How Participatory is Participatory Flood Risk Mapping?Voices from the Flood Prone Dharavi Slum in Mumbai

Subhajyoti Samaddar?Ha Si?Xinyu Jiang?Junho Choi?Hirokazu Tatano

Abstract Participatory flood risk mapping(PFRM)is a well-recognized and widely implemented tool for meaningful community involvement in disaster risk reduction(DRR).The effectiveness of PFRM remains anecdotal.The PFRM exercise has rarely been applied identically in two different places by two different organizations,which produces varied and uncertain outcomes.In the absence of any agreed and comprehensive framework for participatory DRR,existing studies struggle to provide a scientific account of how the structure,design,and process of PFRM ensure the effective participation of local communities.This study,examines what factors and methods make PFRM an effective participatory DRR tool.In this study,we first identified the process-based criteria of participation.Then we briefly introduced a participatory flood risk mapping exercise conducted in a flood-prone informal settlement in Dharavi,Mumbai.The exercise was carefully designed to meet the process criteria of effective participation.Finally,using qualitative research methods,we evaluated the effectiveness of our PFRM from the local community perspective.The findings show that ensuring community livelihood security and true involvement of marginalized groups,preparing an action plan,and incorporating fun and cultural connotations into the facilitation process are critical components that enhance community participation through PFRM in DRR.

Keywords Dharavi?Evaluation?Mumbai flood?Participatory disaster risk reduction?Participatory flood risk mapping?Processbased criteria of participation

1 Introduction

Participatory flood risk mapping(PFRM)is ostensibly the most successfully applied participatory tool in disaster risk reduction(DRR)(Osti et al.2008;Gaillard and Maceda 2009;White et al.2010;Liu et al.2018).Participatory flood risk mapping generally entails a process that enables the local community to physically locate on maps the impending flood risks in communities,their resources,and the countermeasures they can adopt(Cadag and Gaillard 2012).Local communities are the primary victims of the event and also the first to respond when disaster strikes.Therefore,local communities’perception of the spatial distribution of risk plays a critical role in DRR(Chingombe et al.2015).The PFRM also helps obtain information that is otherwise very difficult to extract(Samaddar et al.2011).The PFRM is also considered to be an effective tool to enhance the community’s risk awareness and preparedness intention,as it provides a visual demonstration of risks through maps.

Nevertheless,the success of PFRM remains uncertain and contentious(Chingombe et al.2015;Minucci et al.2020).The claims made in favor of PFRM are seldom empirically tested.Moreover,participatory mapping has rarely been identically carried out in two different places by two different organizations(Liu et al.2018).Hence,the same exercise has varied outcomes and success rates in different places.Several previous studies have claimed that PFRM is the epitome of the participatory DRR process(Osti et al.2008;Gaillard and Maceda 2009;Minucci et al.2020).But these studies did not adequately elaborate how the structure,design,and process of the PFRM would fit the desired criteria of effective participation(Liu et al.2018).There exists no yardstick to evaluate these claims and compare results across regions(Maskrey et al.2019).This seemingly effective,widely implemented participatory DRR tool has remained unclear about how and to what extent PFRM can meaningfully engage local communities in the decision-making process.This opaqueness creates an urgent need to investigate what makes the PFRM tool effective for community participation in DRR.

Our study aims to define the effective process mechanisms of community participation and attempts to show how participatory flood risk mapping can best ascertain these participatory criteria.Further,the study’s objective is to investigate the success of the PFRM as a participatory DRR tool from the perspective of local communities.Working towards these objectives,we first systemically identify the factors affecting effective community participation through the PFRMand propose a process framework for successful community participation in DRR.Then,we briefly introduce the PRFM exercise carried out in Mumbai,India.Finally,we evaluate how and to what extent this PRFM meets the process-based criteria of effective community participation.

1.1 Challenges for Effective Participatory Flood Risk Mapping(PFRM)

Like other participatory DRR tools,designing effective PFRM and its successful implementation remain uncertain because of the following factors.

