Disaster Risk Awareness: The Turkish Migrants Living in Northern Italy

In this study, we analysed the socio-demographic characteristics and disaster risk awareness of the Turkish migrants living in northern Italy. We initiated the study with an extensive face-to-face questionnaire with 544 individual respondents. With the help of the questionnaire, we gathered information on the socio-demographic structure of the Turkish community living in the area and the immigrants’ disaster experience, their level of disaster preparedness and disaster risk awareness, and their potential behaviour during an emergency. Additionally, we conducted focus group meetings in Milan, Lecco, Como and Varese with 49 migrants living in the region. In the focus group meetings, we discussed the migrants’ awareness of disasters and potential behaviour patterns during emergencies. We collected the informative booklets and past event reports prepared by civil protection centres and municipalities and used them in focus group meetings to collect participants’ opinions. The results show that the migrant communities’ disaster risk awareness is low, but their capacity to adapt to suddenly changing conditions is higher than presumed.


Introduction
Scholars have long studied the variety of reasons that people migrate from one place to another. Human mobility has a long history and understanding migration patterns has been central to migration studies for decades. The most significant drivers are often structural (e.g., economic development in the countries of origin) [1]. However, there is no simple explanation such as the pull and push factor, and the migration decision depends on complex interactions of many factors [2]. UNHCR [3] states that migrants and receiver countries benefit from migration, and it enriches their current situations. UNHCR [3] also indicates the situations that leave migrants in vulnerable conditions, and "situational vulnerability" is one of them. Migrants' situational vulnerability is higher than that of the inhabitants for many reasons, the primary one being a lack of knowledge of the local risks and language.
Despite the importance of the issue, many studies on migration and disaster risk management are very limited; they mainly focus on non-European countries and revolve around the specific impact of the hazard, e.g., the effect of storms on migrant communities [4,5]. There are a wide range of studies on post-disaster migration [6]. However, there are few on migrants' perception and awareness of risk and preparedness level [7]. The concepts of risk perception and preparedness have often been associated in many disaster studies not necessarily focused on migrants and framed along the lines of "prediction". Namely, risk perception is a predictor of preparedness behaviour [8].
Migrants are considered the "(in)visible victims of disasters" whose unique needs are often overlooked in disaster planning [9]. Unequal access to disaster preparedness resources is coupled with a poor understanding of migrants' risk perceptions, which Around twenty-five years ago, Susan Cutter asked 'are societies more vulnerable to environmental hazards?' and pioneered the first comprehensive social vulnerability index, which generally includes qualitative indicators rather than quantitative [14]. Social vulnerability is mostly described by individual characteristics of people, such as age, race, health, income, type of dwelling unit and employment [15] (p. 243). The other factors that increase the social vulnerability of a community are a lack of access to resources such as information, knowledge and technology, limited access to political power, absence of social capital, beliefs and customs, deficiency of physical environment, individuals with disabilities, and type and density of infrastructure [16][17][18][19] (cited in [20], p. 245).
Findings from qualitative research, in fact, show that people may be simultaneously vulnerable and resilient [20]. This knowledge led to an important conceptual shift, as it challenges ideas of migrants as passive victims to emphasise their potential role as resourceful agents [20] (p. 6). Studies have also highlighted the key role of social capital in resilience, especially in environments with cultural and language barriers [21].
Additionally, with regard to preparedness, it is a well-known fact that people living in hazard-prone areas are often unprepared, and they fail to take precautionary measures to reduce the impact of a disaster [22,23] (cited in [24]). Communication is one of the many reasons for this. How risk communication is conceived by actors involved in disaster risk management can make a difference [25]. As highlighted by the Intergovernmental Panel on Climate Change [26], taking into account the specific needs of different societal groups is key. Communication processes should be as inclusive as possible, meaning that local risk perceptions and the local framing of risks and needs cannot be ignored since, for instance, language skills influence the levels of disaster preparedness [27]. As Fielding pointed out, different people and different locations require additional warnings [26]. Targeting group-specific information based on the heterogeneity of citizens is crucial [28].
Other relevant terms are place attachment [29], sense of community [30] and sense of place [30]. The study by Misshra et al. [29] asked the question "Does place attachment and the consequent emotional connections and ties that people have with environments affect their preparedness for natural disasters, such as floods?" The authors addressed the research question by considering the three attachment types: "economic, genealogical, and religious". The results show that there is a strong correlation between place attachment and flood preparedness. Place attachment can be considered one of the differences between migrants and local inhabitants. Regarding the sense of community, it requires further studies to investigate the impact of "sense of community" [30] and "sense of place" [31] on the preparedness and awareness of migrants.
We started this study with the problem being that the migrants are invisible victims whose unique needs are not included in disaster planning. Another challenge is that they are often labelled as vulnerable, and their capacities are overlooked in disaster risk studies. In this study, we collect information on the socio-demographic characteristics of migrants and their access to resources such as information, knowledge and technology that helps us to understand the social vulnerability of migrants. The results of the study will help decision makers to adjust disaster risk planning by considering the unique needs of the migrants based on their socio-cultural and economic conditions. It is worth considering that it would be misleading to frame the risk awareness and preparedness of migrants only through the concept of "vulnerability" without considering their capacities as well.

