Risk and Resilience: A Case of Perception versus Reality in Flood Management

: Canada’s vast regions are reacting to climate change in uncertain ways. Understanding of local disaster risks and knowledge of underlying causes for negative impacts of disasters are critical factors to working toward a resilient environment across the social, economic, and the built sectors. Historically, ﬂoods have caused more economical and social damage around the world than other types of natural hazards. Since the 1900s, the most frequent hazards in Canada have been ﬂoods, wildﬁre, drought, and extreme cold, in terms of economic damage. The recent ﬂood events in the Canadian provinces of Ontario, New Brunswick, Quebec, Alberta, and Manitoba have raised compelling concerns. These include should communities be educated with useful knowledge on hazard risk and resilience so they would be interested in the discussion on the vital role they can play in building resilience in their communities. Increasing awareness that perceived risk can be very di ﬀ erent from the real threat is the motivation behind this study. The main objectives of this study include identifying and quantifying the gap between people’s perception of exposure and susceptibility to the risk and a lack of coping capacity and objective assessment of risk and resilience, as well as estimating an integrated measure of disaster resilience in a community. The proposed method has been applied to ﬂoods as an example, using actual data on the geomorphology of the study area, including terrain and low lying regions. It is hoped that the study will encourage a broader debate if a uniﬁed strategy for disaster resilience would be feasible and beneﬁcial in Canada.


Introduction
The impacts of geo-hydrologic hazards in the last two decades of the 20th century were felt by three-quarters of the population worldwide [1]. Since the 1900s, the most economically damaging disasters from natural causes in Canada were floods, wildfires, drought, and extreme cold. Recent research has linked specific flooding events, as well as a general rise in the intensity of wet weather in the northern hemisphere, to the effects of rising greenhouse gas levels and global climate change [2]. Past studies intended to focus on disaster response founded on a top-down approach, but the focus has now shifted to a community-based approach [3][4][5][6][7][8][9] that stresses resilience building. After the Hyogo Framework for Action (HFA), 2005-2015 [10] launched a movement for building the resilience of nations and communities to disasters. The Sendai Framework for Disaster Risk Reduction (SF-DRR) 2015-2030 [11] serves as a continuum to the commitment supported by the United Nations Office for Disaster Risk Reduction (UNDRR). The SF-DRR notes the need for improved understanding of disaster risk in all its dimensions of exposure, vulnerability, and hazard characteristics, as well as the strengthening of disaster risk governance [11]. Potential variations in the understanding of risk, i.
Enhance whole-of-society collaboration and governance to strengthen resilience. ii.
Improve understanding of disaster risks in all sectors of society. iii.
Increase focus on whole-of-society disaster prevention and mitigation activities. iv.
Enhance disaster response capacity and coordination and foster the development of new capabilities. v.
Strengthen recovery efforts by building back better to minimize the impacts of future disasters.
The EM strategy supports the FPT governments' vision to strengthen Canada's EM capabilities to prevent/mitigate, prepare for, respond to, and recover from disasters, to reduce disaster risk and increase the resiliency of all individuals and communities in Canada. Frequent flooding is a serious concern in Canada, as was evident from the recurring spring floods in Ontario, Quebec, and Toronto Islands during 2017-2019. Previous recent significant floods include the 2013 southern Alberta floods, the 2013 Toronto urban flood, the 2014 Saskatchewan and Manitoba floods, the 2011 Manitoba flood, and the 2017 British Columbia flood. A potential flooding concern for Toronto Islands is in place in 2020 as Lake Ontario levels may rise in spring.
Although the importance of community perception of risk in the decision-making process has been extensively discussed in the literature [4,8,[40][41][42][43][44], the concept has not been applied in practice in a significant way across disciplines. Some of the reasons for that include difficulty in measuring the perception of different actors in the system, assessing the impact of the inclusion of opinion in risk assessment and disaster risk reduction (DRR) policy, and, most of all, designing and collecting relevant data. Perceived resilience is key to comprehend and estimate as it relates to how people perceive risk (exposure to hazards), vulnerability (susceptibility), and their capacity to cope including the institutional support. In survey-based qualitative research, it has been established that post-disaster experiences of affected communities show their preference for coping actions align with their individual personalities [3,[45][46][47][48]. It is clear that pre-disaster preparedness and capacity building very much depends on how people "feel" about the probability of another disaster to occur in their community [3,49]. A recent study has used a questionnaire survey to evaluate risk perception for risk awareness and to increase resilience in schools [50]. It is worth noting here that people's participation in community-based disaster management must not merely be an illusion of inclusion [51,52].
