Using photovoice to investigate the impact of place attachment on community resilience in Iceland

Abstract

Purpose

Globally, climate change is exacerbating the impacts of climate-related, natural hazards including avalanches. However, there is limited knowledge about how small and remote communities are affected by and perceive the effects of a changing climate with hazards that increase in intensity and/or frequency. Consequently, there is a call for more actionable and interdisciplinary climate adaptation research, which takes its starting point in understanding the local concerns of people living in small remote communities.

Design/methodology/approach

This paper test the photovoice method to gather respondents’ perceptions of the place in which they live and the hazards they face through personal narratives of photographs.

Findings

Despite its limitations, the photovoice method was found to be a suitable tool for gaining valuable insights into the communities while ensuring comfort and enjoyment for both participants and the researcher.

Research limitations/implications

The findings also show that despite its limitations photovoice is a useful method for shedding light on risk perception, place attachment and resiliency in communities facing the risk of natural hazards. The study found that place attachment is an important factor to consider in disaster risk management, policy and decision making.

Originality/value

The paper adds to a growing body of literature surrounding the relationship between place attachment and community resilience to climate-related natural hazards. The authors examined the impact of place attachment on community resilience, focusing on two rural and remote villages located in the Westfjords in Iceland – Patreksfjörður and Flateyri. Societal aspects of natural hazards have hitherto been hardly addressed in Iceland and to our knowledge the applied method has not been tested before in such a setting. The photovoice method is tested to gather respondents’ perceptions of the place in which they live and the hazards they face through personal narratives of photographs.

Citation

Lyons, A., Gísladóttir, J. and Kokorsch, M. (2024), "Using photovoice to investigate the impact of place attachment on community resilience in Iceland", Disaster Prevention and Management, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/DPM-01-2024-0030

1. Introduction

1.1 Capacity, resilience, place attachment

Climate change has been identified as a cause of increasing and differentiated exposure to natural hazards. Factors such as socio-economic status, geographical location and access to resources can cause exposure to differ between people, communities and regions, with marginalised, low-income and certain geographically located communities facing increased exposure (IPCC, 2021). Iceland is no exemption, where calls have been made for continual updates in disaster risk management and scenario analysis (Björnsson et al., 2023). Recent reports on climate change adaptation in Iceland found that resilience is a vital part of adaptation (Icelandic Climate Council, 2020; Ministry of Environment, Energy and Climate, 2023). It is pointed out that climate change induced hazards, such as increased precipitation, have various societal impacts through damage to infrastructure and housing, as well as the need for added perspectives in planning.

Snow avalanches (henceforth “avalanches”) are an example of a natural hazard which has the potential to damage infrastructure, services and human wellbeing (Strapazzon et al., 2021; Ballesteros-Cánovas et al., 2018; Decaulne, 2007). The threat of avalanches in Iceland has increased since the twentieth century due to population distribution and improper zoning causing buildings to be poorly located, with little information about how these areas could be affected by avalanches (Decaulne, 2007). Avalanches and the risk of avalanches have many impacts on communities, including environmental, economic and social impacts. Of which road closures, interruptions in the movement of goods, services and people, and in worst-case scenarios, burials and deaths are important factors in the context of Iceland and this study.

Decaulne (2007) highlights the lack of studies in relation to avalanche hazards in the Westfjords of Iceland prior to 1995. Furthermore, studies that did exist focused mainly on the technical aspects of avalanche events and largely ignored social issues, such as risk acceptance. The avalanches in 1995 in two communities in the Westfjods, Flateyri and Súðavík, exposed the extent to which Iceland was unprotected and underprepared for avalanches of this nature – which resulted in the deaths of 34 people. Following the events, public awareness regarding avalanche risk in Iceland grew, sparking efforts to better prepare for any future events (Decaulne, 2007) which consequently increased the amount of research on avalanches in Iceland (Bernhardsdottir, 2001; Finnsdottir and Elklit, 2002; Thordardottir et al., 2015; Fjeldsted, 2020; Kokorsch and Gisladottir, 2023).

