3.1 The context of the collectives

We now consider the first case study that involves a group of women’s collectives that engage in a range of activities aimed at reducing vulnerability to a range of social and environmental risks. The collectives are located near the town of Nepalganj in the Banke district of the Terai of Nepal in the Karnali Basin approximately 8km from the Indian boarder. Over the past 15 years, Nepalganj has experienced considerable socio-economic changes including rapid urbanisation and population increases. These are to a large extent the result of regional conflicts and changing work migration patterns [25] . The majority of the participants of these women’s collectives are low caste urban poor women without a regular income. Nepalganj’s climate is controlled by the Asian monsoon. In this region climate change is expected to impact the Asian monsoon by increasing interannual variability [12] . This in turn is likely to manifest in decreased rainfall during the winter, an unpredictable start date to the monsoon rains, and less frequent and more intense monsoon rainfall [12] .

The women’s collectives in Nepalganj were initially formed through the facilitation of a local NGO. Between 1998 and 2005, female community members from certain sections of the town of Nepalganj were invited by a Nepali NGO to participate in a group action process. This participatory development method encouraged the women to think about the challenges they faced and potential solutions. The women identified many issues relating to the weather and climate, particularly the impacts of floods and droughts on their livelihoods. For example, in recent years there has been less agricultural work available due failed crops and the late onset of the monsoon and related cultivation processes. The result of this group process was the formation of collectives that supported a range of activities aimed at reducing vulnerability to these environmental risks. These include: savings and loans schemes that made available finance as a buffer in times of hardship such as drought, collective agriculture initiatives that provided food to the group members and collective business enterprises, such as a mushroom farm, that provided additional streams of income. All these actions increased the resources that the group members could draw upon in general, and during times of environmental shocks and surprises.

3.2 Functioning of the collectives

The collectives functioned initially with a relatively uniform operational structure. The facilitators were employed by a local NGO and they led the initial formation and functioning of the groups. Each collective consisted of 6 to 35 members. Over time the collectives elected a leader and an assistant leader from within their group. These leaders would then be responsible for organising the collective in their geographical area. The NGO facilitator would serve as a resource and guide for the group with a gradually diminishing presence. The facilitators would attend group meetings for the first few years and assist by taking minutes and developing the skills of other group members. As the number of collectives increased, regional committees were established, in coordination with the facilitating NGO, and these were attended and run by the leaders of the local collectives. These committees provided additional coordination and links between the groups and increased their capacity to engage issues on a different scale, such as regional health care, and facilitate larger projects such as raised cement roadways. The regional committees provided an additional level of social network that enhanced networking social capital.

One of the main functions of the collectives was to establish and manage a credit and saving scheme. This capital available for loans was raised and managed by the members themselves. The NGO acted as a facilitator of the process rather than a source of credit. Meetings were held regularly to administer loans and repayments. Money for the group was raised by its members in the form of savings with a low interest rate. This could be loaned by individuals or drawn upon for the purpose of group projects when necessary. Through a consensus decision making system groups autonomously decided on the priorities for expenditure of these funds. Examples of this include the financing of labor for the installation of donated toilets and loans to those in the group who needed the assistance most urgently [24] . As the collectives became more established they increasingly began to lobby local government representatives to provide better services to the community.

3.3 Contribution to adaptive capacity

This case study illustrates four important contributions of collective action to the strengthening of adaptive capacity. First, collective action facilitated the establishment and strengthening of social networks. These networks serve as communication channels for new knowledge relating to environmental changes and risks, planning processes, and emergency information during times of environmental stress and hazards. One development actor involved in the project stated: “ the development of new relationships and networks among the participants was just as valuable as the practical, or tangible, outcomes of the project ” . Given the uncertainties of climate predictions and the range of potential surprises faced by communities, networks that can rapidly disseminate new and updated information are crucial in enabling effective adaptation responses. Secondly, collective action improved the economic resources of the members of the women’s collectives. By increasing the resources of the participants, the vulnerability of the participating individuals and communities was reduced. A savings scheme created a financial reserve that individuals could draw on in times of hardship. In terms of climate risks, these funds were sometimes used to help people after a crop had failed due to adverse weather conditions. The capacity to adapt to uncertain environmental risks was also enhanced by the micro-credit and savings scheme. The administration, ownership and management of this scheme was organised by the collectives. The joint operation of these saving schemes by the collective is likely to further strengthen social networks.

