At Risk While on the Move—Mobility Vulnerability of Individuals and Groups in Disaster Risk Situations

Abstract

Vulnerability is often analysed as a static condition of residents at a location, exposed to disaster and other risks. Studies on individual aspects of mobility and vulnerability exist, but comprehensive studies or guiding frameworks are lacking. The paper’s unique contribution compared to existing vulnerability models lies in emphasising vulnerability not only at fixed places, but also during transit, movement, and temporary phases. This paper highlights the current state of research on mobility vulnerability within disaster risk contexts. Through a systematic literature review, the study discovers a lack of research analysing specific vulnerabilities during mobility. Additionally, existing vulnerability frameworks are improved by incorporating (i) disaster risk and impact scenarios, (ii) different types of movements and mobilities linked to disaster risk situations, (iii) multiple localities, modalities, and temporalities, as well as multiple risks during sequences of movement and stationary phases, (iv) daily and occasional hazards, and (v) emic and etic perspectives on vulnerability. The findings of this study aim to inform future research on risk and vulnerability, supporting more effective responses amidst the changing dynamics of disaster situations.

1. Introduction

Disaster risk research addresses scenarios, potential damages, and losses related to human lives, health, and other environmental impacts. It is driven by events of natural hazards, such as earthquakes, floods, or wildfires, which, when they occur, often result in human casualties, health impacts, and economic damages [1]. A substantial amount of research has been conducted to better understand disaster risk [2]. This notion has been promoted on the international stage by organisations such as the United Nations [3]. However, it is based on insights gained from humanitarian aid and development research and practice, as well as insurance, risk management, and other areas interested in understanding risk assessment to predict better future outcomes, including both opportunities and potential losses [4,5]. Despite decades of research, scholars have observed that a deeper understanding of disaster risk has not led to a reduction in losses [6]. This paradox can partly be explained by the so-called ‘Dyke effect’, which refers to locations behind flood protection dykes that receive attention after a disaster, such as a flood, and subsequently implement structural measures for protection [7]. This, in turn, creates a false sense of safety for residents living behind those dykes and leads to even further urban development. The dyke effect refers to the connection between well-intended improvements through flood protection measures, which, however, lead to a counterproductive effect of low hazard awareness levels, thereby increasing the overall risk. Place-based approaches are a hallmark of disaster risk research because disasters are context-specific, and local conditions—such as those of the people, environment, and various other features—must be taken into account [8,9]. Place-based approaches have also gained traction in risk research under the vulnerability paradigm, leading to numerous studies that provide a deeper conceptual understanding of the impact side of risk, within the framework of vulnerability and its components [10,11]. This marks a significant enhancement of traditional risk models, which primarily focused on hazards, probabilities, and related damages in linear relations constrained to general categories [1]. The components of vulnerability—exposure, susceptibility, and related capacities—are also closely connected to resilience in disaster contexts, which has fostered further conceptual development and innovation in disaster risk research [9].

Understanding disaster risk continues to be prioritised in United Nations declarations, for instance [3]. Nevertheless, there are also processes involving alternative approaches that focus on loss and damage calculations about climate change, aimed at illustrating the need for action by small island states facing challenges such as sea level rise and potential relocation [12]. Research on climate change either emphasises mitigating hazards or adapting to their consequences, presenting a somewhat different perspective from traditional disaster risk [13]. While many efforts aim to align definitions and conceptual notions, this remains an ongoing process, fraught with challenges in integration [14].

Given this background, this study seeks to advance the understanding of vulnerability and its conceptual dimensions. While place-based approaches and conceptual frameworks are essential, a common critique highlights their limitations due to their spatial and often static nature [15,16]. Dynamic aspects need to be incorporated, often understood as interrelations between components or locations, which are covered by feedback loops or system dynamics, for example [17,18].

Disaster-induced migration: Mobility-related disaster studies have a long tradition, such as on migration in the aftermath of disasters such as hurricane impacts [19], people, or ‘involuntary movers’, who do not return after disasters [20], or people displaced after disasters such as the 2004 Indian Ocean Tsunami [21] and related relocation problems, also affecting daily commute and mobility in this sense [22]. An area that has gained traction in studies is mobility-related aspects, such as migration and relocation due to climate change or other factors [23,24].

Daily and health-related mobility: The specific term ‘mobility vulnerability’, is currently applied in the contexts of transportation or COVID-19 studies [25,26]. However, this term and the idea of mobility encompass more than just transportation, infectious diseases, or migration; they include a multitude of mobility processes, such as commuting for jobs or education, tourism, and travel related to destinations, specifically incorporating evacuations or the risks faced by people when they are away from home. What is missing despite numerous studies on individual aspects of mobility in the context of vulnerability is a comprehensive overview and conceptual approach.

The objective of this paper is to identify the current state of research on mobility vulnerability in the context of disaster risk. The information regarding existing concepts and usages will inform the synopsis of mobility, particularly concerning dynamic components of vulnerability that are not yet captured in existing frameworks. The study will initially conduct a systematic literature review, aiming to encompass as many aspects of mobility and vulnerability as possible to delineate the landscape of research in this field. Subsequently, the study will develop conceptual components that encompass key elements of a framework for mobility vulnerability. Key findings, as well as limitations, will also be discussed.

2. Methodology

The study’s methodology comprises two parts. The first part is a systematic literature review, and the second part involves the development of a conceptual framework.