Place attachment is a field of study which examines the bond people have with places in relation to certain social, physical or functional aspects of the place (Lewicka, 2011; Bonaiuto et al., 2016; de Dominicis et al., 2015; Scannell and Gifford, 2010). Place attachment is cited in environmental disaster literature as a possible deciding factor for community members to stay in an area despite environmental hazards (Blondin, 2021; Scannell and Gifford, 2010; de Dominicis et al., 2015). Bird et al. (2011) found that emergency response plans must be developed in consultation and collaboration with local communities to ensure applicability. Bonaiuto et al. (2016) in agreement with de Dominicis et al. (2015) indicate a lack of information surrounding the relationship between place attachment and risk perception—finding just 31 social science works regarding the topic from 1996–2016.

1.2 Research question and objective

Understanding the relationships between risk perception and place attachment in communities facing the risk of natural hazards can provide insights into community resilience. Considering the complexity and sensitivity of the topics of avalanche risk and place attachment as well as the differentiated impacts of natural hazards, it was necessary to approach this study using methods that amplify the voices of local communities and capture the nuances involved in disaster studies. Photographs were considered an appropriate medium to conduct such a study supported with personal narratives and stories. This approach is known as photovoice which seeks to provide contextual understanding and community engagement in social research studies such as this one (Sutton-Brown, 2014). The research objective was to explore the extent to which photovoice can be useful for shedding light on risk perception, place attachment and community resilience in areas living with natural hazards by addressing the research question “how effective is photovoice for investigating the effect of place attachment on community resilience to natural hazards?”.

1.3 Conceptual framework

Resiliency encompasses persistence, adaptation and transformation, it is about the capacity to withstand external shocks (Davoudi, 2018). Recently, resiliency has become an important concept in studies relating to vulnerable settlements (Kokorsch and Gisladottir, 2023). The concept of community resilience appears in the literature regarding climate change adaptation and measuring populations’ ability to deal with the effects of climate change (Kruse et al., 2017). Hovelsrud et al. (2018) highlight a gap in research regarding the impact of local and social perceptions of risk on avalanche preparedness. Amundsen (2012) notes in her analysis of community resilience, that resilience is dynamic and changes over time. Further, the rate at which climate change is causing new and irregular hazards strains the capacity of a community to remain resilient. The concept resilience can be criticised on many fronts, in particular when the political causes – for example a neoliberal status quo – that make resilience building measures necessary are not scrutinized (Kokorsch, 2022; Wilson, 2017). Notwithstanding, the concept deems useful for the given context and case study sites, applying the definition of Amundsen (2012, p. 46): “Community resilience is the ability of a community to cope and adjust to stresses caused by social, political, and environmental change and to engage community resources to overcome adversity and take advantage of opportunities in response to change”.

The framework for the identification and characterisation of risk perception (adopted from Raaijmakers et al., 2008) has four themes; (1) Control: when an individual or group feel as though they have control over the risk as a result are more prepared. (2) Risk reduction: An individual or group of people who are very aware of the risk but feel badly prepared. These people will demand risk reduction and considers the risk imposed upon them. This group will often assume the responsibility for preparing the population for a hazard lays in the hands of authorities, instead of taking individual action. (3) Safety: an individual or group of people who believe that the risk is acceptable or small, therefore feels safe and will not be prepared for a risk. (4) Ignorance: an ignorant individual is not aware of and will therefore not be prepared for the risk.

