Discover the dynamics An intersectional analysis of overt and hidden vulnerabilities to flood risk in urban Denmark

an


Introduction
This paper explores the concept of vulnerability in urban affluent societies focusing on the complex intersections and various forms of vulnerabilities that arise in high-resource countries at risk from climate change.In affluent societies, high levels of prosperity, socioeconomic stability and security create favourable conditions for reducing vulnerability and adapting to a changing climate.Nevertheless, we argue that vulnerability exists in any society, often taking different shapes, and requires greater recognition and attention in discussions of climate adaptation and disaster risk management.To overcome this challenge, we present a conceptual framework that approaches vulnerabilities as dynamic and contextual.
Vulnerability is defined as the conditions determined by physical, socioeconomic and environmental processes, which increase the susceptibility of a community or individual to the impacts of climate change (adapted from UNDRR, 2023).Disaster risk reduction and climate change scholars tend to focus on vulnerability among low-resource, disadvantaged and marginalised urban communities.Despite many solid reasons for focusing on those affected the most by climate change, understanding the drivers of vulnerabilities may provide an important lens to assess climate risk and climate justice in affluent, digitalised and privileged urban settings.The purpose of the paper is to offer such an understanding.To achieve this, we suggest a framework that approaches vulnerabilities as the result of social disparities related to accessibility, connectivity, mobility and diversity.
Because of the high levels of wealth, equality, and highly efficient social services, vulnerabilities in urban affluent settings are frequently overshadowed by the success of their socio-economic models and, as a consequence, difficult to identify by local disaster management authorities.As such, Frederiksberg is a critical case in the sense it is considered one of the most affluent places in the world.It is one of the wealthiest municipalities within the welfare state of Denmark, private and state-based insurance schemes are in place, the technological advancement is high and the climate adaptation infrastructure is highly developed.
Our approach builds on an intersectionality perspective, which supports a dynamic understanding of vulnerabilities to climate change.Although intersectionality has been applied both to the Global North and South, less evidence of the usefulness of this approach to unravel hidden vulnerabilities has been proved in the wealthy urban 'enclaves' existing in affluent societies like Frederiksberg.
The case study pays special attention to digitalised spaces and societies with an established digital communication infrastructure and aims at supporting researchers and organisations working with disaster risk management in defining, assessing and including dynamic vulnerabilities in climate adaptation processes.The conceptual approach is based on existing literature and adapted based on data from focus groups and a qualitative interview study carried out in Frederiksberg.

An intersectional and dynamic understanding of vulnerabilities to climate change
Vulnerability is a foundational concept in environmental and social justice discourses, and more recently it started to be used in the climate change debate (Oppenheimer et al. 2014).This is expressed in the climate justice movement calling for the protection of vulnerable people (interpreted as those experiencing the highest impacts of climate change and with less access to resources to deal with), and their inclusion in adaptation decisions (Fulton, 2002).
The understanding of disparities is central to the discussion on vulnerability and climate change.Both environmental and climate justice approaches claim that the system should work to reduce disparities to give everyone the same resources to cope with risks (Schlosberg and Collins, 2014).Vulnerability in environmental studies is used to discuss justice issues of distributive inequality (e.g.inequity of impact), lack of societal recognition in the decision-making system, marginalisation and weakening of capabilities (Schlosberg, 2007;Taylor 2000).
Recently the intersectionality approach to vulnerability has been introduced as a method in the climate justice debate (i.e., Thomas et al., 2019;Amorim-Maia et al., 2022;Gutterman, 2022) and in intersectional risk theory (Nygren et al., 2020).Coming from social justice studies (Ahmed, 2017;Crenshaw, 1989) and pointing to issues of race, class and gender (Davis, 2008), intersectionality has been used to open questions about 'identity ', 'power', 'subjectivity', and dominance (Chisty et al., 2021;Schmitt, 2013;Werner and Zimmermann, 2003;McCall 2005) to understand what characteristics are relevant in producing vulnerability on a case-by-case basis.This is particularly important in unravelling the unequal production of climate vulnerabilities.According to the intersectionality approach, individuals are not equal and treating them as fixed social groups could reduce the capacity to read the internal variability of these groups such as the complexity of the contextual disparities (Ranganathan and Bratman, 2019;Collins and Bilge, 2020).Vulnerability can be related to how the system advancement process can reduce inclusiveness rather than increase it.Individual markers are read in relation to other individual or societal markers to understand individual vulnerability.These factors are related both to individual characteristics and especially to the system's capacity to support all the members of a community.Thus, intersectionality can be described as 'how intersecting power relations influence social relations across diverse societies as well as individual experiences in everyday life' (Collins and Bilge, 2020n.p).
Most recently, intersectionality has been used also in disaster studies as a way to read the dynamicity of vulnerability over time (e.g., Prohaska, 2020;Kuran et al., 2020;Orru et al., 2022).These studies are based on the idea that the system perpetuates social disparities that are at the basis of levels of diversification of people's exposure, sensitivity and resilience, producing vulnerability (Fordham et al., 2013).Accordingly, vulnerability is a structural property that can change across time and place.In this perspective, individual characteristics play a role in building vulnerabilities as a result of societal disparities and the inefficiency or unwillingness of the system to address them.Race, gender and class become significant in a society that makes them factors of discrimination and builds (and perpetuates) its own relational system of power on these discriminating elements.Thus, the dynamicity of vulnerability is the result of the shaping of the system over time and of its contextual characteristics.

