There is a growing consensus that flood risks in urban areas are increasing due to climate-induced changes to precipitation patterns, rising sea levels, higher peak discharges of rivers and canals, and increased urban development (Mitchell 2003, IPCC 2012). These changes challenge the flood resilience of urban areas. In accordance with the ideas of the Resilience Alliance (https://www.resalliance.org/resilience), we consider urban areas to be flood resilient if they have the capacity to absorb or withstand floods in such a way that the urban system remains within the same regime, essentially maintaining its structure and function. Governance systems are needed to enhance the flood resilience of urban areas.
Flood-resilience-enhancing governance systems tend to be multilevel. Activities within the policy cycle and in learning processes occur at spatially different administrative levels (Pahl-Wostl et al. 2013). Because administrative borders often do not align with the boundaries of the water systems involved, multiple administrative units will be involved in flood risk governance. The elaboration and implementation of flood risk management measures by these units requires coordination to avoid problem shifting from one unit to another. In addition to this requirement, the involvement of multiple levels of policymaking may also be the result of institutionalized solidarity between areas. The latter implies that money needs to be transferred from one region to another in cases in which risks faced by the receiving areas are generally considered as unacceptably high (Jongman et al. 2014). In both cases, higher governance levels may play a coordinating role. However, this role may differ between centralized and decentralized systems. Within a multilevel governance system, centralization is high if strategic goals and policies are only formulated at the highest level, and lower levels implement them according to directions from above. In centralized systems, actors from the highest level take a leading role in the implementation of measures on both large and small geographical scales. In contrast, decentralization of power makes multilevel governance systems more polycentric. In these systems, decentralized authorities develop their own strategic goals and tailored policies. Centralized regimes tend to have a lower adaptive and transformative capacity than do polycentric systems, provided that in the latter case, the decentralization of power is combined with effective coordination and a balance between bottom-up and top-down processes (Pahl-Wostl et al. 2013).
Thus, coordination mechanisms are required to organize the activities that occur at the different levels, both in centralized and more decentralized flood risk governance systems. It is, however, not clear what characteristics these mechanisms have. So far, literature points to the multilevel aspects of managing urban flood resilience in rather general ways, by indicating that these aspects are relevant (Bulkeley 2010) or that decentralized flood risk governance approaches have limitations (Hanssen et al. 2013, Johannessen and Hahn 2013, Stevens and Hanschka 2014), or by including multilevel aspects as elements of a framework for profiling the characteristics of risk governance (Walker et al. 2014). To address this knowledge gap, we analyze which multilevel coordination mechanisms are required to enhance the flood resilience of urban areas, and what factors account for their importance.
We first develop a framework in which we conceptualize urban flood resilience and theorize on which strategies and which multilevel coordination mechanisms will be needed for managing urban flood resilience. We then assess six cases of managing urban flood resilience in urban agglomerations in six different European Union (EU) countries to analyze the use and characteristics of these mechanisms. Finally, we discuss the ways in which the coordination mechanisms differ among the cases and what roles are played by different contextual factors.
To enhance flood resilience, urban areas need the capacity to resist, to absorb and recover, and to transform and adapt (Hegger et al. 2016). Following Hegger et al. (2016), we define the capacity to resist as the ability not to be adversely affected by floods, by raising the threshold above which floods can cause harm. More specifically, this capacity to resist implies a timely implementation of effective structural or upstream retention measures that allow resistance to floods. Urban areas have a capacity to absorb and recover if they remain functioning, respond to a flood, and recover without shifting to a different system state. This capacity can be realized by timely implementation of effective measures that allow for the response to and recovery from floods: presence of flood awareness, flood mitigation measures, insurance systems, forecasting and warning systems, and crisis management. To be able to transform and adapt, urban areas must have the ability to adjust to external drivers that affect the exposure of people and economic assets to floods, moderate potential damages, take advantage of opportunities, make deliberate small-scale changes, and cope with consequences. These abilities require the presence of institutionalized mechanisms for learning (e.g., learning-action alliances, strong knowledge institutes) and evidence of the capability of (local) actors to adopt new flexible policies.
The implementation of the necessary capacities is challenging because it involves a multiplicity of flood risk management strategies (FRMSs; Hegger et al. 2014). For instance, flood defense, by constructing dikes or upstream retention areas, may be complemented by flood risk prevention (spatial planning measures), flood risk mitigation (flood-proof building), flood preparation (emergency plans), and flood recovery (rebuilding and insurance) strategies. This multiplicity is challenging compared to situations in which only a single strategy is chosen to address flood risks; additional actors are involved, and more activities need to be coordinated. Improving flood prevention, for example, requires cooperation between actors responsible for dikes and watercourses and actors responsible for spatial planning. Formal instruments requiring the consideration of water issues in spatial planning can contribute to such cooperation by clarifying the flood risks of planned urban developments (Immink 2005, Wiering and Immink 2006, Hartmann 2009, Hartmann and Driessen 2017). Improving flood warning systems and evacuation planning will necessitate the involvement of emergency planners. In cases of flood recovery, the number of actors broadens still further, including residents and insurance companies. Raising overall flood awareness will be a precondition for involving these different participants in flood risk management. Apart from these factors, it must be specified what strategies will be implemented where, and with what level of ambition. We argue that these challenges may be better addressed if certain coordination mechanisms are present (Dieperink et al. 2016).
First, policy entrepreneurs are needed to initiate and coordinate the processes at different levels of policymaking that aim to enhance urban flood resilience (Brown and Clarke 2007, Huitema and Meijerink 2009, Partzsch and Ziegler 2011). They must provide for visionary, entrepreneurial, and collaborative leadership (Van den Brink et al. 2011, Rijke et al. 2013) and must involve relevant stakeholders. To avoid too much complexity, some authors argue that stakeholders’ activation should be selective (Rowe and Frewer 2005, Green and Penning-Rowsell 2010); however, the literature also suggests that addressing problems from different perspectives necessitates the proactive involvement of a broader community (Garrelts and Lange 2011, Brugnach and Ingram 2012, Rijke et al. 2013). Apart from inspiring policy entrepreneurs, processes to enhance urban flood resilience also need dedicated, neutral, and skilled process managers. These managers must be able to organize meetings in such a way that participants are kept at the table and remain interested, and that mutual learning is promoted (Dieperink et al. 2012).
Second, managing urban flood resilience in a multilevel governance context may require the development of bridging concepts (Greiving and Angignard 2014), especially in cases where actors aim for a more diversified set of FRMSs. Bridging concepts may refer to abstract terms (such as resilience or sustainability) that facilitate communication across disciplinary borders by creating a shared vocabulary (Star and Griesemer, 1989), although parties may differ on the precise meaning of the term in question (Brand and Jax 2007). In contrast, more elaborated plans and visions may also serve as bridging concepts. Bridging concepts must connect different levels of policymaking by addressing a variety of policy frames and solutions (Van den Brink et al. 2011, Brugnach and Ingram 2012, Rijke et al. 2013) and may thereby inspire actors to find common problem definitions (Brugnach and Ingram 2012) and align and balance multiple objectives (Rijke et al. 2013). Bridging concepts must contain inspiring and convincing narratives that define a common ground for future cooperation, focusing, for instance, on acceptable risks, safety standards, the division of costs and benefits, and related funding principles (solidarity or user pays). The EU Floods Directive can be considered as an example of such a bridging concept.