1.1.1 Level of Participation

There exists no agreed,universal definition of participation in general(Arnstein 1969;Rowe and Frewer 2000).Disagreements and differences are often recounted while dealing with the scope and principle of community participation in DRR(Scolobig et al.2016;Samaddar et al.2017).For example,for some,the community’s involvement should be limited to the role of a passive recipient of information,whereas others have suggested community opinions and concerns should be included in the risk management process(Fazey et al.2010;Rollason et al.2018).Yet,this perspective of participation has been criticized for limiting the community’s involvement to merely a risk assessment(Gladfelter 2018).Recently,participation has been conceptualized as a collaborative knowledge development process between local communities and experts,including researchers and local governments(Okada et al.2013;Tozier de la Poterie and Baudoin 2015;Samaddar et al.2017).Given the varied and conflicting opinions of participation,it is critical to illustrate to what extent and in what ways PFRMwould facilitate community participation.

1.1.2 Scope and Purpose

Some studies contend that the majority of the participatory exercises in DRR are designed merely to educate participants about risk,including the types and nature of disasters,hazard sources,factors and consequences of disasters,and so on(Aalst et al.2008a,2008b;Okada et al.2013).Very few participatory tools deal with the training and tutoring of communities for disaster preparedness(Samaddar,Choi et al.2015).

1.1.3 Facilitation Process and Power to Control the Exercise

Participatory tools often vary in terms of the role and power of the facilitators in conducting the exercise,such as who will be involved,when,and to what extent?Yamori(2009)argues that in most participatory disaster risk management(PDRM)exercises,the facilitators highly control the process.As a result,the actual participation of the community is compromised,and powerful,local elites indirectly reinforce their authority through the exercise(Gladfelter 2018).

1.1.4 Art and Skills of Conducting Participatory Exercise

Conducting participation exercises is considered an art.The success of a participatory exercise depends significantly on the facilitators’knowledge,experience,knack,and skills(Rowe and Frewer 2005).Several contextual and temporal factors influence the art of facilitation(Maskrey et al.2019):(1)Exercise location:local places are preferred over places outside of a locality;(2)Language used:a local language is always prioritized over a foreign language that experts might be comfortable with;(3)Time constraints(long and short):a long-lasting exercise destroys participant motivation and attention;a short one does not allow time to discuss issues in depth.The variation of the process mechanism entails varying outcomes and success.

The manifold objectives of PFRM,varied operational methods and facilitator skills,and untested claims may create confusion among practitioners,hindering successful PFRM implementation.

1.2 Process-Based Criteria for Community Participation

The participatory process defines the means and quality of participation.The process describes whom to involve,as well as when,how,and to what extent.The process mechanism also underlines the conditions of effective community involvement,such as fairness,transparency and accountability,time effectiveness,community empowerment,and so on(Chess and Purcell 1999;Rowe and Frewer 2000;Samaddar et al.2019).There is a lack of a comprehensive framework for community participation in DRR(Vallance 2015;Maskrey et al.2019).The lack of framework for participatory DRR coerces us to explore and assess the participation process parameters depicted by studies in planning and management discourses,including risk management,natural resource management,and environmental planning.Table 1 lists the process-based criteria for successful community participation as derived from our literature review.

2 Mumbai Flood Risk Management and Community Participation

Mumbai,India’s financial capital,experienced a catastrophic flood in 2005 that killed more than 900 people and caused an estimated total loss of USD 5 billion(Samaddar et al.2012;Patankar and Patwardhan 2016).Additionally,2000 residential buildings were substantially damaged,another 50,000 residential buildings were partially damaged,and 40,000 commercial establishments suffered heavy losses(Gupta 2007).Despite the absence of major floods since 2005,Mumbai is frequently affected by local floods each year(Sherly et al.2015).In the core city,an average of more than 70 incidents of waterlogging are reported every year(Patankar and Patwardhan 2016).In August 2020,for example,a large section of south Mumbai was inundated and flooded for several hours following heavy monsoon rains(Cappucci 2020;Deshpande 2020).