Surpassing the static perspective in urban affluent societies
Another step is required to understand the dynamic approach to vulnerability, starting with the recognition of the limits of a static approach.The static approach runs the risk of potentially producing new risks rather than reducing them as it overlooks the variability of those external factors that interfere with individual markers that create vulnerabilities.Consequently, a society could be unable to produce measures that adapt to this variability.In particular, in affluent societies, the static approach falls short of capturing the complex interdependencies between social, spatial and temporal aspects of climate change impact that open new (hidden) nuances of vulnerability.Furthermore, we sustain that the static approach adopts a deterministic view of vulnerability as it tends to classify people based on fixed and immutable markers.This poses a risk of not recognising the resilience capacity of those individuals classified as vulnerable.People that usually experience high levels of vulnerability, similarly also develop greater adaptive capacities and resilience as is the case of refugees (Uekusa and Metthewmann, 2017;Pulvirenti and Mason, 2016) and people with disabilities (even though this should mostly be read as the result of a situation of constrain and not without costs for these subjects).Consequences are that response systems may be unable to recognise who could become vulnerable in some situations.Moreover, this perspective could blind disaster management organisations when vulnerabilities appear to be either existing or non-existing according to pre-identified markers, ultimately generating hidden vulnerabilities.Fordham (2007) argues that a context-specific analysis is required in studying vulnerability "before assigning groups the potentially stigmatising label of vulnerable" (p.2).Accordingly, the intersectionality approach is applied in this paper as a result of contextual analysis, introducing a new approach to intersectionality in disaster studies.
Such a dynamic and contextual assessment of vulnerabilities to climate and disaster risks is paramount in urban affluent societies where 'vulnerability' rarely follows conventional markers.First, because affluent societies are associated with prosperity and stability, vulnerability is difficult to identify, recognise and measure.The markers of vulnerability as often applied in vulnerability assessments rarely apply in highly affluent settings (Orru et al., 2022).The implication is that some individuals could be overlooked because they are treated homogeneously in disaster risk management efforts without consideration of the intersectional conditions that could produce hidden vulnerabilities.Second, because vulnerability is a complex and multifaceted concept influenced by social, economic, political and environmental factors, effective measures are difficult to identify, implement and scale.Third, material capacities (infrastructure, GDP and technological advancement) are not enough to adapt to the impacts of climate change.We continue to see an "ongoing fixation of material coping capacities" (Eriksen et al., 2020, np), however, even well-developed adaptation infrastructure and advanced technology can lead to segregation in affluent communities.In some instances, affluence itself may lead to 'lock-ins' or the creation of disaster risk due to the highly interdependent and complex system interactions (Beck, 2007).In climate adaptation strategies, an increasing role has been assumed by online geotechnologies, especially for providing information, supporting the response system and helping to coordinate the efforts (Harrison and Johnson, 2016).The increasing role of online technologies in disaster risk communication sparks concerns about how vulnerabilities and their spatial distribution are shaped by technologies, especially in terms of accessibility and inclusion.