Third, the existing system of formal and informal rules should be clear. Ideally, rules do not only aim to guarantee basic rights and equity (Van den Brink et al. 2011); constitutional and specific laws also define the autonomies and shared responsibilities of actors at different levels. Subsidiarity will be a guiding principle in the division of autonomies and responsibilities over different levels of governance. According to the sustainability principle, flood risk policies will have to be developed at the lowest appropriate level. Of course, what this means in practice is often contested (Dieperink and Driessen 2007). Ideally, rules clarify which actors at which levels have authority and policy freedom to use policy instruments such as standard setting, issuing permits, subsidizing, and tax raising. Rules may also institutionalize top-down hierarchical command-and-control interactions, as well as horizontal interactions in which formal positions between actors (operating on different levels or not) are more equal. Covenants may be used to initiate or reinforce the results of such multistakeholder processes (Edelenbos et al. 2009, Dieperink et al. 2012), for instance, by formalizing the status of developed strategic visions (Rijke et al. 2013).
Fourth, and related to the above ideas, attempts to enhance urban flood resilience will only be successful if multiple levels of governance provide for the required resources. Funds must be allocated to organize meetings and enable participation processes facilitating the connections between different actors and across different levels of governance (Dieperink et al. 2012). Teams must be established for monitoring and evaluating policy experiences (Van den Brink et al. 2011) and community building (Fleischhauer et al. 2012, Rijke et al. 2013). Managing urban flood resilience also necessitates the development of a shared knowledge base concerning systems’ functioning (Van Herk et al. 2011) and the mobilization and connection of different forms of knowledge (Brugnach and Ingram 2012, Hegger et al. 2014). Apart from these process-related resources, funding is needed for the actual implementation of the strategies. Someone must pay for the improvements in diking systems or (re)construction of buildings or sustainable urban drainage systems. The enhancement of urban flood resilience may benefit from a formal agreement concerning the funding required and those levels of governance that are responsible for resourcing.
To conclude, managing urban flood resilience will require the elaboration of multiple FRMSs, which implies that multiple levels of governance will be involved. Coordination between these levels is needed and can be achieved by proactive policy entrepreneurs, bridging concepts, clear rules, and the provision of necessary resources.
To further analyze what are the required characteristics of multilevel coordination mechanisms and what factors account for these characteristics, we conducted six in-depth case studies of flood-prone urban agglomerations: Dordrecht, the Netherlands; Kingston upon Hull (Hull), UK; Geraardsbergen, Belgium; Karlstad, Sweden; Wrocław, Poland; and Nice, France (Table 1). Since 1950, several flood events, i.e., submersions of parts of the city, have taken place in each city. All of the cities are susceptible to both fluvial and pluvial flooding. Nice and Hull are also vulnerable to coastal flooding. Wrocław seems to be the most vulnerable of the six cities. Within these urban agglomerations, (mixes of) FRMSs are discussed and implemented, requiring not only the engagement of local authorities, but also of actors operating at higher levels of governance. Four main administrative levels are involved in managing urban flood resilience in the Netherlands, UK, and France; five in Belgium; and three in Sweden and Poland. Apart from public authorities, companies and individuals are also involved.
The six cases have been studied in the context of the STAR-FLOOD project funded by the EU Seventh Framework Programme. Each case study is based on 15 (on average) semistructured in-depth interviews with key informants (civil servants, flood risk management professionals, academics, nongovernmental organizations [NGOs], and neighborhood groups) and a systematic analysis of policy documents. Relevant documents were selected after consulting the key informants. The resulting case descriptions were discussed in workshops at the national level and several times within the STAR-FLOOD Consortium to ensure case comparability. The case descriptions can be found at http://www.starflood.eu/country-reports/.
We found that coordination mechanisms comprising proactive policy entrepreneurs, bridging concepts, and the clear allocation of authorities and responsibilities are present in all six cases (Table 2). In all cases, different levels provide relevant knowledge and the necessary funding for flood risk management. We will elaborate on this in more detail for each case.
Most of the Island of Dordrecht (119,000 inhabitants) is protected by dikes (Hegger et al. 2014). However, the historical port, parts of the historical center, and more recently developed housing and business areas are all located outside the dike-protected area. If flooding occurs, it will be deep and rapidly endanger 1910 people (Veenstra-Huisman 2014). Material losses may be as high as €7 billion (Hegger et al. 2014).
Because the probability and impact of flooding in Dordrecht is expected to increase, the local authority of Dordrecht initiated a debate on managing flood risk by diversifying FRMSs (Kaufmann et al. 2015). Together with the regional water authority, they elaborated on the idea to become a more self-reliant island. Cooperation was sought with Sheffield and Rotherham (UK), Hannover (Germany), Bergen (Norway), and Seattle (USA) through the EU-funded MARE project (Managing Adaptive Responses to Changing Flood Risk in the North Sea Region, the North Sea Region Programme 2007–2013: http://archive.northsearegion.eu/ivb/projects/details/&tid=95). This project focused on developing a transnational methodology to implement urban flood risk management plans in four countries. In addition, the national government began the Dutch Delta program which, among other things, explored the feasibility of a multilayered safety (MLS) approach. The MLS approach requires the consideration of different measures in flood risk management: probability reducing measures (e.g., dikes), sustainable spatial planning, and emergency management. Water authorities were critical about the MLS approach and questioned whether measures in one layer could replace those in another, what this approach would mean for existing protection levels of the dike system, and who would consequently be responsible for flood risk management. After a societal cost-benefit analysis made clear that Dordrecht’s protection level was twice as low as economically optimal (Kelder et al. 2013:20), the Minister transferred the MARE project into a pilot study to implement the MLS approach. The pilot study was carried out by a consortium comprising the local authority, a consultancy firm, and Delft University of Technology, and proposed a diversified set of FRMSs.
First, tailor-made reinforcements of primary flood defenses (dikes) are envisaged. Second, restoring and strengthening old dikes will compartmentalize the island, which may reduce economic damage by a factor of five to ten and reduce to zero the expected number of victims. Third, the use of movable flood defense barriers to protect the entrances of existing houses is suggested, while the floor levels of new houses will be raised. In addition, the construction of floating houses is another option that is being considered. Fourth, it is argued that pro-active spatial planning should reduce urbanization in the south of the island, while compartmentalization will create a safe area in the northeastern part of the island, providing citizens with a place to go if evacuation is needed. Apart from redesigning existing buildings such as schools and hotels, additional new smart shelters will be constructed, and vital infrastructure in zones outside the main dike ring will be raised or protected. It is expected that 75% of the residents in the western and southernmost compartments of the island can be evacuated to areas within the island in case of a flood (Kaufmann et al. 2015).
The local authority of Dordrecht clearly acted as a policy entrepreneur. It had to play this role because it was dependent on other governance levels for the realization of its ambitions. The local authority only has a formal mandate for rain and sewer water collection. Other governance levels must also provide relevant resources. The improvement of secondary dikes must be financed by the regional water authority. The national level is not only involved in flood forecasting and visioning, but also in planning, financing, and implementing the flood protection program for primary flood defenses. The national level is also responsible for managing the Maeslantkering, a storm surge barrier, which also protects Dordrecht. At the regional level, the province can influence land use in unembanked areas via spatial planning, provincial building by-laws, and the provincial water plan. The latter addresses waterlogging, safety, and crisis management. The regional safety authority is responsible for implementing crisis management and has established a warning system and clarified its responsibilities and those of the regional and national water authorities. In crisis situations, a regional operational team will take the operational lead. These divisions of authority and responsibility appear to be clear.