Collective action can also provide a space for community members to voice, discuss and solve problems. This is the third contribution of collective action to adaptive capacity we identify from this case study. The group action process facilitated the identification of both the problems and solutions. The ability to downscale modeled climate change projections to regional and local scales remains limited to-date but in many cases adaptation responses will nevertheless need to be identified and implemented by local actors. The development of more robust decision making frameworks enhances the capacity of local actors to adapt to a range of environmental changes and risks. Finally, collective action contributes to individual and collective empowerment that can establish and strengthen relationships with local government actors and lead to stronger advocacy. Many of the members of the women’s collectives noted that they and their families felt more confident in the community as a result of their participation in the collectives. Because of this empowerment members of the collectives felt more confident to meet with local government officials and to request better community services, such as improved access to clean drinking water and the cleaning of drains. This empowerment is a manifestation of what Adger [1] defines as the networking capital form of collective action.

3.4 Interaction of women’s collectives with local governance processes

In general people in the community, and members of the collectives, did not trust the government due to widespread and long-term issues of corruption. An NGO actor noted “ it is generally understood by Nepalis that government is corrupt and filled with nepotism ”. This situation has been exacerbated by the absence of local elections in recent years due to regional conflicts and instability in the national government. However, despite these issues we identified several notable examples of positive interaction between the collectives and local government actors.

Being a part of a collective enabled their members to engage in advocacy with their local government. All members of the collectives were part of the marginalised section of the community as a result of a range of intersecting factors such as landlessness, gender and unemployment. The government engaged in only a very limited way with this demographic for a range of cultural and social reasons, including the legacy of the caste system and differing religious backgrounds. The members of the collectives expressed that on their own they felt voiceless and that they were initially reluctant to approach the government. Through the initial group action process many of the collectives decided that some of the challenges they were facing should be addressed by the government. For example, inadequate and blocked public drains frequently resulted in local flooding during monsoonal rains. Many of the collectives hence started to lobby the government to provide better services. Members of the collectives reported that they felt more confident to talk to their local government officials to request funds and services as a result of the collective process. After several years these actions resulted in a range of tangible positive outcomes including improved drainage systems, roads and the provision of additional land for agriculture. Several members reported that as their sense of empowerment increased, their perception of themselves changed and they became bolder. Collective action challenged the experience of disconnection between communities and local government.

3.5 Enabling and constraining factors of collective action

We now identify enabling and constraining factors for collective action in Nepalganj. Factors that enabled the continuing functioning and growth of collectives included a long-term commitment from the local NGO, the capacity of key actors, and the development of social networks. The long-term support of the local NGO provided secure employment to the local facilitators and enabled the program to develop and improve methods over time. Hancock [25] argues that the level of commitment of facilitators and group leaders provided a long-term and committed engagement with the communities that ultimately precipitated the formation of the collectives. Many participants observed that the development of social relationships between members of the collectives and to other collectives both strengthened the continuing functioning and the formation of new collectives [24] [25] The enhancement of social networks in the community enabled the collectives to include more participants and to help a larger number of people. Some of the collectives also extended their services to men.

A range of constraining factors limited the success of the collectives. These mostly pertain to culture, religion, politics, governance, capacity, and resource availability. In some cases, different religious and political views and affiliations hampered the relationships between collectives and the wider community and local government actors. For example, Worboys [25] reports that some government representatives refused to acknowledge or meet with some collectives because their members were affiliated with a different religion or a different political party. A patriarchal system within the society provided an impediment for some women being able to participate in the collectives. Gender also created a barrier for some women in trying to meet with male government staff. Members of some collectives reported that certain government representatives refused to meet with them unless they were accompanied by men. One local government employee reported that he had lost social status by interacting with the women collective (p. 40) [25] . Finally, an overall lack of available resources, both individually and from local government, provided a consistent challenge for many of the collectives. In addition to scarce financial resources Hancock [24] noted that low levels of literacy sometimes created difficulty in recording savings and loans and day-to-day challenges such as sickness and poor access to healthcare reduced attendance at meetings.