A conceptual framework to analyse vulnerability in urban affluent societies
The LINKS H2020 European project has developed a dynamic vulnerability framework to analyse vulnerability production/reduction in affluent societies.It plays special attentional to communication technologies as their potential has been recognised by the disaster literature in terms of increasing the capacity of information, reliability and interaction between different stakeholders (Jaeger et al., 2007), the support for people in making decisions and the resilience of local communities (Lindsay, 2011).
The framework starts by considering how individual and societal factors influence different individuals and their resilience capacity.The combination of these factors is responsible for the dynamicity of vulnerabilities.The approach uses intersectionality to assess urban affluence across four macro-categories: accessibility, connectivity, mobility and diversity.These categories are based on an extensive literature review (Bonati, 2021;Morelli et al., 2022).Thus, the framework has been tested and refined as follows: -In disaster scholarship, accessibility is defined in relation to access to resources and it depends on the power relations that guide a society (Bryant, 1998).Accordingly, people with limited access to resources have fewer opportunities to respond adequately to a disaster and receive attention in the disaster risk management system; -Connectivity is a spatial concept that describes the level to which a system can connect places and people but also how people experience connections.With connections we mean both material connections represented by infrastructures and non-material ones like social and political connections (Foley 2020); -Mobility refers to the movement of people, goods and ideas that disasters can activate or disrupt.Power and inequality have strong consequences on the capacity of people to move and react in front of shocks opening issues of mobility justice (Sheller, 2018); -Diversity in disaster studies has been mainly associated with the gender dimension, however, in this framework, the concept is understood from a more general perspective.With diversity, we include personal markers (Dominey-Howes et al., 2014), capabilities, resources, skills and knowledge (see Becker, 2002).Diversity is divided into subdimensions that intersect with the levels of accessibility, connectivity and mobility, as summarised in Table 1, and that can be defined as follows: • physical/sensorial: the individual physical markers (e.g.age, sex and race) including physical and sensorial disabilities • material/infrastructural: the availability of material resources (e.g.economic), infrastructure and the possibility to use them • social/cultural: social and cultural aspects like gender power relations/hierarchies and cultural belonging, but also individual skills and knowledge related (e.g.educational level) • relief/organisational: the existing organisational system and the capacity of the system to represent all community members and involve them in the decision-making.
The same sub-dimensions identified for diversity are also used to describe the sub-levels of accessibility, connectivity and mobility as described in Morelli et al. (2021).Thus, e.g., accessibility could be physical, sensorial, cultural, social and organisational.Some examples of the potential aspects of intersectionality to identify vulnerability are provided in Table 1.

Frederiksberg as a critical case
The Municipality of Frederiksberg is the smallest and most densely populated municipality in Denmark.It takes up 8,7 square metres and has a population of around 105.000 people.Frederiksberg serves as a 'critical case' defined as a "case having strategic importance in relation A.B. Nielsen et al. to the general problem" (Flyvbjerg, 2006, p. 229).As such, we strategically selected Frederiksberg Municipality as a "tough test" (Levy, 2008) based on the following logic: If vulnerabilities to climate change exist at Frederiksberg, they exist in other affluent landscapes.
Frederiksberg is highly exposed to flooding in densely populated and low-lying parts and in areas where infrastructure is not adequately designed to lead the water away from properties and streets.Like the rest of the country, climate change affects the intensity and frequency of extreme weather events.Denmark receives a larger amount of rain in autumn, winter and spring and a higher intensity of precipitation in the summer months in the shape of "cloudbursts" and extreme rain.These events will become more frequent and more intense towards 2100 depending on climate mitigation efforts (DMI, 2023).Fig. 1 below shows the flood risk at Frederiksberg in 2112 in the context of a 100-year event.
In recent years, Copenhagen and Frederiksberg Municipalities have experienced a range of extreme rainfall events.The largest and most impactful was the 2011 event that flooded large parts of the cities and led to damage to the infrastructure, homes and businesses.The total amount of damage from that event has been calculated to be almost 1 billion euros.The 2011 event highlighted the necessity of adapting to climate change and led to city-wide adaptation strategies in both Copenhagen and Frederiksberg municipalities (COWI, 2012).
Despite the prognosis, the majority of Danes do not worry much about flooding (Beredskabsstyrelsen, 2022).Denmark is a welfare state with strong public institutions, similar to the other Nordic countries.Disaster response is a public responsibility dealt with by public organisations, fire brigades, municipalities etc. and the response capacity is considered high.There are several definitions of a welfare state: one of the central definitions is a society with a solidarity-based redistribution of common goods that accommodates the citizens most vulnerable and in need, which depends on a high tax on income for most groups (Esping- Andersen, 1990;Bruun et al., 2017).This consequently leads to a population that in general has high expectations of the welfare state to take care of vulnerabilities (Andersen et al., 2017).The principles do however undergo changes, and in recent years there is an increasing expectation of citizens to act as self-dependent and citizens need to prove 'worthy' to require assistance (Bruun et al., 2017;Gullestad, 1992).
Frederiksberg's population holds resources in many ways: residents have a higher average income than the average Danish population and the people living in the surrounding Copenhagen Municipality (Danmarks Statistik, 2019).Frederiksberg residents have a higher level of education compared to the Danish population in general, and private estate is some of the most expensive in Denmark.Only a minor part of the apartments are public housing, and the ownership is dominated by privately owned estates and cooperatives compared to Denmark in general.Due to the urban setting, the municipality is dominated by condominiums (95% of the households) (COWI, 2012), and only a few detached houses and villas.Around 20% of the residents immigrated to Denmark from other countries or are descendants of immigrants.This is a larger percentage than the country's average (Danmarks Statistik, 2021).
Denmark is one of the most digitised countries in the world.The European Commission's Digital Economy and Society Index (European Commission, 2022) compares member states on the parameters of 'human capital', 'integration of digital technology', 'connectivity', and 'digital public services' and Denmark holds high scores on all four parameters.There are no particular studies of the degree of digitalisation in Frederiksberg but 70% of the Danish population have what is defined by DESI as 'basic digital skills'.Communication and information from Danish national and local authorities to citizens are to a large extent digital on either digital platforms, mail, or by text messages, also in crises.This tendency is likely to increase even further in the years to come.