There is considerable interest in disaster resilience as a mechanism for mitigating the impacts on local communities, yet the identification of metrics and standards for measuring resilience remains a challenge [53]. By measuring baseline characteristics of communities, changes in disaster resilience over time can be monitored by estimating the individual drivers of the disaster resilience (or lack thereof)-social, economic, institutional, infrastructure, and community capacities [53]. Conventionally, quantitative methods for disaster risk assessment are better understood and also preferred as they tend to be based on numbers and indices [19,[54][55][56][57]. However, interactions between the makeup of the communities, their priorities, and general vulnerabilities, and the geomorphology of the region are natural and must be integrated holistically in risk and resilience assessment [12,[58][59][60][61][62][63][64]. On the other hand, qualitative methods are generally based on surveys, focus groups, questionnaires, and interviews. The data collected from these methods is then examined, thematized, and analyzed for the understanding and interpretation of a variety of phenomena [65][66][67]. Qualitative techniques are useful for measuring the impact of policy, needs-assessment of communities, and understanding of people's behaviour and preferences during and post-disaster [68,69]. While the most common approach in research is to apply one of the two methods, quantitative or qualitative, this study integrates both techniques with the intent to capture people's perception as well as the reality on the ground. Therefore, the focus of this study is to explore how to evaluate and incorporate people's perceptions of risk, exposure, susceptibility, and coping capacity to realize gaps between perceived disaster resilience and objectively assessed disaster resilience. Once these gaps are identified, mitigation measures, coping capacity building efforts, and adaptivity initiatives can be developed and implemented with greater success. Based on these principles, a national strategy must be considered and prepared for a more comprehensive application for disaster risk management. The following sections outline the data used in the study, the methodology developed and used on a Canadian city, discussion of the results, and concluding remarks.

Materials and Methods
The method proposed in this study is an adapted and extended version of the world risk index (WRI) method [19]. While the WRI measures disaster resilience only objectively, the technique proposed here evaluates people's perception of disaster resilience (function of exposure, susceptibility, and coping capacity) and incorporates their opinion into the assessment of perceived resilience. Previous studies have shown that a questionnaire survey is a useful tool to capture risk awareness and perception of the population to help plan future risk management efforts and encourage a resilience culture in the community [47,50]. In this study, objective and perceived resilience are being evaluated and compared, gaps between perceived and actual resilience are identified, and an integrated resilience is calculated, as well. Community participation is essential in effective and successful disaster management in communities, and this point has been validated by various studies that are based on simulation-based planning tools, basically to engage stakeholders in a user-friendly manner [55,70].
Perceived measurements-In order to incorporate people's perceptions, we have used questionnaire surveys to collect data from four different locations in the City of Brampton in the Greater Toronto Area (GTA) of Canada ( Figure 1). The selection of the survey sites was made based on a good representation of the community. Specifically, a sports and community centre (survey location 1) that also houses the public library and a swimming pool, a multicultural community centre (survey location 2), a church (survey location 3), and a restaurant (survey location 4), are places regularly used by the community. We received 100 responses to the questionnaire consisting of 29 questions designed to reflect people's perceptions of threats from natural hazards, how they would cope in emergencies, their background, and how they engage in their local environment. We explained the nature and intent of the study to each participant, including what is meant by risk, how it is a function of different elements, including exposure to hazards. However, it should be noted that at the time of the survey, the exact buffer zones to estimate exposure had not been determined and, therefore, were not communicated to the survey participants. The precise nature of the questions and how they relate to model parameters is given later in this section. People's perceptions are treated as representative of the entire city for demonstrating the methodology leading to the perceived assessment of community resilience (Table 1).
