a closer look at each category. This stage was guided by a focus on the interviewees’ views and experiences of the different technologies. In the analysis phase, we continuously iterated between empirical data, emerging observations, and the existing literature [54,55].

6. Public Warning Systems: Citizens’ Reflections

In the interviews, the participants discussed several potential channels for obtaining emergency information and public warnings. The recognition of many different channels is essential since people often need to confirm the information in warnings from multiple sources before acting [56]. Some interviewees emphasised the importance of their senses of smell and sight in the recognition of potential health hazards in their own living environments.

“You smell it or see [the light flashing on] the sirens.”

(Male, 20+ years, Germany)

However, most interviewees reflected on three distinct types of technologies for delivering public warnings. These were: firstly, communication media including conventional radio, telephonic and television transmissions, as well as digital search engines and social media; secondly, sirens; and thirdly, mobile emergency applications. We analysed findings regarding these different groups of channels.

6.1. Communication Media: Trustworthiness and Serendipity

Participants talked a lot about the importance of different conventional communication media such as telephones, radio, and television in the delivery of public warnings. Many seemed to rely on their social network, meaning that people expect that someone close to them, e.g., a family member, friend, or colleague, will learn about the situation and inform them, either at home or by phone.

“If I think about a real situation, I do not recall that there has been one, but I think I would get information from a colleague or from [my husband]. Unfortunately, I’m not good at following these kinds of things. So, most likely someone else would warn me.”

(Female, 50+ years, Finland)

“Maybe a friend would call me.”

(Female, 80+ years, Finland)

“Asking someone more specialised than me. If not, I would look around, depending on where I am, and call people I know.”

(Male, 50+ years, Greece)

In addition to conventional telephone calls, many mentioned radio and television transmissions as important channels for public warnings. At the same time, their ability to reach the public was considered to be haphazard, as people are only able to hear the public warnings if they have their devices switched on. Regarding radio, many interviewees mentioned that they seldom listened to the radio and most often only while driving.

“I think the advice is to turn on the radio and listen for information. Whatever. I never actually listen to the radio.”

(Female, 50+, Finland)

“In the car, we often have the radio on. … At home, we don’t listen to the radio very much.”

(Female, 65+, Finland)

Regarding television, the Finnish interviewees mentioned the red banner that appears in the middle of a programme to inform and warn the public about acute, severe danger in the area, for example, a large industrial fire or a bear sighting in a residential area. This type of television warning is often emphasised with a beeping sound that some interviewees found annoying. One interviewee posed a question that reveals why radio or television alone are not sufficient warning systems for the public:

“If there is some dangerous chemical in the air, can we assume that people have their television or radio on?”

(Female, 20+ years, Finland)

Some of the communication channels embedded in the television transmission simply seemed outdated to interviewees.

“I don’t know if anyone in the 2020s uses teletext, but you can probably find the information there as well.”

(Female, 30+ years, Finland)

In addition to conventional media, interviewees also talked a lot about the importance of various digital media (e.g., search engines and social media) in emergency communication and the delivery of public warnings.

“Almost everyone has the Internet on their phone. Therefore, you just type in the search term ‘emergency situation’.”

(Female, 20+, Finland)

“On my mobile phone, if I swipe to the left, there is a list of news updates. Or, on YouTube, I see something like that too.”

(Male, 20- years, Germany)

In addition to experts, local and national news, as well as news from the authorities, were considered to be reliable sources of information on the Internet.

“News from the police.”

(Female, 25+, Germany)

“I will turn to an expert, depending on the type of risk there is.”

(Female, 50+ years, Greece)

Unsurprisingly, a stable Internet connection was considered to be critical for receiving information.

“On the Internet. But it is scary if the Internet fails. Then we are in real trouble.”

(Female, 65+ years, Finland)

Interestingly, many expected to receive a warning via a text message on their mobile phones.

“Nowadays, when everything is so advanced, there must be a text message to everyone’s phone.”

(Male, 20+ years, Finland)

“I think that nowadays there would be a text message.”

(Female, 50+ years, Finland)

“Text messages are perhaps the best technology to deliver the information.”