Methods
The empirical analysis in this article derives from 22 expert interviews and six focus groups conducted between November 2021 and August 2022.The 22 interviewees were strategically selected to represent disaster management organisations in Denmark that work with flooding in the capital region and include authorities, civil society, industry, the media and research organisations.The interviews served to understand how disaster management organisations operate to support disaster risk management using technology in different European countries (Nielsen et al., 2021).Questions for both the interview study and focus groups were developed by a Danish team of researchers and tailored to explore flood management in Frederiksberg Municipality.The overall themes for both methods were decision-making procedures (e.g.efficiency, perceptions of existing processes, coordination between actors, the inclusion of actors), vulnerability (definitions and understandings, inclusion of citizens, role of citizens) and information (e.g.digital communication, false information, credibility and quality).
The six focus groups were all conducted with citizens living in Frederiksberg Municipality.The invitation stated a focus on "cloudbursts, flooding, water contamination and other emergency incidents" and 37 persons ended up participating in one of the six focus groups.It was an explicit aim to construct heterogeneity in the groups to secure dialogue across the different participants' experiences, perspectives, and concerns (Stewart and Shamdasani, 2015).Four to nine persons joined each group, in total twenty-one women and 16 men, from the age of 25 to 80 with an average age of 60 years.The relatively high average age is a well-known phenomenon in Denmark when recruiting participants for research projects.A couple of the participants had physical disabilities.The choice of focus groups as the methodological frame is not done to obtain as close a representation of demographic parameters as possible, but to explore experiences and sensemaking (Stewart and Shamdasani, 2015).
The analytical process was abductive (Blaikie, 1993) as we searched for known and predefined categories such as experiences of previous floods, preferences of media and communication, types of vulnerabilities and concerns for future incidents.Parallel to this, we were aware of upcoming and new themes that appeared in the empirical material.These categories were: social networks, the feeling of being safe or unsafe, digitalisation, and the responsibility of civil society.The transcripts of the interviews and focus groups were first coded in NVIVO using predefined categories.The following interpretation of the participants' experiences, sensemaking and discourses in relation to these codes (Brinkmann and Kvale, 2015) offer two overall insights.First, the data contain accounts of previous events, sensemaking of previous and current disaster risk management processes as well as own experiences of managing flooding in organisations working with disasters and among residents living at Frederiksberg.In addition, the focus group data reflect discourses and attempts to position oneself and others regarding resilience and vulnerability.Second, the focus group data can be regarded as enactments of social life (Halkier, 2010), since the interactions, perspectives and engagement provided in the focus groups are expected to be identified outside the focus group setting.All dimensions provide insight into the social practice of living on Frederiksberg and dealing with potential hazards in an urban setting in a welfare state with low exposure to emergencies and disasters.

Analysis
Using the vulnerability framework presented in section 3, the following presents an assessment of vulnerabilities in an urban affluent setting, the metropolitan municipality of Frederiksberg.The assessment follows the diversity dimensions of physical/sensorial, material/infrastructural, social/cultural and relief/organisational and shows how issues related to accessibility, connectivity and mobility intersect with each of these dimensions helping to identify the hidden vulnerabilities in the context of Frederiksberg.