Materials and Methods
In the study, we applied various research methods at three different urban scales: regional, community, and household. The study started with a face-to-face questionnaire conducted during the National Parliamentary Elections in May 2015 in Milan. We administered the questionnaire to 544 individuals. The respondents were selected randomly at the entrance of the Consulate General of the Republic of Turkey in Milan by researchers. The questionnaire gathered information on the Turkish community's socio-demographic characteristics, their disaster experience, disaster preparedness, disaster awareness, and their potential behaviour during an emergency. To collect further information and gather in-depth knowledge on the awareness of disaster, we decided to conduct focus group meetings with various socio-cultural groups, including students, expats, religious minorities, and the members of a religious-political movement. Additionally, we conducted literature research regarding past natural hazards in northern Italy to inquire about visual and written resources during focus group meetings. The risk maps and reports that were prepared by the civil protection authorities and municipalities were examined on a regional scale. We collected the informative booklets and past event reports prepared by civil protection authorities and municipalities to analyse details and share them with participants during focus group meetings to learn more about the participants' experiences and opinions.
Due to the exploratory nature of this study, we started with some generalisations about the socio-demographic characteristics of the Turkish communities living in northern Italy. We presumed that most Turkish communities living in Italy are composed of workers in the food and construction sectors and students. The former has rapidly increased in recent years. The Italian education system became an option for students who do not speak Italian with the launch of English graduate programs. A small portion of the first incoming students decided to work or continue their doctoral or post-doctoral training. The rate, which was significantly small in the first few years, continues to increase every year.

The Survey Area
In Italy, the majority of the Turkish population lives in northern Italy; therefore, the study covers nine administrative regions in the service area of the Consulate General of the Republic of Turkey in Milan, located in the north of Italy. These regions are (1) Lombardia, (2) Valle d'Aosta, (3) Liguria, (4) Piemonte, (5) Veneto, (6) Trentino Alto Adige, (7) Emilia Romagna, (8) Marche, (9) Friuli Venezia Giulia ( Figure 1). According to the information received from the Turkish General Consulate at the beginning of the project, in May 2015, approximately 29,000 citizens had been at the consulate for various consular procedures; however, it is not possible to obtain a concrete number of citizens residing in the functional area of the General Consulate. However, this number was estimated to be approximately 30,000 people by the employees of the Consulate General. The net number of Turkish citizens recorded in the voter roll through address declaration to the Consulate General of the Republic of Turkey in Milan was 10,373 (18 years old or older) as of May 2015, when we started the study. The regions that they were living in and the respective resident numbers of the 19 ject, in May 2015, approximately 29,000 citizens had been at the consulate for various consular procedures; however, it is not possible to obtain a concrete number of citizens residing in the functional area of the General Consulate. However, this number was estimated to be approximately 30,000 people by the employees of the Consulate General. The net number of Turkish citizens recorded in the voter roll through address declaration to the Consulate General of the Republic of Turkey in Milan was 10,373 (18 years old or older) as of May 2015, when we started the study. The regions that they were living in and the respective resident numbers of the 19