As well as climate change and sea-level rise,anthropogenic factors,such as land reclamation,land cover changes,clearing of vegetation,and haphazard urbanization,have contributed greatly to Mumbai’s increased flood risk(Bhagat et al.2006;Pathak et al.2020).In its origins,Mumbai was an island city that was later expanded with land reclamation from the sea,resulting in an increased risk of flooding(Sherly et al.2015).About half of the city’s core and a quarter of its suburbs were built on reclaimed land(Patankar and Patwardhan 2016).Most of the city,including major commercial and residential areas such as the Banda Kurla Complex,the airport,and Worli,is built on reclaimed land that is barely above sea level,making them very vulnerable to high tides and storm surges(de Sherbinin and Bardy 2015;Parthasarathy 2016).Rapid urbanization and land reclamation have increased the builtup area and decreased the amount of forest and wetlands,which have indirectly increased the vulnerability of the city to flooding(Butsch et al.2016).As one example,built-up areas in Mumbai increased by nearly 114%from 1995 to 2005,while natural areas such as forests and wetlands,including mangrove forests,decreased by 35%(Bhagat et al.2006).Even though there are no specific data available,scholars have maintained that the same trend has continued(Boyd et al.2015;Sherly et al.2015).As a result of rapid urbanization in recent decades,the natural hydrological cycle of the city has been dramatically altered,which has resulted in a large increase in runoff(Pathak et al.2020).Mumbai’s drainage system dates back to 1863 when it was designed by the British government.Although the capacity of this system is 25 mm per hour,in reality,it is required to discharge 50 mm per hour during the monsoon season(Sherly et al.2015;Patankar and Patwardhan 2016).The Brihanmumbai Storm Water Disposal System(Brimstowad)project,originally proposed to improve the drainage system of Mumbai in 1993,has not been completed because of a lack of funding and other factors(Parthasarathy 2016).

Flooding in 2005 and on an annual basis has been most severe in the informal settlements or slums(Chatterjee 2010;Parthasarathy 2011),which represent 50%of the city’s total population(Bardhan et al.2015).Generally,slums are located in the most hazardous and disaster-prone areas,including low-lying abandoned places,wetlands,along railway lines,dumping grounds,footpaths,and Nullah(stream)sides.Slums along the Mithi River were the most affected area by the 2005 flood.The majority of people living in these slums are migrants involved in informal economic activities,which include household and small commercial enterprises.Researchers (Chatterjee 2011;Samaddar et al.2012;Patankar 2015)have found that the livelihoods and economies of the slums are particularly susceptible to annual flooding and have the least ability to recover.Slum communities do not receive any government assistance for disaster recovery,nor do they have insurance coverage(Patankar and Patwardhan 2016).In fact,even the local government’s official assessment of flood losses did not include the losses in the informal sectors,which are mainly concentrated in slum areas(Patankar and Patwardhan 2016).The 2005 Mumbai flood divulged that the local government could not reach the affected communities immediately after the catastrophe.It was the local slum communities themselves that helped each other in rescue and relief operations(Bhagat et al.2006;Samaddar et al.2014).These studies reported that community mutual help continued untill the post-disaster

recovery phase,such as providing job information,loans for business recovery,social support for child-rearing for working women,and so on(Chatterjee 2010).It is already evident that local participatory platforms and communitydriven groups,such as ALM(advanced locality management),played a significant role in Mumbai disaster and environmental risk reduction activities(Surjan and Shaw 2009).However,existing disaster management initiatives are largely top-down,and vulnerable,marginalized communities living in slums are not meaningfully involved in the disaster management and planning process(Parthasarathy 2011;Samaddar and Tatano 2015).Parthasarathy(2016)demonstrated that the local community’s participation in flood risk management remains elusive in Mumbai due to the Balkanization of institutions and governance mechanisms involved in disaster risk management.The lack of political and economic power has escalated the disaster vulnerability of the marginalized slum communities(Butsch et al.2016).The clear inference drawn from the above facts is that there is an urgent need to improve the resiliency of local communities.The community,particularly the marginalized slum community,should be meaningfully involved in the disaster risk management process(Parthasarathy 2016).The new challenge for the local government is to execute a participatory program that would enable meaningful community participation in disaster management.

