Strategies for Pre-Flood Risk Management - FLOODsite

Integrated Flood Risk Analysis and Management Methodologies

Strategies for Pre-Flood Risk Management CASE STUDIES AND RECOMMENDATIONS

Date March 2007

Report Number T13-07-04 Revision Number 1_0_P4 Deliverable Number: D13.1 Due date for deliverable: February 2007 Actual submission date: March 2007 Task Leader IOER

FLOODsite is co-funded by the European Community Sixth Framework Programme for European Research and Technological Development (2002-2006)

FLOODsite is an Integrated Project in the Global Change and Eco-systems Sub-Priority Start date March 2004, duration 5 Years

Document Dissemination PU Public PU PP Restricted to other programme participants (including the Commission Services) RE Restricted to a group specified by the consortium (including the Commission Services) CO Confidential, only for members of the consortium (including the Commission Services)

Co-ordinator: HR Wallingford, UK Project Contract No: GOCE-CT-2004-505420 Project website: www.floodsite.net

FLOODsite Project Deliverable D13.1 Contract No:GOCE-CT-2004-505420

T13_07_04_pre_flood_RM_strategies_D13_1_v1_0_p04.doc 02 04 2007 ii

DOCUMENT INFORMATION Title Strategies for Pre-Flood Risk Management Lead Author G. Hutter Contributors L. McFadden, E. Penning-Rowsell, S. Tapsell, M. Borga Distribution Public Document Reference T13-07-04

DOCUMENT HISTORY

Date Revision Prepared by Organisation Approved by Notes 02/04/07 1_0_P4 G. Hutter IOER - - 04/04/07 J Bushell HRW Final formatting for publication and

change of name from �T13_07_01_Strategies_D13_1_v1_0_p04.doc�

ACKNOWLEDGEMENT The work described in this publication was supported by the European Community�s Sixth Framework Programme through the grant to the budget of the Integrated Project FLOODsite, Contract GOCE-CT-2004-505420. DISCLAIMER This document reflects only the authors� views and not those of the European Community. This work may rely on data from sources external to the FLOODsite project Consortium. Members of the Consortium do not accept liability for loss or damage suffered by any third party as a result of errors or inaccuracies in such data. The information in this document is provided �as is� and no guarantee or warranty is given that the information is fit for any particular purpose. The user thereof uses the information at its sole risk and neither the European Community nor any member of the FLOODsite Consortium is liable for any use that may be made of the information. © FLOODsite Consortium


T13_07_04_pre_flood_RM_strategies_D13_1_v1_0_p04.doc 02 04 2007 iii

SUMMARY The report �Strategies for Pre-Flood Risk Management � Case Studies and Recommendations� proposes (1) a theoretical framework to analyse the content, process, and context of strategies for reducing flood risk within catchments. (2) Three case studies illustrate why researchers and practitioners alike can benefit from using the framework to better understand the process dimension of strategies for pre-flood risk management (which is, in this report, mainly long-term planning of combinations of structural and non-structural measures). (3) The report formulates six recommendations to practitioners how to improve flood risk management through shifting attention. (1) In European Member States, managing floods is shifting from protecting against floods to Flood Risk Management (FRM). Simply put, this requires that decision makers responsible for different components of FRM and people (citizens) take the full range of possible floods and their consequences as well as probabilities and uncertainties into account. This kind of rational decision making places heavy demands on decision makers and citizens alike. Up to now, it is clear that research and practice should go in the direction of flood risk management. How this is possible, however, requires - among others � a more systematic understanding how strategies as combinations of contents (aims, measures, and so forth), process patterns (e. g. strategic planning, learning), and societal context conditions interact. The framework for analysing strategies for pre-flood risk management makes suggestions with regard to defining strategy, using linear and adaptive processes of strategy making, strategic planning at different spatial levels and under uncertainty and with reference to the well-known notion of learning. Thereby, the report is grounded in current debates (e.g., �Learning to live with rivers�, �Strategic planning of cities and regions�) and attempts to suggest some fruitful discussions for FRM research (e.g., seeing learning as complex process of exploitation what is already known and exploration what might come to be known in the future). (2) The report is based on an extensive literature review (risk management, FRM, strategy and management research, organizational learning, spatial planning). To ground argumentation, the report refers to three case studies: i) The London study shows that a series of important steps have been taken towards the integrated management of a highly complex physical and social environment. However, it is also clear that the current strategy process still has significant limitations if assessed from the perspective of the framework proposed here. A primary conclusion emerging from this work is that translating the vision of an adaptive strategy process into the reality of flood risk management remains a challenging task. ii) The Dresden study comes to similar conclusions against the background of a recent flood disaster that came like a �bolt from the blue� (Weisseritz river flood in August 2002). The study shows that long-term planning does not only deal with distant futures. Planning itself qualifies as a long-term endeavour. The Adige river study underlines this conclusion. In sum, case study results motivate to further define a common European approach to developing strategies for pre-flood risk management in the long run. (3) To specify its conclusions, the report attempts to formulate recommendations to practitioners of pre-flood risk management. Perhaps the most important result is the recommendation that we should carefully deploy the idea that inclusive (all relevant actors) and open (without restrictions in content) dialog is the �one best way� to effective flood risk management for dealing with uncertainty issues of long-term planning in a rational way. Uncertainty is certainly a top scientific topic. However, people (researchers included) strive for certainty in daily-life practices and when it comes to recovering from a flood disaster. This has some implications for how flood risk managers can learn effectively about uncertainty of flood risk in long-term planning when power to implement plans is limited and resources are scarce. One implication is that learning from the past and learning for the future should not be seen as antipodes. They are different learning processes with the possibility of mutual reinforcement.


T13_07_04_pre_flood_RM_strategies_D13_1_v1_0_p04.doc 02 04 2007 iv


T13_07_04_pre_flood_RM_strategies_D13_1_v1_0_p04.doc 02 04 2007 v

CONTENTS Document Information ii Document History ii Disclaimer ii Summary iii Contents v

1. Introduction ...................................................................................................................... 1 1.1 Background.......................................................................................................... 1 1.2 Purpose of the Report .......................................................................................... 1

2. Framework to Analyse Strategies for Pre-Flood Risk Management................................ 2 2.1 Aim of the Framework for Strategy Analysis...................................................... 2 2.2 Definition of Strategy and Overview over the Framework ................................. 3 2.3 The Content Dimension of Strategies.................................................................. 7 2.4 The Process Dimension: Strategic Planning and Learning................................ 20 2.5 The Context Dimension: External and Internal Constraints .............................. 46

3. Case Studies ................................................................................................................... 48 3.1 Aim, Approach, and Research Design............................................................... 48 3.2 Case Study Results: Overview........................................................................... 57 3.3 The Thames Estuary Study................................................................................ 58 3.4 The Weisseritz River Study ............................................................................... 64 3.5 The Adige River Study ...................................................................................... 72