Kingston upon Hull (> 250,000 inhabitants), situated on the east coast of northern England, at the confluence of the River Hull and the Humber Estuary, is naturally susceptible to fluvial, tidal, and pluvial flooding. Estimates suggest that as many as 44,000 of the 99,000 properties in the city are at risk from a 1% flood (River Hull Advisory Board 2016). This case study is indicative of the situation in England, where all FRMSs have been present for many years and in which overarching policy is often established at the national level but delivered at the local level. Additionally, working in partnership, a strong discourse in England, is also apparent in Hull, in the development of programs of measures, including Flood Alleviation Schemes (FASs).
Defense and land drainage are the dominant FRMSs in this case, and investment continues to maintain and strengthen fluvial and coastal defenses within the city. However, the significant threat posed by pluvial flooding was highlighted by the summer floods in 2007. This threat led to a reemphasis of the need to employ a diversified approach to managing flood risk, recognizing the natural context of this case’s flood risk, and ensuring coordination between upstream and downstream authorities. To mitigate pluvial flood risks, several FASs are currently at different stages of design, consultation, and construction, and involve different actors and interventions at multiple governance levels (European, national, local, and individual). The most advanced scheme, the Willerby and Derringham Flood Alleviation Scheme (WaDFAS) is based on the principle of upstream flood storage to reduce the risk of downstream flooding to > 8000 properties (HCC and ERYC 2018). Because of a lack of space in Hull, the development of the scheme was initiated jointly by the neighboring Hull City Council and upstream East Riding of Yorkshire Council in their capacity as Lead Local Flood Authorities, and the creation of a joint strategy. Although UK flood risk management policy has promoted a partnership approach for many years, this case still highlights the fundamental role of policy entrepreneurs, the recognition within both Lead Local Flood Authorities of the need to be creative, and the establishment of an innovative approach to flood risk management.
Furthermore, the involvement of additional actors was key, with public participation activities involving local citizens (e.g., parish meetings, public exhibitions, and a “flood bus”) being used to facilitate public dialogue and to demonstrate the benefits to both communities. Additionally, expropriation by a national legal instrument (the compulsory purchase order) was used for the first time in a flood context in England. This instrument was used because it was considered to be the best mechanism for gaining the necessary control of the large-scale flood storage area, although an agreement was reached with the landowner to lease back the land to retain its agricultural use.
The importance of sustainable urban drainage is also acknowledged in Hull City Council’s surface water management and flood mitigation investment plans (HCC 2009, 2011). Although retrofitting and property-level measures are not fully supported, there are efforts to mitigate risk exposure by enforcing designated floor heights according to a property’s location, and by accounting for flood risk standing advice provided by the Environment Agency, which operates at the regional level (HCC 2007). In addition, Hull City Council is exploring options for small-scale flood storage areas at recreational sites such as parks and football fields. To improve flood risk preparation, the Environment Agency has implemented a new flood warning service for Hull East, West, and City Centre using new modeling data to enhance the accuracy of flood forecasting, and communicating warnings via text, email, phone, or fax.
Despite a growing role for the local level, the implementation of the WaDFAS still draws on multiple levels of governance. Apart from a locally generated levy and other public monies generated at the local level, funding for the WaDFAS has come from the European Regional Development Fund and national flood risk management grant-in-aid funding. Other flood risk management schemes have also benefitted from more regionally generated financing through the Humber Local Enterprise Partnership. This situation of multiple funders has not only ensured that sufficient resources have been allocated to the schemes and to the whole process of their development, but to some extent, it has also acted as a facilitating bridging mechanism drawing the different levels of governance together to try to achieve the same goal. In general, the goals of economic development (this area has a high degree of economic deprivation) and flood mitigation have been tackled together. A major challenge in securing funding for the scheme relates to the methodology of the cost-benefit analysis upon which prioritization and funding is based. Despite being at risk from multiple sources of flooding, a property can only be counted once and is required to be assigned to a single flood alleviation project. To tackle this, the River Hull Integrated Catchment Strategy (River Hull Advisory Board 2016) was produced to test multiple solutions and, when implemented, to maximize benefit-cost ratios. This strategy is perceived as an important step forward in terms of facilitating joint prioritization of investments and a sequenced approach to defense and mitigation works (Alexander et al. 2016).
Geraardsbergen (32,800 inhabitants) is located on the River Dender, a 69 km long tributary of the River Scheldt. The city is protected by dike and sluice infrastructure. Following past flood events (in 1995, 1999–2000, 2002–2003), small-scale flood defense measures were first implemented (CIW 2009), but these could not completely prevent two submunicipalities (Overboelare and Zandbergen) from being flooded. Hence, it is generally acknowledged that the Dender basin needs an integrated and diversified approach, which not only relies on flood protection but also on prevention and preparation measures (Mees et al. 2016). Adopting a diverse approach requires cooperation between actors at four different government levels.
The division of responsibilities in Flanders’ flood risk governance is clear. The Flemish government is responsible for the management of the Dender and its main tributaries (Mees et al. 2016). Until recently, the City of Geraardsbergen authority was responsible for smaller “category 3” watercourses, but in 2014, it transferred this competence to the province. Spatial planning has many governance levels, but the City of Geraardsbergen has a key position because it grants building permits. The provincial and regional water managers do, however, offer guidance and expertise through water assessment advice and development of flood cartography. Moreover, the Flemish government defines the legislative framework on spatial planning. The city is also the main actor in emergency planning. In this task, it is supported by the province and the federal level through legislative prescriptions, guidance, training, and financial and material resources. In terms of funding, the management of watercourses is financed by the provincial and regional level; crisis management is the financial responsibility of the city, but with support from the federal level. Flood recovery is funded through a private insurance system, and noninsurable damage can, in certain cases, be refunded by the Flemish disaster fund. Until 2014, the latter used to be a federal fund, but is now regionally provided.
Since the beginning of the 2000s, a discourse has been prevalent among actors within the Flemish and provincial government that “water needs space,” and thus, increasing importance is given to the field of spatial planning (Mees et al. 2016). The 2003 Decree on Integrated Water Policy initiated the water assessment, according to which local authorities must consult with provincial and national water managers before they can grant building permits. Before the 2010 flood, the application of the water assessment in Geraardsbergen was rather loosely implemented because the advice given by flood risk managers was not binding.
The 2010 flood turned out to be a trigger event for enhanced multilevel cooperation. The water assessment is still not binding, but it is very rarely deviated from. The flood also demonstrated deficiencies in the communication between local emergency planners and the water management sector. Today, contact details are shared between the different domains and levels in a collectively developed manual. In general, communication and cooperation between the governmental levels involved has improved through regular meetings and personal contacts (Mees et al. 2016). The local authority acted as policy entrepreneur and has appointed a coordinator on integrated water policy to organize activities between actors within and outside the municipal administration. Apart from this, the provincial governor united all relevant actors within the Dender sub-basin board in a series of intergovernmental meetings. The Flemish government is currently drafting water action plans at the local level based on prevention, protection, and preparation (CIW 2015). In a first phase, a cost-benefit analysis has been conducted to define the most appropriate measures. In a next phase, there will be negotiations with local governmental actors regarding which flood risk management measures will be implemented, and by whom.