Table 1 Process based criteria for community participation in flood risk mapping exercises

Table 1 continued

3 Methods

In this study,we first conducted participatory flood risk mapping in a community in Dharavi,Mumbai.Once the risk mapping process had been completed,we asked participants how they perceived its success.Our empirical investigation was qualitative in nature.We describe the study area,the steps,and the process involved in participatory flood risk mapping.We also discuss the methods used to collect data for evaluating the effectiveness of community participation.

3.1 Case Study:Participatory Flood Risk Mapping in Rajiv Gandhi Nagar,Dharavi

The participatory flood risk mapping was carried out in a flood-prone slum community,namely Rajiv Gandhi Nagar in Mumbai,India early in 2017.Rajiv Gandhi Nagar comes under Dharavi(Fig.1),the largest slum in Mumbai city,located on the bank of the Mithi River.Dharavi,which now houses 1 million people in a 2.1 km2area,was once a tiny fisherman community on the city’s outskirts.After the textile industry boom during the 1950s and 1960s,many jobless rural laborers migrated from different parts of India to Mumbai,and slums such as Dharavi sprang up.These poor,rural migrant communities,lacking access to the exorbitant formal housing stock,had to encroach on barren,environmentally fragile,and hazardous low-lying lands adjacent to the railway track,sewage lines,and river basins(Sharma 2000).The study area,Rajiv Gandhi Nagar(Fig.2),is the newest addition to this squatter settlement.A population of approximately 25,000 occupies roughly 0.9431 km2of land.The area has been designated as a highly flood-prone,low-lying settlement by the Municipal Corporation of Greater Mumbai(MCGM).

3.2 Steps and Process of Participatory Flood Risk Mapping

The flood risk mapping in Rajiv Gandhi Nagar was carried out in February 2017 by researchers and affiliates of Kyoto University’s Disaster Prevention Research Institute in collaboration with the local city government(MCGM)and other academic institutions,including the School of Planning and Architecture(SPA),New Delhi,and the Sir J.J.College of Architecture,Mumbai.The objective of flood risk mapping was to identify and locate the flood risk areas and find possible countermeasures to enhance their adaptive capacity.A wide range of survey tools(Fig.3),including town watching,group discussion,open-ended interviews,and observations,were used to collect flood risk information.We collected information that included flood height and duration,annual water logging depth and period,building and property damage,perceived vulnerable areas,existing and changing land-use patterns,and so on.Local communities led the exercise,yet the contributions from all stakeholders as facilitators and opinion providers,including academic institutions,nongovernmental organizations(NGOs),and local governments,were also sought.

Fig.1 Location and land-use map of Rajiv Gandhi Nagar in Dharavi,Mumbai,India

Figure 3 describes the steps and procedures pursued in the exercise.Table 2 shows how the exercise ensured local community participation through various activities.For example,like the 2005 flood map(Fig.4),the local communities developed a series of maps related to flood risk issues in the area,such as a vulnerability map,an annual waterlogging map(Fig.5),household flood damage and loss map,and so on.

Fig.2 An elevated overview of Rajiv Gandhi Nagar in Dharavi,Mumbai,India

Local communities through the PFRM informed facilitators that in the 2005 flood,a vast area of the slum had‘‘neck height’’water for 48 hours(Fig.4).The houses closest to the drain and creek,inhabited by the poorest of the poor and the most recent in-migrant dwellers,were the most severely affected.More than half of the population had a 20,000 INR(USD 290)total loss due to the flood.Before 2005,people had not experienced floods at that scale,so a majority of the inhabitants did not foresee the impact and refused to evacuate to higher places.However,dwellers reported the absence of an early warning,of a reliable information source,and of adequate evacuation shelter also contributed greatly to the residents’unwillingness to evacuate.

Fig.3 Steps of flood risk mapping in Rajiv Gandhi Nagar,Mumbai,India

3.3 Methods of Evaluation

This study evaluated the success of participatory flood risk mapping with the following objectives:(1)to find the performance of the PFRM in meeting the criteria of an effective participatory process;(2)to examine the merits and demerits of the PFRM from a community perspective;and(3)to identify any missing aspects of the participatory risk mapping design to ensure a more meaningful future involvement of local communities.In addition,the participants themselves evaluated their level of participation and the effectiveness of the exercise.