4. Recommendations .......................................................................................................... 78

5. References ...................................................................................................................... 82

6. Task-Specific Glossary .................................................................................................. 91


T13_07_04_pre_flood_RM_strategies_D13_1_v1_0_p04.doc 02 04 2007 vi

Tables Table 1 Local Planning at Framework and Project Level 16 Table 2 Local Spatial Plans at Framework and Project Level - Examples 17 Table 3 The Linear and the Adaptive Process Model of Strategy 21 Table 4 Matching Process Models of Strategy Making to Context Conditions 24 Table 5 Matching Knowledge Problems and Options or Process Management 31 Table 6 First, Second, and Third Order Effects of Strategic Planning 39 Table 7 Episodic and Continuous Change 43 Table 8 Four types of Organizational Change Processes and their Outcomes 44 Table 9 Context Conditions for Pre-Flood Risk Management in European Member States 47 Table 10 A Multilevel Approach to Conducting the Three Case Studies 49 Table 11 Comparing the Case and the Subunit of a Case Study 50 Table 12 Allocation of Data Collection Resources 52 Table 13 Comparing the Content, Processes, and Contexts of the Cases 54 Table 14 Assessing the Validity and Reliability of the Results 55 Table 15 High-Level Strategic Alternatives of Flood Protection and Flood Risk Management 68 Table 16 Deploying Strategic Planning at Different Spatial Levels of FRM 69 Table 17 Water Resources Management Institutional Framework in Italy 75 Figures Figure 1 Three Dimensions of Strategies for Pre-Flood Risk Management 5 Figure 2 Components of Flood Risk Management 8 Figure 3 Rational Planning Model 9 Figure 4 Political Decision-Making Model 10 Figure 5 Structural and Non-Structural Measures 12 Figure 6 Structural and Non-Structural Measures in the Downstream Basin 12 Figure 7 Classifying Measures and Policy Instruments 15 Figure 8 Learning to Link Plans at Framework Level with Strategic Projects 27 Figure 9 Matching Strategic Planning Modes to the Uncertainty of Context 29 Figure 10 Integrating Spatial Scales, Policy Issues, and Planning Horizons 33 Figure 11 The Relevance of Projects for Strategy Making: Six Propositions 36 Figure 12 Exploitation, Exploration, and Rational Decision Making 40 Figure 13 Process Theories of Organizational Development and Change 45 Figure 14 The River Thames and London 58 Figure 15 Flood Risk Management Plan Development and Review Process 60 Figure 16 Real Land-use within the Weisseritz River Area 65 Figure 17 Inundated Areas in August 2002 and Statutory Flood Plain 66 Figure 18 High Involvement of Critics in the Formation Period of Innovations 71 Figure 19 The Adige River Basin close to Trento 72


T13_07_04_pre_flood_RM_strategies_D13_1_v1_0_p04.doc 02 04 2007 1

1. Introduction 1.1 Background Extreme flood events like the Weisseritz flash flood within the Elbe river basin in August 2002 and their adverse impacts on people and properties have urged practitioners (e. g. LfUG 2004) and scientists (Schanze 2002, DKKV 2003) to argue that strategic planning should be deployed more systematically for avoiding unexpected impacts in flood-prone areas. For instance, strategic spatial planning offers measures which could effectively channel new development to areas not exposed to flooding, thereby discouraging development on floodplains. Spatial planning could serve as means for integrating stakeholders into the decision making process (Burby et al. 2000) for assessing current and future risks of different land use possibilities. In terms of reducing the vulnerability in flood-prone areas and increasing the preparedness of actors, the need is apparent for a planning-based strategy to pre-flood risk mitigation. However, developing a planning-based strategy is a difficult task for various reasons. Directly after a severe flood event the need for an effective strategic approach to pre-flood risk management is usually widely acknowledged. Nevertheless, after some time memories of the event and its causes fade and it is not easy to maintain political support for strategies of pre-flood risk management. As a consequence, in political decision making at local level flood risk issues are often outweighed in the discussion about strictly prohibiting built-up areas on floodplains (Böhm et al. 1998, Fleming 2002, DKKV 2003). Furthermore, a strategic approach requires constant co-operation of water authorities, municipalities, and regional bodies. But, co-operation is costly (e.g. direct costs in terms of time, financial and human resources to maintain dense communication). Therefore, the call for a strategic approach to pre-flood risk management does not suffice. It should be demonstrated how such an approach can be implemented under real-world conditions (Penning-Rowsell & Fordham 1994, Parker 2000, Hutter 2006).

1.2 Purpose of the report Task 13 generates a multidimensional framework to analyse strategies for pre-flood risk management. It specifies this framework with regard to the key topic of reducing vulnerability in flood-prone areas through strategic planning at local level. To realise this aim, existing content knowledge referring to the question what structural and non-structural measures should be linked with regard to the recurrence, speed of onset, and impacts of flood hazards (e.g. Hall et al. 2003, Hooijer et al. 2004) is combined with how-to knowledge. How-to knowledge refers to the process of deploying local spatial planning at the framework and project level (European Commission 1997). From a methodological point of view, questions of how to manage a process should be empirically analysed by conducting case studies. Case studies enable the investigation of process patterns which are too complex to be included into surveys (Yin 2003). The purpose of this report is to describe the framework for analysing strategies for pre-flood risk management, make explicit its rationale, and to specify this framework with regard to the key problem of reducing vulnerability in flood-prone areas through strategic planning at the local level (Chapter 2). Then, the report describes the main features of the case study design and summarizes the case study results (Chapter 3). Conclusions come in the form of six recommendations to practitioners of pre-flood risk management. Guiding themes are: Adaptive process management, strategic planning, and learning.



2. Framework to Analyse Strategies for Pre-Flood Risk Management

�The comment, ´Great strategy; lousy implementation´, gives unjustified credit to the

strategist. If the strategy has been designed without taking account of the organizations

capacity for implementation, it´s a lousy strategy.� (Grant 2005, p. 187)

Strategy seems to be a simple term. Within daily life we use the term to refer to a set of activities to influence the world around us in a specific way. Strategy directs and guides behaviour. Furthermore, strategy refers to intentionality. Strategies are consciously chosen. Notwithstanding this daily-life understanding, over the years, strategy research has developed a more complex concept of strategy as multidimensional phenomenon. This concept takes implementation and the challenge to link strategy with day-to-day activities into account. Attempts to deploy pre-flood measures for risk mitigation can benefit from adopting this multidimensional understanding of strategy. To show this, section 2.1 states briefly what the aim of developing such a framework for strategy analysis is (not least of all, to avoid disappointments). Section 2.2 provides an overview over the framework. Thereby, strategy is defined as multidimensional phenomenon encompassing content, process, and context elements. Sections 2.3 to 2.5 deal with specific dimensions of strategy making (content, process, and context).