Water 2020, 12, x FOR PEER REVIEW 4 of 19 various studies that are based on simulation-based planning tools, basically to engage stakeholders in a user-friendly manner [55,70]. Perceived measurements-In order to incorporate people's perceptions, we have used questionnaire surveys to collect data from four different locations in the City of Brampton in the Greater Toronto Area (GTA) of Canada ( Figure 1). The selection of the survey sites was made based on a good representation of the community. Specifically, a sports and community centre (survey location 1) that also houses the public library and a swimming pool, a multicultural community centre (survey location 2), a church (survey location 3), and a restaurant (survey location 4), are places regularly used by the community. We received 100 responses to the questionnaire consisting of 29 questions designed to reflect people's perceptions of threats from natural hazards, how they would cope in emergencies, their background, and how they engage in their local environment. We explained the nature and intent of the study to each participant, including what is meant by risk, how it is a function of different elements, including exposure to hazards. However, it should be noted that at the time of the survey, the exact buffer zones to estimate exposure had not been determined and, therefore, were not communicated to the survey participants. The precise nature of the questions and how they relate to model parameters is given later in this section. People's perceptions are treated as representative of the entire city for demonstrating the methodology leading to the perceived assessment of community resilience (Table 1). We have used the Thiessen polygons method to aggregate the surveyed areas for the entire city by dissemination areas (DAs)-areas of equal density of population, for perceived assessment of the parameters used in the methodology ( Figure 2). The Thiessen polygon technique appropriately assigns areal significance to each survey site by constructing perpendicular bisectors to the lines joining each site with those immediately surrounding it. These bisectors form a series of polygons, each polygon containing one site. The data collected at a site gets assigned to the whole area covered by the enclosing polygon. We have used the Thiessen polygons method to aggregate the surveyed areas for the entire city by dissemination areas (DAs)-areas of equal density of population, for perceived assessment of the parameters used in the methodology ( Figure 2). The Thiessen polygon technique appropriately assigns areal significance to each survey site by constructing perpendicular bisectors to the lines joining each site with those immediately surrounding it. These bisectors form a series of polygons, each polygon containing one site. The data collected at a site gets assigned to the whole area covered by the enclosing polygon. It is worth noting here that two survey locations, BBQ and Tabernacle church, shown in Figure  1b, were lumped together as one entity, "BBQ and Tabernacle" for calculations in GIS. The reason being, the total number of responses collected from the two locations combined was rather small (14) compared to the other sites.
As demonstrated in Table 1, three parameters, namely exposure, susceptibility, and lack of coping capacity, are appropriately extracted from the responses to the survey questions by assigning binary values to them. Each parameter is comprised of several variables, representative of the parameter [19,[71][72][73]. Although we have presented a select set of variables to represent each of the three parameters to demonstrate the method, these should not be taken as exhaustive. Disaster types, geomorphology, landuse of the region, and demographics of the population must be taken into account in the determination of these variables. The three parameters are allocated weights according to their assumed influence in this particular case study [55]. The weights can vary depending on the impact of each parameter on community resilience. For example, a known dangerous environment (high exposure), a prosperous and educated neighbourhood (low susceptibility), and an accessible network of emergency services (high coping capacity) must guide how the parameters can be weighted. For example, the residential development in the Barker reservoir in Houston, Texas, that got flooded during Hurricane Harvey [74], will have a high exposure. Table 1 summarizes the proposed process, including a guide to assigning binary values to individual variables as part of the methodology developed to account for people's input in the process of resilience assessment. For example, the parameter exposure is based on dangerous locations such as: The transportation network (risk from derailment, explosion, oil spill)  Chemical plants and hazardous industries  Transmission lines (elevated health risk from high voltage corridors)  Oil and gas pipelines (toxic spills) It is worth noting here that two survey locations, BBQ and Tabernacle church, shown in Figure 1b, were lumped together as one entity, "BBQ and Tabernacle" for calculations in GIS. The reason being, the total number of responses collected from the two locations combined was rather small (14) compared to the other sites.