The physical and sensorial dimension of vulnerability to flooding
In the case of Frederiksberg, this dimension evolves around vulnerabilities related to physical ageing and health.When asked who is vulnerable in case of cloudbursts at Frederiksberg, the focus groups perceive 'the elderly' as vulnerable to floods and as someone who should be paid special attention to in flood risk management.Interestingly, many of the participants above 65 described themselves as resourceful and resilient and in contrast to 'old people' who they deemed vulnerable to flooding.Physical ageing only becomes an issue when associated with issues of mobility, as mobility is a prerequisite to taking preventive measures against flooding, preparing flood response measures, and if one needs to evacuate.
People with disabilities were also identified as vulnerable in relation to mobility issues.One focus group participant uses a wheelchair and once experienced that the lift was out of order for four hours where she lives: "I couldn't get from the attic down to my apartment on a lower floor (…) I called for help, and they answered if I could get help from people on the floor where I was stuck?That raises the question: What happens in case of a power failure?There are other residents dependent on wheelchairs in my building.The other day one of them told me about being stuck in the apartment for eight days and not being able to leave since the lift was out of order" (Frederiksberg resident, 2022).
Descriptions also include experiences of relatives, acquaintances, and neighbours who help reduce the vulnerability to flooding in various situations where residents lack the physical ability to cope during the 2011 flood.Moreover, focus group participants refer to episodes where residents lacked information or/and flood warnings from the authorities or the cognitive ability to fully grasp such information.A participant tells how a neighbour needed help and care to enter his home: "One person with disabilities came back from vacation, but our lift was out of order due to all the water and it didn't work for three months.We had to carry him up the stairs to his apartment, and then we hoped that the municipality would assist him" (Frederiksberg resident, 2022).
As highlighted in this quote, acknowledging and addressing physical/sensorial conditions are essential for ensuring mobility in a disaster.Similarly, connectivity to authorities and neighbours becomes paramount for the ability to handle an emergency if one has physical/ sensorial problems.In an interview, a government official asks rhetorically "what they [the government] should do about it" referring to the complex problem of mobility and information access creating vulnerable conditions for some citizens when they intersect.The official answers the question by adding: "it is probably a combination of how we get information out to them and how we get there to help them" (Government official, 2022).This is a complex issue related to both questions of mobility justice (Sheller 2018;Lord et al. 2009) and information access in disaster response (Morelli et al., 2021), which challenges disaster response in urban affluent societies as it demands a new way of looking at vulnerabilities and responding as an emergency organisation.

The material and infrastructural dimension of vulnerability to flooding
In an urban affluent setting like Frederiksberg, the material/infrastructural dimension is key for understanding what drives vulnerability before, during and after a flood.Frederiksberg is a densely built and densely populated municipality within a capital region and the highquality infrastructure (e.g.water, electricity, district heating, sewage system, road, bike paths, streetlights, internet cables, blue and green infrastructure) makes the city run smoothly daily.Similarly, the municipality is implementing a large-scale flood management plan, where the core is a new layer of city infrastructure that delays guides and catches the water to prevent flooding in low-lying areas (Frederiksberg Kommune, 2012;2017).This speaks to an "ongoing fixation of material coping capacities" (Eriksen et al., 2020, np), which often characterises climate adaptation and disaster risk reduction efforts in the Global North.However, as we argue in the following sections, material fixation may have a flipside and create disaster risks.
Flood risks are unequally distributed across Frederiksberg.In particular, the geographical location, type of housing and floor level (in multi-story buildings) matter.People living in low-lying areas of the city, on the ground and first floor and people with storage facilities in the cellar, are at risk of being flooded directly from the street or through the sewage system.Residents who own their homes tell about their experiences from the 2011 flood: "The water ran down the basement well and through the cat limb, and one moment later it entered through the letterbox.We closed the door to the hall and did what we could to clear the rooms and place all the stuff on the tables.We thought 'now, we have secured this'.But the water kept rising and it felt like we were in an aquarium, the water got nearer the windows, and it kept rising.My wife said 'we have to jump out of the windows, I'm afraid what happens if we open the door'" (Frederiksberg resident, 2022).
Another focus group participant witnessed the same flood close by but the location of her apartment made her experience very different: "Well, I don't think that anything like it will happen in my home.I live on the third floor.That is convenient" (Frederiksberg resident, 2022).For the first resident, the material wealth in the shape of a house on one of the most expensive addresses in Copenhagen became a vulnerability because of its low-lying location and lack of flood preparedness.
Flooding does however affect more than the interior of one's home, floor, walls and furniture.City-wide infrastructure was equally a concern among both disaster management organisations and citizens when discussing experiences from past floods.One example is the access to digital information and the ability to connect to others being at risk even though the location of the home seems safe.A focus group participant tells about some of the consequences of being out of electricity in the 2011 flooding: "We felt disconnected from the world around us.We were unable to obtain any connection to the internet, and all electrical installations in the house were flooded" (Frederiksberg resident, 2022).
Information from authorities and news media is available and comprehensive during crises in Denmark.However, when critical infrastructure breaks down it influences the possibility to respond to a disasterin this case following the electronic media in a situation where information is paramount as it is used to decide how to react (Lai et al. 2018) and an important source of psychological support (Sutton et al. 2014).
A local government representative similarly reflects upon how the break-down of infrastructure limited the mobility of people in the context of the 2011 flooding and how the infrastructure design and location create vulnerabilities in a flood situation: "What we did not really think about during the cloudburst in 2011 (…) was that basically, people could not get to the hospital, because all roads were flooded.So, people could probably not go to other hospitals either, so it is important to think how we can ensure infrastructures continue to function" (Government representative, 2022).
The same government official provides an example of how the infrastructure itself became the source of vulnerability during the 2011 flooding.The sewage system could not handle the amount of rain that fell and as a consequence water started entering people's homes through drains and sewer covers (Government representative, 2022).Additionally, the pressure from the water made sewage covers on roads and streets "blow off" making it unsafe to be on the streets of Copenhagen and Frederiksberg.Thus, as the interviews showed material and infrastructural diversity was mainly connected to the feeling to be isolated, alone and disconnected in one's home/city.