2.1 Aim of the Framework for Strategy Analysis The literature on strategy and management that could be important for FRM, especially pre-flood risk management, is vast and difficult to overview (there are some good handbooks, e.g., Pettigrew et al. 2002, Easterby-Smith & Lyles 2003, Poole & Van de Ven 2004). It is the aim of the framework for strategy analysis in the context of pre-flood risk management to point to important themes and their implications for FRM research and practice. The framework follows two somehow more specific aims: • To define strategy in a way that process patterns of strategy making are recognized of utmost

importance for shifting effectively from a pure flood protection strategy to flood risk management. Research on floods is moving, but so are other research and policy fields too. Coming to more integration between, for instance, spatial planning, flood protection, and measures to increase preparedness of citizens requires paying ample attention to how these linkages can be build.

• To show that strategic planning and learning are important �boundary spanning concepts� between different policy fields as well as between different disciplines, for instance, strategic planning in FRM and in spatial planning.

It is not the aim of the report to come to conclusions that are ready for context-specific implementation. The framework is more about formulating conclusions how to allocate attention and to propose issues that are worth attending to. Strategy making is a lot about defining priorities for scarce time and attention (Weick 1993/2001). Furthermore, strategy making is not only about giving �right� answers. It is about posing fruitful questions too. After synthesizing important literature and applying the framework to three case studies (Chapter 3), we will suggest in form of six recommendations where it leads to if one adopts the strategy research perspective proposed in this report.



2.2 Definition of Strategy and Overview over the Framework The term strategy is not new to flood risk management researchers (e.g. Penning-Rowsell & Peerbote 1994, p. 5, Kundzewicz 2002, Hooijer et al. 2004). Seldom is the term explicitly defined (an exception is Hooijer et al. 2004). Even when explicit definitions are given, the findings of strategy research (e.g. Pettigrew & Whipp 1991, Burgelman 2002, Pettigrew et al. 2002) are not taken into account. Looking at the research on strategy and management (see the Handbook of Strategy and Management, edited by Pettigrew et al. 2002) fosters a comprehensive, multidimensional understanding of strategy and the forces that influence strategy. A comprehensive understanding of strategy is necessary to improve the management of problems that are difficult to solve. For example, it is widely acknowledged that the aim to reduce vulnerability in flood-prone areas by discouraging development on floodplains and by channelling demands for housing and infrastructure to areas unlikely to be flooded is no easy-to-solve management problem. Yet, rarely the conceptual tools and empirical findings from strategy research are used to shed light on this management challenge from a new perspective. (New, after all, means new to flood risk management research.) The aim of this section is to demonstrate that strategy research has something to offer for pre-flood risk management. To show this, we start with what could be seen as common ground for using the term strategy.

Origins and Meaning of the Term Strategy The term strategy derives from the Greek word strategia, meaning �general-ship�, itself formed from stratos, meaning �army�, and �ag, �to lead� (Grant 2005). Military strategy and strategies in other fields of societal development (business economics, public affairs, and so forth) share a number of common concepts and principles, the most basic being the distinction between strategy and tactics (Grant 2005, Burgelman 2002). Strategy is the overall solution for deploying resources to establish a favourable position; a tactic is a scheme for a specific action. Whereas tactics are concerned with winning battles, strategy is concerned with winning the war. Strategic decisions, whether in military or other fields, share three common characteristics (Grant 2005, p. 14, see also Ghemawat 1991):

• They are important. • They involve a significant commitment of resources. • They are not easily reversible.

Especially the third characteristic is important to understand how flood risk management can benefit from strategy research. If external pressures are dominant, there is only very little decision space for an actor to choose his strategy (Schreyögg 1984, Burgelman 2002). Therefore, strategic decisions imply a significant degree of freedom for choice (Hrebiniak & Joyce 1985). But, strategic decisions are fundamental decisions that influence the long-term development of an actors welfare. They refer to the interface of internal and external context conditions (Grant 2005). For example, making strategic decisions a business organisation chooses in which industries it will compete against other enterprises. The decision to compete in a specific industry heavily influences the deployment of resources and the building of distinctive competencies in the medium and long term. Once a strategic decision has been made fundamental decision possibilities decrease and one can concentrate on delivering the net benefits the strategy is promising. But, context can change. Therefore, simultaneously or in specific time periods contexts conditions have to be monitored to assess if they still support the strategy chosen. This tension between a) defining strategy to determine resource deployments and to build distinctive competencies on the one hand and b) to preserve flexibility and fundamental decision possibilities on the other cannot be avoided, but, has to be handled within the process of strategy making (Volberda



1998, Van de Ven & Poole 1988). Especially within large organisations strategy takes the form of strategy making across multiple levels of decisions to address changing external contexts and internal resources and capabilities (Burgelman 2002). We conclude, that strategy is a multidimensional phenomenon that encompasses the dimensions of content (�Deciding what to do�), process (�Deciding how to do it�), and context (�Aligning strategic decisions to internal and external conditions�). Shona Brown and Kathleen Eisenhardt elucidate this meaning of the strategy concept referring to two questions which have to be answered at a time (1998, p. 4): �Where do we want to go?� and �How do we get there?�

Defining Strategy as Multidimensional Phenomenon Against this background, a multidimensional understanding of strategy adopting the following definition is proposed for FRM (see Hutter 2006 as well as FLOODsite 2005): A strategy for pre-flood risk management is defined as a consistent combination of long-term goals, aims, and measures, as well as process patterns that is continuously aligned with the societal context. The rationale for this definition is as follows: Changing from the paradigm of flood protection to flood risk management raises challenging questions of formulating and implementing strategies within society. In particular, reducing vulnerability and increasing preparedness require a comprehensive understanding of pre-flood risk management.

Related, But Different Definitions of Strategy Strategy as multidimensional phenomenon is distinct from other possibilities to define strategy. Firstly, the definition of strategy used in this report is distinct from the classic definition from business economics: The classic approach defines strategy as combination of measures and necessary resources for actions to implement the basic long-term goals of a business organisation (see Whipp 2001, p. 15151). This definition of Alfred Chandler (1962) is closely linked to strategic planning as long-term planning under conditions which are largely predictable (Volberda 1998, p. 37, p. 206). The classic definition does not explicitly comprise process patterns of strategy making. But, under conditions of increasing uncertainty process becomes more important to consider different views on a complex phenomenon, to exploit different possibilities of formulating and implementing strategies, and to adapt swiftly to unforeseen conditions. Uncertainty is an increasingly relevant topic for modelling and managing flood risk (Sayers et al. 2002, Hall et al. 2003, FLOODsite 2005). Hence, process patterns should be taken into account in a systematic way and not ad-hoc. Secondly, Hooijer et al. see strategy as �a consistent set of measures, aiming to influence developments in a specific way� (2004, p. 346). This definition is restricted to the content dimension of strategies (e.g., aims, measures, scenarios, and so forth). Process and societal context issues are not included. In this report the term strategic alternative is used for combinations of structural and non-structural measures aiming to influence developments in a specific way. Strategic alternatives are tactics for pre-flood risk management. Thirdly, in the flood risk management literature you can find a simple daily live definition. In this case, a strategy is defined as a statement indicating the direction of using structural and non-structural measures (e.g. �Do-nothing strategy�, �Do-minimum strategy�, �Use new measures to enhance flood risk management standard�). We will consider this daily live definition as content within empirical investigation. Put differently, strategy in the sense of simply stating the direction of various measures will be interpreted as empirical data. It should not be confused with the scientific understanding of strategy as multidimensional phenomenon combining content, process, and context.