The systematic literature review adheres to the PRISMA standard [27,28] and is based on predefined search term strings researched within recognised scientific data platforms: Web of Science, Scopus, and PubMed. The research was conducted as the first step in the Web of Science on one day on 13 June 2025, to maintain a consistent sample and temporal snapshot. Scopus and PubMed were analysed to obtain an additional overview and check for any missing general directions in research. All databases were analysed, employing a stepwise approach in selecting topic research, including abstracts, keywords, and titles, alongside identifying review papers within this sample. Review papers provide a concise overview of a field and serve as a valuable source for streamlining the findings. Furthermore, the same research was restricted to title searches under the assumption that key terms found in the title would yield articles that focus on this more than if any of the key search terms were only captured in the abstract or keywords.

The PRISMA approach was followed, with duplicate records being removed and studies from platforms lacking abstracts excluded. In the subsequent screening step, records that did not fit the study’s scope were excluded, and the exclusion criteria were detailed in a diagram (Figure 1). Additional studies were removed if their content was redundant and provided no new information beyond that already captured by the identified studies.

Finally, those papers identified as best fitting the scope were analysed in detail through in-depth reading and summarising the key results in a search results table. From the search results, initial studies were selected that contained the key search terms in the title and were review papers. If no relevant findings were present, studies with the key terms in the title were selected. This was followed by review papers that captured the search terms in the title, abstract, or keywords, and then by the general group of findings related to the selected keywords.

The conceptual framework development in the second part is based on the theoretical components of vulnerability found in the literature, which is documented and detailed in the respective results section. This theory-informed approach was developed under expert judgement, stemming from the author’s subjective experience and balanced by the systematic results of the literature review and the documented theoretical background in the literature.

3. Results

3.1. Results of the Systematic Literature Review

The retrieved results have been documented with numeric details in the search results,

Table 1. Search results table with amounts of studies and exclusion.

Search Term	Title-Abs-Keys	Title-Abs-Keys Review Papers	Title Search	Title Search Review Papers	Excl No Abs	Excl Dupl	Excl No Fit	Excl Redund. Info
General topic
Vulnerab * mobility	4579	288	89	2			1
Social Vulnerab * mobility	1261	78	12	0	3		5	2
“Mobility vulnerability”	7	0	2	0			6
Social Vulnerab * mobility framework	171	19	0	0			17
Group type
Vulnerability mobil * individual	748	49	1	0			50
Vulnerability mobil * person	180	12	0	0		1	11
Vulnerability mobil * group	848	52	2	0			54
Vulnerable mobil * group	1604	183	5	0			4
Vulnerability mobil * crowd	31	0	0	0			31
Process
Vulnerab * evacuation	1250	68	58	1	2	5		49
Vulnerab * relocate *	1012	59	13	0			13
Vulnerab * transit	1116	40	57	1		5	1	49
Livelihood
Vulnerab * commuter	131	1	2	0			2	1
Vulnerab * farmer	4596	333	169	3
Vulnerab * mobility farmer	59	7	1	0			8
Vulnerab * flight crews	19	2	1	0			2	1
Vulnerab * Fisher	1196	46	19	0			19
Vulnerab * Sailor	11	0	1	0			10	1
Vulnerab * Shepherd	34	1	0	0			34
Vulnerab * Nomad	49	1	2	0			3
Vulnerab * Occupation	1746	106	13	0			12
Vulnerab * labor mobility	270	8	4	0	2		2
Leisure
Vulnerab * Tourist	938	31	23	0	3		20
Vulnerab * Tourist evacuation	28	0	0	0			28
Vulnerab * Traveller	356	38	9	0			9
Disaster terms
Vulnerab * Displaced	1397	119	23	0	7		16
Vulnerab * Homeless	1880	154	120	6			6
Vulnerab * Migrant	4374	291	270	7			6
Vulnerab * Refugee	2750	278	134	4			4
Situations
Vulnerab * mobility daytime	10	1	0	0
Vulnerab * mobility nighttime	10	1	0	0			1	8
Vulnerab * mobility unfamiliar *	13	0	0	0			13
Vulnerab * mobility situate *	260	10	0	0			10
Vulnerab* mobility situatedness	1	0	0	0			1
Vulnerab * mobility disaster	274	13	0	0			13
Vulnerab * mobility hazard	295	16	1	0			16

The asterisk * symbol was used in the search strings to identify variants of a word.

and

Table A1. Additional search results table with amounts of studies.

Search Term	Title-Abs-Keys	Title-Abs-Keys Review Papers	Title Search	Title Search Review Papers	Excl No Abs	Excl Dupl	Excl No Fit	Excl Redund. Info
Location
Vulnerab * home	12,140	859	296	7
Vulnerab * travel	3235	177	50	0
Vulnerab * destination	1464	78	20	0
Hub
Vulnerab * airport	338	14	6	0
Vulnerab * border	2841	154	69	1
Vulnerab * train station	150	0	0	0
Vulnerab * mobility infrastructure	344	22	0	0
Means of transportation
Vulnerab * bicycle	393	16	4	0
Vulnerab * bus	1432	16	14	0
Vulnerab * car	1335	88	15	0
Vulnerab * subway	282	5	12	0
Vulnerab * taxi	118	3	3	0
Vulnerab * train	14,041	906	125	4
Vulnerab * truck	294	14	12	0
Vulnerab * plane	1657	54	4	0
Vulnerab * boat	398	16	4	0
Vulnerab * ferry	57	2	1	0
Vulnerab * ship	1101	57	57	0

The asterisk * symbol was used in the search strings to identify variants of a word.

in the Appendix A. These results reveal that the search terms are associated with varying numbers and types of research, reflecting certain areas of focus while indicating a lack of existing studies in others. Under the cluster of search terms related to the general topic of this study, the specific term “mobility vulnerability” is represented by only seven studies, in contrast to “social vulnerability and mobility,” which yields over 1000 studies (Table 1). In the subsequent research cluster concerning group type, the number of findings is comparable, except for the term “crowd,” with only 31 studies in total, significantly lower than the several hundred findings for the terms “individual” or “group.” It is intriguing that the exact spelling of “vulnerability” or “vulnerable” produces different results; the term “vulnerable mobile groups” nearly doubles the number of studies found.