(Male, 70+, Finland)

However, sending text messages can be costly and time-consuming, often requiring a lot of human effort. For these reasons, text messages are not used for emergency communication in every country.

6.2. Sirens: Various Uncertainties

Many interviewees reflected on the importance of sirens in emergencies. In many cities in Finland, the functioning of the sirens is tested at regular intervals: for example, at noon on the first Monday of every month. This testing practice was not familiar to all interviewees.

“I don’t know if I have lived somehow… I did not know about [the sirens]. It was only about six months ago, when I was at home. … There was a terrible alarm siren and I learned that it was only a test.”

(Female, 20+ years, Finland)

Those who were aware of the testing practice reflected on the real situations in which sirens would be used. One interviewee pointed out that the siren sound can be inaudible in some locations. This can happen for many reasons, such as geographical location. In general, hearing the siren at an unusual time was mentioned as an important indication of a warning for the public.

“You hear the sound and recognise it. On the other hand, you are not alarmed by the sound as you learn that there is always a test at the same time. If you hear the siren at an unusual time, you know that you must become alert.”

(Female, 50+ years, Finland)

Importantly, the siren sound is only an indication of some sort of emergency for the public. It was not clear to all of the interviewees what situations the sirens would be used in.

“They test the alarm system like Mondays at noon, but I have been thinking, in what real situations would these be used? If I wasn’t following any media and then a nuclear plant exploded somewhere, would there be an alarm that everybody would hear?”

(Female, 20+ years, Finland)

In addition to awareness and hearing issues, another matter that raised questions and uncertainties was people’s reactions to the sirens. Some interviewees doubted that people would react to sirens and were unsure whether hearing the siren would influence people’s behaviour in any way in a real crisis. Some suggested that spoken information delivered in connection with the siren sound could be used to clarify the situation for the public. Not everyone was convinced by this suggestion, and they felt that it would be difficult to understand spoken words delivered through a loudspeaker.

“Well, it is a test alarm. It doesn’t cause any reaction. It probably wouldn’t cause any reaction even if it was a real alarm. Because you would only hear some metallic mumbling.”

(Male, 60+ years, Finland)

Thus, although sirens are traditional public warning systems, there seem to be many ambiguities and uncertainties among members of the public regarding their use and functioning.

6.3. Informed Ignorance of Mobile Emergency Applications

Warning apps are growing in importance as a channel for emergency communication in many countries. 112 Suomi is a mobile app produced by the Emergency Response Centre Agency under the Ministry of the Interior in Finland [49]. The comparable warning app in Germany is called Warn-app NINA [50]. In Greece, the forest fire risk warning application of the General Secretariat for Civil Protection serves as a channel for issuing public warnings about forest fires [51]. The Greek application takes advantage of text messages as a channel for delivering such warnings.

Many but not all of the interviewees in our study had heard about the existing warning apps in their home countries. Many even had positive images of these warning apps, but they had not downloaded them themselves.

“This is the first time I’ve heard about this application.”

(Female, 20+ years, Finland)

“I don’t have it but I know that it is really good.”

(Male, 20+ years, Finland)

“No, I have not [heard of it]. Nina, just like the name ‘Nina’?”

(Female, 25+ years, Germany)

“I have heard of it; I do not know it.”

(Male, 55+ years, Greece)

Some interviewees were aware of the warning app but were unsure about whether they had installed it or not.

“I don’t think I have it. I do know that it exists, but I probably don’t have it installed.”

(Female, 65+ years, Finland)

“It is familiar to me. Well, I wonder if I have it in this new phone.”

(Male, 60+ years, Finland)

Interestingly, those who had downloaded the application did not necessarily know how to use it properly. It seemed they had not spent time getting familiar with the application’s content or usage, but instead had only assumptions or even hopes regarding its functionalities.

“As a matter of fact, I am not that familiar with it.”

(Male, 70+ years, Finland)

“I think I know the essentials [of its usage], but luckily, I have never had to use it.”

(Male, 60+ years, Finland)

“I do have the 112 application on my phone, and I hope that [warnings] would come through that. It sometimes notifies me about very general things, but I haven’t really checked. I just imagine it would work like that.”