Photovoice is a Community-based Participatory Research method (CBPR) (Sutton-Brown, 2014; Hergenrather et al., 2009; Freedman et al., 2014) – a style of social research whereby change is sought through respectful engagement with communities allowing for the empowerment of participants (Wilson, 2019). Photovoice is a qualitative method in which participants use photographs and stories about the photographs to share memories and provide detailed information on lived experiences (Nykiforuk et al., 2011; Börner et al., 2015). This method allows participants to visually express ideas that are hard to express in words – this provides a nuanced insight into community issues and is thus a tool, that helps disaster studies to be informed by local realities. Nykiforuk et al. (2011) highlight the major advantage of using photovoice in social research – its ability to benefit numerous stakeholders such as the participants, the researcher, the community and decision/policy makers. The method allows for active engagement and emphasises the participant’s roles as indispensable members of the research team. Börner et al. (2015) used a photovoice style method to explore perceptions of environmental health risks, finding that this method is well suited for studies analysing these kinds of risks and risk perceptions by allowing for the capture of complex information as well as overcoming language barriers. However, obstacles such as access to equipment and technical ability of participants and researchers must be accounted for. Using the photovoice method of community-based participatory research promotes bottom-led, top-fed research, meaning the researcher is merely facilitating the project and the issues and interventions are identified by the people whom the project will affect. As the participants are free to discuss anything they like, if there is a community issue which has not been anticipated by the researchers then it will be noted and discussed. The result of the photovoice method is an innovative solution to a community issue on a local level (Nykiforuk et al., 2011).

2. Methods, data collection and analysis

The primary method used was photovoice, with a case study approach used to carry out several semi-structured in-person interviews with members of two study locations in the Westfjords, Flateyri and Patreksfjörður. In addition, a “Strengths, Weaknesses, Opportunities and Threats” analysis was applied to evaluate the effectiveness of the photovoice method.

2.1 Case study locations

The study locations were chosen based on specific characteristics associated with their remoteness which creates both challenges and advantages when dealing with natural hazards. Challenges include coping with the hazards while isolated and with a lack of resources, and advantages include the historical experience of dealing with hazards in such an independent way (Kokorsch and Gisladottir, 2023). Current risk and historic avalanche events were also factors. The northwest of Iceland, due to its topography and climate, is prone to avalanches – 65% of people living in this region are threatened by debris flow and avalanches. In the 20th century in Iceland, 166 people died as a result of avalanches with 90 of these deaths occurring in the Westfjords (Decaulne, 2007).

Patreksfjörður is the largest community in the southern Westfjords, with 840 inhabitants (Statistics Iceland, 2022). The community’s primary industry is fishing, which includes commercial fishing, aquaculture and tourism. In the winter months, the community is at risk of avalanches and accessibility becomes an issue when the roads periodically close for several days at a time. Avalanche modelling carried out by Tracy and Jóhannesson (2003) found a total of nine different avalanche starting zones in the mountain above the inhabited area of Patreksfjörður. Furthermore, in January 1983, the community was hit by two slush avalanches which caused flooding and damage to sheep sheds, fences and 16 houses. Six people were injured and four people lost their lives (Fjeldsted, 2020). Since then, avalanche defence barriers have been constructed above the village. In January 2023 another “slushflood” occurred in Patreksfjörður, although no one was injured the event stood as a reminder there is still avalanche danger to the community.

Flateyri is a community of 257 people located in the northern part of the Westfjords (Statistics Iceland, 2022) and is at severe risk from avalanches (Sigurðsson, 2004). The community relies mainly on fishing, aquaculture and tourism industries. Flateyri also has a folk school which attracts young Icelandic and foreign students. The main services and inhabited areas are located towards the top of the spit, however, some residential areas as well as the harbour are located within avalanche risk zones (IMO, 2020). In October 1995 an avalanche hit Flateyri, destroying 22 homes, burying 45 people under the snow and 20 people were killed (Finnsdottir and Elklit, 2002). As the weather was particularly bad on that day and given the remote nature of the community, emergency assistance from neighbouring communities was delayed and restricted. Since then, major avalanche barriers have been constructed above the community. Nonetheless, in January 2020, two large avalanches hit the community in one evening. Although most of the snow was deflected from the village by the barrier, a mass of snow overflowed the defence wall causing one teenage girl to be rescued from her house which was filled with snow. There was also significant damage to the harbour, including several boats. In January 2023 another avalanche fell from another mountain just outside the village, which stopped about 40 meters above the road.