A Multidimensional Framework for Strategy Analysis To systematise this multidimensional understanding of strategy, Figure 1 encompasses the dimensions of content, process, and context. Important categories within the dimensions are listed.

Process:� Strategic planning mode:

programming, scenario-based planning, etc.� Model of formulation and implementation:

linear, adaptiv � Learning processes at different levels:

individual, group, organisation, network

Internal context:� Politics� Resources� Responsibility� Culture� Capabilities

Content:� General aims and specific targets� Strategic alternatives as combinations of measures� Structural and non-structural measures� System analysis: controllable, not controllable variables

External context:� Political� Legal� Social� Economic

Figure 1: Three Dimensions of Strategies for Pre-Flood Risk Management (Adapted from Pettigrew & Whipp 1991, p. 26)

Firstly, strategies encompass a content dimension which refers to a complex hierarchy of flood risk management aims, targets and combinations of structural and non-structural measures. In contrast to operational decisions, strategic decisions involve the evaluation of measures with regard to a complex system of general aims and specific targets and the capability of dealing with trade-offs. Thereby, the social, environmental, and economic dimensions of sustainable development should be considered (Hooijer et al. 2004). Secondly, strategies reflect and should match the societal context within specific catchments (e. g. political conflict potential, resources, cultural �world views�). Thirdly, the process dimension describes how strategies are formulated and how they can be implemented. This dimension refers to questions of how planning can be effective under increasing uncertainty and how learning for pre-flood risk management can be fostered. It is crucial to recognise that planning and plans are elements within this framework of strategy. They are not themselves the strategy. Usually, especially within the public sector, planning is necessary for �winning the game�, but, it is not sufficient (Mintzberg et al. 1999, Wiechmann & Hutter 2007). Planning has to be complemented by implementation and learning processes on various levels of societal development (de Geus 1996, Chakravarthy & Lorange 1991). Sim Sitkin (1996, p. 554) shows well the close relationship between well-planned actions and learning. Process is about learning how to deal with diverse political interests, resource scarcity, existing responsibility of actors, cultural �world views� (Thompson et al. 1990, Hooijer et al. 2004 for an adoption to flood risk management), and limited capabilities to act and interpret the often complex processes of strategy and especially strategic change. Process patterns refer to recurrent interactions of using the linear and/or adaptive model, planning modes, and different types of learning.



As indicated above, one fundamental function of strategies is to reduce the variety of possible decisions and actions of the strategy maker(s). A strategy aligns different variables within the content, process, and context dimension, and, therefore, necessarily excludes other possible combinations. Changing strategies requires time, resources, changing power structures, and capabilities to learn under enabling and inhibiting factors for learning. Sanchez & Heene (2004) argue that strategy analysis should concentrate on difficult decisions that imply trade-offs and tensions. This notion has been elaborated in empirical research on new forms of organising and strategising within contexts characterised by increasing uncertainty and complexity of social interactions (Pettigrew et al. 2003). In line with these suggestions Hooijer et al. (2004) claim that pre-flood risk management should address the question of how to evaluate pre-flood measures with regard to multiple aims and targets (Hooijer et al. speak of objectives). Pre-flood risk management should be seen as a contribution to an integrated water management on catchment scale and part of sustainable development (see also Mileti 1999). For example, floodplains can be used for urban development, nature conservation issues, and the construction of flood protection assets. Municipalities that are responsible for delivering a bundle of very different goods for citizens (e. g. social security, housing, open spaces, protection against natural hazards) can have problems in evaluating how to use the floodplain with regard to the complex set of aims they have to consider. Hence, one could argue that especially municipalities face the double challenge of making clear strategic choices in the present and preserving flexibility to make decisions in the future with regard to future generations (de Bruijn & Klijn 2002, Hooijer et al. 2004).



2.3 The Content Dimension of Strategies A flood can be understood as a temporary covering of land by water. To take the evaluative societal context into account, one could emphasise that a flood is a �unwanted� temporary covering of land by water (FLOODsite 2005). Thus, a flood is a temporary phenomenon. Therefore, it seems useful to draw a distinction between pre-flood measures that are deployed before a flood, event management, and post-flood measures. This report focuses on spatial planning as pre-flood measure of non-structural �nature�. It is shown that deploying spatial planning for pre-flood risk management requires a strategic approach for combining multiple planning measures over time. Flood risk management aims at a holistic societal analysis, evaluation and reduction of flood risk (Schanze 2005; cf. Sayers et al 2002, Hall et al 2003). The flood risk system comprises the whole causal chain constituting flood risks, ranging from the flood hazards as sources over various pathways to the receptors with the consequences of floods (SPRC-Model; Kundzewicz & Samuels 1997). The model proposes that any flood event starts with an initiating event (= Source, e.g. meteorological event, technical failure). Water is conveyed to exposed values (= Receptor, e.g. properties, population) through specific flood routs (Pathway). In coincidence of flood event and exposed values losses (Consequences) may occur. Given that measures are implemented with the aim of flood risk reduction, those can be directed towards any component related to the genesis of risk. For the risk formula this means that hazard and vulnerability, both including the elements which they are composed of must be understood as potentially subject to intervention in terms of flood risk reduction. The following risk components can be addressed:

• The hazard, nature and probability of which are influenced by the source (extent of initiating event), condition (natural, humane made) of the pathway and the physical pattern of the receptor area (e.g. conveying capacity in a developed area)

• The value of the receptor (number and type of properties, people, other values) exposed to the hazard (value at risk)

• The susceptibility of receptors (values) to the hazard • The coping capacity in the receptor area.

In developed (�formal�) societies, flood risks are primarily managed by politicians and officials representing public institutions. Scientific institutions and non-profit organisations may play an important role. Flood risk management takes place on different spatial and related administrative levels of decision making (e.g. local, regional, national) involving various fields of policy making (e.g. water authority, spatial planning authority). Flood risk management differs depending on types of waters (e.g. mountainous river, lowland river), flood types (e.g. flash floods, plain floods), land uses (e.g. rural, urban), planning and administrative systems (e.g. according to catchments or communities) as well as natural (e.g. geomorphology, regional climate) and societal (e.g. social, economic) conditions and trends (Schanze 2005). To structure the overall approach of flood risk management, it seems appropriate to draw a distinction between three components: flood risk analysis, risk assessment, and risk measures (see Figure 2).