(Female, 40+ years, Finland)

The general principle of these apps is that the user receives public warnings and other public notifications when they have both installed the application on a mobile device and enabled location services and notifications. If these functions are not enabled, the user does not receive any warnings or notifications from the app. If the user wants to see and hear new notifications immediately in a pop-up window, they may need to enable them separately in the settings section of their mobile phone.

“I have not noticed that it would give any warnings.”

(Male, 70+ years, Finland)

Location services is a relevant aspect of mobile emergency applications [9] and was an issue that some of the interviewees in this study also reflected on. One interviewee said that the app would be useful if someone went hiking or berry picking and lost their way in the woods, for example. This could be an emergency, at least in Finland, and yet the interviewee had not installed the app on their mobile device. Another interviewee thought that the location information provided by the app was not correct, which was the reason she had not installed it. These examples show that the sense of place [28,29] provided by location services is a relevant factor in people’s intentions to use a mobile emergency application, but not necessarily an important one.

Another issue raised by one interviewee was the sound signal used by the app to alert users to a warning notification. The interviewee’s complaint about the quietness of the sound signal is another example that demonstrates people’s lack of knowledge of how to change settings in their mobile applications.

“At the beginning I paid it no attention at all, because when you hear the beeping, it is different from a text message. … It is the kind of beeping that you need to be pretty close to hear it.”

(Female, 65+, Finland)

Among the interviewees, few knew how to take advantage of the emergency mobile application in full. They were aware that the 112 Suomi app also contains instructions for action in various emergency situations, and that Warn-app NINA also provides useful information regarding current restrictions relating to the COVID-19 pandemic. In Greece, some interviewees were aware that the application sent text messages warning of severe heat and prohibiting entry to forests on windy days due to the risk of forest fires spreading quickly.

7. Discussion

Safety is an essential aspect of quality of life in cities and requires continuous development efforts. In this study, we explored citizens’ perspectives and the ways in which people in three countries (Finland, Germany, and Greece) think about obtaining warnings and information on emergencies involving health risks. Our aim was to gain an overall understanding of people’s perceptions of current public warning systems. However, in the spirit of smart city development, we also specifically investigated people’s experiences of the existing warning apps used for emergency communication in the countries studied (112 in Finland, NINA in Germany, and the forest fire risk warning application of the General Secretariat for Civil Protection in Greece).

The results suggest that people’s knowledge of public warning systems is inadequate. People put a lot of trust in their social network, receipt of text messages, and ability to use web-based search engines in order to learn about emergencies. The idea of text messages as a public warning system seemed a little peculiar; due to the resources required (time, money, and human effort), text messages are not used as a means of delivering public warnings or emergency communication in all countries (e.g., in Finland).

The use of conventional radio and television transmissions for delivering public warnings was considered relevant but haphazard. Additionally, the use of sirens in crisis situations raised many questions and uncertainties, particularly regarding whether the sound signals could be heard and whether the use of sirens was able to produce appropriate reactions. In this way, this study confirms the results of Douvidet et al. [8], who found that even if the public considers sirens to be a relevant warning system, it is uncertain whether they would hear the siren or know how to react appropriately.

More generally, the study emphasises aspects of audio output in alerting the public, as this topic was raised in discussions concerning not only sirens but also public warnings delivered via conventional television broadcasts and mobile warning applications. The study confirms the results reported by Tan et al. [43] that people want audio signals for public warnings to be distinguishable, of an appropriate volume, and not to cause any annoyance. From the perspective of smart city progress, this raises the challenge of developing audio output for public warnings that advance safety in cities in crisis situations. Here, people’s knowledge and acceptance of sound signals for warnings also play an important role.

The study adds to the discussion about mobile emergency communication [9,10], suggesting that people show informed ignorance regarding existing mobile emergency applications. In other words, many people in our study knew of the existence of mobile emergency apps and even recognised their advantages, but still they either had not installed the app or had not learned how to use it properly. In Greece, citizens’ awareness and knowledge of the forest fire warning app may have changed in the time since our study was conducted. During extreme weather or forest fire situations since the summer of 2021, the app has been widely used as a public warning channel. However, the potential increase in people’s knowledge and experience with mobile warning apps, whether in Greece or elsewhere, requires further investigation.