As demonstrated in Table 1, three parameters, namely exposure, susceptibility, and lack of coping capacity, are appropriately extracted from the responses to the survey questions by assigning binary values to them. Each parameter is comprised of several variables, representative of the parameter [19,[71][72][73]. Although we have presented a select set of variables to represent each of the three parameters to demonstrate the method, these should not be taken as exhaustive. Disaster types, geomorphology, landuse of the region, and demographics of the population must be taken into account in the determination of these variables. The three parameters are allocated weights according to their assumed influence in this particular case study [55]. The weights can vary depending on the impact of each parameter on community resilience. For example, a known dangerous environment (high exposure), a prosperous and educated neighbourhood (low susceptibility), and an accessible network of emergency services (high coping capacity) must guide how the parameters can be weighted. For example, the residential development in the Barker reservoir in Houston, Texas, that got flooded during Hurricane Harvey [74], will have a high exposure. Table 1 summarizes the proposed process, including a guide to assigning binary values to individual variables as part of the methodology developed to account for people's input in the process of resilience assessment. For example, the parameter exposure is based on dangerous locations such as: Objective measurements-We have used the 2011 census of Canada for demographic information, Municipal Property Assessment Corporation (MPAC) average property values, slopes and terrain of the region, and landuse. The landuse data is valuable for determining the location of critical infrastructure and critical facilities for the assessment of objective parameters. The GIS software, ArcGIS, is used for data processing and analysis according to the dissemination area (DA) map of the study area. Appendix A lists all the data sources. Figure 3 shows the municipal boundaries and dissemination areas in the GTA. Figure 4 presents datasets, cropped for the City of Brampton, used for parameters estimation proposed in this method.   The landuse data in Figure 4b includes residential (green), commercial (dark purple), government and institutional infrastructure (green), industrial (dark blue), parks and recreational (red), and open area (blue). Similarly, in Figure 4f, emergency services include fire stations (red), police stations (dark blue), and hospitals in green. It should be noted that none of the dumpsite buffers fall within the boundary of the city and, therefore, do not contribute toward exposure in this case study. Table 2 summarizes the process of quantifying individual variables within each of the three parameters as part of the methodology. Real datasets explained and illustrated in Figure 4 earlier  The landuse data in Figure 4b includes residential (green), commercial (dark purple), government and institutional infrastructure (green), industrial (dark blue), parks and recreational (red), and open area (blue). Similarly, in Figure 4f, emergency services include fire stations (red), police stations (dark blue), and hospitals in green. It should be noted that none of the dumpsite buffers fall within the boundary of the city and, therefore, do not contribute toward exposure in this case study. Table 2 summarizes the process of quantifying individual variables within each of the three parameters as part of the methodology. Real datasets explained and illustrated in Figure 4 earlier The landuse data in Figure 4b includes residential (green), commercial (dark purple), government and institutional infrastructure (green), industrial (dark blue), parks and recreational (red), and open area (blue). Similarly, in Figure 4f, emergency services include fire stations (red), police stations (dark blue), and hospitals in green. It should be noted that none of the dumpsite buffers fall within the boundary of the city and, therefore, do not contribute toward exposure in this case study. Table 2 summarizes the process of quantifying individual variables within each of the three parameters as part of the methodology. Real datasets explained and illustrated in Figure 4 earlier have been used in the measurement of the model parameters: exposure, susceptibility, and lack of coping capacity. Some of the processed data are shown in Figure 5. For example, exposure is determined based on whether or not a resident is located in proximity of (within a kilometer of) a highway, railway, river, industrial site, pipeline, transmission line, oil and gas facility, or dumpsite. Similarly, objective assessment of susceptibility is derived from census data, including the residence type, the age of its construction, the property value, language skill, employment status, and disability rate. The lack of coping capacity is derived from census and GIS data; it includes, income, education, a distance of more than one kilometer from emergency services, a fire station, police service, or ambulance service. See Figure 5 for the visuals of the buffer zones. Table 2. List of parameters and variables for the objective assessment of community resilience. The total of all assigned weights W i = 1.  The overall resilience is calculated using Equations (1) to (6), derived from the WRI method [19]. Precisely, Equation (1) calculates the Lack of Resilience using objective data for the three parameters that are individually normalized. The parameters can be weighted using weights thought to be appropriate for this case study for demonstration purposes. The weights ( 1 , 2 , 3 ) are open to adjustment in individual cases depending on the potential influence of the parameters, as well as the objective of the study. For example, a vulnerable community is perceived as more susceptible in comparison to an affluent neighbourhood. Residential development well outside of flood zones indicates a lower level of exposure even though the population may be regarded as vulnerable. If the community is at a substantial distance from a healthcare facility, it may reflect a lack of coping capacity in emergencies. Therefore, in this scenario, it will make sense to allocate a lower percentage of weight to "Exposure" and higher percentage to "Lack of Coping Capacity" and "Susceptibility." For the application and demonstration of the methodology, we have assigned weights to the parameters as given in Equation (2) for both objective and perceived calculations.