Within the cluster of process-related search terms, findings are similar for “evacuation,” “relocation,” and “transit.” For the cluster of livelihood-related search terms, there are varying numbers of studies. The highest counts are observed for studies on farmers, followed by those on occupation and fishers. Studies on sailors, shepherds, flight crews, or nomads are also covered, though to a much lesser degree. Commuters and labour mobility have been studied in over 100 studies. Under the cluster of leisure activities, there are more studies on the vulnerabilities of tourists than on travellers.

Regarding disaster-related search terms, over 1000 findings are retrieved for “displaced,” “homeless,” “refugees,” and “migrants.” In the cluster of situation-related search terms, there is only one finding for the term “situated,” but over 200 for “situation” and similar terms. Quite low numbers are returned for “day- or nighttime mobility” or the term “unfamiliar,” with only about 10 findings. Studies on “mobility and disaster” or “mobility on hazard” fall within the range of 200 studies.

In the cluster of locations, high numbers are retrieved for “destination” or “travel,” with the highest being for “home,” which has over 10,000 studies. Within the cluster of hubs, the highest numbers pertain to border studies, with studies on train stations, airports, or infrastructure mobility numbering in the hundreds or more.

In the final cluster concerning means of transportation, a heterogeneous pattern emerges. The lowest numbers are recorded for “ferry,” followed by “taxi.” However, it is noteworthy that for most other means of transportation, there are over 100 studies. Notably, studies on “vulnerability” and “train” yield the highest number of results in this table, with 14,000 findings.

After searching the Web of Science, Scopus, and PubMed, they were also analysed to check whether the directions of research had not been captured yet. As this was not the case, and directions of research in epidemics, medicine, transportation, and health sciences, as well as disaster risk, were also repeated, it was not followed up further.

As an interpretation of these numerical findings, they show which specific terms and related fields already have higher coverage. This can help identify existing research gaps and inspire new studies, while also informing which terms and directions can be used for more systematic, quantitative, and in-depth literature analysis, such as biometric analysis.

Exclusion criteria are provided in Figure 1, and further details for further screening and analysis are provided in Appendix section “Details on Exclusion for Further Screening and Analysis”.

3.2. Findings Within Selected Studies

The detailed search results of 16 selected studies are documented in

Table A2. Details from the search results.

Search Term	Social Group	Mobility Context	Hazard or Driver	Research Field	Source
Vulnerab * mobility	Households	Transportation environment	Terrorism	Urban geography	[29]
Social Vulnerab * mobility	Communities	Income, transportation	Transportation challenges	Occupation	[26]
Social Vulnerab * mobility	Domestic workers, flight crews, and sailors	Travel	COVID-19	Public health	[30]
“mobility vulnerability”	Census blocks	Human mobility, mobile phone data	COVID-19	Travel	[25]
Vulnerable mobil * group	Households, elderly, lower socioeconomic status	Foot traffic	Snowstorm	Natural hazards	[31]
Vulnerab * evacuation	Household, females, seniors	Mass evacuation,	Flood	Transport	[32]
Vulnerab * evacuation	Elderly	Host communities	Wildfire	Society	[33]
Vulnerab * evacuation	Drivers	Driving during emergency evacuation	Emergencies	Planning	[38]
Vulnerab * transit	Neighbourhoods	Access and waiting	Heat	Transport	[34]
Vulnerab * transit	Passengers	Bus ride	Black carbon exposure	Transport	[42]
Vulnerab * transit	Vulnerable commuters, riders	Public transit	Crime	Security	[43]
Vulnerab * Occupation	Professions	Job activities	Climate change	Sustainability	[35]
Vulnerab * Migrant	Migrants	Crossing a mountainous rainforest region	Migration, abuse, exploitation, malnourishment	Medicine	[40]
Social Vulnerab * mobility framework	People	Climate mobility	Climate extreme events	Geography	[41]
Vulnerab * mobility nighttime	Urban population	Diurnal migration in day and nighttime	Heat, flood	Health, Communications	[36,37]

The asterisk * symbol was used in the search strings to identify variants of a word.

in the Appendix A. These have been analysed in more depth through reading and represent a picture of the plethora of mobility aspects related to vulnerability and relevant to disaster risk. Reading the table through the columns, it becomes clear that different social groups have been covered; however, most studies focus on households or communities and are often captured in broad studies rather than empirical research. The mobility contexts are quite diverse. They include studies on the transportation environments of households, with an emphasis on the vulnerabilities of households concerning current political and security-related conditions [29]. Income is a factor determining transportation options, and related challenges are analysed within the field of occupational studies for communities [26]. COVID-19 studies investigated the public health aspects of flight crews, sailors, and domestic workers [30]. Such studies have analysed travel behaviour influencing exposure to the infection, but also as a general pattern of human mobility between census blocks, using mobile phone data [25].