A limitation of this study is that it has explored citizens’ perspectives of public warning systems in only three countries (Finland, Germany, and Greece) with the largest number of interviewees being in Finland. It must be noted here that interviewee recruitment was not an easy task in any of the countries. We suspect that one reason for these recruitment difficulties might have been our interview guide in which our aim (to gain an understanding of citizens’ perspectives of public warning systems) was part of our larger research project on chemical hazard and risk information. Many people seemed hesitant to participate and said that they had insufficient knowledge of chemical hazards. It took a lot of time and effort from us to convince people that no prior knowledge was needed and that we were interested in their views and experiences. Our suggestion is that future studies pay careful attention to the preparation of interview guides since gathering a range of topics in one interview may have consequences for the interviewee recruitment process, even if the topics were connected to each other.

An additional challenge was posed by the COVID-19 restrictions, which led to the situation that all interviews had to be held online. This meant that participation required the technology necessary to participate, thereby having consequences for the inclusivity aspects of our research design. Only one of our focus group participants was a senior citizen who did not have the required technology but was able to participate with the help of relatives. All other participants in both focus groups and individual interviews were either students or well-educated professionals and pensioners. In this light, the result about mobile warning apps seemed somewhat surprising that even though many interviewees knew of the existence of such mobile emergency apps and even recognised their advantages, they still either had not installed the app or had not learned how to use it properly. Age did not appear to play any role here. However, future studies about public warning systems ought to be designed to include the perspectives of marginalised groups too.

In data collection, we used two different methods: focus groups and semi-structured interviews in Finland; focus groups in Germany; and semi-structured interviews in Greece. The challenge in focus groups can be that overly talkative people dominate the discussion. Through small group size and well-structured interviews, we paid careful attention that all focus group participants were able to express their views on all questions. The advantage of focus groups is the comments made by participants to each other. By keeping the conversational tone in individual interviews, we enabled interviewees to elaborate on their perspectives even though group interaction was not possible. Despite different data collection methods, we think that these measures in interview planning and conduct resulted in data that allowed us to proceed with the analysis and comparison.

Despite the largest amount of empirical data being from Finland, the analysis provided similar results in the three countries. However, more empirical research would be needed to conduct a more detailed comparison in the future. Additionally, more research on citizens’ perspectives of public warning systems and people’s experiences of the existing warning apps should be conducted in other countries. Nowadays, many countries have their own warning apps used by authorities for emergency communication. The advantages and disadvantages of a range of public warning systems in various countries, different warning application features as well as peoples’ knowledge and use habits of the warning apps could provide important insights for the further development of both conventional and digital emergency warnings. An important future research avenue would also be to examine the inclusivity aspects of public warning systems in general and warning apps in particular. The results would feed into the further development of smart and safe city initiatives.

8. Conclusions

The results reveal that even in times of digitalization, citizens still perceive that conventional warning systems play a relevant role in emergency communication. This is understandable since in cases of network connectivity disturbances, access to the Internet and digital devices would be disrupted, which worried some interviewees. Of conventional public warning systems, siren-driven alert systems have implications for urban development. As indicated by this study, the audibility of alarm sounds can be a problem. However, at the same time, it is important to recognize that optimising the warning siren system placement is not necessarily easy [38]. This study suggests that another issue that requires more consideration is people’s awareness of siren use situations and reactions. Here, citizen involvement in testing of the systems and their training about and in emergencies could be some examples of the ways to advance safe city development efforts.