The field surveys were conducted three months after the exercise.A total of 44 PFRM participants,including 19 Chawl(neighborhood)leaders and 25 citizens,were interviewed.Originally,two Chawl leaders from every Chawl participated in the PFRM exercise,and we selected one Chawl leader from each Chawl for this evaluation survey.The reason for selecting Chawl leaders is that they were the most active participants in the PFRM exercise.Chawl leaders participated from the beginning to the end of the exercise to design the program,disseminate information,mobilize local residents for the exercise,arrange mapping surveys and group meetings,and provide vital information.We interviewed 25 local residents whoactively participated in the exercise and were willing to participate in the interview survey for the PFRM project evaluation.We had only four women participants and three youth respondents for the survey.In general,the participation of the youths and women was not a significant feature of the PFRM project in Rajib Gandhi Nagar.

Table 2 Participatory process criteria for the flood risk mapping in Rajiv Gandhi Nagar,Mumbai,India

Fig.4 The 2005 flood map of Rajiv Gandhi Nagar,Mumbai,India derived from the participatory mapping exercise

Fig.5 Annual waterlogging in Rajiv Gandhi Nagar,Mumbai,India

The field site is densely populated,and study respondents preferred to give interviews somewhere within the locality instead of going outside the community.Therefore,we opted for local tea stalls,club rooms,and religious buildings as the sites in which to conduct face-to-face interviews.Often,outside the main agenda of the interview,respondents were welcome to express their other concerns and interests ranging from personal and family matters to community well-being and livelihood concerns—outliers to flood risk management.The free and intimate dialogues with respondents became instrumental for quick rapport building and getting authentic information from respondents.The interview included both structured and open-ended questions,as shown in Table 3.Interviews were mainly carried out only during the weekend or sometimes in the evenings of weekdays so that the survey would not clash with the participants’work schedules.Because of the very informal and open-ended interview style,data collection was slow and lengthy.The interviews were conducted mainly in Hindi,which is the language spoken commonly across this multi-linguistic community.

We used the grounded theory research method for data collection and analysis because this method was successfully employed earlier on similar topics(Moore 1996;Tuler and Webler 1999).This method is considered very effective in obtaining participant perspectives.Unlike quantitative research methods,grounded theory does not restrict respondents’opinions and responses within a set of questionnaires predetermined by the researchers.Further,if the respondents have less formal education,this survey method allows researchers to obtain respondents’true perspectives using a storytelling exercise,describing key events,and using photographs.Data are categorized with respect to similar relevant characteristics in a process called coding.First,a relatively large number of categories are developed;then,through iteration,these categories are grouped into more abstract categories of conceptual relevance to the analysis.Finally,categories are grouped according to their relationships with each other.

4 Results of the Evaluation

In general,the participants rated the risk mapping to be quite high as a participatory disaster management tool.In total,the exercise received 7.6 points out of 10.But the detailed testimonials,as reflected in Fig.6 and Table 4,reveal that the flood risk mapping exercise did not meet all of the criteria for effective community participation.

4.1 Motivation for Participation

Engaging communities from the inception of the project was the prime goal,but participants believed it was very challenging and minimally achieved.They argued that the involvement of local people at the beginning of the project was negligible and that only local leaders,including Chawl leaders,political and ethnic leaders,and religious gurus,participated.Participants presumed that this is due to the following reasons:

(1)A flood mapping exercise has never been carried out before in Rajiv Gandhi Nagar.Local communities had no idea about its effectiveness in flood risk management.Therefore,they were not motivated to participate;and

(2)Local communities had been receiving overwhelming invitations from several organizations to participate in similar so-called participatory programs and projects.Often,these projects ostensibly appeared very promising but,in reality,offered nothing for the community’s livelihood development.The inadequacies of these training programs increase the local community’s initial apathy and skepticism towards the risk mapping project.

A middle-aged participant who is an auto-rickshaw driver stated the following:

Almost every month,NGO staff knock on your doors.Sometimes they come for vaccination,TB[tuberculosis eradication],women training[skill up-gradation],and green and clean slum….You don’t know…every time they put some new reasons.We joined many training programs and realized that NGOs actually wanted to use us.If people join their programs,they[NGOs]get more funds and fame.So they use us for their success.