Comprehensive Questionnaire: The Size of the Sample
There was no information on the socio-demographic status of the overall Turkish community. According to information obtained in May 2015, 10,373 residents out of 19,936 are registered voters in the Consulate General of the Republic of Turkey in Milan. Based on this number, the sample size was calculated as 544 individuals ( Table 1). The number of families living in the region is unknown, so we used the number of registered voters to decide on the size of the sample. While preparing the questionnaire, the questions were designed to understand the socio-demographic characteristics of migrants, their disaster experience, their awareness of disaster risk and their preparedness level. The questions in the last section of the questionnaire were prepared to gather more information about the Turkish community's socio-demographic characteristics, such as the gender, age, educational status, and language skills of the sample who participated in the comprehensive questionnaire study. A set of

Comprehensive Questionnaire: The Size of the Sample
There was no information on the socio-demographic status of the overall Turkish community. According to information obtained in May 2015, 10,373 residents out of 19,936 are registered voters in the Consulate General of the Republic of Turkey in Milan. Based on this number, the sample size was calculated as 544 individuals ( Table 1). The number of families living in the region is unknown, so we used the number of registered voters to decide on the size of the sample. While preparing the questionnaire, the questions were designed to understand the socio-demographic characteristics of migrants, their disaster experience, their awareness of disaster risk and their preparedness level. The questions in the last section of the questionnaire were prepared to gather more information about the Turkish community's socio-demographic characteristics, such as the gender, age, educational status, and language skills of the sample who participated in the comprehensive questionnaire study. A set of questions were designed to understand the citizens' experience of disasters, disaster preparedness and mitigation actions. The third set of questions were designed to learn more about the participants' preferences for communication media and how often they use them. We wanted to select the most used communication tool to raise awareness about the risks of disasters in their regions of residence by sharing information leaflets and video messages.
The classification used in the survey is as follows: For the classifications with "*", the unit of analysis is individual. For "socio-demographic analysis", the unit of analysis is family. The last page of the questionnaire was composed of questions about the socio-demographic characteristics of the family members living in the same house, such as the number of people living in the house, their ages and education levels, and the languages that the family members speak at home to communicate. During the questionnaire, researchers were present at the site to help the individuals to fill in the questionnaires. In addition to questionnaire forms, we prepared visuals to support the respondents in understanding the questions. Five hundred and forty-four individuals filled out the questionnaires, and 525 families were represented in the study. When the individual disaster experience was different, we let more than one family member fill out the questionnaire. We stapled the questionnaires together when multiple family members filled out the questionnaire, and only one family member filled out the last page. When we calculated the number of family members, we found that we had the socio-demographic data of 1785 individuals.

Focus Group Meetings
We conducted focus group meetings in four different locations (Milan, Lecco, Como and Varese), considering the location and diversity of the Turkish migrants. In total, 49 migrants attended the focus group meetings (for details, please see Appendix A). We decided the contents and locations of the focus group meetings considering the results of the questionnaires. The purpose of the focus group meetings was not to compare the results with the questionnaire, but to gain in-depth knowledge on the awareness, needs, feelings, beliefs, behaviour patterns in a possible emergency, and priorities of various groups. At the beginning of each meeting, the primary investigator (PI) welcomed the participants, introduced herself and the project and the setting of the focus group meetings, mentioned the rights of the participants, and participants signed the consent forms that included information on the study and the participants' rights. Then, the participants were asked to introduce themselves. During focus group meetings, participants were asked ten questions. The meeting began by asking the participants to define "what is a disaster according to them". After these ten questions, participants were asked if they wanted to share anything else or whether they had questions for the researcher or not.
The focus group questions are as follows: • What is a disaster? Please specify. • Do you think that an environmental disaster, such as a flood or earthquake, will happen to you? During the focus group meetings, various hazard maps, disaster photos, and newspaper columns regarding the past events were shared with the participants. The severity and probability of the reoccurrence of these events were discussed to a large extent. In this way, we aimed to attract the participants' attention and carry out a more collaborative and interactive discussion. Furthermore, they had been encouraged to enhance their resilience in disaster risk management. Two MSc students assisted the PI during the focus group meetings. The focus group meetings were as follows; for more details, please consult Appendix A.