Digital technologies such as warning apps are increasingly being used by authorities to distribute public warnings. The results of this study show that even if people consider such applications to be useful, they do not necessarily install the app or learn to use it properly. These results imply that it is not sufficient to build digital infrastructure; instead, the development of smart, safe cities requires education to make people aware of novel digital technologies for emergency situations, e.g., warning apps, and to enable them to use and take advantage of these technologies. Such education could involve the use of advertising campaigns or training delivered within schools, but it should be carefully designed, and its effectiveness investigated. New and innovative ways of educating the public in disaster awareness and preparedness are needed [36]. One example could be the development of engaging mobile games for training purposes about emergencies involving health risks. Overall, our results mean that hard and soft infrastructure-oriented strategies [1] cannot be separate efforts. In fact, the development of smart, safe cities requires continuous work and the integration of both strategic aspects, i.e., technological infrastructure development including digitalisation and the advancement of knowledge and participation of people.

Author Contributions

Conceptualization, S.Y.-K.; methodology, S.Y.-K. and M.S.; formal analysis, S.Y.-K. and M.S.; investigation, S.Y.-K., M.S., A.Z., O.A., H.H., M.T.K., M.K. and J.P.; writing—original draft preparation, S.Y.-K.; writing—review and editing, S.Y.-K., J.P. and P.V.; project administration and funding acquisition, P.V. and H.H. All authors have read and agreed to the published version of the manuscript.

Informed Consent Statement

Before the interviews, all interviewees received a data protection notice in their native language containing information about the processing of personal data. Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data not available due to privacy restrictions.

Acknowledgments

We would like to thank all the interviewees who participated in this study. We would also like to thank the academic editor and two anonymous reviewers for their feedback and useful comments for the improvement of the manuscript. Additionally, we would like to thank the European Union’s Horizon 2020 FET Open programme for funding this research (grant agreement No 899261). We also would like to thank “Open access funding provided by University of Helsinki”.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Table

Table 1

Interviewees in Finland, Germany, and Greece.

	No. Interviewees	Males	Females	Age Range	Students	Employees or Pensioners
Finland	40	11	29	21–84	25	15
Germany	11	7	4	18–33	4	7
Greece	23	15	8	20–61	5	18

References

1. Angelidou, M. Smart city policies: A spatial approach. Cities; 2014; 41, pp. S3-S11. [DOI: https://dx.doi.org/10.1016/j.cities.2014.06.007]

2. Su, K.; Li, J.; Fu, H. Smart city and the applications. Proceedings of the IEEE International Conference on Electronics, Communications and Control (ICECC); Ningbo, China, 9–11 September 2011; pp. 1028-1031.

3. Lacinák, M.; Ristvej, J. Smart city, safety and security. Procedia Eng.; 2017; 192, pp. 522-527. [DOI: https://dx.doi.org/10.1016/j.proeng.2017.06.090]

4. Giffinger, R.; Fertner, C.; Kramar, H.; Kalasek, R.; Pichler-Milanovic, N.; Meijers, E. Smart Cities: Ranking of European Medium-Sized Cities; Centre of Regional Science, Vienna University of Technology: Vienna, Austria, 2007; Available online: http://www.smart-cities.eu/download/smart_cities_final_report.pdf (accessed on 16 January 2023).

5. Ristvej, J.; Lacinák, M.; Ondrejka, R. On smart city and safe city concepts. Mob. Netw. Appl.; 2020; 25, pp. 836-845. [DOI: https://dx.doi.org/10.1007/s11036-020-01524-4]

6. de Haan, F.; Butot, V. Finding safety in the smart city: A discourse analysis with strategic implication. International Security Management—New Solutions to Complexity; Jacobs, G.; Suojanen, I.; Horton, K.E.; Bayerl, P.S. Springer: Cham, Switzerland, 2021; pp. 225-242.

7. Fedorov, V.; Robnik, A.; Alexey, T. “Safe City”—An open and reliable solution for a safe and smart city. Elektroteh. Vestn. Engl. Ed.; 2012; 79, pp. 262-267.