A shop owner stated this:

Enough!People are frustrated joining this and that training programs.Flood mapping…you are lucky that the word[was one]we never heard before.That’s why at least a few were willing to join and try....

Therefore,although the leaders participated throughout the process in the mapping exercise,citizen involvement was seen only at the survey and mapping phases.Often,the local neighborhood leaders had to force their fellow citizens to participate.

4.2‘‘To Survive’’as the Source of Motivation

The participants added that the inhabitants in slum areas are generally poverty-stricken and have to toil hard to ensure the bare minimum for survival.Hence,unless there are visible and substantial outcomes that contribute to their community’s day-to-day life,they are less motivated to participate.A participant added the following:

If you tell these folks when a flood will come and how it would affect their job,they will be very interested.People left their homeland,belongings and came here[Mumbai]!Only because to survive...if they see threats coming to their jobs,they become alert and want to fix it.

Table 3 Survey questions for evaluation of a participatory flood risk mapping exercise in Rajiv Gandhi Nagar,Mumbai,India

Fig.6 Performance score of participatory risk mapping by project objective categories

The participants argued that the risk mapping exercise should make a visible connection between flood risk impacts and how‘‘to survive.’’Participants reported that‘‘to survive,’’they need jobs,wages,food,shelter,education for their children,and protection from lifethreatening events.Dwellers in Rajiv Gandhi Nagar consider the flood risky only when it impacts their basic survival needs.

4.3 Fairness,Trust,and Accountability

In this evaluation,issues related to mutual fairness,trust,and accountability apparently received a very high score(Table 4)because respondents speculated that the participants of risk mapping did not take the exercise seriously as their vested interests were not involved(Fig.7).Citizens were aware that the outcome of the projects would not affect their tenure rights,housing ownership,employment,and governmental compensations because the government did not sponsor the project.Therefore,the seemingly higher mutual trust score,as shown in Fig.7,does not reflect the truth of the intercommunity dynamics existing in Dharavi.Apparently,in real life,prevailing mutual trust and cooperation would fade away,replaced by confrontations and distrust,when projects deal directly with jobs,redevelopment,and property-related matters,such as what happened in the Dharavi Redevelopment Projects.The participants believe that various organizations had

historically exploited the local communities in Dharavi in the name of redevelopment,slum up-grades,community well-being,rehabilitation,and so on.Over a long period,accumulated social and political discrimination and deprivation have created deep social distrust between local communities and both government and nongovernmental organizations.On occasion,external agencies ignited conflicts and distrust among various ethnic and cultural groups in Dharavi.Building trust again would require prolonged and persistent initiatives.

Table 4 Merits and demerits of flood risk mapping to ascertain each process-based criterion of participation

Table 4 continued

4.4 Participation as Co-learning and Co-creating

The participants,however,affirmed that the PFRM exercise boosted citizens’motivations and willingness to work together.It also helped them understand the shared interest and concerns of all.The exercise increased their flood risk awareness and self-confidence to manage flood risks.

Fig.7 Participants’speculation on the development of false trust appeared in the flood risk mapping exercise

The participants were quite satisfied with the facilitation process(6.93)and the time of the exercise.They appreciated the presence of students in the training session because they were friendly,open to taking queries,and skillful in demonstrating the problem simply and easily.Participants suggested incorporating more visual demonstrations and local knowledge in the exercise to enhance the learning process.More intensive participation of external agencies,particularly field engineers and technical staff from the Municipal Corporation of Greater Mumbai authority(GNorth Ward,MCGM),was sought after because of their wide knowledge about flood problems and their countermeasures.The participants indicated that conducting the exercise only on holidays and in the evening allowed more people to join.