Flood risk management

Hazarddetermination

Vulnerabilitydetermination

Current / futureflood risk

Pre-floodmeasures

Flood eventmanagement

Post-floodmeasures

Riskperception

Tolerableflood risk

Risk analysis Risk measuresRisk assessment

Figure 2: Components of Flood Risk Management (Source: Adapted from Schanze et al. 2005)

Risk Analysis and Risk Assessment Flood risk analysis determines current or � based on proposed activities or uncontrollable trends � future risks. It is based on the determination of the flood hazard and the vulnerability and their combination considering exposure. Hereby, the flood hazard can be defined by probability, magnitude and duration of the hazardous event (Plate 1999). Vulnerability is influenced by the values exposed and their susceptibility to the hazard (damage potential / values at risk) and by the existing capacity to economically, socially, politically and ecologically withstand or recover from threats and damages connected with the hazard (coping capacity / resilience). With other words, vulnerability describes the potential for harmful consequences which can be caused by a hazard. For vulnerability determination exposure and susceptibility of elements of risk need to be analysed, including among others an examination of land use pattern and change, inquiry of immobile and mobile elements at risk and determination of their value, estimation of potential direct and indirect flood effects, social structure of population and identification of vulnerable groups with major impediments to recover from flood events. Up to now, the scope of most investigations on vulnerability is quite narrow and mostly restricted to direct economic losses (Messner & Meyer 2005). Risk assessment covers the risk perception and the decision regarding the toleration of a certain risk. Fundamental is the understanding of risk as the coincidence of a hazard with potentially damageable values. In a simple formula risk may be described as a function of hazard and vulnerability: risk = hazard * vulnerability. Risk perception means an overall view of risk held by individuals or groups depending on values, experiences and feelings. The decision to tolerate a certain risk needs a weighing of costs and benefits in monetary and non-monetary terms. It is based on the assumption that a risk element has both, costs by potential damages and (current or potential) measures on the one hand and the potential economic yield and other benefits for using a flood-prone area on the other hand. The weighing is dedicated to current risks with a certain degree of reduction as well as to alternative mitigated risks. It takes place on different levels from individuals to the actors of floor risk management of international river basins. Assessment is the most subjective subtask within flood risk management. It strongly depends on societal context.



Goals, Aims, Targets, and Actions The content dimension comprises goals, aims, targets, as well as actions to deploy structural and non-structural measures. Referring to the content dimension of strategies one has to answer the question what problem of pre-flood risk management is being addressed and how it can be specified (e.g. specification of the aim to reduce vulnerability with regard to defining as product of susceptibility and value, see FLOODsite 2005). The content dimension of strategies for pre-flood risk management will be elaborated below with regard to the problem of controlling and reducing vulnerability in flood-prone areas through local spatial planning. The following concentrates on more formal differences between goals, aims, targets, and strategic alternatives.

Goals

Aims

Targets

Actions

Figure 3: Rational Planning Model (Source: Bryson 2004, p. 18)

The distinction between goals, aims, and targets is based on the common scientific assumption that it is necessary to differentiate explicitly between analysing cause-effect-relationships on the one hand and addressing questions of evaluating these relationships on the other. Whereas daily life normally does not sharply differentiate between causal and evaluative statements and related actions, scientific inquiry has to explicitly address the question which statements refer to reality independently from the subjective evaluation of the researcher. This report is based on the decision that its research is mainly descriptive. Its purpose is, as mentioned before, to systematically gather descriptive data referring to strategies for pre-flood risk management, and to interpret these data with regard to the theoretical framework and a specific management problem for focussing scientific analysis. Statements referring to the evaluation of human decisions, actions, and their consequences, as well as the consequences of flood hazards, are understood as components of reality. With regard to the evaluative dimension of strategies, this report distinguishes between goals, aims, and targets. (Other approaches are possible.) Goals are defined as long-term goals of an actor or a set of actors (individual, group, organisation, network of organisations). They primarily are grounded in institutional external conditions of strategies. They refer to the identity of the actor(s): For instance, in democratic political systems politicians pursue the goal of winning elections for different reasons (e.g. political belief that a specific strategy for pre-flood risk management should be adopted, acquiring reputation and political power, increasing income, and so forth). This example illustrates that goals can be adapted to different interests of actors. Compared to aims and targets, they are a more abstract entity. They serve as a stable basis for evaluating different tactics despite changing societal context conditions. Whereas goals refer to conditions with long-term stability (e.g. institutional constraints and definitions of responsibility), aims reflect conditions that can be changed in the medium term. (The terms aim and objective are interpreted to be synonyms.) More than goals, aims refer to the desire of actors to change the outside world. For instance, just to mention a hypothetic case, reacting to a recently experienced local flood event, politicians can be urged by citizens and organised stakeholders to pursue the aim of controlling development on floodplains through spatial planning. Politicians continue trying to win



elections. Because of changed external context conditions to win elections, the aim of avoiding a further increase of development in flood-prone areas becomes more important for reaching this political goal. Examples of aims are: Prohibiting further urban development on floodplains and relocating settlements from flood-prone areas to non-risk areas within the territory of a municipality; increasing the speed and effectiveness of communication between local authorities, water authorities, and emergency institutions; increasing the preparedness of households with regard to flash floods. Targets specify aims. Targets can be defined temporally, spatially, and with regard to the possibility of quantifying the desired effects of flood risk management. For instance, the Environmental Agency (EA) states that key targets of its strategy for flood risk management in the first three years (beginning in year 2003) are to • �have no loss of life through flooding, • reduce risk of flooding to life, major infrastructure, environmental assets and some 80,000

homes� (EA 2003, p. 6) Targets are preconditions for programming specific actions, for answering the questions if overall aims were realised, and for allocating responsibility with regard to intended consequences. Despite further benefits of targets for precise and transparent strategy making they are not always to be found in policy practice. Formulating targets increases the risk that unsuccessful policies can be identified. Therefore (and for other reasons), politicians as well as representatives from agencies sometimes prefer formulating more abstract intentions in the form of aims. Figure 4 expresses this in a more systematic way. Up to here, the report assumes implicitly that goals, aims, targets, and actions are ordered hierarchically for rational decision making. In the �real world�, however, it can be difficult to observe such rational decision making. Hierarchies of aspirations (goals, aims, and so forth) are produced as outcome of effective strategy processes. Seldom do such processes function in accordance to hierarchies. Strategy, after all, is about finding new and important solutions of high complexity and considerable uncertainty. John Bryson (2004) proposes that decision making relevant for strategic planning in the public sector proceeds more like a kind of issue- and action focused process. In this process, decision makers start with issues (which are important and sometimes fundamental questions a decision maker can do something about), comes up with actions, and then looks for aims to justify existing solutions. There is great latitude in the public sector in linking aims, targets, issues and actions. For this reason it is important to distinguish systematically between process and outcome (March 1994).