Appendix shows avalanche hazard mapping of Patreksfjörður and Flateyri created by the Icelandic Metrological Office (2003, 2020). The maps show the risk zones in the communities which are based on the yearly probability of an individual being killed by an avalanche (Ágústsson et al., 2003). In both communities there is infrastructure, services and residential housing located within Zone C, which contain the area’s most at risk. In Patreksfjörður this includes several houses, the harbour and the hotel. In Flateyri this includes several houses as well as the harbour.

2.2 Step-by-step of the photovoice method

The photovoice method can have many different approaches (Hergenrather et al., 2009). For the purposes of this study, a step-by-step photovoice method was adapted from Sutton-Brown (2014).

Firstly, a group of participants was recruited. The desired sample selection was members of the community including those who have lived in the community for a long period of time, newcomers to the community, those with a foreign background, young people, older people and all genders. The location of the residence of the participant within the community was also important, particularly those living within a hazard zone. Participants who were members of volunteer organizations that have a role in disaster response in the communities, such as the Red Cross and Search and Rescue team, were also desired.

Local personnel were selected through personal contacts, suggested contacts and through snowball sampling. Two people of foreign backgrounds were specifically approached by the researcher and asked to participate. The perspective of foreigners was important to include as in the Westfjords 22.3% of the population are immigrants or second-generation immigrants, which is the second highest proportion of immigrants in the country (Statistics Iceland, 2022). The exact number of interviewees for the study was seven participants in Patreksfjörður and six in Flateyri. The participants included both men and women of different ages. While under 18-year-olds were excluded intentionally for ethical reasons, the participants were mostly middle-aged and Icelandic-born. The participants interviewed were of a similar demographic which does not represent accurately the populations of either community.

A brief description of the photovoice method was provided to the participants which explained why it was being used in the context of this research. Instructions for participants regarding the photographs were also provided. These instructions were communicated to participants ahead of interviews via phone call or email and offered participants a chance to ask any questions. The requirements for the photographs were as flexible as possible. It was requested that each participant brings with them 3 photographs, one per theme. The photographs should depict the themes as best they can, but in any way they want. The themes for the photographs were as follows:

• (1)

  What Flateyri/Patreksfjörður means to you.

• (2)

  The meaning of avalanche risk to Flateyri/Patreksfjörður.

• (3)

  A picture of something that reflects the future of Flateyri/Patreksfjörður for you.

The photographs could be from the past or present and could be presented in digital or paper forms. The reason for not supplying participants with a camera, that is usually done using the photovoice method, and allowing them to bring previously taken photographs, were of practical as well as a financial nature. It has also been done in other studies, such as Burles and Thomas (2014). As we found it important to allow participants to share photographs of their lived experiences in communities living with avalanche risk, we wanted to create the space for them to bring photographs of past events.

The above themes were chosen to gather information regarding the sense of place and attachment to place, as well as perspectives on avalanche risk, risk perception and resilience. Participants were not involved in generating the guiding questions, but the questions were informed after focus groups had been conducted in both communities on climate change adaptation. The third category concerning the future was chosen as a chance for participants to bring up thoughts on resilience or the issue of climate change – if they chose to do so. The term climate change was not directly referred to, to leave the category as open to interpretation as possible. Depending on the consent obtained, the digital photographs were sent to the researcher by email and the paper photographs were scanned onto the researcher’s phone using a photo scanning app. A consent form was created which included options to consent to the interview, audio recordings, the results being shared in a publicly available dissemination. In addition, ethical approval was obtained for this study in June 2022 by the thesis advisory board of the University Centre of the Westfjords.

The participants were given the instructions and photograph categories in advance of the interview taking place to provide appropriate time to find or take photographs. It should be noted however, that it was not always possible to provide instructions more than a day in advance as sometimes participants confirmed their participation in the research and attendance on the same day as the interview was conducted. This step is important to note when undertaking a photovoice approach, as traditional interviews often do not require any preparations. This is added work for the participant who are volunteering their time for an interview and could act as a deterrent, and the need to prepare photographs in advance was identified as a barrier for participation. Three potential participants declined to partake in the photovoice part of the interviews, and one respondent did not have time to find photographs and instead described what they might have brought.