Ethical Considerations and Data Management
Ethical aspects were at the centre of our study. We obtained the necessary permissions from the General Consulate of the Turkish Republic in Milan and the Turkish Republic Supreme Election Council to conduct the questionnaires during the parliamentary elections. We ensured honesty and transparency towards research subjects involved in several stages of the study, such as face-to-face questionnaires and focus group meetings. Participants voluntarily engaged in the study, and they were given the project's informed consent form and detailed information sheets in advance. The consent forms were in Turkish. We had two participants who required translation of the documents to Italian, and we translated all the documents for them to Italian. The consent form explicitly stated that participation is voluntary. Anyone has the right to refuse to participate and to withdraw their participation, samples or data at any time without any consequences. Participants gave their consent by signing a separate form from the questionnaire, as the questionnaires were anonymous. We did not collect more data than were necessary to reach the research goal. All data were handled in a manner that respected the rights specified in the agreements (informed consent and transfer of intellectual property).

Results of the Questionnaire
The results of the questionnaire set out the socio-demographic characteristics of the Turkish migrants living in northern Italy. First, the majority of the participants (60% men and 40% women; sample size 544 individuals) reside in the Lombardy (Milan, Como, Lecco, and Varese), Emilia Romagna (Modena and Bologna) and Liguria (Imperia, Turin) regions ( Figure 2). As for the age group of the participants, the highest number of participants was in the 25-34 age group with 38%, and this was followed by the 35-44 age group, with 29% ( Figure 3). As for the employment status, 53% of the participants were employed full-time, 11% were students, and 17% were housewives, of which 95% came to Italy due to marriage (Figure 4).  • Socio-demographic characteristics As for the age group of the participants, the highest number of participants was in the 25-34 age group with 38%, and this was followed by the 35-44 age group, with 29% ( Figure 3). As for the employment status, 53% of the participants were employed full-time, 11% were students, and 17% were housewives, of which 95% came to Italy due to marriage (Figure 4).

•
Socio-demographic characteristics As for the age group of the participants, the highest number of participants was i the 25-34 age group with 38%, and this was followed by the 35-44 age group, with 29% ( Figure 3). As for the employment status, 53% of the participants were employed full-time 11% were students, and 17% were housewives, of which 95% came to Italy due to marriag (Figure 4).  Regarding the educational status, the majority were high school graduates, with 30% followed by elementary school graduates, with 27%. Some participants had never been t elementary school or had left elementary school. Among them, we encountered two illit erate women (sample size: 544 individuals).
The participants were asked questions to comprehend their level of linguistic skills It was observed that all of the participants can communicate in Italian to various extents Overall, 17% expressed their capability to handle daily tasks with the level of Italian tha they speak, whereas 45% of the participants were confirmed to have a good understand ing of the Italian language. Two participants were observed as being as non-Turkis speakers during the questionnaire. More than 40% of the participants can speak one mor European language in addition to Turkish and Italian. The majority indicated English a  Regarding the educational status, the majority were high school graduates, with 30%, followed by elementary school graduates, with 27%. Some participants had never been to elementary school or had left elementary school. Among them, we encountered two illiterate women (sample size: 544 individuals).
The participants were asked questions to comprehend their level of linguistic skills. It was observed that all of the participants can communicate in Italian to various extents. Overall, 17% expressed their capability to handle daily tasks with the level of Italian that they speak, whereas 45% of the participants were confirmed to have a good understanding of the Italian language. Two participants were observed as being as non-Turkish speakers during the questionnaire. More than 40% of the participants can speak one more European language in addition to Turkish and Italian. The majority indicated English as the most widely spoken language among them. French, German and Spanish followed English in this classification (sample size: 544 individuals).
Furthermore, 79% of the participants declared that they speak another language, such as Kurdish or dialect, apart from Turkish, Italian and another European language (sample size: 544 individuals). The participants stated that they speak Turkish, Italian and Kurdish sequentially in their homes. They strongly support the idea of multilingualism by bringing up multi-lingual children who can speak Turkish, Italian, Kurdish and at least one other European language ( Figure 5). • Disaster experience (sample size: 544 individuals) Regarding disaster experience, 31% of the participants confirmed that they had experienced earthquakes and 4% had experienced floods in Italy to ranging extents. In all, 3% of the participants stated that they experienced both disasters ( Figure 6). In particular, the participants from Modena and Milan had incurred monetary and property losses due to earthquake and flood disasters, respectively. One family mentioned that they did not ask for funding from the Italian government as they were not aware of such a mechanism. One person from Modena declared that many families living in Modena returned to Turkey after the occurrence of the Modena Earthquake in 2012.  Anxiety about natural hazards was identified in 63% of the participants. While 11% of the participants declared "excessive anxiety", 52% of them expressed "anxiety" in characterising their level of concern against disasters.