Goals

Aims

Targets

Issues and actions

Figure 4: Political Decision-Making Model (Source: Bryson 2004, p. 19)



Pre-Flood Measures, Event Management, and Post-Flood Measures Risk Measures encompasses structural (�hard�) and non-structural (�soft�) measures for reducing current and future flood risk (Kundzewicz & Samuels 1997, Kundzewicz 2002) until a tolerable level has been reached from the point of view of political representatives and experts within the policy field of flood risk management (Schanze et al. 2005). Using the term reduction as general term for all structural and non-structural measures corresponds to current European and national Water Policies. Thereby, reduction comprises all measures with a potential to reduce risks related to flooding. The term does not imply that risks have to be reduced to zero or that considering a measure already implies a decision to use it for risk reduction. • Measures of pre-flood risk management are deployed before a flood event occurs. Two

approaches can be distinguished: (1) Preventing floods and the adverse impacts of flooding (e.g. developing spatial plans to discourage development in flood-prone areas and to allocate development to non-risk areas). (2) Preparedness is the ability to ensure effective response to the impact of hazards, including the issuance of timely and effective early warnings and the temporary evacuation of people and property from threatened locations (ISDR 2004). To develop this ability requires time and resources. Therefore, it should be seen as part of pre-flood management which is closely linked to event management.

• Event management comprises all activities at the beginning and during flood events (e.g. forecasting the event precisely, giving effective early warning to the public). Compared to pre-flood management, event management decisions and actions have to deal above average with unexpected events and unforeseen situational contingencies (Tonn et al. 2000). In this case, the capability to self organise (Comfort et al. 2001) becomes crucial for combining formal and informal approaches to event management (e.g. combining official and unofficial flood warning systems, Parker & Handmer 1998).

• Post-flood risk management encompasses all decisions and actions which occur after a flood event. Often, post-flood management focuses on reinstalling the �status quo� that was in place before the flood event occurred through the deployment of financial compensation mechanisms and structural measures to reconstruct flood protection assets.

Pre-flood risk management, event management, and post-flood measures are distinct categories from an analytical point of view. In the real world they are interwoven. For instance, financial insurance mechanisms not only remedy the damages of past flood events, they have an influence on the expectations of actors about future insurance payments for flood damages. They can discourage or encourage development on floodplains. Hence, they influence the conditions for pre-flood risk management.

Physical Measures and Policy Instruments (≈ Structural and Non-structural Measures) To gain a more detailed picture what measures are available to reduce flood risk within catchments, the classic distinction between structural and non-structural measures can be considered (see Figure 5). Generally, structural measures involve some engineering work (Penning-Rowsell & Peerbolte 1994). They are capital-intense investments that require maintenance over a long period. Non-structural measures encompass all measures that do not involve engineering work. They are used for controlling land use, insuring flood losses, and informing people about the risk of flooding, to mention only a few examples. Figure 5 corresponds to the notion that flood risk management encompasses measures to control the flood hazard (the water that produces the flood), vulnerability by controlling land use (e.g. discouraging new development on floodplains trough spatial planning).



Dams, reservoirs, retarding basins

Channel modifications

Levee banks

Flood-proofing

Catchment modifications

Schemes of drainage andflood protection

Flood forecasting, flood warning &emergency planning

Planning controls

Acquisition & relocation

Flood insurance

Public information & education

Structuralmeasures

Non-structuralmeasures

Watercontrol

measures

Financial relief&

loss reduction

Land usecontrol

measures

Figure 5: Structural and Non-Structural Measures (Source: Penning-Rowsell & Peerbolte 1994, p. 6)

The classification of Hooijer et al. (2004) identifies four main groups of �room for river� measures (Figure 6): a) dike relocation; b) flood bypasses; c) retention areas; d) cyclic floodplain rejuvenation. Furthermore, it distinguishes between six groups of structural and non-structural measures for damage prevention: a) flood hazard maps and flood risk maps; b) spatial planning; c) building regulations for flood-resistant and flood-proof design of buildings; d) financial instruments: insurance, subsidies, fees, and other economic incentives; e) communicative instruments: between authorities, civil protection and emergency management agencies, but also directed towards raising awareness to flood risk in general and the preparedness of the population at risk in particular, f) structural measures for flood control and management: areas, barriers, (sets of) dikes.

�Room-for-Rivers� Measures

DamagePrevention

Structural and Non-Structural Measures in the Downstream Basin

DikeRelocation

FloodBypasses

Retention Areas

CyclicFloodplain

Rejuvenation

FloodHazard/RiskMaps

SpatialPlanning

Financial Instruments

CommunicativeInstruments

StructuralMeasuresto Protect

Areas at Risk of

Flooding(e.g. Dike

Raising)

BuildingRegulations

Figure 6: Structural and Non-Structural Measures in the Downstream Basin (adapted from Hooijer et

al. 2004)



Traditionally, managing floods and flood risk focused on water control through structural measures like dykes, dams and reservoirs. Currently, we can see a shift from flood control to more holistic approaches for managing flood risk (Schanze et al. 2005) within a European perspective on integrated governance and water basin management (Bressers & Kuks 2004). The risk-based approaches underline the importance of considering land use and spatial planning. Spatial planning for controlling development in flood-prone areas is a key topic, especially with regard to vulnerability associated with extreme flood events (see Hooijer et al. 2004). However, in this report, spatial planning is analysed within a holistic understanding of the whole range of possible determinants, assessment aspects, and measures. We should remember that planning for reducing vulnerability is one possible non-structural measure, not the �one best way� for managing flood risk. Strategic alternatives are combinations (�bundles�) of physical measures and policy instruments for pre-flood risk management. Specific management problems of pre-flood risk mitigation are due to various causes (e.g. tendency to develop built-up areas on floodplains is facilitated by development plans that do not address flood risk as a strategic problem, limited preparedness of citizens, no participatory planning process at local level, and so forth). These problems have to be addressed by consistent bundles of various measures. We already mentioned that strategic alternatives are tactics for using pre-flood measures.