4.5 Power to Influence the Decision Making Process

The participants believed that there are two factors strongly associated with the power to influence the decision making process(Fig.8):resource availability and community empowerment.Local residents often come up with innovative ideas for community development,but they habitually fail to prosecute those innovations because of the lack of financial resources.A shop owner said the following:

We have many plans and ideas.We often discuss them when we meet each other,causally in Chawls[neighbors]committee meetings and club events.But we cannot take them seriously.The main reason is no funds at all.You again look into help from some NGOs or municipalities.Outsiders treat us like beggars.You have no respect when you ask someone for help.

Likewise,the local community also feels an implicit imperative to refrain from negotiation when funded and supported by outsiders.Otherwise,the funding would be renounced.As a result,local resident groups hardly execute their own ideas alone.Mostly,there would be proposals from the government and NGOs to do certain things,and citizens would have to join them by making comments and suggestions for minor revisions.Participants mentioned that although the risk mapping idea was a good proposal,the original plan came from the researchers.A 55-year-old railways worker had the following question:

The mapping exercise was originally your idea[referring to the research team].You had money to run them.We also had found the idea may be good to give a try.We knew it would not cost us much.I would have been happy if the plan was made and created by Jhoporpatti(slum)people.Why should we put our legs into your shoes?

The participants suggested that empowering residents,particularly local youths,through education,short-term training programs,or skill up-grades would enable them to become independent with their decision-making authority.

Fig.8 Slum dwellers frequently illustrate how their struggle for survival(‘to survive’)disempowers them,reframing them from the decision making and plan implementation

4.6 Missing Link to Participation

The participants were asked to point out the missing components in the flood risk mapping process whose absence prevents the exercise from becoming truly participatory.Participants suggested adding four more components to the process of participatory flood risk mapping as shown in Table 5:(1)fun;(2)recognition;and(3)preparation of an action plan.

Participants believed that some fun needs to be added to the risk mapping process.Initially,the exercise was very serious as many people from external agencies participated.Participants assumed that a very serious environment during the workshop would demotivate citizen participation.Therefore,they also suggested conducting this event along with the existing cultural events or sports of the community.A participant pointed out this:

If we do this mapping during community sports and cultural events,people will listen to you.People seek fun here.Life is already full of tension.We always deal with serious issues.So you will see—‘‘more fun,more involvement.’’

The participants believed that the leaders had a significant role in convincing people,assembling them,and organizing the mapping exercise.All that work was done voluntarily;yet,leaders received criticism from various groups of the community.Hence,the project should recognize their hard work—by organizing events and gestures such as mega-success events,community gatherings,or posters to acknowledge leaders.In addition,social recognition would foster new leaders,particularly youngsters,to organize such collective initiatives.

The participants suggested that in order to grow the local community’s interest in the PFRM,the exercise should go beyond simply understanding the risk.It is imperative to identify the strengths of local communities in order to implement innovative flood prevention methods.As people see that they can control and manage the flood,they show interest in the exercise.However,community members often feel that identifying and collecting information is a waste of their time and energy,because it does not show any way out of their current situation.

5 Conclusion

The participatory flood risk mapping exercise(PFRM)has received wide recognition as an effective tool that meaningfully engages local communities in DRR.As a result,many studies also have been done on PFRM in recent times.But these studies(Osti et al.2008;Gaillard and Maceda 2009;White et al.2010;Liu et al.2018;Ardaya et al.2019),other than giving a descriptive account of the steps and procedures of a specific exercise,do not illustrate how the structure,design,and process of the exercise lead to successful community involvement.Other than a few sporadic initiatives(Bajek et al.2008;Mercer et al.2008),the literature lacks any comprehensive study of the process criteria for community participation in DRR.Consequently,it has been difficult to understand what works,where it works,and why it works.In the past,a few PFRM

Table 5 Additional process factors of effective community participation in the Mumbai participatory mapping exercise

studies(Reichel and Fro¨mming 2014;Liu et al.2018;Ardaya et al.2019)were carried out to examine the extent to which the PFRM exercise increases the success rate of DRR projects in local communities.The success of DRR projects does not necessarily mean local communities are effectively involved(Samaddar et al.2019).Similarly,effective community participation alone does not necessarily ensure effective,let alone comprehensive,DRR.Given this challenge,our study undertook a pioneering initiative to systematically and empirically evaluate the process-based criteria of PFRM,and to engage local communities in DRR from the local community’s own perspective.The study has the following key findings and planning implications for DRR.