•
Disaster preparedness (sample size: 544 individuals) The majority of the participants were opposed to being self-prepared for disasters, propounding the lack of self-preparedness in Italian society. Even if the participants were quite conscious of the drawbacks of unpreparedness, surprisingly, the overwhelming majority were reluctant to take preventive actions. The participants who had been exposed to disasters in Turkey were perceived as being more susceptible and more predisposed towards the behaviour of "preparedness".
Only 23% of the participants informed their children about how to act during a natural hazard. Overall, 71% declared their ignorance about how to use "Fire Extinguisher" equipment. Meanwhile, 92% of the participants stated being self-conscious to switch on/off the gas, electricity, and water valves. In all, 87% expressed that they keep their important documents such as passports, insurance and deed papers in somewhat safe places. Overall, 83% of the participants admitted not having an "Emergency Kit" in their home, while 17% do have a "First Aid Kit". More than 50% of those maintaining a First Aid Kit confessed their ignorance in keeping the necessary medical supplies up to date.

•
Potential behaviour of the respondents during disasters (sample size: 544 individuals) Overall, 83% of the respondents declared not having planned where to reunite in case of an emergency. As a response to the "Where would you prefer going if you were supposed to leave Italy in case of a disaster?" question, while 64% of the participants indicated "Turkey", the remaining 36% answered "other cities of Italy or Europe" based on the relocation of their extended families.

•
Disaster awareness (sample size: 544 individuals) We presented seven disaster scenarios, including earthquake, flood, drought, snowstorm, pandemic, climate change and fire, to the participants. They were asked to classify them from the most probable (1) to the least (7). Participants declared "flood" as the most likely disaster to occur and "drought" as the least likely one in categorising the disasters for the area of interest.

•
The most used communication media tool With the help of this study, we wanted to raise the awareness of Turkish citizens about the risks of natural hazards in their vicinity and enhance their resilience in disaster risk management. For this reason, we asked participants a couple of questions to better understand the most common means of communication to convey "awareness-raising" messages. The responses of the participants showed that not everyone has a smartphone and continuous internet connection. The best means of communication was found to be "SMS" to deliver messages. The participants were asked which social media networks they use the most. More than 400 participants declared having a Facebook account and using it actively in their everyday lives. Therefore, a Facebook account was activated to inform the participants about the recent developments on the topic. We kept the Facebook account active for three years.
During the questionnaire, special topics on which the participants lacked sufficient information in disaster management were revealed, and informative leaflets were prepared to provide accurate information regarding these topics. The leaflets were distributed to the public in the General Consulate of Turkey in Milan. In addition to that, the researchers are currently in collaboration with "Search and Rescue Association" (AKUT) in Istanbul, Turkey, to provide the most accurate responses to the questions such as, "What is a family disaster plan? What are the essential components of an Emergency Kit? How to act during a flood?". The responses obtained from AKUT Team experts are published periodically as a series of videos via the project's Facebook page.

The Results of the Focus Group Meetings
The questionnaire revealed the spatial dispersion of the participants in northern Italy. Most of the population had been identified as settling down in Lombardy (Milan, Lecco, Como and Varese), Emilia-Romagna (Modena and Bologna) and Liguria regions (Imperia). Therefore, we decided to conduct focus group meetings in the Lombardy Region.
During the focus group meetings in Milano, Como, Lecco and Varese, all participants actively participated in the group discussion. We started each focus group meeting with the question of what a disaster is. The generally agreed on definitions are "loss of property", "loss of life", "material loss or damage", and the need for evacuation. During the focus group with women, they defined the disaster as "migration itself is a disaster" and "being prone to Islamophobia" ( Table 2). Table 2. Summary of focus group meetings.