Parameter Criteria to Assign Binary Values to Variables
Equation (3) gives the estimate of Resilience using objective data, Equation (4) estimates Resilience using perceived data, and Equation (5) determines the combined Resilience by summing up the perceived and objective Resilience estimates. We propose here that the combined Resilience is a way to account for perceived and real measures of community resilience. The overall resilience is calculated using Equations (1) to (6), derived from the WRI method [19]. Precisely, Equation (1) calculates the Lack of Resilience using objective data for the three parameters that are individually normalized. The parameters can be weighted using weights thought to be appropriate for this case study for demonstration purposes. The weights (W 1 , W 2 , W 3 ) are open to adjustment in individual cases depending on the potential influence of the parameters, as well as the objective of the study. For example, a vulnerable community is perceived as more susceptible in comparison to an affluent neighbourhood. Residential development well outside of flood zones indicates a lower level of exposure even though the population may be regarded as vulnerable. If the community is at a substantial distance from a healthcare facility, it may reflect a lack of coping capacity in emergencies. Therefore, in this scenario, it will make sense to allocate a lower percentage of weight to "Exposure" and higher percentage to "Lack of Coping Capacity" and "Susceptibility." For the application and demonstration of the methodology, we have assigned weights to the parameters as given in Equation (2) for both objective and perceived calculations.
Equation (3) gives the estimate of Resilience using objective data, Equation (4) estimates Resilience using perceived data, and Equation (5) determines the combined Resilience by summing up the perceived and objective Resilience estimates. We propose here that the combined Resilience is a way to account for perceived and real measures of community resilience.

Lack of Resilience Objective
where The estimated total resilience has been calculated by normalizing the combined resilience estimates using Equation (6): It is noteworthy that Equations (3) and (4) were calculated using normalized and unweighted individual parameters in Equation (2), leading to an adjusted maximum total value of perceived (and objective) Resilience. The objective and perceived resilience calculations are done using the same formula.

Results
The findings of the research are illustrated through Figures 6-8. Objective and perceived measures were made for the survey sites. The actual datasets on the geomorphology of the study area, including rivers and low lying regions, indicate that the method has been applied to a flooding scenario as an example. As mentioned earlier, two survey locations, namely, BBQ and Tabernacle church, were lumped together as one entity, "BBQ and Tabernacle" for calculations in GIS due to the small number of responses from the two locations. Each of the three parameters was calculated separately to achieve perceived and objective estimates of them for each survey location. Figure 6 shows objective measures of the three model parameters, namely, Exposure, Susceptibility, and the Lack of Coping Capacity calculated for survey location 2, the Community Centre. Similar maps were obtained for perceived measures of the three parameters. Also, similar maps were developed for other survey locations. In total, 12 maps were obtained for objective and perceived estimates of the three parameters for the three survey locations. Although it may be desirable for the readers to visualize the step-by-step development of these maps, to avoid overcrowding of illustrations, selected graphs are presented here.  Figure 7 shows the integrated (perceived plus objective) measure of Resilience for the three survey locations, separately. Figure 8 is a picture of the perceived and objective Resilience measures for the entire City of Brampton, and Figure 9 represents the integrated resilience map for the city. Each of the three parameters was normalized before being weighted in Equation (1). Therefore, as can be seen in Figure 8, the two resilience maps show estimates of up to 63 for perceived Resilience and up to 89 for the objective Resilience. It is noteworthy that the method suggests the ideal resilience value to be 100. Therefore, objective resilience gives a better level of resilience in the study area as compared to perceived measure of resilience. In essence, the maximum value of objective resilience is higher than perceived resilience, and the higher the value of resilience, the better in each case scenario.   Figure 7 shows the integrated (perceived plus objective) measure of Resilience for the three survey locations, separately. Figure 8 is a picture of the perceived and objective Resilience measures for the entire City of Brampton, and Figure 9 represents the integrated resilience map for the city. Each of the three parameters was normalized before being weighted in Equation (1). Therefore, as can be seen in Figure 8, the two resilience maps show estimates of up to 63 for perceived Resilience and up to 89 for the objective Resilience. It is noteworthy that the method suggests the ideal resilience value to be 100. Therefore, objective resilience gives a better level of resilience in the study area as compared to perceived measure of resilience. In essence, the maximum value of objective resilience is higher than perceived resilience, and the higher the value of resilience, the better in each case scenario.  Figure 7 shows the integrated (perceived plus objective) measure of Resilience for the three survey locations, separately. Figure 8 is a picture of the perceived and objective Resilience measures for the entire City of Brampton, and Figure 9 represents the integrated resilience map for the city. Each of the three parameters was normalized before being weighted in Equation (1). Therefore, as can be seen in Figure 8, the two resilience maps show estimates of up to 63 for perceived Resilience and up to 89 for the objective Resilience. It is noteworthy that the method suggests the ideal resilience value to be 100. Therefore, objective resilience gives a better level of resilience in the study area as compared to perceived measure of resilience. In essence, the maximum value of objective resilience is higher than perceived resilience, and the higher the value of resilience, the better in each case scenario.  