8. Douvinet, J.; Serra-Llobet, A.; Bopp, E.; Kondolf, G.M. Are sirens effective tools to alert the population in France?. Nat. Hazards Earth Syst. Sci.; 2021; 21, pp. 2899-2920. [DOI: https://dx.doi.org/10.5194/nhess-21-2899-2021]

9. Aloudat, A.; Michael, K.; Chen, X.; Al-Debei, M.M. Social acceptance of location-based mobile government services for emergency management. Telemat. Inform.; 2014; 31, pp. 153-171. [DOI: https://dx.doi.org/10.1016/j.tele.2013.02.002]

10. Bean, H.; Liu, B.F.; Madden, S.; Sutton, J.; Wood, M.M.; Mileti, D.S. Disaster warnings in your pocket: How audiences interpret mobile alerts for an unfamiliar hazard. J. Contingencies Crisis Manag.; 2016; 24, pp. 136-147. [DOI: https://dx.doi.org/10.1111/1468-5973.12108]

11. Tan, M.L.; Prasanna, R.; Sock, K.; Hudson-Doyle, E.; Leonard, G.; Johnston, D. Mobile applications in crisis informatics literature: A systematic review. Int. J. Disaster Risk Reduct.; 2017; 24, pp. 297-311. [DOI: https://dx.doi.org/10.1016/j.ijdrr.2017.06.009]

12. Wirtz, B.W.; Becker, M.; Schmidt, F.W. Smart city services: An empirical analysis of citizen preferences. Public Organ. Rev.; 2022; 22, pp. 1063-1080. [DOI: https://dx.doi.org/10.1007/s11115-021-00562-0]

13. Sharifi, A.; Allam, Z.; Feizizadeh, B.; Ghamari, H. Three decades of research on smart cities: Mapping knowledge structure and trends. Sustainability; 2021; 13, 7140. [DOI: https://dx.doi.org/10.3390/su13137140]

14. Nam, T.; Pardo, T.A. Conceptualizing smart city with dimensions of technology, people, and institutions. Proceedings of the 12th Annual International Digital Government Research Conference: Digital Government Innovation in Challenging Times; College Park, MD, USA, 12–15 June 2011; pp. 282-291.

15. Siokas, G.; Tsakanikas, A.; Siokas, E. Implementing smart city strategies in Greece: Appetite for success. Cities; 2021; 108, 102938. [DOI: https://dx.doi.org/10.1016/j.cities.2020.102938]

16. Haque, A.B.; Bhushan, B.; Dhiman, G. Conceptualizing smart city applications: Requirements, architecture, security issues, and emerging trends. Expert Syst.; 2022; 39, e12753. [DOI: https://dx.doi.org/10.1111/exsy.12753]

17. Graham, S. Bridging Urban digital divides? Urban polarisation and information and communications technologies (ICTs). Urban Stud.; 2002; 39, pp. 33-56. [DOI: https://dx.doi.org/10.1080/00420980220099050]

18. Appio, F.P.; Lima, M.; Paroutis, S. Understanding smart cities: Innovation ecosystems, technological advancements, and societal challenges. Technol. Forecast. Soc. Chang.; 2019; 142, pp. 1-14. [DOI: https://dx.doi.org/10.1016/j.techfore.2018.12.018]

19. Alfouzan, F.A.; Kim, K.; Alzahrani, N.M. An efficient framework for securing the smart city communication networks. Sensors; 2022; 22, 3053. [DOI: https://dx.doi.org/10.3390/s22083053] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35459038]

20. Hollands, R.G. Will the real smart city please stand up?. City Anal. Urban Trends Cult. Theor. Pol.; 2008; 12, pp. 303-320. [DOI: https://dx.doi.org/10.1080/13604810802479126]

21. Mansoor, A.; Chandra, K. Becoming a smart city: Best practices, failures and practical challenges. Proceedings of the CeDEM Asia 2018: International Conference for E-Democracy and Open Government; Yokohama, Japan, 12–13 July 2018; pp. 1-7.

22. Ammara, U.; Rasheed, K.; Mansoor, A.; Al-Fuqaha, A.; Qadir, J. Smart cities from the perspective of systems. Systems; 2022; 10, 77. [DOI: https://dx.doi.org/10.3390/systems10030077]

23. Amin, A. Lively infrastructure. Theory Cult. Soc.; 2014; 31, pp. 137-161. [DOI: https://dx.doi.org/10.1177/0263276414548490]

24. Kourtit, K.; Pele, M.M.M.; Nijkamp, P.; Pele, D.T. Safe cities in the urban world: A comparative cluster dynamics analysis through machine learning. Sustain. Cities Soc.; 2021; 66, 102665. [DOI: https://dx.doi.org/10.1016/j.scs.2020.102665]

25. Khan, S.; Zaman, A.U. Future cities: Conceptualizing the future based on a critical examination of existing notions of cities. Cities; 2018; 72, pp. 217-225. [DOI: https://dx.doi.org/10.1016/j.cities.2017.08.022]

26. Van den Berg, L.; Pol, P.M.J.; Mingardo, G.; Speller, C.J.M. The Safe City: Safety and Urban Development in European Cities; Euricur: Ashgate, UK, 2006.