Natural hazards are also drivers of mobility studies in this context, and snowstorms, as an example of how households affect the pedestrian mobility of specific groups, such as elderly people or those with lower socioeconomic status, have been analysed [31]. Floods are another topic that affects households, especially females and elderly people, and leads to mass evacuation [32]. Wildfires are another natural hazard affecting elderly people and their vulnerability, arriving in host communities [33]. Heat is another natural hazard that impacts neighbourhoods through transport research, affecting mobility phases such as accessing transport and waiting times [34]. Climate change in sustainability studies is analysed as a driver of certain professions to seek mobility in their job activities [35]. Studies on heat or floods use migration patterns of day and nighttime populations in urban areas to determine different exposures [36,37].

Evacuation is a major topic in many emergency situations and is also related to the vulnerabilities of drivers [38]. There are informative studies on the risks faced by evacuated people, such as elderly people, when arriving at shelters in the event of wildfires [33] or nuclear accidents [39].

Migration is a greater topic, with migrants facing many problems during their often hazardous journeys, such as abuse, exploitation or malnourishment [40]. Climate change and related extreme events are a driver of climate mobility in a more general sense, studied in geography for people and their behaviour [41].

Bus rides expose passengers to black carbon exposure in different segments of the vehicle and are an example of exposure to additional hazards, including daily hazards, during mobility [42]. Crime is another example from security research for public transit of commuters and vulnerable riders [43].

3.3. Summary of Key Findings

The overall objective was to identify the state of the art in mobility and how it is already covered in vulnerability research. The results of the first scoping part of the study revealed that it is a field with a large number of studies in diverse directions (Table 1). However, some patterns emerged: vulnerability is mostly examined as a static condition of a place. This branch of study investigates mobility as one of the many aspects of people’s livelihoods and does not examine mobility vulnerability in its entirety, but rather highlights aspects such as dependency on public transport that characterise socio-economic conditions. Additionally, numerous studies have found that mobility is also a factor related to physical mobility-related medical problems. There are also studies with mobility in the title, but they are not related to the scope of the research, such as those on mobile phones. However, specifically, mobile phones and mobility data are a major means of mobility research in general. Analysing such data can include information about the day- or nighttime distribution of populations, commuting activities, and multi-localities, which are within the scope of this study.

Migration studies and relocation due to climate change encompass a broader field, where only a part of the studies analyse vulnerability during the mobility phase. In contrast, others focus on vulnerability at the place of origin or destination, or in temporary shelters.

As an overall finding, studies on natural hazards, or hazards that can occur daily, such as crime or pollution, reveal that they add additional stress loads. However, stress and problems are also created by job or other travel needs, including related distances and traffic conditions, leading to vulnerability. Certain social vulnerability profiles, such as elderly people, females, or individuals with lower socioeconomic status, are typical aspects also found in many other static social vulnerability studies. Prominent topics include individual global events, such as COVID-19, as well as migration and evacuation, which are specific disaster risk-related topics, alongside natural hazards.

3.4. Development of a Mobility Vulnerability Framework

For the conceptual advancement of vulnerability, different additional conceptual components of mobility are developed and suggested by the following graphics and their descriptions.

3.4.1. From Static to Situational

From the literature, a critique of the lack of dynamic aspects in vulnerability assessment and frameworks is identified [44]. This is reflected in many studies that focus solely on one location, where vulnerability is assessed. Instead of presenting this as a static picture, it is essential to incorporate dynamic aspects of mobility and changing situations. The first enhancement component, therefore, is the description of a disaster risk situation. A situation is captured in existing definitions of vulnerability when a spatial exposure overlaps with people, their belongings or environments [8,45,46,47]. This overlapping situation occurs statically when houses are built in a flood zone. However, these situations occur dynamically when people enter an area that is prone to flooding. This means that the temporal aspect of exposure must also be considered in a comprehensive disaster risk assessment. The following graphic’s (Figure 2) novelty lies in adding the dynamic phase into existing frameworks by illustrating the dynamic process, involving a transect and transit of persons or groups as they walk, drive, or are moved through the exposure situation.

An example is when people enter a building, open space, or board a vehicle, and in this process of moving, they are already in a disaster risk situation as soon as they enter the disaster exposure zone and time. People then move through this window of exposure, either while travelling physically or when time passes. There is also a differentiation between the risk and the impact phase. People are exposed to risk when they are exposed to a specific hazard. Before they are exposed, they still face a risk of such an exposure. When people have entered an exposed situation, their risk can turn into damage, loss, or other negative impacts. It is also possible that people leave the exposure zone without harm, as reflected in the figure. However, people are always exposed to some degree of accident risk when driving a car in traffic.

Additionally, people may occasionally be exposed to driving in a car during a flash flood or a hailstorm. Therefore, there is a general risk associated with certain single hazards, including daily risks or extreme events, which often occur and are prevalent in many places where people travel. However, we suggest investigating mobility vulnerability specifically in relation to hazard exposure, conceptually and focusing on its particular characteristics within that situation.