Karlstad (89,000 inhabitants) is in the southwest of Sweden and is vulnerable to both fluvial and pluvial flooding. A 1% flood is estimated to cost about 10.5 billion SEK (€1.1 billion) and flood approximately 25,000 homes. In contrast with the national level in Sweden, where flood risk management is almost nonexistent, flood risk governance is placed high on the political agenda in Karlstad.
The priority of flood risk management is manifested through a local flood management program, which was developed after the 2000–2001 floods and was also inspired by a national investigation of major climate change-induced risks and vulnerabilities (Swedish Commission on Climate and Vulnerability 2007). The program constitutes a comprehensive approach to flood risk management, pointing at the necessity to develop a portfolio of different FRMSs. So far, mainly temporary defenses have been used in Karlstad, but the flood management program proposes some permanent defense structures to be developed (Karlstad Local Authority [Karlstads kommun] 2010). The pursuit of diversification is visible also in the Karlstad Water and Sewerage Plan, which expresses the need to develop sustainable stormwater management and increased reliance on solutions that try to mimic nature’s way of handling stormwater (Ek et al. 2016). Flood prevention and mitigation are primarily handled within the local authority, whereas emergency services are organized in cooperation with four other municipalities.
Although flood risk management in Sweden is mainly a matter for the municipalities, higher governance levels do play a role. The national level provides limited funding for implementing measures to reduce flood risk that the municipalities may apply for, but the local level acts as a policy entrepreneur because it carries the main responsibility for planning, financing, and implementing the flood management program. The local level has the authority to levy taxes. A municipal flood risk manager is responsible for coordinating the work related to the implementation of the program and oversees the financial resources earmarked for flood risk-related investments (Karlstad Local Authority [Karlstads kommun] 2012). The flood risk manager is also responsible for contributing to knowledge building (Karlstad Local Authority [Karlstads kommun] 2010). In addition to the limited funding, the national level sets rules and guidelines and provides support and knowledge, e.g., through building regulations and guidelines on how mitigation measures can be implemented (National Board of Housing, Building, and Planning [Boverket] 2010b). The building regulations constitute a bridging mechanism between the local level where the prevention and mitigation strategies are implemented and the national rule-setting level. The regional level is represented by the County Administrative Board of Värmland, which is a regionally located national agency that ensures that national rules and guidelines are implemented and provides support and recommendations to the local level (County Administrative Board of Västra Götaland and County Administrative Board of Värmland [Länsstyrelsen i Västra Götalands och Värmlands Län] 2011). Municipalities willing to allow developments in flood-prone areas need to clarify in a detailed plan what measures must be taken to secure a building (National Board of Housing, Building, and Planning [Boverket] 2010a), and the County Administrative Board must approve these municipal plans before development can commence. Although the County Administrative Board holds the right to revoke the plans (Planning and Building Act, chapter 10, page 10), this is rarely done, which may be because of the involvement of the authority at earlier stages of the planning process.
Wrocław (631,377 inhabitants) is one of Poland’s largest cities. It lies in the flood-prone area of the Odra basin and has suffered from discharges from the Odra and its tributaries, the Oława, Ślęza, Bystrzyca, and Widawa. The length of all rivers within the city borders is approximately 100 km, with possible losses from a 1% flood expected to be > €700 million. Following the 1903 flood, significant defense structures were constructed, and for a period of 50 years, new development was prohibited on the most endangered areas; however, this restriction was later relaxed. An unusually high level of precipitation (Dubicki et al. 1999, Kundzewicz et al. 1999) and a lack of adequate protection structures and capacities made Wrocław the biggest victim of the 1997 “millennium flood.” In 2010, another serious flood hit the city, although losses were smaller because the city was better prepared.
The traumatic event of 1997 not only brought devastating losses but also stimulated spontaneous citizen involvement during the crisis and in recovery actions. The mayor played a key role in creating a sense of community in the city (Sitek 1997, Kubicki 2010). In the aftermath of the flood, Wrocław became a center for discussions on flood management, including environmental advocates such as the World Wildlife Fund and other NGOs. Community awareness-raising activities were established, and a local flood leaders’ system was established. Intervention and evacuation plans and crisis communication were reconsidered, and small-scale investments were made in sand bags and pumps.
The 1997 flood had impacts beyond the local level of policymaking. It accelerated the creation of a new Water Act (finalized in 2001) and initiated substantial reconstructions of the preparation strategy (finalized with an Act on Crisis Management in 2007). A new system of emergency management was established, with a significant role for the local authorities to cooperate with the main operational force, the fire brigades. Flood forecasting and warning systems were also improved. In addition to this, a Program for the Odra River was elaborated by the World Bank. This large collaborative investment program facilitated infrastructure investments, which not only improved the defense system of the Wrocław Water Junction, but also strengthened the position of the flood defense system on the regional policy agenda. Apart from aiming to coordinate investments in flood defense measures, inland waterways, and afforestation initiatives, it also addressed water quality, spatial planning, and insurance incentives (Pelnomocnik Rzadu do Spraw Programu dla Odry 2011). The Program for the Odra River offered a window of opportunity for local flood management actors to improve their cooperation and bargaining skills and obtain experience in raising external funds. However, despite an initially diversified approach, the program eventually evolved into a hard-defense and infrastructure-oriented action plan. In addition to modernization of the Wrołcaw Water Junction, upstream retention reservoirs were also included (such as in Raciborz). The 2010 flood resulted in the construction of an additional dike.
The inconsistency of the structural dominance of the Program for the Odra with the more environmentally oriented EU Water Framework Directive (that Poland was required to implement following EU accession in 2004) led to the program’s discontinuation by the end of 2014. Poland’s accession to the EU not only made additional funds available, but also increased transparency in water and flood risk management and the importance of environmental concerns. In addition to the EU, the International Commission for the Protection of the Odra River also has influence. This was later established after the millennium flood (which also affected other countries) and offers a platform for the discussion of the transboundary aspects of flooding with fellow German and Czech stakeholders.
Following the decentralization and administrative reform of 1999, the maintenance of levees and infrastructure became the responsibility of the regional governments. However, some ambiguities exist concerning the responsibilities for smaller watercourses. Moreover, because of a lack of competences, regional planners have difficulty in resisting municipalities’ ambitions to develop in flood-prone areas. At the same time, local-level advocacy work of some environmental NGOs resulted in the demolition of some levees in the Odra basin. Another challenge concerns the instability of financing and competition for funds. The state Institute for Meteorology and Water Management, the main provider of data and expertise, is underfinanced and needs to focus on offering commercial services.
Like many other Mediterranean cities, Nice has a high urban density and an economy highly dependent on tourism as well as on housing, business, and transport systems. The risk of flooding in Nice (343,304 inhabitants), located in the Region Provence-Alpes-Cote-d’Azur in the southeast of France, mainly comes from the River Var, although concerns about marine submersion are rising. In the recent past, flooding has caused high amounts of damage and even casualties (Larrue et al. 2016). The Var has the typical characteristics of Mediterranean rivers, which are subject to sudden, violent, torrential flooding. In the case of the Var, upstream infrastructure works have resulted in an acceleration of river flow, thus increasing flood risks.