Spatial planning for Reducing Vulnerability The theoretical framework to be outlined in this chapter fosters a comprehensive analysis of strategies for pre-flood risk management. The framework shows that developing a pre-flood strategy on catchment scale is a complex endeavour. It requires considering a wide range of possible flood events (e.g. extreme events), different spatial scales and planning horizons as well as various actors with different responsibility, capabilities, goals, aims, and measures. Probably, within many catchments a fully fledged pre-flood risk management strategy will become reality in the long term. To demonstrate the applicability and implications of the framework, it seems useful to analyse a specific management problem. Solving the problem in the real world should be crucial for developing a holistic approach to pre-flood risk management on catchment scale (Schanze et al. 2005). Reducing vulnerability through spatial planning is such a problem. Many practitioners and scientists claim that spatial planning is a promising non-structural measure for controlling the exposure of people and property in flood-prone areas and for channelling urban development to non-risk areas. Local planning is crucial for influencing day-to-day decisions of developers, land owners, and households that significantly influence in sum the flood risk system of catchments. A participatory planning process could involve all stakeholders at local level, especially citizens. In this case, it provides a platform for discussing the long-term effects of developing high- and low-risk areas, thereby systematically raising public awareness about flood risk to enhance their ability of coping with floods. In what follows, the idea of reducing flood risk through spatial planning is further outlined. Existing contributions to flood risk management research sometimes vaguely refer to planning as a non-structural measure. In this report, it is demonstrated that a spectrum of relevant plans exist at local level. Therefore, planning for pre-flood risk management requires a strategy to combine multiple measures over time. There is a growing consensus within the European research community as well as within related policy fields (e.g. water management) that pre-flood risk management requires deploying spatial planning at the local and regional level more intensely and systematically than before. Scientific contributions are, for example, Hooijer et al. (2002, 2004), Böhm et al. (1999), Böhm et al. (2002), Müller (2003b), Janssen (2005). Policy documents from different European countries point into a similar direction (LAWA 2004, EA 2003). It is even argued that spatial planning is the most promising non-structural measure for building disaster-resilient communities (Burby et al. 1998, 2000, Mileti 1999). This consensus is based on the assumption that the �principle cause of the adverse effects of flooding has always been the urban development on flood plains that followed the expansion of towns and cities away from naturally protected locations into areas where flooding was infrequent but not



unknown. Many towns and cities were built on bridging points over major rivers, but have expanded massively into adjacent areas, including onto flood plains.� (Penning-Rowsell 2001, p. 108) Based on a strong and enduring political will to reduce flood risk through reducing vulnerability, spatial planning could channel additional demands for housing, industry estates, and infrastructure to areas not at risk of flooding and could discourage urban development in flood-prone areas. Of course, development in flood-prone areas would not be strictly prohibited. But, in case of creating new buildings these would be technically adapted to different possible flood events. If necessary, retro-fitting of existing buildings would take place. The safety standard and the condition of flood protection assets would be taken into account. Furthermore, spatial planning could foster an open discussion at the local level about the risks of using an area for different land use purposes. Thereby, planning could provide a communication platform for motivating and informing people in flood-prone areas how to develop the necessary knowledge and behaviour potentials to cope with floods. Up to now, there is only limited empirical evidence that these ideas about using spatial planning for pre-flood risk management systematically will be applied in practice. In practice, reducing vulnerability through spatial planning at local and regional level is not necessarily a straightforward matter, despite possible disastrous consequences of floods on tangible and intangible assets in flood-prone areas. On the contrary, recent studies investigating measures for the Rhine river basin argue that the most cost-effective possibilities of deploying spatial planning for controlling urban development are underused and that it is not at all clear that reducing vulnerability through planning will be an issue on the local and regional policy agenda in the future (Böhm et al. 2002, Hooijer et al. 2004). Hence, it remains an open question if spatial planning actually will become a �true� non-structural measure of pre-flood risk mitigation (for a similar appraisement with regard to pre-flood risk management through spatial planning in England and Wales see Penning-Rowsell 2001, Howe & White 2002). Unfortunately, statements about the uncertain future of spatial planning as mitigation measure within Europe have to be formulated on a relatively weak empirical basis. The relationship between pre-flood risk management and spatial planning has been analysed primarily from a normative point of view and with regard to the content dimension of strategies (e.g. How & White 2002, Olshansky & Kartez 1998 summarise the findings of implementation research in the USA). For instance, Böhm et al. (1999) investigated how spatial planning could and should be used at different spatial levels to consider flood risk issues within decision making. They found that especially local planning authorities often do not use the wide spectrum of possibilities within the given German spatial planning system. But, the reasons for this finding and possibilities to overcome this unsatisfying current state of spatial planning are not investigated further. Hence, there is a lack of concrete knowledge showing why local planning authorities do not systematically use spatial planning in advance for reducing flood risk and how planning-based strategies could be fostered in practice. Content knowledge about effective and efficient strategic measures is a necessary condition for pre-flood strategies (Hooijer et al. 2004). Yet, is should be complemented with studies that analyse the process and context dimension in some detail. Often, translating land-use-planning terms accurately and unambiguously is not possible. Each term relates to other terms. For instance, the term �regional� can have very different meanings whether defined from a European perspective or from the view of local authorities. To bridge the diverse terminology with regard to land use planning systems across Europe, the European Commission invented the term spatial planning (European Commission 1997). The term spatial planning encompasses all public policies and actions intended to influence the distribution of activities in space and the linkages between them. It will operate at EU, national and local levels and embraces land use planning and regional policy (European Commission 1997, p. 156). The term spatial development refers to changes in the physical world. Spatial development comprises changes in the distribution of activities in space and the linkages between them through the conversion of land and property uses (European Commission 1997, p. 156). The terminology outlined in �EU Compendium of Spatial Planning Systems and Policies� serves to classify planning measures. However, the terminology in the compendium has to be expanded to



consider recent findings in spatial planning research (Mastop & Faludi 1997, Salet & Faludi 2000, Albrechts et al. 2003, Healy 2003, Wiechmann & Hutter 2007). The compendium addresses mainly statutory (formal) plans. Non-statutory (informal) concepts are excluded (European Commission 1997). This seems to be a too narrow approach to spatial planning in general, and with regard to local planning in particular. Local informal planning approaches are widely practiced at framework level. In theory (Healy 2003), but also in practice strategic spatial planning is seen as a social process enhancing the capability of all stakeholders to think about and act for a desired overall spatial pattern of urban and regional development (Albrechts et al. 2003). It supplements rather than supplants statutory plans (Wiechmann & Hutter 2007). Enhancing the ability of stakeholders to consider spatial development as long-term endeavour is an important topic to understand and improve strategies for pre-flood risk management. Therefore, it seems appropriate to take non-statutory plans into account (e.g. spatial development concepts for the whole territory of a municipality). Olfert and Schanze developed a classification scheme that explicitly takes formal and informal spatial planning as policy instruments for pre-flood risk management into account (see Figure 7).