3.4.2. Sequences of Stationary and Mobility Vulnerability Patterns

The next graphic illustrates different types of mobility vulnerabilities for typical movements of persons or groups (Figure 3). It shows that the previous graphic was limited in its general idea and must be adjusted to specific movement types in daily life situations. In daily life situations, several stages of resting at a place and moving typically occur. This must be added as a sequence of situations to enhance the previous idea of disaster exposure zones and time. Starting with the upper line as the first class or example, it is commuters who travel from home to work, school or elsewhere. Some vulnerabilities need to be identified at their point of origin, during travel, and at intermediate stations, such as bus stations using public transport, filling stations for cars, stop-and-go traffic, or similar locations. Commencing their travel, again, there are phases of travel and mobility until people reach their destination. A job place, kindergarten, school or supermarket is a temporary stay. After that stay, which also has its vulnerabilities, people return home, repeating the travel only in the opposite direction.

This example already demonstrates that this is a sequence of place-based vulnerabilities, which can be measured using traditional static approaches for homes, stations, temples, state destinations, and other locations. Day- and nighttime population studies are an example, but also travel-time or service area studies are a field of application [48,49,50]. What is less evident in the literature studies is the mobility phases during travel. Some studies examine the exposure and vulnerabilities of people while commuting, driving, or walking (Table A2, Appendix A). It is necessary to combine the stationary or temporal stations of places with movements that can also occur in vehicles, which involve a combination of a place situation with other people. However, there is also a movement component that adds additional hazards and stress, such as the risk of car accidents or navigating through different exposure zones. Home and other destination places are different because they typically do not move. But the hazards, such as extreme weather or contamination clouds, move over them. This aspect of the dynamic mobility of people travelling, the locations of vehicles, and the hazard-movement pattern should be added to the conceptual framework and understanding of mobility vulnerability.

The next line in Figure 3 illustrates a pattern for tourism, which follows a sequence of home, travel, interim stations, travel, and temporal stay. However, what is already different is that tourists can face more unfamiliar environments during their travel. This cannot be generalised since some tourists visit the same vacation spot repeatedly and become accustomed to it. This would be similar to a daily commute situation, which over time becomes familiar with travel, movement, and various situations. However, as an assumption, many tourists also travel to certain stations and destinations for the first time or a few times. Therefore, overall, it is characterised in the figure also by the symbolisation of unfamiliar environments. In some places, such as interim stations like airports or train stations, the unfamiliar environment also includes unknown persons.

While commuters and tourists typically return home after their temporary stay, this is not always the case for other patterns related to disaster risk. Evacuation is an example where people, in most cases, are involuntarily expelled from their homes and must conduct a spontaneous transit to a safe place or shelter. In many cases, people can return home from their shelter. However, when the home is inhabitable or destroyed, additional travel to a new, unknown residency must be undertaken. In comparison to tourism, these places transport people into similar, unfamiliar environments. Still, to a higher degree, they are unfamiliar due to a lack of choice and because evacuations occur less frequently than planned tourism. The number of unknown persons they meet during transit, shelter, or in their new residences after arrival is also different from that encountered during tourism or a daily commute.

After a while at the new residency, familiarity with the environment will develop. This is also the case for migrants, some of whom have a mix of planned behaviour and their own decisions, while others are forced to move or relocate and have less time for planning. Migrants typically do not return home, at least not immediately, and must face the challenges of settling into new residences, similar to those faced by people who have been evacuated.

This addition to the conceptual framework illustrates different stages of transit and travel, where places and movements are not bound to a specific location, but rather to a corridor or transect that people move along.

3.4.3. Personal and Situational Vulnerability Components

A critique from literature is taken up, that vulnerability is only characterised for a place and group of people from outside perspectives and features [51]. What is also referred to as the etic component should be supplemented by a personal, emic component that views vulnerability from the perspective of the individuals themselves. For example, there is a difference in how evacuees perceive vulnerability (emic) versus how planners measure it (etic). Planners might measure it by square metres allocated to each evacuee. In contrast, the evacuees actually perceive their vulnerability by their personal situation of their family being dispersed across a refugee camp. This is more difficult to capture by general statistical or other data and is also more related to personal behaviour and psychological studies. We believe this needs to be combined more effectively to address the gaps between individual behavioural studies and multi-criteria characteristics, conditions, drivers, root causes, and other factors contributing to individual or group types of social resilience [52].

The separation of dynamic phases of origins or root causes of vulnerability in society, and is propagations through drivers, leading to conditions that render people vulnerable or at risk and exposed to hazards, has been captured already by the known conceptual frameworks stemming from development, research and related frameworks, depicting different phases of risk and its drivers [10,11]. This notion is also partly captured in disaster cycles or similar representations [53]. There are similar frameworks that also depict social amplifications of risk [54]. More frameworks exist, depicting multiple layers of single and multi-hazard risks, pathways and impact chains [55,56,57]. In some respects, therefore, dynamic phases and components have already been captured. However, they only partly capture mobility vulnerability in the aspects described above. For example, multi-locality is captured in principle in some frameworks, but often only as a sidenote, and without studies following this empirically, because the conceptual framework does not detail it [58,59]. Multi-modality is also already captured in transportation studies in terms of different vehicle types for transportation, and is also used increasingly as a term for various types of information sources or remote sensing sensors [60,61,62]. However, a multi-modality of situational vulnerability is still missing as a framework.

The next figure (Figure 4) illustrates the separation of emic and etic, as well as individual and group-related aspects, in the situational dimensions and conceptual components of vulnerability [63].