Questions Answers and Reactions
What is a disaster? Please specify. Loss of property, loss of life, material loss or damage, the need for an evacuation, being a migrant, Islamophobia Do you think that an environmental disaster will happen to you?
Most of them said no. There was a difference between the ones who had already experienced an earthquake or flood event and the ones who had never experienced one During an emergency/disaster, how do you reach the information that you need?
Asking my neighbour/friend and family member or calling 112. Participants were not aware of any of the information websites that we shared with them (Information resources covering the regions that they are living were shown to the participants in terms of visual materials and was asked) Do you know these sources of information?
All of them said no How would you be aware of these resources? How do they reach you? Social media (Facebook) and SMS What can you do to protect yourself and your family from a disaster?
The participants in the Varese focus group meeting were very well prepared If you experience a disaster/catastrophe, will you go back to Turkey?
The participants discussed this question, and their final answer was a "yes".
Most of the participants had experienced an earthquake or a flood event in Turkey or Italy. Most of the participants stated that a disaster could happen at any moment; some had a fatalistic approach. The participants in Varese experienced the 1999 Izmit earthquake in Turkey, and one of them was in the earthquake's epicentre. They were still feeling the impact of the event. This group's awareness level was the highest, and they conducted several emergency drills at their home with their children.
It was clear that the priorities during an emergency and reactions to the situation change according to gender, age, and family presence. The first reaction of women was bringing the family together; the first reaction of men was to understand what's happening and the extent of the disaster. On the other hand, all students said that the first thing they would do is reach out for their passports and cash.
Most of the women are dependent on their husbands and do not speak Italian. This linguistic incapability creates a barrier for adaptation, isolates them from local society and increases their vulnerability. They seek word of mouth information and communicate with their neighbours or friends who speak the same language. The focus of mothers is their children. They told us that, first, they would seek their children, and after finding them, they would call their husbands for help.
On the other hand, migrants are tightly connected. Their social network is the main resource, especially those isolated due to the language barrier. However, it is still not possible to conclude that the strong sense of community provides resources that make them resilient in the long run, as in some cases, being isolated might be a barrier to reaching out for essential information and resources.

Discussion
The "City, Migration and Disaster" study was set out to explore the environmental disaster risk awareness of the Turkish community living in the region, as well as their socio-demographic structure. Indirectly, in practice, the study raised awareness of the Turkish migrants and referred them to sources to increase their knowledge and awareness of disaster risk.
Similar to findings from other research, our study confirms a lack of preparedness and a more general lack of interest in preparedness actions. This is aligned with the "invisible" framing arguments [9,10] and can be related to a lack of involvement in disaster decision making processes. This causes a low level of awareness despite the participants living in Italy, which is a country prone to natural disasters, for 10-20 years. Notwithstanding low interest in preparedness, the level of risk awareness with regard to natural hazards seems quite high, since floods and earthquakes were deemed as the most probable risks.
Additionally, our study confirms the role of past experiences in disasters, since the participants who were exposed to disasters in Turkey are perceived being as more susceptible and more predisposed towards the behaviour of "preparedness". However, as shown by Becker et al. [32], the experience-preparedness relationship is a complex one, and may differ in relation to hazards and the socio-economic status of the person. Nonetheless, the importance of past experiences cannot be underestimated, as it is a determinant of future actions and of reslience as well [33][34][35]. The importance of having a multi-faceted approach was also confirmed, as gender and cultural differences seemed to emerge: for instance, concerning gender differences and priorities during the response phase. Moreover, as stated in the introduction, such an approach would be a pre-condition for ensuring proper inclusion in disaster risk reduction (DRR). This seems to be corroborated (even if indirectly) by answers to questions about disasters in general. For "what is a disaster", most of the respondents did not mention a specific hazard but rather referred to, e.g., "migration", "Islamophobia". DDR policies should take into account socio-cultural differences in perceiving disasters.
In line with the findings of this study, we do not seek to label migrants as 'vulnerable', as they have unique capacities that could increase their resilience. The results confirm the studies of Fussell et al. [12] and Guadagno et al. [21], proving that social networks can be a resource in a disaster for migrants. If social capital is important, social competence is also crucial to enhance resilience. In the focus groups, social compentence emerged in relation to the priorities and needs during an emergency, since male participants would rightly call 112. The discussion in the focus group meetings was in line with the Uekusa and Mattheweman [20] study that stated that struggling with the existing inequalities in their daily lives makes migrants resilient. Overall, the results of our study show that there is a high potential for resilience that seems to emerge through some key resilience dimensions that vary from prior experiences with disasters to social capital and competence.
It is also possible to relate the findings of this study with coping mechanisms for trauma. During the focus group meetings, we observed that participants tend to make decisions based on their previous experiences and having a family or not. Additionally, not being attached to the place provides them with the freedom to move in the case of a disaster, but being a part of a close-knit community is one of the main mechanisms to cope with disasters such as floods and earthquakes. More studies can be conducted to relate the findings further with the "Nudge Theory" to improve the resilience of migrants [36][37][38]. The findings from our study suggest that understanding cultural barriers is key for disaster preparedness. Without proper linguistic skills, it is impossible to ensure disaster preparation across all phases of the disaster cycle (mitigation, preparedness, response and recovery).