Urban flood risk management in Nice is undertaken by different actors at different levels of government. Although national legislation on flood prevention has existed since the 1990s, the Flood Risk Prevention Plan on the River Var was only finally approved by the Prefect (the local representative of the state) in 2013. The departmental level is responsible for the maintenance of defence infrastructure and, together with the municipal level, is also in charge of preparation actions and the funding of the fire brigade. The local and intermunicipal levels thus play a central role in prevention and mitigation at the basin level. Water management and development plans are being developed to coordinate departmental and intermunicipal activities in the field of flood risk.
The main financing tool for flood risk management at the level of the river basin, the Action Program for Flood Prevention (PAPI) for the Var River, must integrate all FRMSs (apart from recovery) and is the main instrument for coordination among state, departmental, and local levels. Inclusion in a PAPI is a requirement for FRMSs’ funding. However, the cohesiveness of PAPIs is questioned, as in practice they turn out to be a collection of measures (to be financed) instead of a cohesive and integrated multistrategic flood risk management program. In reaction to this fragmentation, a group headed by the Conseil Départmental of Alpes Maritimes (provincial level) has operated as a policy entrepreneur by initiating an informal dialogue about a local strategy. This strategy must coordinate actions and actors within the territory of Alpes Maritimes. The Conseil Départemental has a central position in water governance of the Var basin. Apart from being in charge of the local strategy for flood prevention and of the main Water Planning Framework, it manages defense infrastructure and plays a mediating role among the different local actors (municipalities, the Prefect, the water police). It also provides technical knowledge, particularly on defense and mitigation.
The mayor of Nice has acted as a policy entrepreneur by initiating the so-called Eco-Valley project, which was translated in 2009 into the Operation of National Interest Var Plain, which was launched by the national government together with the intermunicipal authorities. The Eco-Valley master plan is a bridging concept that aims to develop the Var Plain. The Eco-Valley project aims to construct a future city consisting of a business center; a multimodal transport hub; a high-tech manufacturing, food, and horticultural production center; and an ecoresidential quarter in the western part of Nice. An assessment procedure (Hydraulic Consistency and Global Development Scheme) has been set up to certify that the system of defense is suitable for new development in the area. However, flood risk management is not a priority of the master plan, but a subsidiary action. Local associations have criticized the project for both negligence of flood risks and a lack of transparency.
The Etablissement Public d’Aménagement (EPA), the leading authority of the Eco-Valley project, gathers representatives from national, regional, departmental, and metropolitan levels and is also responsible for balancing the national flood policy requirements as laid out in the PAPI and the development ambitions of the project. The EPA executive board is headed by the metropolitan level. This provides for a coordination mechanism for flood risk management because of the proximity of policy sectors within the city administration (urban planning, the Mayors’ police power, the management of protection infrastructures and mitigation initiatives).
In all six cases, the capacities to resist, to absorb and recover, and to transform and adapt have been further developed. Thus, all cases, to a certain degree, exemplify successful multilevel flood risk governance. The cases make clear that managing urban flood resilience can be undertaken by the elaboration of different mixes of FRMSs. These different mixes of FRMSs are elaborated by a multiplicity of actors operating at different, mutually dependent, levels of governance. Flood risk governors operating at the urban level are dependent on higher levels because the elaboration and implementation of FRMSs often takes place at a higher geographical level. Higher levels are dependent on lower levels to implement tailor-made solutions and portfolios of measures (from all FRMSs), to learn from experiences, and to upscale good urban practices. Policy entrepreneurs, bridging concepts, clarity of responsibilities, and resource provisions are all in place to coordinate the different levels. Their key characteristics appear to be highly context specific. Apart from this, the cases also made it clear that some additional coordination mechanisms must be incorporated in our framework.
Different actors can play the role of policy entrepreneurs (Table 2). They promote initiatives, bring actors together, and collect finances and resources. Most policy entrepreneurs tend to be public actors (especially local governments), but in Wrocław, an NGO operated as a policy entrepreneur. This organization secured support at different levels for its ambition to introduce environmental concerns into the flood risk management debate. The cases also indicated that multiple policy entrepreneurs can be active at the same time. In Nice, the entrepreneurial role of the Conseil Départemental mainly provides coherence among the different actors and FRMSs at a local level, whereas the city Mayor and Metropolitan Council opted for an integration of flood risk management and development ambitions.
In all of the cases, bridging concepts are used, but their geographical scales and the amounts of freedom they offer to the actors involved differ. The MLS and “space for water” approaches are nongeographically bounded visions, whereas catchment plans for the Var (PAPI) or Odra, Karlstad’s flood management program, or project development plans have a more area-specific focus. In Hull, joint strategic flood risk management planning bridges upstream communities and the urban area, but this is not the case in Wrocław. Contrary to the Var plan, in which all FRMSs are addressed, the Odra plan shows a dominance of the flood defense strategy. Because of a lack of expertise, nonstructural FRMSs have little potential in Wrocław. The prefix “Eco” in Nice’s Eco-Valley project evokes the dual nature (or ambiguity) of this master plan because it can mean either “economic” or “ecological” to bridge actors of all levels and natures (e.g., public, private, and potentially from civil society). Also, in Hull, the funding for flood risk management approaches was linked to the generation of economic development and urban renewal.
In general, the division of authorities over different levels is clearly formalized in our cases. Subsidiarity seems to be the leading principle in the division of responsibilities over the different levels. This also implies that actors at different levels provide urban areas with the required resources for managing flood resilience. Flood risk governance is indeed funded from different sources from different levels. However, in Karlstad, the municipality has the main responsibility for funding flood risk management measures, and it also has the authority to collect taxes. Karlstad, like all other municipalities in Sweden, can apply for limited financial support from a national fund. In all of the cases, scientific research on flood forecasting and innovative technologies, as well as the actual weather and flood forecasting, is provided at the national level. However, local research units tend to work closely with local administrations. The different levels involved in Dordrecht share their knowledge through policy documents, reports, and strategy-specific working groups in which all levels participate. The Dordrecht case also shows that the local authority can act as a knowledge provider for higher levels of governance because the MLS approach is explicitly established as a learning project. The EU also operates as resource provider. It enhances mutual learning, as shown in the Dordrecht case by the EU-funded MARE project, and provides for additional funding, as took place in Hull and Wrocław. In the case of Wrocław, additional funding for large-scale retention measures was provided by the World Bank.
However, our case studies also made clear that our framework was incomplete. Based on the analyses, we argue that two additional coordination mechanisms must be added to our framework. In addition to a clear division of tasks and responsibilities, we argue that the presence of formal hierarchies also contributes to coordination. Even in Karlstad, which was the most decentralized case, local-level plans must be approved by actors operating at a higher level. We think that such an approval mechanism must be considered to be a separate coordination mechanism. We also found that apart from formal hierarchies, coordination can also be achieved by the development of specific coordination bodies in which actors from different levels exchange their ideas and try to find synergies and win-win situations. This mechanism must also be added to our framework.