Figure 4 is organised along the conceptual components of vulnerability, exposure, susceptibility and capacities. The terminology and exclusion of certain components are topics of discussion in many studies [64]. We follow a basic definition that captures those three, but acknowledge that, depending on context, it might make sense to exclude exposure when it is already captured under the hazard or other risk components. Capacities are also excluded in some recent definitions because, in practice, there is often a problem separating a negative version of a feature describing susceptibility from the prevalence of its positive opposite when it is then termed capacity [65,66,67]. We attempt to address this by categorising all aspects within a person as internal susceptibilities and those from external individuals, resources, or the environment as external capacities. We also acknowledge that there is some discussion or dispute as to whether capacities are what is also named resilience [9]. We leave this aside and suggest that researchers applying this approach decide whether resilience is regarded as a measurable item that does not need to overlap with vulnerability or susceptibility, or whether it is used as an umbrella term for the entire study [68].

The figure shows that the upper row captures personal situational characteristics. During movements and mobility, temporal aspects need to be incorporated into existing frameworks that currently focus solely on spatial exposure. The duration at a location is a factor that influences exposure; for example, dosages of contamination, being immersed and affected by a flood, avalanche, or earthquake are similar. The respective useful time separations need to be adjusted to each specific hazard. Short-, medium-, and long-duration events can be separated to identify different types of risks for individuals. We consider all the components in this first row of personal situation characteristics, which are not concerning a hazard, but rather being familiar with the environment and situation. When a person enters an unknown location, most of the situation and environment will still be rather unfamiliar in the first moments. When people become stuck in a subway station, for example, or at an airport for several hours, they come to know their environment. It is essential to distinguish this from longer durations, as the situation then transitions into a place-based vulnerability assessment that can be compared to existing static vulnerability assessments, which generalise conditions for longer residencies. The duration at a location is only one factor; another is repeated visits to the same location that can also determine how and whether people familiarise themselves with it. Another addition is the situational change that is also related to the type of environment. For example, some stations, such as bus or train stations, provide orientation if exits and driving directions primarily go in one direction. This can become more confusing for orientation when it is a hub with many floors, or when it is a hub like an airport with a high frequency of people and traffic changes. As an opposite example, with calmer situations, a building or destination could be identified with low frequencies of change.

For personal situation characteristics, certain susceptibility factors also need to be included. Some are skills that people possess all the time, such as being informed about disaster risks and how to cope with them. Communication skills, such as languages that may be needed when travelling, are another essential skill, as well as spatial orientation or being able to recognise people’s behaviours or communication styles. Other attributes need to be added, such as disaster personalities that might lead to different reactions under stress than during regular working times, as studies on the World Trade Centre attack, for example, have identified [69]. A person also undergoes different situational vulnerabilities due to their psychological composition and other factors during the day. This should also be considered concerning the internal personal situation or vulnerability.People can also carry gadgets with them, but typically, mobile phones and others are dependent on charged batteries. Mobile devices often come with warning apps already installed. Gadgets or tools are a matter of conceptual discussion as to whether they could also be placed into the category of external capacities because they can become lost, be dysfunctional or not be carried along. External capacity, however, consists of other persons, but of course, even under personal situational characteristics in interplay with any other external capacity, which is found in the lower row of the chart. External capacities can be a barrier for communication when people around are not known, or in hostile environments. Dependency on others is a factor that can be a pro or a con. Close relationships can be good for identifying people to trust and be familiar with. The same is true for pets, but on the other hand, pets, as well as other friends and family members, can also represent a type of dependency that requires taking care of them during evacuation or other aspects, which can slow down the evacuation process.

The second row in the chart illustrates the vulnerability of the situational place, as depicted in the previous figure. At the situational level, it has already been raised whether a single hazard creates a certain exposure situation. This is different if compounding or multiple hazards exist at the exposed place, and during the sequence of movement. Compounding events are not only related to natural hazards or climate change extremes, but can also indicate additional or secondary problems or threats, such as crime or infectious diseases.

The acceptability of a situational place can resemble many of the typical components captured in existing vulnerability studies of place. However, some are specific to depicting the situation, such as accessibility, which is not only dependent on the general road layout and evacuation routes, but also on questions of timing and whether fire doors are left open for better convenience during daily use. It is also dependent on crowds or traffic jams, blocking accessibility. The situation is also different in its susceptibility, whether considering a building, a street, or a vehicle, as outlined already in Figure 3. The susceptibility of a place is also bound to the profile of the persons and groups at this place. Dependency on capacities is a wide range, and specifically in certain situations, it should be considered that mobile phone reception is a key factor. However, people in such places are also dependent on the governance and management of those in charge of the transportation system, as well as on emergency management. This is part of what is also termed basic or critical infrastructure, and includes utility companies for power, information, heat, food, and many other essential services that people rely on during travel or at places of stay [3,70,71]. It also includes a warning, which is disseminated not only technically via cell phones but also via sirens and other means. Finally, it is also dependent on the capacities of society, which is characterised by different cultures of solidarity and cohesion in daily life. Still, these vary in disaster or stressful situations [11,72,73].

As a summary, the enhancement of existing vulnerability frameworks lies in these conceptual aspects (Figure 5):

• Integrating
• Disaster risk and impact situations
• Classes of movements and mobilities related to disaster risk situations
• Multi-localities, multi-modalities, multi-temporalities, and multiple risks during sequences of movement and stationary situations
• Daily and occasional hazards
• Emic and etic characteristics of vulnerability

The following figure summarises the previous conceptual diagram in various aspects. It is necessary to restrict the discussion to additional dimensions and representativeness that are also discussed in the text above, to remain focused on the clarity of key components, such as hazard interacting with the vulnerability dimension, and to outline mobility and its interactions and sequence in this study.