2.3 Interviews

A semi-structured format was chosen for the interviews and they took place from September to December 2022. The interview had two parts, part one gathered information about the participant’s sense of belonging in the community and included such things as how long the participant had lived in the community and the positive and negative aspects of living in the community. In this part of the interview, participants were asked to mark on a paper map provided, their home or place of work and the places or services they value most in the community. The map added a visual reference point for the interviews. The second part of the interview was the photovoice section. Participants were asked to show the photo they chose for each of the categories. Many photo voice studies follow a pre-defined set of question, as outlined in Hergenrather et al. (2009), however in this study, the participant was free to talk about the photo as they wished. This ensured that the interview was participant-led and prioritised participant perspectives (Nykiforuk et al., 2011). If needed, the researcher added prompting questions such as when the photo was taken, where the photo was taken, who took the photo and why. The participants once again used the paper map to place a sticker where each of the photographs were taken. Each part of the interview took approximately 20 min.

2.4 Methods of analysis

It has been ascertained that taking the photos and narratives together is imperative to avoid misinterpretations, with the essence of photovoice being prioritising the participant’s perspective (Nykiforuk et al., 2011). As using photovoice in studies regarding place attachment is a relatively under researched field, the method of analysis used was created for this study. To characterise how the communities of Patreksfjörður and Flateyri perceive natural hazards – using photovoice, the interviews and photographs were coded using Raaijmakers et al. (2008) characteristics of risk perception. Therefore, the top-level coded segments were Control, risk reduction, safety and ignorance. MAXQDA software was used to create the coding system, identify overlapping themes and avoid duplicate themes. A “photomap” was created for both Patreksfjörður and Flateyri showing the photographs that respondents brought relating to avalanche risk and where the photographs were taken.

To examine how place attachment shapes community resiliency in Patreksfjörður and Flateyri a qualitative approach was taken to study the impact of place attachment on community resilience. A framework of analysis was adapted from Kokorsch and Gisladottir (2023). Firstly, the aspects of place attachment which exist within the communities, if any, are identified. This includes the different aspects of the place (social, physical and functional) as well as the different bonds to the place (affective, cognitive and practical) that appear in both the photographs and the narratives.

To examine the extent to which photovoice be useful to shed light on place attachment in communities facing risk of natural hazards, the successes and short comings of the photovoice method were assessed based on author’s own reflections, structured through an evaluation of the method’s Strengths, Weaknesses, Opportunities, Threats (SWOT analysis). Recommendations for future uses of photovoice in place attachment and risk perception studies were given. The method for reviewing the usefulness of photovoice was largely based on the researcher’s own observations made while carrying out the photovoice method.

3. Results

3.1 Communities perception of natural hazards

Themes identified during photovoice interviews included Control: focused on preparedness, community action and Red Cross as well as Search and Rescue involvement. Both communities demonstrated control characteristics, showcasing preparedness in personal and community contexts, behavioural adjustments in response to risks and active involvement in Red Cross or Search and Rescue groups. Risk Reduction: which included acknowledgement of risk without a sense of control. Participants discussed barriers/defence structures, as well as damage and destruction caused by previous natural hazards. The barriers and defence were discussed frequently by participants from Patreksfjörður as barriers are currently being constructed above the community. Participants noted the trust you must have in the barriers as well as the people that built them. Other participants spoke about the resistance to the walls when they were initially being built. Damage and destruction were commonly discussed by Flateyri participants due to recent avalanche events in the community. Safety (downplay of risk): some participants in Flateyri downplayed risks, feeling secure despite past incidents. No instances of ignorance were noted although this characteristic could exist beyond the study’s scope.