All coordination mechanisms identified in our framework appeared to be present in the cases studied. However, their characteristics differ because of several mutually related contextual factors. Geographic factors not only define what flood risks occur but also what FRMSs are feasible within a certain spatial setting. The Dordrecht case highlights that flood defense measures may be difficult to implement in highly populated and congested urban areas. In the Var valley in Nice, mitigation measures such as flood-control areas are not considered a viable option. In such situations, alternative strategies tend to be necessary, each of which brings an additional network of involved actors whose activities must be coordinated through the mechanism identified. Cultural and economic factors also matter, as well as the general institutional context. Because of its risk-avoiding culture, protection standards for dikes are much stricter in the Netherlands as compared to, e.g., the UK. In the latter, temporary flooding seems to be more accepted. Economic factors also account for specific actor involvement, and this appears to be the case in Hull and Nice, where the realization of co-benefits is required for obtaining the necessary funds. Moreover, a lack of national funding in the Wrocław case paved the way for the involvement of the World Bank and, to a lesser degree, the EU. Finally, we found that the already existing institutional context defines, to a large degree, which actors will be involved in new urban flood risk governance initiatives. To reduce transaction costs, these initiatives will be discussed in already existing coordination bodies, instead of setting up new bodies. In more decentralized flood risk governance systems such as Karlstad, national actors will not play key roles in the coordination mechanism present. Existing institutional contexts tend to be rather long-standing. However, Poland, following the collapse of communism (1989–1990), underwent a rapid and deep transformation of its administrative, economic, and political system, resulting in a decentralization of policymaking competences. The latter necessitates a further elaboration of coordination mechanisms.
Our comparative case study approach was not only challenging but also inspiring. As a result, several topics for future research on coordination mechanisms have emerged. Because “what works, where, when, and how” seems to be highly context specific, a more in-depth analysis of the effects of contextual factors on the coordination mechanisms is required. More in-depth research about the role of policy entrepreneurs might provide, for instance, insight into factors that define the strategies they use to frame common interests or mutually supplementary roles of different policy levels. An in-depth focus on bridging concepts may better clarify what conditions may be needed, at different levels of policymaking, to ensure their effectiveness. For instance, must plans have formal legal status to play a bridging role? And, if so, what conditions matter? In-depth studies of successful cases of managing urban flood resilience might reveal if an overlap in authorities and responsibilities between governance levels may be necessary and what characteristics such an overlap may have. Furthermore, it is relevant to gain more insight into the cultural and political conditions under which hierarchical steering in flood risk governance is accepted by society.
We analyzed which coordination mechanisms are required to enhance urban flood resilience. We conclude that, apart from a clear allocation of responsibilities, the presence of formal hierarchical relations, coordinating bodies, and bridging concepts (visions, water assessment plans and programs) are needed to connect different levels of governance. Policy entrepreneurs must embrace these bridging concepts to stimulate “out-of-the-box” thinking by flood risk managers, spatial planners, and other stakeholders, while also enabling the integration of feasible options into concrete policy plans. Furthermore, coordination can be achieved by the provision of funding and knowledge. The more specific characteristics of these six coordination mechanisms depend on geographical, cultural, and institutional contextual factors.
The work described in this publication was supported by the European Union’s Seventh Framework Programme through the grant to the budget of the Integrated Project STAR-FLOOD, contract 308364. The authors thank Dries Hegger and Peter Driessen as well as three anonymous reviewers for their valuable comments on earlier drafts of this paper.
Alexander, M., S. Priest, A. P. Micou, S. Tapsell, C. Green, D. Parker, and S. Homewood. 2016. Analysing and evaluating flood risk governance in England: enhancing societal resilience through comprehensive and aligned flood risk governance. STAR-FLOOD Consortium. Flood Hazard Research Centre, Middlesex University, London, UK. [online] URL: http://www.starflood.eu/documents/2016/03/wp3-en-final-webversion.pdf
Brand, F. S., and K. Jax. 2007. Focusing the meaning(s) of resilience: resilience as a descriptive concept and a boundary object. Ecology and Society 12(1):23. http://dx.doi.org/10.5751/ES-02029-120123
Brown, R. R., and J. M. Clarke. 2007. Transition to water sensitive urban design: the story of Melbourne, Australia. Report 07/1. Facility for Advancing Water Biofiltration, Monash University, Melbourne, Australia. [online] URL: http://www.monash.edu.au/fawb/publications/final-transition-doc-rbrown-29may07.pdf
Brugnach, M., and H. Ingram. 2012. Ambiguity: the challenge of knowing and deciding together. Environmental Science and Policy 15(1):60-71. http://dx.doi.org/10.1016/j.envsci.2011.10.005
Bulkeley, H. 2010. Cities and the governing of climate change. Annual Review of Environment and Resources 35:229-253. http://dx.doi.org/10.1146/annurev-environ-072809-101747
County Administrative Board of Västra Götaland and County Administrative Board of Värmland (Länsstyrelsen i Västra Götalands och Värmlands Län). 2011. Stigande vatten: en handbok för fysisk planering i översvämningshotade områden. Länsstyrelsen i Västra Götalands och Värmlands Län, Göteborg and Karstad, Sweden. [online] URL: http://www.lansstyrelsen.se/gotland/SiteCollectionDocuments/Sv/Milj%C3%B6-och-klimat/Klimat-och-energi/Stigande-vatten.pdf
Coördinatiecommissie Integraal Waterbeleid (CIW). 2009. Het bekkenbeheerplan van het Denderbekken: integraal waterbeleid in de praktij 2008–2013. CIW, Erembodegem, Belgium. [online] URL: http://www.integraalwaterbeleid.be/nl/bekkens/denderbekken/documenten-denderbekken/7_DENDER_BBP_LR.pdf
Coördinatiecommissie Integraal Waterbeleid (CIW). 2015. Aanpak wateroverlastproblematiek: transitie naar meerlaagse waterveiligheid (MLWV): rapportering svz 2015. CIW, Erembodegem, Belgium. [online] URL: http://www.integraalwaterbeleid.be/nl/publicaties/afbeeldingen/2015-07-23_PhD_4derapp-wateroverlast_bijlage1.pdf/view
Dieperink, C., R. Boesten, J. Hovens, and H. Tonkes. 2012. Sustainable coastal development and open planning? Transferring the integrated area approach to Bulgaria. Sustainable Development 20(1):58-70. http://dx.doi.org/10.1002/sd.464
Dieperink, C., and P. P. J. Driessen. 2007. Gelaagde beleidsvoering: verticale taakverdeling en organisatie in het milieubeleid. Pages 239-264 in P. P. J. Driessen and P. Leroy, editors. Milieubeleid: analyse en perspectief. Coutinho, Bussum, The Netherlands.
Dieperink, C., D. L. T. Hegger, M. H. N. Bakker, Z. W. Kundzewicz, C. Green, and P. P. J. Driessen. 2016. Recurrent governance challenges in the implementation and alignment of flood risk management strategies: a review. Water Resources Management 30(13):4467-4481. http://dx.doi.org/10.1007/s11269-016-1491-7
Direction Régionale de l’Environnement, de l’Aménagement et du Logement Provence-Alpes-Côte-d’Azur (DREAL PACA). 2017. Population et emplois impactés par les surfaces inondables par débordement de cours d’eau. DREAL PACA, Marseille, France. [online] URL: http://www.rhone-mediterranee.eaufrance.fr/docs/dir-inondations/cartes/nice/TRI_NICE_CE_INDICATEUR.pdf
Dubicki, A., H. Słota, and J. Zieliński, editors. 1999. Dorzecze Odry: monografia powodzi, lipiec 1997. Instytut Meteorologii i Gospodarki Wodnej, Warsaw, Poland.