27. Kourtit, K.; Nijkamp, P.; Suzuki, S. Comparative urban performance assessment of safe cities through data envelopment analysis. Reg. Sci. Policy Pract.; 2021; 13, pp. 591-602. [DOI: https://dx.doi.org/10.1111/rsp3.12276]

28. Yuen, B. Safety and dwelling in Singapore. Cities; 2004; 21, pp. 19-28. [DOI: https://dx.doi.org/10.1016/j.cities.2003.10.005]

29. Lang, J. Urban Design: The American Experience; John Wiley and Sons: New York, NY, USA, 1994.

30. Brands, J.; Suojanen, I.; van Doorn, J. What do people talk about when they talk about experiencing safety?. International Security Management—New solutions to Complexity; Jacobs, G.; Suojanen, I.; Horton, K.E.; Bayerl, P.S. Springer: Cham, Switzerland, 2021; pp. 67-81.

31. Shayan, S.; Kim, K.P.; Ma, T.; Nguyen, T.H.D. The first two decades of smart city research from a risk perspective. Sustainability; 2020; 12, 9280. [DOI: https://dx.doi.org/10.3390/su12219280]

32. Yli-Kauhaluoma, S. Reflections on self-tracking routines: Conducting maintenance of digital data. Tecnoscienza Italian J. Sci. Technol. Stud.; 2019; 10, pp. 53-78.

33. Tan, M.L.; Prasanna, R.; Stock, K.; Hudson-Doyle, E.; Leonard, G.; Johnston, D. Enhancing the usability of a disaster app: Exploring the perspective of the public as users. Proceedings of the 16th ISCRAM Conference; València, Spain, 19–22 May 2019; pp. 876-886.

34. Fischer-Preβler, D.; Benaretti, D.; Fischbach, K. A protection-motivation perspective to explain intention to use and continue to use mobile warning systems. Bus. Inf. Syst. Eng.; 2022; 64, pp. 167-182. [DOI: https://dx.doi.org/10.1007/s12599-021-00704-0]

35. Laughery, K.R. Safety communications: Warnings. Appl. Ergon.; 2006; 37, pp. 467-478. [DOI: https://dx.doi.org/10.1016/j.apergo.2006.04.020]

36. Guillot, S.; Jarvis, P.; Powell, T.; Kenkre, J. Knowledge, experience and preparedness: Factors influencing citizen decision-making in severe weather situations. Int. J. Emerg. Manag.; 2020; 16, 60. [DOI: https://dx.doi.org/10.1504/IJEM.2020.110108]

37. Sorensen, J. Hazard warning systems: Review of 20 years of progress. Nat. Hazards Rev.; 2000; 1, pp. 119-125. [DOI: https://dx.doi.org/10.1061/(ASCE)1527-6988(2000)1:2(119)]

38. Siliézar, J.; Aumond, P.; Can, A.; Chapron, P.; Péroche, M. Case study on the audibility of siren-driven alert systems. Noise Mapp.; 2023; 10, 20220165. [DOI: https://dx.doi.org/10.1515/noise-2022-0165]

39. Whittaker, J.; Taylor, M.; Bearman, C. Why don’t bushfire warnings work as intended? Responses to official warnings during bushfires in New South Wales, Australia. Int. J. Disaster Risk Reduct.; 2020; 45, 101476. [DOI: https://dx.doi.org/10.1016/j.ijdrr.2020.101476]