In addition, Photomaps were created (Figures 1 and 2) showing photographs from Patreksfjörður and Flateyri depicted varying perspectives on avalanche risk: Patreksfjörður showcased defence structures, while Flateyri highlighted damage and darker visual narratives, reflecting a greater perceived danger.

The difference in perspectives between the two communities could be in part influenced by the more recent and more deadly avalanche events in Flateyri, showcasing the lack of infrastructure in the village that proved to be a serious problem during the 2020 avalanche, as well as other issues including poor transport links and the vulnerability of the harbour.

3.2 Place attachment and community resilience

The results revealed strong place attachment within the communities, emphasizing shared experiences and connections. In Patreksfjörður, social aspects dominated, evidenced by the research participants active roles in community organizations like the Red Cross and Search and Rescue. Proximity maintenance, that is the desire to remain or return to the place in which you are attached, seen through long-term residence and emotional connections to homes, reinforced these social bonds. Functional ties were also apparent in Patreksfjörður, with discussions revolving around the harbour, fishing grounds and aquaculture’s impact on employment. Participants in Flateyri highlighted the emerging creative industry as a focal point in the community as well as the importance of community events and the local pub. In both communities, physical aspects including the surrounding nature demonstrated strong practical and affective bonds.

Both positive and negative impacts of place attachment on community resilience emerged. Negative aspects included self-protective behaviour, spatial optimistic bias, reluctance to change and underestimation of risk. Those aspects are captured in replies by respondents, such as “I have the bad habit of seeing the positive things. I forget the negative” and “Everywhere there is something, like the earthquakes in Grindavík”, when asked questions regarding risk perception. Positive impacts identified included place-protective behaviour, pro-environmental behaviour, local engagement and high-risk awareness, exemplified in responses such as “I think it also means maybe a little bit of restrictions, because even when you are not thinking about it [avalanche risk] every day, you have to keep it in the back of your head, and you don’t do certain things”.

Quinn et al. (2018) suggest that identifying different place attachments can improve our understanding of how communities respond to risk. This research, in agreement, found that analysing community resiliency through the lens of place attachment deepened understanding of how these communities respond to risk, currently and in the future.

3.3 Assessment of photovoice method

To evaluate how effective the photovoice method is for investigating the effect of place attachment on community resilience to natural hazards, we carried out a SWOT analysis of the application of the method as it was used in this research.

3.3.1 Strengths

Participant comfort and enjoyment: it was noted that participants seemed to enjoy sharing their photographs and telling the stories that went with them. This was evident in the change in body language as participants began talking about the photographs and the accompanying memories. Participant enjoyment was also demonstrated in four interviews when participants initially presented three photographs and as the interview progressed, they thought of new photos they would like to share, and produced several more photos then was required.

Basing information on lived experiences, concerns and perceptions: a fundamental strength of the photovoice method is being able to base any findings in the experiences of people who live in the study areas. Problems, perceptions and ideas are identified by the participants and not by the researcher. After exposing and narrating the photographs he brought, one participant concluded “That is me, these pictures are me” which captures nicely the ability of this method to portray personal aspects of the participants lives.

Facilitated open dialog: the photograph element of the interview was effective in facilitating open dialog as participants often spoke freely about the photograph without the need for continuous questioning from the researcher. This was also useful for gathering personal information about sensitive subjects such as avalanche danger without asking probing questions.

Flexibility of application of method: when reviewing photovoice literature it was apparent that there are different ways to apply the photovoice method. Studies such as Börner et al. (2015) and Sutton-Brown (2014) gave cameras to participants and allowed them a time slot to take the photos. In this study participants were allowed to use photographs they already had. This flexibility enables the method to be used in a wide range of studies with a range of different timeframes and budgets.

3.3.2 Weaknesses

Additional effort required, preparation: this method requires added work of finding photographs ahead of time. As the participants can be busy and volunteering their own time, this added effort could deter those with limited free time from participating in the research.