The social and cultural dimension of vulnerability to flooding
According to our conceptual framework for assessing vulnerability, the social/cultural dimension concerns the normative and cultural aspects of accessibility, connectivity and mobility.This includes what actions and beliefs are considered legitimate and valuable, and who are considered legitimate actors on both societal and individual scales.
In Frederiksberg, a central question related to social/cultural dimensions of vulnerability is that of information: how (fast) information about flood risk is communicated, who receives this information and whether communication on flood risk is trusted and acted upon.In general, Danes have high expectations of the public authorities and their ability to communicate (Andersen et al, 2017;Beredskabsstyrelsen 2022) as well as a high degree of trust in authorities and the information they communicate.The empirical material is rich on accounts from all age groups, and most focus group participants expect information and warnings directly targeted to their needs: One participant expresses the level of expectations in this way: "I must admit that I am quite comfortable.I know it sounds a bit stupid when I tell you about it, but I kind of expect to be informed without being the one who is going to search and find the information.It might be a thing that sticks to my generation" (Frederiksberg resident, 2022).
The quote is from a younger person, who awaits messages targeted and the preferred format is a text message on the phone.
There is not a clear divide between age groups concerning preference towards a certain type of media outlet.However, many of the older participants are reluctant towards social media and social media use in crises opening to issues of digital disability (Goggin and Newell 2003).This is a challenge in an affluent society like the Danish, where social media, e-mails and other digital communication solutions are increasingly used by public sector actors as well as NGOs and legacy media (Eriksen et al., 2020).
Despite the speed of digitalisation in the Danish public sector, and regardless of the generally high level of digital competencies in the population, there is still a request and a preference for applying classical media formats like letters, posters and notes on bulletin boards in the hall/entrance to the staircase by the oldest participants when it comes to information that is not very urgent.The continuous increase in digitalisation, in general, is problematised by several focus group participants: "It is a problem with the elderly residents, who are not digital, it is difficult for them to get information, whether it is from the hospital or the municipality, and if they fall ill, they can't order groceries online and get them delivered directly since they don't have an online account" (Frederiksberg resident, 2022).
As a consequence, people without digital competencies and access are more vulnerable to incidents like cloudbursts since their capacity to seek information on digital media is low.Similarly, a government official points out that "foreigners -people who work in the country or visit as tourists" (government official, 2022) are unlikely to follow the Danish media.Access to information and digital formats seems to be a parameter that causes vulnerability to floods.If individuals with limited access to official information are, however, connected, to people with access it may reduce their vulnerability.There are some examples of civil society volunteering where citizens help citizens overcoming the obstacles of becoming digitalised and concerns about what happens if a crisis occurs: "I have neighbours I help that way!I happen to be quite skilled in IT, so I assist them with all the online requirements from the authorities.One of them, first she needed to scan her passport with her smartphone, and she only has an old fashion telephone, and she can't get down the stairs, and she can't go and see the service desk at the townhall, she can't make it, so I try to help her.I believe she received text messages; I think she was updated during the cloudburst and the water contamination" (Frederiksberg resident, 2022).
Vulnerability is thus also associated with a lack of social networks.Examples below are from two focus groups where participants describe the relationship between vulnerability and social ties.Some people are isolated, and as this quote shows, despite that, they live closely together in condominiums: "Some years back, I experienced a neighbour who rejected moving to a nursing home, and who almost did not have any social contact at all.There was just this one friend who visited her.She lay dead for almost a week in her apartment before anyone realised because when you refuse others to come into your home" (Frederiksberg resident, 2022) People are often believed to assist and support each other in socionatural hazards, and that weak social networks can get strengthened (Quarantelli, 1993).In this affluent urban context, some lack a social network, and this disconnects them.However, as reflected in the quote below, this varies across space and creates different levels of vulnerability.Some do not have the resources to help and assist each other, which this example from the 2011 cloudburst shows: "The solidarity that occurs was obvious in our building since we were all in the same situation [following the cloudburst] but when I went to see my old friend at one of the streets close by, the entire building was covered by black, dirty water, it was bad.It was elderly people, some of them were immobile, and they really needed help, and for some strange reason they were not able to coordinate themselves, it seemed as if they were not able to help each other" (Frederiksberg resident, 2022) The focus groups reflect important insights into vulnerabilities in affluent societies.Vulnerabilities exist in places that fixed markers related to wealth, race and class do not capture.In this case, a group of elderly people in an upper-middle-class neighbourhood without strong social ties.People, who do not experience or perceive themselves as vulnerable, may lack adaptive capacity and be overlooked in response efforts that rely on static and deterministic understandings of vulnerability (Uekusa and Metthewmann, 2017;Pulvirenti and Mason, 2016).