The conceptual framework could be expanded to include various types of hazards and risks that people are exposed to when travelling or in transition. Examples of specific risks that mobile people and groups face that they would not encounter at home include situations on roads and exposure to accident risks; in crowds, exposure to the risk of crushing; and in migration, risks of violence, crime, disease, or drowning.

4. Discussion

4.1. Limitations of This Study

This study is a conceptual development with the main purpose of advancing the conceptual understanding of vulnerability. It is primarily a didactic approach that guides research and policy frameworks to establish a better structure and criteria, helping to detect both dynamic and situational aspects of vulnerability, as well as static ones. This paper does not intend to demonstrate this directly in a concrete situation on the topic, as it has identified a major gap in real-world or empirical research on such dynamic situations. This is despite many other studies having already identified such shortcomings, including a lack of dynamic aspects of vulnerability, in previous studies. Therefore, it seems increasingly necessary not just to critique but, instead, to guide how to frame mobility vulnerability and which criteria and dimensions have great potential to advance existing static approaches.

Actual empirical research has engaged with the reality of vulnerability on the ground for several decades already [10]. Based on this inductive collection of knowledge, frameworks have emerged early on that stress, for example, the role of underlying root causes, such as marginalisation, as background drivers and conditions leading to situations that render people exposed to, and especially vulnerable to, hazards and, hence, risk [10,11]. However, despite the introduction of such notions of situations and dynamic development, distinct studies focusing on different situations of mobility vulnerability are not as common as static approaches. Additionally, similar frameworks that adopted the concept of representing a time arrow, or rather a cycle of conditions, have primarily been understood to conceptually divide into phases before, during, and after a disaster event [10,53]. However, what is still missing in both conceptual underpinning and real-world research is a more systematic assessment of mobility vulnerability.

Despite the advancements suggested by the framework, certain limitations must be stressed. There is a possibility of selection bias in search terms and scope, as it is based on the expert’s selection. The documentation of the method may help clarify which terms have been used and guide fellow researchers on expanding the terms and scope. The systematic literature review also contributed only partially to the intended results in guiding the development of the conceptual framework. This may be due to the incorrect selection of search terms or the breadth of the mobility field, which encompasses many areas. There is a lack of studies specifically focusing on the situational phase or movement that we intended to capture in terms of mobility vulnerability. However, certain aspects could be included in the development of the conceptual framework, including single and multiple hazard contexts, transportation modalities and related groups, vulnerability groups, mobility types, and place-based notions. Existing studies in this direction are promising and highlight important aspects related to hazard exposure, transportation, specific phases of evacuation, social vulnerability and migration, all of which are directly linked to disaster risk [74,75].

The limitations of the conceptual framework must also be emphasised. First of all, many other aspects of mobility from different streams of research, as well as those from other conceptual frameworks, could be added. However, what is still missing are frameworks of vulnerability that not only outline the conceptual dimensions of exposure, vulnerability, or resilience, or that mention mobility as a black box, but that provide concrete details, criteria, and different phases of situations that analyse mobility vulnerability more systematically. This framework is intended as a discussion tool to stimulate future developments in the field and should be refined over time. Its graphical representation is also limited, as it cannot fully visualise all dimensions of mobility, situations, and its dynamic aspects. Although the framework may appear static, it highlights specific aspects of mobility and situations that were not previously captured.

The study emerged as a reaction to the limitations of static data or representations, which also exist in other areas of vulnerability and disaster risk [76]. But it was inspired by the idea that while there is a strong focus on stationary places, there should also be research on non-stationary and movement aspects of vulnerability. Criticism has been raised that this static place-based assessment model does not capture a mobile society: “heightened levels of mobility are theorised to be leading to a so-called ‘placeless society’ and possibly nullifying theories of locality-based risk perception” [29]. The proposed framework does not fully capture the idea of a placeless society, but rather that of a place-shifting society—a society that is highly mobile both spatially and temporally, in both daily and extraordinary situations. It is, however, congruent with the general idea behind a placeless society, in contrast to a simplified and static one. Interestingly, studies also emerge on the idea of captivity of people locked-in at their situations and place-based vulnerability [77]. This can be interpreted as almost the opposite of the idea of a placeless society. It seems, therefore, useful to analyse and capture all the dynamic and situational patterns and variability between these two extreme ends: between people locked into a situation and place versus a placeless society.

We should, however, caution that in reality, it is a mix of stationary places of residence, work, and other activities, as well as movement, and patterns vary according to factors such as personal preferences, income, culture, age, and physical mobility. We should also caution that this cannot be generalised, as has been shown with tourists in the literature and examples above, since they can be following established paths over and again, or explore new places, which is often left to several factors of personal preferences, as well as external conditions of income, culture, or age and physical mobility. A key finding of this study is the need to identify typical classes and patterns of movement, particularly in disaster risk contexts. However, it is also necessary to incorporate this with patterns of people in daily situations. This can better represent how vulnerability overlaps with multiple hazards and their associated exposures, as well as the multiple dimensions of vulnerability that a person or group may have [55,78]. This includes both internal and external attributes of perceptions of the person’s vulnerability and perceptions by others, as well as the external situational environments. These are also liable to change, not only during movement, but also over time, which we also captured in the framework.