Conclusions
In this study, we investigated the socio-demographic characteristics, risk awareness, participation in development, prevention and mitigation strategies, education programs, capacity to invest in mitigation, access to flood information and training/experience of the population, perception and awareness of risk condition, awareness of education programs, individual preparation, and understanding of the ways to access flood information among the Turkish migrants living in the area.
In this study, our target group was legal migrants older than 18 years old. However, some marginal groups might be more vulnerable than our samples, such as illegal migrants and close-knit communities that we could not reach out to to conduct focus group meetings. We completed 544 questionnaires with respondents living in nine regions in northern Italy. We limited the geographical focus to the Lombardy region during the focus group meetings because of the high number of Turkish migrants living in the area. The survey was conducted in 2015; the results presented here might be considered "old", but the results can inform future studies on the Turkish community in other European countries. Researchers may benefit from the methodological approach and the findings. For instance, policymakers may be interested in understanding socio-cultural dimensions that should not be overlooked in DRR processes and policies. Moreover, the results are aligned with the findings of previous studies in other countries.
The project succeeded in drawing great attention from both the affiliated institutions and the public, who voluntarily and actively participated in each project stage. It was carried out mainly in the Lombardy region due to limited time and resources. Nevertheless, the project has the opportunity to be extended to cities such as Modena and Imperia, where a remarkable Turkish population that needs to be informed about earthquake and flood risks in their area of settlement is present. Moreover, the project offers insights for further research on the Turkish communities in other European countries. The recent flood events in July 2021 in Belgium, the Netherlands and Austria proved the importance of conducting such studies in hazard-prone areas with a large number of migrants.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. The consent forms were in Turkish. We had two participants who required translation of the documents to Italian, and we translated all the documents for them to Italian. The consent form explicitly stated that participation is voluntary. Anyone has the right to refuse to participate and to withdraw their participation, samples or data at any time without any consequences. Participants gave their consent by signing a separate form from the questionnaire, as the questionnaires were anonymous.

Data Availability Statement:
We did not collect more data than were necessary to reach the research goal. All data were handled in a manner that respected the rights specified in the agreements (informed consent and transfer of intellectual property). Collected data is anonymized and not open source. The data report was prepared in Turkish and submitted to the funding body at the end of the project.

Acknowledgments:
The authors extend their greatest appreciation to the Republic of Turkey Ministry of Culture and Tourism Presidency for Turks Abroad and Related Communities, the General Consulate of the Turkish Republic in Milan and the Turkish Republic Supreme Election Council for the permission and support during the survey, the migrant community who participated in questionnaires and focus group meetings, and AKUT Search and Rescue Association for their informative videos on various hazards and emergencies. The authors would like to extend special thanks to the four MSc Students who supported the research during the fieldwork. Berrak Balcı and Onur Sagır supported the study by conducting the face-to-face questionnaires. We would like to extend our gratitude and thanks to Burcu Koçoglu and Zehra Irem Turksezer for their dedicated support. They conducted the face-to-face questionnaires, took notes during the focus group meetings, and transcribed the recordings after the meetings. In addition, we thank them for their effort to translate the PI's final report from Turkish to English. We included some parts of the translated document in the manuscript. The authors thank Scira Menoni for her support during the "City, Disaster and Migration" Postdoc project. We thank the two reviewers for their valuable comments on the first version of this article.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the study's design, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results. The content is the authors' sole responsibility, and the views expressed here are of the authors and not the funding institution.