The relief and organisational dimension of vulnerability to flooding
In Frederiksberg, discussions related to the responsibility of flood risk are often guided by the established political and economic structures of the Danish welfare state (see Section 4.1).This implies that flood risk management in Frederiksberg is deeply entangled with an organisational system where the public sector has responsibility for the most vulnerable: those without access, connections and the ability to move.The dominant role and responsibility of the welfare state are also present in both focus group and interview data.When asked who is responsible for fellow citizens, a focus group participant explains that "It is never a stranger who can be responsible", thereby stressing that civil society is less important.An interviewee from the emergency services explains how the tasks and responsibilities held by the public sector support the vulnerable during a flood: "(…) the municipalities know to a very large extent in advance because they are clients and citizens of the municipality.They have home care, they have home nursing and social administration.They are in contact with the municipality in all possible ways."(Representative from the emergency services, 2022).
As described in Section 5.1, the strong belief and trust in the welfare state's ability to reduce vulnerability exist among both citizens and authorities.Nevertheless, it also creates a tendency to oversee or downplay the vulnerabilities that arise when disparities related to accessibility, connectivity and mobility intersect in specific times and places.Vulnerabilities were a theme that most of our interviewees from the authorities had difficulties talking about.At first, they found the concept of vulnerability difficult to define and to provide concrete examples of processes or standards that are sensitive to vulnerabilities or people at risk from flooding.This is reflected in the answer from an interviewee when asked if they work with a dynamic concept of vulnerability: "We have not done that before in terms of being so specific regarding elderly residents nor other characteristics like depending on which roads they live on, or if they have a basement or not (…) Yeah, I sense this question makes me start thinking…" (Government representative, 2022).
Another interviewee answers that vulnerability is not at the forefront of discussions, but expresses the assumption that people have the capacity and resources to navigate themselves: "This is not something we have discussed -in which situations it is that we have to reach these vulnerable groups?(…) But it's an interesting way to look at it, I just need to think about how we can do it".(Government representative, 2022) A representative from an NGO working with disaster risk management in Denmark points to this exact issue with the current system: "I think my point was that the authorities' needs assessment of people who are not already categorised as vulnerable, who are inhome care or are in some way in contact with the municipality.Well, the needs assessment, it does not exist, I mean simply" (representative from an NGO, 2022).
Interestingly, the authorities are aware that their approach to assessing vulnerability is based on fixed understandings of vulnerabilities that may not match the needs of the people living at Frederiksberg.Nevertheless, the interviews with the Danish authorities reflect how current strategies and disaster management practices oversee some of the vulnerabilities to climate change that we identify in this paper.In this way, the organisational set-up in itself may co-create vulnerable outcomes, as its sensitivity to dynamic vulnerabilities is limited yet trusted by the people who rely on its support.
This calls for a different approach to understanding vulnerabilities, which captures disparities across accessibility, connectivity and mobility divides in Frederiksberg.The adapted version of Table 1 below (Table 2) shows what a dynamic vulnerability assessment in an urban affluent Frederiksberg may look like in the concrete context of flood risk in 2022.It points to the fact that several dimensions associated with 'affluence' (e.g.technology, high-quality urban infrastructure, education levels, welfare state structure) create flood risks in this particular situation and should be taken into consideration when flood risks are addressed.