40. Wood, M.M.; Mileti, D.S.; Bean, H.; Liu, B.F.; Sutton, J.; Madden, S. Milling and public warning. Environ. Behav.; 2018; 50, pp. 535-566. [DOI: https://dx.doi.org/10.1177/0013916517709561]

41. Zhou, Z.; Zhang, X.; Ki, E.J. Enhancing the efficacy of instructing and adjusting information: An exploration of interactivity and vividness. J. Contingencies Crisis Manag.; 2023; pp. 1-17. [DOI: https://dx.doi.org/10.1111/1468-5973.12482]

42. Dallo, I.; Stauffacher, M.; Marti, M. Actionable and understandable? Evidence-based recommendations for the design of (multi-) hazard warning messages. Int. J. Disaster Risk Reduct.; 2022; 74, 102917. [DOI: https://dx.doi.org/10.1016/j.ijdrr.2022.102917]

43. Tan, M.L.; Prasanna, R.; Stock, K.; Doyle, E.E.; Leonard, G.; Johnston, D. Modified usability framework for disaster apps: A qualitative thematic analysis of user reviews. Int. J. Disaster Risk Sci.; 2020; 11, pp. 615-629. [DOI: https://dx.doi.org/10.1007/s13753-020-00282-x]

44. Karma, S.; Georgiou, S.; Boukis, I.; Argyris, I.; Pelli, E.; Statheropoulos, M. Multi-hazard Risk Approach and Inclusive Community Engagement in Disaster Risk Management: Experiences & lessons Learned by a Recent Wildfire Event in the Greek Island of Rhodes amidst COVID-19 Pandemic Crisis; European Center for Forest Fires: Athens, Greece, 2021.

45. Xu, D.; Yan, C.; Zhao, Z.; Weng, J.; Ma, S. External communication barriers among elderly deaf and hard of hearing people in China during the COVID-19 pandemic emergency isolation: A qualitative study. Int. J. Environ. Res. Public Health; 2021; 18, 11519. [DOI: https://dx.doi.org/10.3390/ijerph182111519]

46. GIDPROvis. Available online: https://www.helsinki.fi/en/projects/gidprovis (accessed on 6 September 2023).

47. Lobe, B.; Morgan, D.L.; Hoffman, K. A systematic comparison of in-person and video-based online interviewing. Int. J. Qual. Methods; 2022; 21, 16094069221127068. [DOI: https://dx.doi.org/10.1177/16094069221127068]

48. Stewart, D.W.; Shamdasani, P.N.; Rook, D.W. Focus Groups. Theory and Practice; 2nd ed. Sage publications: Thousand Oaks, CA, USA, 2007.

49. Emergency Response Centre Agency. 112 Suomi Application. Available online: https://112.fi/en/112-suomi-application (accessed on 16 January 2023).

50. Bundesamt für Bevölkerungsschutz und Katastrophenhilfe. Warn-App NINA. Available online: https://www.bbk.bund.de/DE/Warnung-Vorsorge/Warn-App-NINA/warn-app-nina_node.html (accessed on 16 January 2023).

51. Ministry for Climate Crisis and Civil Protection. Forest Fire Risk Warning Application of the General Secretariat for Civil Protection. Available online: https://www.civilprotection.gr/en/112-emergency-communications-service (accessed on 16 January 2023).

52. Wilkinson, S.; Silverman, D. Focus group research. Qualitative Research: Theory, Method, and Practice; Silverman, D. Sage: Thousand Oaks, CA, USA, 2004; pp. 177-199.

53. Schreier, M. Qualitative content analysis. The SAGE Handbook of Qualitative Data Analysis; Flick, U. Sage Publications: London, UK, 2014; pp. 170-183.

54. Charmaz, K. Constructing Grounded Theory: A Practical Guide through Qualitative Analysis; Sage Publications: London, UK, 2006.

55. Eriksson, P.; Kovalainen, A. Qualitative Research in Business Studies; Sage Publications: London, UK, 2008.

56. Botterell, A.; Addams-Moring, R. Public warning in the networked age: Open standards to the rescue?. Commun. ACM; 2007; 50, pp. 59-60. [DOI: https://dx.doi.org/10.1145/1226736.1226767]