Although Figure 8 is the key outcome of this study, Figure 9 has been arrived at by combining the two resilience measures, showing the regions of variable resilience. It is useful to see how the two measures of resilience may have aligned with each other, positively or negatively. In future research, it would be helpful to subtract the two resilience measures from each other and visualize where the gaps are between the public perception and the reality on the ground. The shades of green indicate levels of resilience of the community, the darker, the better. The shades of yellow and parts of red are in regions where rivers and creeks flow. The areas shown in red indicate low resilience due to a variety of reasons such as industrial areas, flood zones, low lying areas, and a dense network of watercourses.  Although Figure 8 is the key outcome of this study, Figure 9 has been arrived at by combining the two resilience measures, showing the regions of variable resilience. It is useful to see how the two measures of resilience may have aligned with each other, positively or negatively. In future research, it would be helpful to subtract the two resilience measures from each other and visualize where the gaps are between the public perception and the reality on the ground. The shades of green indicate levels of resilience of the community, the darker, the better. The shades of yellow and parts of red are in regions where rivers and creeks flow. The areas shown in red indicate low resilience due to a variety of reasons such as industrial areas, flood zones, low lying areas, and a dense network of watercourses. Although Figure 8 is the key outcome of this study, Figure 9 has been arrived at by combining the two resilience measures, showing the regions of variable resilience. It is useful to see how the two measures of resilience may have aligned with each other, positively or negatively. In future research, it would be helpful to subtract the two resilience measures from each other and visualize where the gaps are between the public perception and the reality on the ground. The shades of green indicate levels of resilience of the community, the darker, the better. The shades of yellow and parts of red are in regions where rivers and creeks flow. The areas shown in red indicate low resilience due to a variety of reasons such as industrial areas, flood zones, low lying areas, and a dense network of watercourses.

Discussion and Conclusions
The core objective of this study was to develop a methodology to capture community perceptions of flood hazard risk based on their exposure, susceptibility, and lack of coping capacity, and incorporate those perceptions into estimating community resilience. Inspired by the world risk index (WRI), we have successfully demonstrated that the WRI determinants can be used to explain the difference between how risk is perceived and the factual picture presented by the objective data. We estimate objective resilience using actual data related to exposure (dangerous locations), susceptibility (derived from demographic data), and lack of coping capacity (derived from a combination of census and landuse data). The two resilience measures, perceived and objective, were combined by adding them and normalizing the resulting combined resilience. It should be noted that the perceived and objective parameters were normalized at the unweighted stage that justifies the similar treatment of the two (perceived and objective). The study should be used as a decision-making tool to enhance the resilience of communities based on their circumstances.
We used a questionnaire to engage people, but it should be noted that some of the nuances of the methodology were under development at the time of the survey. The randomly selected respondents were residents of the City of Brampton, as reflected by their use of the community centre, sports centre, the church and the restaurant used as survey locations. We acknowledge the small sample size of 100 participants that may have led to a rather underwhelmingly visual difference in the perceived and objective resilience measures. The reason for that to happen could be the fact that a small sample from each survey site was used to represent an entire region determined by the Thiessen polygon method. Furthermore, with a limited representation of the community in the study area, the perceived data estimates are bound to have uncertainties that can be addressed with extensive survey data. The determination of the buffer zones for different exposure variables was made by treating the variables such as highways and rivers in a similar manner. However, this aspect needs to be improved by allowing different buffer zones for various variables.
In summary, this research makes a reasonable preliminary attempt toward achieving a better representative measure of the resilience of a community in the context of disasters and emergencies. Future research is recommended for examining current disaster mitigation policies in Ontario, engaging representative communities to capture their perceptions, and looking at how specific changes can be made in those policies to improve their relevance and outcome for a diverse population. Solutions to disaster risk reduction and preparedness strategies lie in meaningful consultation with stakeholders, no matter how insignificant some may seem. We recommend future research to refine the methodology along with its various aspects presented here. The approach should also be user-friendly for a broader application. In Canada, the most frequent hazard, flooding, and other disasters are managed by the local authorities first; the province assists if local capacities prove to be insufficient; and eventually, the federal government helps if the consequences are too severe. In many cases, the federal help arrives when it is too late due to unclear guidance, inconsistent messaging, political and ideological differences among actors, and conflicting priorities. There is a need, now more than ever before, to develop a national strategy for resilience to all disasters to enable an environment of swift impact assessment of events and allocation of resources across the nation.

Conflicts of Interest:
The authors declare no conflict of interest.