Edelenbos, J., P.-J. Klok, and J. van Tatenhove. 2009. The institutional embedding of interactive policy making: insights from a comparative research based on eight interactive projects in the Netherlands. American Review of Public Administration 39(2):125-148. http://dx.doi.org/10.1177/0275074008317157
Ek, K., S. Goytia, M. Pettersson, and E. Spegel. 2016. Analysing and evaluating flood risk governance in Sweden: adaptation to climate change? STAR-FLOOD Consortium, Utrecht, The Netherlands. [online] URL: http://www.starflood.eu/documents/2016/03/wp3-sw-final-webversion.pdf
Environment Agency. 2010. Hull and coastal streams catchment flood management plan: managing flood risk. Environment Agency, Leeds, UK. [online] URL: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/289265/
Fleischhauer, M., S. Greiving, F. Flex, M. Scheibel, T. Stickler, N. Sereinig, G. Koboltschnig, P. Malvati, V. Vitale, P. Grifoni, and K. Firus. 2012. Improving the active involvement of stakeholders and the public in flood risk management – tools of an involvement strategy and case study results from Austria, Germany and Italy. Natural Hazards and Earth System Sciences 12:2785-2798. http://dx.doi.org/10.5194/nhess-12-2785-2012
Garrelts, H., and H. Lange. 2011. Path dependencies and path change in complex fields of action: climate adaptation policies in Germany in the realm of flood risk management. Ambio 40(2):200-209. http://dx.doi.org/10.1007/s13280-010-0131-3
Green, C. H., and E. C. Penning-Rowsell. 2010. Stakeholder engagement in flood risk management. Pages 372-385 in G. Pender and H. Faulkner, editors. Flood risk science and management. Wiley-Blackwell, Oxford, UK. http://dx.doi.org/10.1002/9781444324846.ch18
Greiving, S., and M. Angignard. 2014. Disaster mitigation by spatial planning. Pages 287-302 in T. Van Asch, J. Corominas, S. Greiving, J. P. Malet, and S. Sterlacchini, editors. Mountain risks: from prediction to management and governance. Springer, Dordrecht, The Netherlands. http://dx.doi.org/10.1007/978-94-007-6769-0_10
Hanssen, G. S., P. K. Mydske, and E. Dahle. 2013. Multi-level coordination of climate change adaptation: by national hierarchical steering or by regional network governance? Local Environment 18(8):869-887. http://dx.doi.org/10.1080/13549839.2012.738657
Hartmann, T. 2009. Clumsy floodplains and the law: towards a responsive land policy for extreme floods. Built Environment 35(4):531-544. http://dx.doi.org/10.2148/benv.35.4.531
Hartmann, T., and P. P. J. Driessen. 2017. The flood risk management plan: towards spatial water governance. Journal of Flood Risk Management 10(2):143-144. http://dx.doi.org/10.1111/jfr3.12077
Hegger, D. L. T., P. P. J. Driessen, C. Dieperink, M. Wiering, G. T. T. Raadgever, and H. F. M. W. van Rijswick. 2014. Assessing stability and dynamics in flood risk governance: an empirically illustrated research approach. Water Resources Management 28(12):4127-4142. http://dx.doi.org/10.1007/s11269-014-0732-x
Hegger, D. L. T., P. P. J. Driessen, M. Wiering, H. F. M. W. van Rijswick, Z. W. Kundzewicz, P. Matczak, A. Crabbé, G. T. Raadgever, M. H. N. Bakker, S. Priest, C. Larrue, and K. Ek. 2016. Toward more flood resilience: Is a diversification of flood risk management strategies the way forward? Ecology and Society 21(4):52. https://doi.org/10.5751/ES-08854-210452
Huitema, D., and S. Meijerink, editors. 2009. Water policy entrepreneurs: a research companion to water transitions around the globe. Edward Elgar, Cheltenham, UK. http://dx.doi.org/10.4337/9781849803366
Hull City Council (HCC). 2007. Hull City Council strategic flood risk assessment. Hull City Council, Hull, UK. [online] URL: http://www.hullcc.gov.uk/portal/page?_pageid=221,578325&_dad=portal&_schema=PORTAL
Hull City Council (HCC). 2009. Hull City Council surface water management plan and aqua green project. Hull City Council, Hull, UK. [online] URL: http://www.hullcc.gov.uk/pls/portal/docs/PAGE/HOME/PLANNING/PLANNING%20POLICY/
Hull City Council (HCC). 2011. Hull Flood Mitigation Investment Plan: report of the City Planning Manager on behalf of the Corporate Director for Regeneration to the Integrated Strategic Drainage Partnership and Planning Committee. Report Number CDR/2011/035. Hull City Council, Hull, UK. [online] URL: https://cmis.hullcc.gov.uk/cmis/Document.ashx?czJKcaeAi5tUFL1DTL2UE4
Hull City Council and East Riding of Yorkshire Council (HCC and ERYC). 2018. Willerby and Derringham flood alleviation scheme (WaDFAS). Civil Engineering Services, East Riding of Yorkshire Council, Beverly, UK. [online] URL: http://www2.eastriding.gov.uk/EasySiteWeb/GatewayLink.aspx?alId=229415
Immink, I. 2005. Established and recent policy arrangements for river management in The Netherlands: an analysis of discourses. Pages 387-404 in B. Tress, G. Tress, G. Fry, and P. Opdam, editors. From landscape research to landscape planning: aspects of integration, education and application. Springer, Berlin, Germany.
Intergovernmental Panel on Climate Change (IPCC). 2012. Summary for policymakers. Pages 1-19 in C. B. Field, V. Barros, T. F. Stocker, D. Qin, D. J. Dokken, K. L. Ebi, M. D. Mastrandrea, K. J. Mach, G.-K. Plattner, S. K. Allen, M. Tignor, and P. M. Midgley, editors. Managing the risks of extreme events and disasters to advance climate change adaptation: summary for policymakers. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK.
Johannessen, A., and T. Hahn. 2013. Social learning towards a more adaptive paradigm? Reducing flood risk in Kristianstad municipality. Global Environmental Change 23:372-381. http://dx.doi.org/10.1016/j.gloenvcha.2012.07.009
Jongman, B., S. Hochrainer-Stigler, L. Feyen, J. C. J. H. Aerts, R. Mechler, W. J. W. Botzen, L. M. Bouwer, G. Pflug, R. Rojas, and P. J. Ward. 2014. Increasing stress on disaster-risk finance due to large floods. Nature Climate Change 4:264-268. http://dx.doi.org/10.1038/nclimate2124
Karlstad Local Authority (Karlstads kommun). 2010. Översvämningsprogram Karlstads kommun. Bibliotekshuset, Karlstad, Sweden. [online] URL: https://karlstad.se/globalassets/filer/miljo/sjoar_och_vattendrag/oversvamningsprogram.pdf
Karlstad Local Authority (Karlstads kommun). 2012. Strategisk plan och budget 2013–2015: ärende 4. Bibliotekshuset, Karlstad, Sweden. [online] URL: https://karlstad.se/globalassets/filer/kommun-och-politik/kallelser-protokoll/2012/kommunfullmaktige/2012-06-18/06_arende_04_strategisk-plan-och-budget-2013-2015.pdf
Kaufmann, M., W. J. van Doorn-Hoekveld, H. K. Gilissen, and H. F. M. W. van Rijswick. 2015. Analysing and evaluating flood risk governance in the Netherlands: drowning in safety? STAR-FLOOD Consortium, Utrecht, The Netherlands. [online] URL: https://www.ris.uu.nl/ws/files/20125622/Analysing_and_evaluating_flood_risk_governance_in_The_Netherlands_08.03.16.pdf
Kelder, E., B. Gersonius, and M. Hulsebosch. 2013. Concept gebiedsrapportage Eiland van Dordrecht. Deltaprogramma Rijnmond-Drechtsteden, Rotteram, The Netherlands. [online] URL: http://edepot.wur.nl/279788
Kubicki, P. 2010. The identity of Wrocław. Tożsamość Wrocławia, czyli jak mieszkańcy Wrocławia konstruują swoje miasto. Pages 203-212 in W. Łazuga, and S. Paczos, editors. Poznań, Szczecin, Wrocław: trzy uniwersytety, trzy miasta, trzy regiony. Libron, Krakow, Poland.