4.1 The context of the collectives

Krabi is one of the southern provinces of Thailand and is located along the Andaman Sea. The Andaman Coast was the most severely affected area of Thailand by the 2004 Indian Ocean Tsunami. The tsunami devastated the provinces of Phang-Nga, Krabi, Phuket, Trang, Satun, and Ranong [30] . In Krabi Province, Phi Phi Island was the worst damaged, and, in particular, the main tourist areas of Ton Sai Bay and Loh Dalum Bay [31] . At least 2,000 people are presumed dead [32] [33] and 15,812 people were directly affected [30] . Following the 2004 tsunami, community-based disaster risk management activities in Krabi Province were initiated by a range of international, national and local organisations, and by the communities themselves. Much of the work of international and national NGOs and government authorities has so far focused on communities that were devastated by the tsunami, and some communities that are perceived as highly vulnerable to future events.

4.2 Functioning of the collective action

Community-based disaster risk management is frequently organised through disaster preparedness committees in which people act as unpaid volunteers for the greater good of the community. For example, the Thai arm of CARE International, the Raks Thai Foundation stated that it is important to implement projects in communities that are already organised [27] . The idea to work in organised communities relates to the notion that it is useful to recognise and build on existing strengths within communities and to work with people that are already actively engaged at the local level. For this reason and because it is difficult to initiate and sustain a new committee for disaster risk management, many NGOs work with existing committees, such as funeral or loan committees. These kinds of committees exist to support members with small loans during times of hardship.

Motivations for participation in a disaster preparedness committee are many and varied. In many coastal communities a principal source of motivation for investing in such a committee is a high awareness of coastal hazards due to the high loss of life during the 2004 tsunami. Another motivation is strongly linked to failures of formal governance responses at the local and a lack of trust in the commitment and capability of government authorities to provide effective disaster risk management. For example, in the tourism communities of Krabi Province emergency aid relief did not reach all eligible recipients; funding was insufficient and available funds were often misappropriated due to corruption and nepotism at the local level [34] . Tensions between communities and local government authorities exist also because of other unresolved issues, such as the use of illegal fishing gear and practices in some communities. Concerns over livelihoods are another important driving force for engaging in community-based disaster preparedness activities. Many communities do not have an interest in disaster risk management per se, but are willing to engage in these activities if they also lead to livelihood improvement.

Collective actions focus on the enhancement of disaster awareness and preparedness of community members and tourists, capacity building for disaster risk management and early warning, and the mobilisation of support from local government and NGOs. Disaster preparedness activities include the collection and dissemination of existing information; the identification of hazards, potential impacts, high-risk areas, safe areas, evacuation routes and those most vulnerable to hazards; the development of public awareness campaigns and school programmes; the preparation of emergency plans; and early warning and evacuation exercises. Activities aimed at capacity building include the recruitment and training of volunteers in emergency response activities such as search and rescue and first aid skills and the development of alternative warning dissemination infrastructure and procedures.

Support for community-based collective action is mobilised in different ways. One mechanism is established social networks with community-based organisations (CBOs) in neighbouring communities. Another mechanism that is becoming increasingly popular with some NGOs is micro-credit schemes. For example, the Raks Thai Foundation uses a Revolving Loan Fund ² as an initial entry point for engaging with communities.