Difficulty in capturing complicated themes in photographs: place attachment in relation to natural hazard risk and risk perception is a complex issue with many layers. The themes that emerge can be hard to capture in photos. We found that the photo narratives mitigated this weakness only to some extent.

Highly personal aspect of photographs: photographs can be highly personal and studies like this one regarding place attachment and avalanches, which affect participants and their families or loved ones, can make it difficult for participants to show photographs. This is especially true for researchers who do not have a prior relationship with participants.

No uniform method of data analyses: as photovoice data can be analysed in multiple different ways, the results of photovoice studies may not be comparable or replicable. In addition, this makes the method of analysis subjective and open to bias.

Presenting the results accurately: as photographs can be interpreted many ways, it is imperative to present the results as accurately as possible. One way to ensure this is to involve participants at all stages of the research including the presentation of results. In addition, post interview surveys can be used to verify whether the data taken from the interviews was interpreted correctly, to gain better insights into the experience of the participant and understand how it could have been improved (Simmons, 2022).

3.3.3 Opportunities

Can overcome language barriers: the photography element of photovoice has the potential to overcome language barriers as photographs are universal. This was evident as all participants (except one) were not native English speakers. Therefore, the researcher was mostly discussing issues with people speaking a different language – bringing the pictures added depth and potentially made it easier for the participants to convey their perceptions and feelings in the interview. Language limitations were mitigated somewhat by facilitating the interviews in both English and Icelandic, but we experienced that the photography element helped to convey feelings, lived experiences and perceptions of participants.

Added methods of analysis: not having a uniform method of analysis also provides opportunities. For example, photovoice allows for different methods of analysis that would not be possible with just text interviews, such as the analysis of colour patterns. In this study as shown on Figure 2 the photographs from Flateyri that related to avalanches were predominantly black and grey with gloomy tones. In contrast, the photographs regarding the future portrayed bright colours-yellows and blues-which instilled a hopeful and happy feeling when viewing them.

Allows for participant creativity: participants were instructed to bring photographs from any source they liked. One participant brought photographs they captured on the way to the interview; another spoke about photographs that were on the walls of her house and from a photo album she made in the 1990s. Another participant used photographs from Facebook and from a book they read. Likewise, Börner et al. (2015) found that participants behaviour suggested that the photography element stimulated their engagement, creativity and enjoyment.

Can capture ideas that are hard to put into words alone: it was noted as a weakness that it can be difficult to capture complicated themes within photographs. However, it can also be a way to capture ideas that are hard to put into words alone. Therefore, we found that combining photographs and narratives together allowed for a comprehensive insight into complex perceptions of participants, as suggested by Börner et al. (2015) and Nykiforuk et al. (2011).

3.3.4 Threats/limitations of method

Accessibility: not everyone has access to photographs or cameras. The issue did not present itself as a problem in this research, which was not surprising, especially since a vast majority of the participants brought photos on their smart phones. Access to technology and the Internet is high in international standards in Iceland (OECD, 2019). However, this can act as a constraint for researchers using this method in communities that either have less access to technology or have had less access or opportunities to photograph their daily lives and experiences using cameras and print the photos.

People not having the technical skill to operate a camera: in addition to the accessibility of photography equipment, having the technical ability to operate the equipment is not a given, and could present issues in the future. To combat this threat, in this study, participants were allowed to bring photographs from any source.

4. Discussion

The results of this study show that there is a future use for photovoice in the field of place attachment and natural hazard studies. Photovoice can be used in combination with other CBPAR style methods to create a multi-disciplinary, multi-faceted approach to place attachment studies whilst amplifying the voices of those affected. Studies using the photovoice method use for example workshops and photo exhibitions in the communities for collective interpretation and dissemination, as highlighted by Liebenberg (2018) and Wang (2003).