Kundzewicz, Z. W., K. Szamalek, and P. Kowalczak. 1999. The Great Flood of 1997 in Poland. Hydrological Sciences Journal 44(6):855-870. http://dx.doi.org/10.1080/02626669909492285
Larrue, C., S. Bruzzone, L. Lévy, M. Gralepois, T. Schellenberger, J. B. Trémorin, M. Fournier, C. Manson, and T. Thuilier. 2016. Analysing and evaluating flood risk governance in France: from state policy to local strategies. STAR-FLOOD Consortium, Tours, France. [online] URL: http://www.starflood.eu/documents/2016/03/wp3-fr-final-webversion.pdf
Matczak, P., J. Lewandowski, A. Choryński, M. Szwed, Z. W. Kundzewicz. 2016. Analysing and evaluating flood risk governance in Poland: looking for strategic planning in a country in transition. STAR-FLOOD Consortium, Utrecht, The Netherlands. [online] URL: http://www.starflood.eu/documents/2016/03/wp3-poland-final-webversion.pdf
Mees, H., C. Suykens, J.-C. Beyers, A. Crabbé, B. Delvaux, and K. Deketelaere. 2016. Analysing and evaluating flood risk governance in Belgium: dealing with flood risks in an urbanised and institutionally complex country. STAR-FLOOD Consortium, University of Antwerp and University of Leuven, Antwerp, Belgium. [online] URL: http://www.starflood.eu/documents/2016/03/wp3-be-final-webversion.pdf
Mitchell, J. K. 2003. European river floods in a changing world. Risk Analysis 23(3):567-574. http://dx.doi.org/10.1111/1539-6924.00337
National Board of Housing, Building and Planning (Boverket). 2010a. Klimatanpassning i planering och byggande: analys, åtgärder och exempel. Regeringsuppdrag (6) M2009/4802/A (DELVIS). Boverket Publikationsservice, Karlskrona, Sweden. [online] URL: https://www.boverket.se/globalassets/publikationer/dokument/2011/klimatanpassning-i-planering-och-byggande-webb.pdf
National Board of Housing, Building and Planning (Boverket). 2010b. Mångfunktionella ytor: klimatanpassning av befintlig bebyggd miljö i städer och tätorter genom grönstruktur. Boverket Publikationsservice, Karlskrona, Sweden. [online] URL: https://www.boverket.se/globalassets/publikationer/dokument/2010/mangfunktionella_ytor.pdf
Pahl-Wostl, C., G. Becker, C. Knieper, and J. Sendzimir. 2013. How multilevel societal learning processes facilitate transformative change: a comparative case study analysis on flood management. Ecology and Society 18(4):58. http://dx.doi.org/10.5751/ES-05779-180458
Partzsch, L., and R. Ziegler. 2011. Social entrepreneurs as change agents: a case study on power and authority in the water sector. International Environmental Agreements: Politics, Law and Economics 11(1):63-83. http://dx.doi.org/10.1007/s10784-011-9150-1
Pelnomocnik Rzadu do Spraw Programu dla Odry. 2011. Program dla Odry - 2006 - aktualizacja. Project Report. Ministry of Environment, Wrocław, Poland.
Rijke, J., S. van Herk, and C. Zevenbergen. 2013. Towards integrated river basin management: governance lessons from room for the river. Pages 1033-1043 in F. Klijn and T. Schweckendiek, editors. Comprehensive flood risk management: research for policy and practice. CRC Press, Boca Raton, Florida, USA. http://dx.doi.org/10.1201/b13715-149
River Hull Advisory Board. 2016. River Hull integrated catchment strategy: strategy document. Final report. East Riding of Yorkshire Council, Beverley, UK. [online] URL: http://www2.eastriding.gov.uk/EasySiteWeb/GatewayLink.aspx?alId=678050
Rowe, G., and L. J. Frewer. 2005. A typology of public engagement mechanisms. Science, Technology, and Human Values 30(2):251-290. http://dx.doi.org/10.1177/0162243904271724
Sitek, W. 1997. Wspolnota i zagrożenie: Wrocławianie wobec wielkiej powodzi. University of Wroclaw, Wroclaw, Poland.
Star, S. L., and J. R. Griesemer. 1989. Institutional ecology, ‘translations’ and boundary objects: amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology, 1907-39. Social Studies of Science 19(3):387-420. http://dx.doi.org/10.1177/030631289019003001
Stevens, M. R., and S. Hanschka. 2014. Multilevel governance of flood hazards: municipal flood bylaws in British Columbia, Canada. Natural Hazards Review 15(1):74-87. https://doi.org/10.1061/(ASCE)NH.1527-6996.0000116
Swedish Commission on Climate and Vulnerability. 2007. Sweden facing climate change - threats and opportunities. Ministry of the Environment and Energy, Stockholm, Sweden. [online] URL: http://www.government.se/legal-documents/2007/12/sou-200760/
Swedish Civil Contingencies Agency (Myndigheten för samhällsskydd och beredskap). 2012. Översvämningar i Sverige 1901–2010. Swedish Civil Contingencies Agency, Karstad, Sweden. [online] URL: https://www.msb.se/RibData/Filer/pdf/26098.pdf
Van den Brink, M., C. Termeer, and S. Meijerink. 2011. Are Dutch water safety institutions prepared for climate change? Journal of Water and Climate Change 2(4):272-287. http://dx.doi.org/10.2166/wcc.2011.044
Van Herk, S., C. Zevenbergen, R. Ashley, and J. Rijke. 2011. Learning and action alliances for the integration of flood risk management into urban planning: a new framework from empirical evidence from The Netherlands. Environmental Science and Policy 14(5):543-554. http://dx.doi.org/10.1016/j.envsci.2011.04.006
Veenstra-Huisman, C. E. 2014. Veiligheid Nederland in kaart 2: overstromingsrisico dijkring 22 Eiland van Dordrecht. Ministry of Infrastructure and Water Management, The Hague, The Netherlands. [online] URL: https://www.helpdeskwater.nl/publish/pages/135901/dijkring_22_eiland_van_dordrecht.pdf
Walker, G., F. Tweed, and R. Whittle. 2014. A framework for profiling the characteristics of risk governance in natural hazard contexts. Natural Hazards and Earth System Sciences 14:155-164. http://dx.doi.org/10.5194/nhess-14-155-2014
Wiering, M., and I. Immink. 2006. When water management meets spatial planning: a policy-arrangements perspective. Environment and Planning C: Politics and Space 24(3):423-438. http://dx.doi.org/10.1068/c0417j