Important examples of collective action in Thailand include the “Mister Tuan-Pai" (“Mr Early Warning”) project at the national level, the Save the Andaman Network (SAN) at the sub-national regional level and the “One Tambon One Search and Rescue Team (OTOS) at the local level. The "Mister Tuan-Pai" project was established under the Disaster Prevention and Mitigation Office (PDPMO) to recruit volunteers to monitor flash flood hazards using rainfall gauges and to provide early warnings to villages in areas at risk. SAN is an informal network of NGOs and CBOs that was established in response to a perceived lack of coordination by the international NGOs in post-tsunami recovery. OTOS was established by the Department of Disaster Prevention and Mitigation (DDPM), the Department of Local Administration, the Health Insurance Office, the Office of Health Promotion and Support Fund, and the Thai Red Cross to 1) ensure the safety of life, and rapid and efficient search and rescue operations; 2) establish efficient search and rescue teams at every Province, District and Tambon (Sub-district) in the country; 3) enhance the capacity and efficiency of search and rescue teams through technical training and drills; 4) build up the self-confidence of search and rescue teams; and 5) provide first aid treatment and rapid transfer to the appropriate medical establishment. OTOS is under the administration of the Local Administration Organisation, responsible for controlling traffic during evacuations.

4.3 Contribution to adaptive capacity

The insights derived from the participatory assessment of early warning system - community linkages [27] show that stakeholder agency and collective action are important elements of the adaptive capacity of communities. Building capacity to cope with environmental (as well as socio-economic) shocks and surprises is an important step towards adapting to climate change. An important aspect of adaptive capacity is what people can do to help themselves through collective action.

The evidence from Krabi Province suggests that local government agencies in some cases lack the capacity to support collective action and can therefore represent a considerable barrier. This is an important governance issue that is also relevant in the context of adaptation projects. Recently, progress has been made in two areas: the first one is a transition in the approach of many governments and NGOs from a focus on post-disaster emergency response to addressing longer-term development that links disaster risk reduction with livelihoods, natural resource management and poverty reduction efforts. The second is the integration of disaster risk reduction and climate change adaptation. This is a subject of considerable current debate but a number of synergies have been put forward by various authors (see e.g., [2] [3] [4] ). We argue that in addition to these two integrative steps, adaptive capacity could be enhanced by better defining and coordinating the roles and responsibilities of government, NGO and private sector actors because this would remove tensions, competition and the duplication of efforts.

In the online dialogue on early warning [29] responses on the theme ‘technology versus community’ indicated that efforts to strengthen disaster risk reduction and early warning are heavily biased towards technology. There is an urgent need to go beyond such technological approaches and to recognise the importance of investing in the capacities of communities by developing and better utilising social networks. The role of social networks has not been recognised sufficiently in the current debate despite it’s potential as an important part of future disaster risk reduction and climate change adaptation strategies. While response capability depends largely on the community’s own capacity to manage risks, the engagement of government and NGO actors to inspire and support collective action is crucial.

This case study demonstrates that community-based action is positioned in the context of multiple needs and interests and that there is a challenge in coordinating multiple stakeholder agendas. Community-based disaster risk management, therefore, needs to be integrated in strategies that address wider community priorities, such as improving and diversifying livelihoods and building capacity for community-based natural resource management. Addressing these concerns helps to build adaptive capacity to climate change because the underlying causes of vulnerability to shocks and surprises are reduced.

4.4 Interactions with local governance processes and other organisations

In many cases, there is a lack of financial and staff capacity for disaster risk management activities at the lower levels of government. Many stakeholders shared a general concern over a lack of human resources, knowledge, experience and skills relating to disaster risk management and a lack of government initiative from sub-national authorities. Contributing to the lack of capacities at sub-national levels are high staff turn-over and a lack of political will to engage in disaster risk management due to many other responsibilities and the prioritisation of other issues that are considered more important. There is also a lack of trust amongst the public in government institutions due to poor public services provision and to corruption. In Khao Lak and Phi Phi Island Calgaro et al. found that “emergency aid relief did not reach all eligible recipients” and the “available funds were often misappropriated due to corruption and nepotism operating at the local level” (p. 47) [34] . Many communities therefore have very little trust in the government’s commitment and capacity to develop effective disaster risk reduction strategies.

Additional challenges exist in the collaboration and communication between government and other stakeholders, such as universities, NGOs and CBOs and the private sector. Local government representatives are not trained to facilitate processes to engage with communities. NGOs tend to be much better at communicating with communities and have well-established methods. However, some local government authorities are reluctant to facilitate NGO initiatives in communities because they don’t want to relinquish authority to the NGOs. Despite this, NGOs often play an important role in supporting communities to initiate planning for community-based disaster risk management, to engage with the local government, to access information and guidance, and to receive financial support.