(1)Flood risks under livelihood risks:More than the facilitation process and style,the agenda and scope of the exercise were found critical for participants’motivation to participate.Local leaders found it extremely challenging to engage people in PFRM activities,especially in the initial phase,because people’s priorities were livelihood survival needs rather than the flood risk issue.Therefore,PFRM activities must illustrate the direct connection between the flood and the community’s jobs,income,shelter,and health problems.The study findings reiterate the importance of sufficiently integrated disaster risk management such that disaster risk is not treated alone,but rather is analyzed in relation to other life risks,including employment risks,financial risks,health risks,and so on.Previously,few studies have indicated the importance of livelihood security for integrated disaster risk management(Gopalakrishnan and Okada 2007;Amendola et al.2008;Hochrainer-Stigler et al.2020).This is especially true in a developing country context(Parthasarathy 2015)although these studies did advance understanding of how livelihood security could enhance local community disaster resiliency.The present study has shown the importance of livelihood security to enhanced local community participation in DRR through the use of PFRM.When the livelihood of local communities is interlinked with a PFRM exercise,communities at risk have more incentive to participate in the program.This helps to build social resiliency and to transform local communities.

(2)Hope and options for flood risk reduction:Participants reported that over time local communities have developed a fatalistic attitude toward flood risks.Because flood became an annual phenomenon,and people did not yet find any options to reduce the incurred loss,solely knowing the risk is not adequate for the community’s sustainable participation.Participatory flood risk mapping can guide the community to possible flood countermeasures they can adopt to reduce loss and protect livelihood.Previous studies have also found that conventional participatory disaster management tools designed to enhance people’s risk awareness remained ineffective,because poorer community members did not possess sufficient funds to adopt any options and alternatives to tackle the disaster(Na et al.2008).

(3)Self-reliance:Local communities found themselves helpless to negotiate with external agencies to pursue their visions of DRR because of a lack of financial,technical,and human resources.Local communities suggested that local youths could be trained to carry out PFRM independently.That would empower them to implement their action plan without external help.(4)Self-trust and self-recognition:Social trust is very fragile among the stakeholders due to several sociopolitical reasons in slums areas.Local communities suggested celebrating the contribution of local volunteers and leaders through advertising their names,celebrating their achievements,and so on.This would increase the mutual trust within the community to cooperate and collaborate for community well-being in activities like the PFRM initiative.Furthermore,the involvement of different stakeholders,such as municipal engineers and government officials,was suggested to increase the credibility of the PFRM process.

(4)Social and cultural viability:Local communities considered the PFRM to be very intensive in terms of time and labor.Leaders and volunteers had to devote lots of time and resources to organize the activities.Therefore,local leaders should be recognized through organizing social and cultural activities.Linking the PFRM with cultural activities or sports as a side event would create higher social acceptance and interest in the exercise.

(5)Sustainability:Social activities like PFRM terminate easily after a while.The local communities argued that some of the plans of the PFRM must be tried to enhance people’s motivation.Visible results can encourage people to keep participating in collaborative activities.

AcknowledgmentsThe authors wish to express their thanks to the leaders and residents of Rajiv Gandhi Nagar,Dharavi who have enthusiastically supported this study for a very long time.Not only did they participate in the risk mapping exercise,but they were engaged throughout the process and played a key role in guiding it.Thank you also for the tremendous assistance provided by the students of Sir J.J.College of Architecture,Mumbai,and School of Planning and Architecture,New Delhi,in carrying out this risk mapping exercise.This project was supported by Future Development Research Funding Program FY 2017,Kyoto University Research Coordination Alliance.

Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License,which permits use,sharing,adaptation,distribution and reproduction in any medium or format,as long as you give appropriate credit to the original author(s)and the source,provide a link to the Creative Commons licence,and indicate if changes were made.The images or other third party material in this article are included in the article’s Creative Commons licence,unless indicated otherwise in a credit line to the material.If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use,you will need to obtain permission directly from the copyright holder.To view a copy of this licence,visit http://creativecommons.org/licenses/by/4.0/.