The photovoice method can be a valuable tool in mitigating research fatigue by offering a more engaging and participatory approach to data collection as well as fostering creativity by using a novel method. However, researchers using this method, in particular in the field of disaster studies, should be aware of the ethical implications, as suggested by Guillemin and Drew (2010), since visual approaches in research have the capacity to elicit complicated emotions related to traumatic memories. As this research topic was sensitive where individual’s past trauma was reflected in the results, and the participants had been ensured anonymity, ethical reasons guided our reasoning for not including public dissemination such as a photo exhibition. If the time and resources allowed, this study would be followed by photo discussions to allow individual participants to share and discuss the photographs they took, to facilitate critical dialogue about community strengths and concerns. As pointed out by Liebenberg (2018), collective interpretation of the images produced in a photovoice study are important for dissemination, as knowledge can inform action and be a catalyst for change. Therefore, including follow-up interviews with the participants could have contributed to capacity building in the communities on an individual level, having implications for community resilience and disaster risk management.

Overall, we found that photovoice has multiple advantages for disaster studies. It provides a way to engage the participants as they choose and bring the photos, meaning that they get to decide what is discussed. In line with Burles and Thomas’s (2014) and Nikiforos and Karakitsou’s (2020) findings, we discovered that this type of research can lead to a deeper understanding of lived experiences. Therefore, it can provide valuable insights regarding peoples sense of place and risk perception in communities living with a natural hazards. As place attachment has been cited as a possible deciding factor for whether community members choose to stay in a hazardous area or not according to Blondin (2021), Scannell and Gifford (2010) and de Dominicis et al. (2015), studying communities using photovoice has the potential to inform decision makers of their resilience. We argue that this method is particularly suitable for the principles of the “Disaster Studies Manifesto” since it allows for an appropriate articulation of local realities. In addition, it was the participants who shared their views, local priorities and potential critiques. The project in Iceland was an example of research being done by, with and for the local communities, based on the main goal of improving understanding of disaster risk management, benefiting the communities at risk. The method was culturally grounded and tailored to local circumstances.

Considering the high proportion of immigrants in the communities, the lack of input from immigrants in this study, there is a need for a study regarding risk perception and place attachment amongst immigrants in small remote communities in Iceland. This group is at greater risk and there are established benefits of photovoice when studying minority groups (Hergenrather et al., 2009). Therefore, photovoice offers an opportunity to advance understanding of place attachment, risk perception and community resilience. It should be noted however that these results reflect the usefulness of photovoice in an Icelandic context. Literary rates, access to technology including cameras and other socio-economic factors could impact the application of this method.

5. Conclusion

This study investigated community members risk perception and place attachment in two small remote villages in the Westfjords, Patreksfjörður and Flateyri, using a photovoice method to conduct a series of qualitative interviews. The findings indicate that risk perception in relation to avalanche risk is high in both communities. The communities largely feel as though they are prepared for and have control over the risk. In addition, strong place attachment was evident which had both positive and negative impacts on community resiliency. The results also show that despite its limitations photovoice is a useful method for shedding light on risk perception, place attachment and resiliency in communities facing the risk of natural hazards. The study found that place attachment is an important factor to consider in disaster risk management, policy and decision making. Overall, we therefore conclude that photovoice provided a nuanced insight into community issues and fostered a more distinct understanding of risk perception in this case, which has relevance for disaster studies. However, the Icelandic context of the study is likely to have affected the applicability of the method.

Figures

Figure A1

Avalanche hazard zoning map Patreksfjörður

Figure A2

Avalanche hazard zoning map Flateyri

Appendix

Figure 1

Photographs relating to avalanche risk, Patreksfjörður showcasing (a) a recent avalanche, (b) a sunny winter day, (c) the memorial for the victims of the 1983 avalanche, (d) the new avalanche barrier, (e) a “snow angel” on top of the mountain and (f) a snowy day in Patreksfjörður in 1983

Figure 2

Photographs relating to avalanche risk, Flateyri, showcasing (a) a crack in the snow cover, (b and e) the aftermath of the 1995 avalanche, (c) a snowy day, (d) snow during the day of the 2020 avalanche and (f) the aftermath of the 2020 avalanche into the harbour