4.5 Key enabling and constraining factors

There were a number of factors that enabled collective action. Our research demonstrates that strong leaders with good social networks are an important enabling factor for community-based disaster risk management. Leaders include those who may not have a formal position but who nevertheless have influence in their communities [29] . For example, in the village of Koh Panyee, the Rescue Team draws on experience from the village health committee that was already well established. The chairman of the health committee has a key role in the Rescue Team, because he is perceived as competent due to his university education and computer skills [35] . Religious leaders can play an important role in building disaster preparedness because they can disseminate information on hazards and disaster preparedness initiatives in their services.

Many young people volunteer their time to these activities because of incentives such as free services such as health care, training in language and other skills that are beneficial for seeking employment and advancing careers. Volunteering also enhances social status and supports political advancement through local electorates in Tambon Administrative Organisations. Participation and ownership of procedures and early warning systems by the community is also an important enabling factor.

Another enabling factor is the integration of disaster risk management into strategies that address wider community priorities, such as improving and diversifying livelihoods and building capacity for community-based natural resource management. In our experience in Krabi Province, the ‘framing’ of activities is crucial in the process of partnership building and bringing people onboard, where the approach has to be expressed in terms that are relevant for the partners.

We observed a range of barriers to collective action on disaster risk management in Krabi Province. Because of limited resources and capacities of sub-national government actors, the different priorities and lack of political will of some local authorities, and insufficient coordination between local government and NGO actors [28] [34] , local government frequently represent a barrier to collective action.

While community leaders often play a critical role in enabling and facilitating action within their communities they can also hamper collective action. For example, some village leaders show no interest in disaster risk management despite interest of the community and this can create conflict and may lead to isolation from NGO and government activities. In Thailand, the village headman plays an important role because he is elected as a representative of the central government and he is elected for life. The political context at the community level is therefore strongly shaped by the politics at the national level and there have been cases of bribery and allocation of funds to relatives of the headman.

A lack of resources at the local level negatively affects the ability of committees to act and to induce positive change in their communities. Volunteers also often lack authority. For example, many young volunteers don’t feel comfortable telling older people what to do and many older people do not take them seriously. Many CBOs have limited legal status and are not recognised by government authorities as legitimate stakeholders in disaster risk reduction and early warning system planning and implementation processes. Whilst there is an increasing emphasis on participatory planning , participatory practices have not yet been mainstreamed into humanitarian action [36] [37] .

In some communities affected by the 2004 tsunami, the importance of disaster preparedness is not fully understood because there is no history of disasters. The Thai philosophy of life “Mai Pen Rai” (English equivalent “Not to worry” or “Never mind”) could be interpreted as complacency. In Muslim society, religious and cultural beliefs about predetermined destiny (fatalism) are often difficult to overcome.

Some government authorities are concerned about the negative image disaster preparedness activities might shed on tourism communities as safe and pristine destinations. The value of community-based disaster risk management and community empowerment is contested; sometimes it is difficult to convince people that investing in disaster risk management is as important as investing in livelihoods and that activities might create co-benefits for both.

Despite the dedication of volunteers, there is concern about the longer-term sustainability of disaster preparedness efforts that rely to a large extent on volunteerism because even volunteers require basic financial support for operational logistics such as transport, food, and compensation for the loss of income. The high turnover of volunteers and the need to continuously recruit and train new people puts a considerable strain on organisational capacities.

Guidance for community-based disaster risk management is not always available to communities or directly useful in the local environmental and socio-economic context of a particular community. In Ban Tha Klong, Thailand, the village committee described that the government provides information and seminars on tsunamis, landslides, and sea-level rise but that there was a lack of access to information on natural resource management and experience, good practice and technical guidance on disaster risk reduction. Several villages committees told us that in order to plan disaster risk management activities, they require detailed information about the community, including infrastructure, population distribution and density, location of vulnerable social groups, geographical maps of the terrain, disaster areas, and tourism areas. These data are usually held by government authorities and requests from village committees to obtain such information are not always successful. However, one could question whether such data-driven planning is useful for local action.