Conclusion and final remarks
The paper addresses the need for adopting a concept of vulnerability in urban affluent societies, which supports better measures for disaster risk reduction and climate change adaptation efforts.The purpose was to provide a framework for assessment, which allows for a greater understanding of "hidden" vulnerabilities in urban affluent places where economic wealth, high-quality urban infrastructure, use of information technology and a large welfare state reduce the risk of climate change.We argue that there is a lack of understanding of what vulnerabilities look like in such affluent urban places and identify a need to assess and address them as dynamic and contextual.Applying an intersectionality approach to the case of Frederiksberg, a small metropolitan municipality within the city of Copenhagen, we suggest a conceptual framework for assessment, which considers physical, material, social and organisational aspects of risk creation and relates them to interacting vulnerability factors related to accessibility, connectivity and mobility.Our analysis shows that vulnerabilities are associated with individuals and groups that live in certain parts of the city prone to flooding, where critical infrastructure is maladapted, where people have weak social ties and reduced ability (physical or cognitive) to engage with internet communication and communicate in local languages.The analysis directs the attention towards elderly citizens, who struggle with decreased awareness, and physical disability as well as individuals who lack the digital competencies to follow the flow of information provided on the internet.The vulnerability framework presented intends to increase awareness among disaster management organisations of hidden and overt vulnerabilities in a context like the Danish.It is an attempt to develop a tool that considers well-known parameters that also represent a dynamic perspective.The assessment of vulnerabilities in Frederiksberg provides insights into three dimensions of risk management and climate adaptation efforts, which is worth discussing further.First, it continues to be a fundamental paradox that some of the wealthiest and welfare-concerned societies with considerable response capacity lack thorough insights into the lack of accessibility, mobility and connectivity among subgroups of citizens.Their sense of vulnerability and attention towards vulnerabilities in various forms can be argued to be unvarnished.As the climate crisis is starting to affect affluent cities like Copenhagen, there is an increasing need to understand the complex interaction between wealth, welfare, climate change and disasters in urban affluent settings.Despite the risk and impact beingfundamentally different from what is experienced in many Global South cities, a different vocabulary, toolset and point of departure are needed to deal with climate impacts in urban affluent societies and to secure climate justice for current and future citizens.This is shown in the assessment done in this paper where a set of vulnerabilities related to technology, age and connectivity is made visible by using a dynamic approach.
Second, and in relation to the aforementioned point, the technology optimism and fixation seen in an urban affluent setting like Frederiksberg becomes a risk in itself.While this understanding of risk is wellestablished in the literature (Beck, 2007), we continue to see a gap in how urban planning and climate change adaptation is addressed in practice.In this paper, we show how material capacities easily become vulnerabilities if not assessed and incorporated as risks in prevention and preparedness efforts.Future climate change-related incidents and the expected impact on residents urge a changing approach, also in urban affluent settings.
There are several arguments for further research to tune the methods applied to be even more precise in identifying categories and mapping the mechanism of intersecting in relation to disaster resilience research.This also concerns the analytical challenge, that the most vulnerable are often the most silent.The study based on focus groups and interviews applies proxy voices of authorities and more resourceful citizens to identify who is vulnerable, these representations are however not sufficient for providing a full or representative picture.Additional methods need to be developed.
A.B.Nielsen et al.

Fig. 1 .
Fig. 1.Flood map of Frederiksberg.The map shows the flood risk at Frederiksberg according to a 100-year event in 2112.Source: Frederiksberg Utility Company.
Access to resources (e.g., free Wi-Fi, high-speed internet connection) Access to technological supports (e.g., smartphones and laptops) able to connect with service systems Availability of home internet connection Level of knowledge/intelligibility of the information (e.g., multi-language text available) Levels of digital disability Use of a child-friendly language Use of the most used channels of communication according to the different age-groupsAvailability of informative materials for different kinds of people (e.g., the LGBTQ + community, refugees, homeless, etc.) A.B.Nielsen et al.

Table 2
Vulnerability assessment -the case of flood risk in Frederiksberg.Reduced mobility of people living in flood-prone areas Missing evacuation assistance for disabled and elderly people A.B.Nielsen et al.