4.2. Can You Leave ‘Your Own Vulnerability’ Behind

Such a question could inspire future research to analyse the vulnerability of individuals at home and the characteristics they carry with them when travelling or moving. For example, people can behave differently when they are free from conditions at home, or they can still be restricted in their behaviour and skills, exhibiting the same limitations they would have at home. Research on this topic so far has covered aspects of elderly people at home versus when they are exposed to disasters [79]. However, the topic should be analysed in many more areas of vulnerability and exposure. It also needs to be analysed whether people can change their identity and weaknesses in exposure situations. The idea that people can change their characteristics or ‘weaknesses’ in response to exposure is speculative and warrants further research to generate robust evidence. Psychological research analysing the disaster personalities of people in crises, such as terrorist attacks, indicates that roughly a third of individuals can change their behaviour to take action to evacuate, while the remainder either remain incapable of changing their weaknesses or require an authority figure or person to trigger action [69]. Other studies have analysed self-efficacy in greater detail, guided by frameworks from psychology, such as protection motivation theory, social cognitive theory, and the transtheoretical model, among others [80,81,82]. However, the integration of psychological and perception frameworks and knowledge into natural hazard-related vulnerability frameworks is a major gap in research [83].

On the other hand, there are many situations or spaces where individuals cannot escape their environment or communities. An example is trains or other forms of public transport, which create communities of ‘common destiny’ [84], where collective and individual behaviours are tightly coupled or entangled; for instance, if one passenger falls seriously ill, the whole trip might need to be stopped. The link between individual and collective vulnerability in confined spaces (such as trains) implies interdependence that is not always evident or measurable. Generalising collective behaviour can be problematic and requires complementing it with research on individual behaviour. This has been a common area of research in disaster-related fields, such as disaster sociology, and also occurs in many related fields of psychology and social theory [72,85,86].

There is a significant amount of unknown territory in understanding how vulnerability can be captured at different stages, including stationary and temporal places, as well as movement corridors [87]. Which of the characteristics of vulnerability, either established at home or collected while moving or travelling, are recovered and propagated by persons or groups? [54]. Finally, it would be interesting to understand which of those characteristics people can apply in a disaster situation [69]. The conceptual framework also highlighted the need to separate situations more clearly in disaster risk research. Exposure to a hazard is not just a set condition. It is a moving window, either in space or time, when it presents a moving hazard, such as weather or a contamination cloud. It is also a moving window in terms of time, and different individuals entering the same open space, building, or vehicle can be exposed and potentially harmed or not. The concept of a “mobile window” of exposure is conceptually challenging and not well understood in social vulnerability research. To operationalise this, studies need to measure vulnerability in motion versus vulnerability in fixed locations. Methods from spatial assessments using moving windows of modelling of land use patterns could be an inspiration [88], as well as models used in disease or epidemic modelling, such as cellular automata, e.g., the game of life [89,90].

The study, therefore, leaves much room for future applications as well as enhancements to questions of mobility of both hazards, places, and vulnerabilities. Finally, any framework or approach claiming to measure vulnerability needs to be verified or validated. Especially because vulnerability characteristics can vary in specific disaster situations. Validation of vulnerability has been identified as a major gap in this line of research [91]. It requires further advancement in terms of methods, frameworks, data, and benchmarks from real disaster situations.

5. Conclusions

The study’s findings reveal that numerous studies have explored vulnerability to mobility topics; however, it also finds a scarcity of studies analysing specific vulnerabilities during mobility. The initial findings of a systematic literature review indicate that the term ‘mobility vulnerability’ is not yet widely used. Numerous studies on vulnerability assessments cover various aspects of mobility, including transportation, migration, and, to a lesser extent, evacuation situations.

However, there are a few studies identified that specifically analyse how the specific situation during mobility, being on the move, travelling, or in transition, imposes vulnerabilities on specific social groups or individuals. In contrast to the numerous studies on static place-based approaches, which primarily focus on identifying groups of people or individuals at their homes or other static locations, this study advocates for more research into mobility vulnerability.

i.

Research recommendations: a focus on science and future vulnerability assessments, which include studies of vulnerability not only at specific locations but also in between, as well as for the specific variables and indicators related to mobility. As the second part of this study, it presents conceptual criteria that could be considered in such an assessment of mobility vulnerability: first, recognising the situation and the moving window of time and exposure. Second, to integrate different studies on the location of origin and destination, and the stations in between, as well as the corridors of movement of individuals and groups of people. To address this, it is also necessary to consider the typical conceptual dimensions of vulnerabilities, including exposure, susceptibility, and capacities.

ii.

Policy/practice recommendation: recognise the necessity of planning for risks while people are travelling, in transition, or especially during disaster situations, such as evacuation. Risks, exposure, and specific susceptibilities must be planned for in advance, and assistance provided during such mobilities, as well as in the recovery process and aftercare.

Mobility vulnerability should therefore be at least conceptually integrated into future vulnerability and risk assessments. Strategic documents, such as those calling for policy actions within international frameworks like the United Nations, should mention this specific part more explicitly. Given the rise in mobility-related disaster risk research on migration or transportation, this should blend in easily. Furthermore, there is a need for more research on specific disaster situations with heightened extreme risks, such as evacuations. The paper is therefore an example of theoretical work in ‘disaster riskology’ [92] and aims to advance the theoretical aspects of vulnerability and risk, as other authors have identified this as a major persistent gap in disaster studies [93].

In principle, this would also prepare future risk and vulnerability studies to better address the momentum of change, which is another factor that needs to be considered in disaster situations, unfolding events, and processes. Not only can hazard compositions and threats change quickly, but also interactions of communication, groups of people, and other interdependencies involving people, infrastructure, and the environment.