Measures and Instruments for Reducing Flood Risk(Pre-flood Risk Management, Event Management)

Stimulation instruments

Communicative instruments

Fina

ncia

l inc

entiv

es

Risk

and

loss

dis

tribu

tion

Regulation instruments

Info

rmal

spat

ial p

lann

ing

Com

mun

icat

ion

/ Dis

sem

inat

ion

War

ning

/ In

stru

ctio

n

Form

al w

ater

man

agem

ent

Form

al e

nviro

nmen

tal p

rote

ctio

n

Fina

ncia

l disi

ncen

tives

PHYSICAL MEASURES POLICY INSTRUMENTSFo

rmal

spat

ial p

lann

ing

Riv

er c

hann

el a

nd c

oast

al

man

agem

ent

Floo

d pr

oofin

g of

bui

ldin

gs a

nd

tech

nica

l inf

rast

ruct

ure

Land

man

agem

ent

Control measures

Adaptation measures

Retreat measures

Evac

uatio

n of

hum

an li

fe

and

life

stoc

k

Evac

uatio

n of

ass

ets

Compensation instruments

Dra

inag

e an

d pu

mpi

ng s

yste

ms

Cha

nnel

con

veya

nce

and

capa

city

Floo

d w

ater

stor

age

Floo

d w

ater

tran

sfer

Coas

tal a

lignm

ent

Coa

stal

ene

rgy

acco

mm

odat

ion

Figure 7: Classifying Measures and Policy Instruments (Source: Olfert & Schanze 2006)

Local Planning at Framework and Project Level Local authorities can be understood as multi-product organisations. They attempt to accomplish a complex bundle of tasks with regard to a well-defined territory. Usually, this territory covers only a part of a basin. The bundle of tasks encompasses social, economic, and ecological aims (e.g., developing dwellings for low-income people, subsidising the local economy, and preserving open spaces). Local authorities primarily formulate and implement strategies to increase the welfare of a local community, not the overall welfare which results out of all decisions and actions relevant for a catchment. The strategies of local authorities are of growing interest for pre-flood risk mitigation. In most European countries, they are responsible for regulating land use changes. By deploying spatial planning local authorities can restrict or allow development in flood-prone areas. At the framework level (the term is explained below), they can attempt to induce development in non-hazardous areas and discourage built-up areas on floodplains. Of course, they have no absolute power. The decisions of local authorities are dependent on external political factors, regulations from higher political and administrative levels (regional, state, federal state). They depend on internal context conditions (e.g.



the political will to learn from flood disasters and to continuously adopt an effective pre-flood risk management approach). Spatial planning at local level is multi-layered. In all European countries there is a distinction between planning for the whole territory of a municipality which takes regional context into account one the one hand and formulating site-specific regulations to control land use on the other hand. To understand the homogeneous features of local planning as well as its diversity within Europe, it is crucial to distinguish between planning at framework level for the whole territory or large parts of it and site-specific approaches at project level (Table 1).

Object

Interaction

Future

Time

Outcome

Effect

Decisions/Actions

Continuous

Open

Central issue

Continually updated

Frame of reference

Spatial development

Until implementation

Closed

Limited to phasing

Blueprint-like

Well-known

Framework Level Project Level

Table 1: Local Planning at Framework and Project Level (Adapted from Mastop 2000, p. 148)

Project plans provide blueprints of the intended end-state of the physical environment, including the measures necessary to achieve that state. The only important social interaction is when the plan is being adopted. Thereafter the plan forms an unambiguous guide to action precisely because the measures to be taken are routine so that the outcome can largely be known in advance. The time dimension in project planning is restricted to simple phasing of works. A project plan is expected to have a determinate effect. After finishing the project it is possible to evaluate if the proposed outcomes were produced and if the planned measures were implemented. Planning at framework level is different (Mastop & Faludi 1997, Salet & Faludi 2000). Spatial plans at framework level are generally long term and comprehensive, bringing together social, economic and spatial considerations. They deal with the coordination of a multitude of actors from diverse institutional contexts (regional bodies, developers, non-profit organisations, citizens, and so forth). Such coordinating is a continuous concern. As all actors want to keep some of their options open, timing is of central importance. Thereby, we have to consider that spatial planners directly control only a limited set of decisions relevant for spatial development. Spatial planning has no absolute power to impose its strategy on others. Its financial resources are limited. Hence, in all European countries co-operation between spatial planners and stakeholders became an important issue. Spatial planning at framework level usually leads to plan documents encompassing text passages and maps to visualise spatial development patterns. They serve as a framework for planning at project level, identifying the broad patterns of spatial development, allocations or zoning, and sometimes for prescribing detailed land use regulations (e.g. piano regolatore generale in Italy, see below). The plan serves as a frame of reference for negotiations and is indicative. Compared with project planning, uncertainty is higher that plans at framework level influence action and implementation at project level. It is not clear if the messages of spatial planners are received and adopted as proposed at project



level. It follows that the relation between planning at framework level and subsequent action is crucial. Departures do not necessarily indicate ineffectiveness. A strategic plan may be interpreted freely by the plan user (which can be the plan maker herself) for adapting to contextual conditions that were not foreseen when the plan was formulated. This comparison of framework and project planning stresses the differences between the two sorts of planning. One can see these differences as extreme points on a continuum. Obviously, there are large projects that can resemble planning at framework level, for example, planning and building the public infrastructure for conducting the Olympic game in a city region. Furthermore, we can think of spatial planning at framework level under stable and relatively simple context conditions which would justify the expectation that the plan has a determinate effect on subsequent decisions and actions. Notwithstanding this more flexible conceptualisation, the aim of comparing framework planning and projects is only to make clear that in this report the term local spatial planning at framework level is used in a broad sense. Planning processes and plans that cannot be classified as project planning in the sense indicated above in Table 1 are classified as planning at framework level (examples are given below). Often, in papers on planning and flood management project-based planning practice is criticised for not embedding development control within a catchment-wide planning approach to pre-flood risk management (for the UK see, for example, White & Howe 2002). Local actors are neglecting strategy within a catchment-wide view (�strategic neglect�). But, we have to consider that detached strategic planning is not a viable alternative. Analysis can paralyse and over-planning is a threat (�plethora of plans�, see Royal Geographical Society 2001) to identify true strategic issues (Volberda 1998). We conclude that both planning at framework and project level are necessary for effective strategy making. It is crucial to connect both levels through continuous learning.

Local Spatial Plans: Examples from Germany, England, and Italy Spatial planning is especially diverse at national and regional level (European Commission 1997). Statutory plans can be described in comparison, but, it is difficult to understand and to evaluate them because complex traditions and informal planning approaches exist. Especially the informal dimension of spatial planning requires deep contextual knowledge about a planning system and its relationship to pre-flood risk management measures. Spatial planning at local level is a bit more homogeneous. A primary concern is to control development proposals from private actors influenced by more or less regionalised market forces (housing market, different industries, and so forth). Various types of spatial plans should be considered. Table 2 shows examples from Germany, England, and Italy with regard to developing planning and development control.

Level of local spatial planning

Germany

UK (England)

Italy

Framework level

- Stadtentwicklungskonzept - Flächennutzungsplan - Städtebaulicher Rahmenplan

- Community strategies - Spatial development concepts - Unitary development plan

- Locale Agenda 21 - Piano regolatore generale

Project level

- Bebauungsplan - Vorhaben- und Erschließungsplan

- Control of development proposals

- Piano particolareggiato - Piano di lottizzazione

Table 2: Local Spatial Plans at Framework and Project Level � Examples

Regional spatial planning is diverse within Germany because of different institutions and regulations at state level (Wiechmann 1998). Notwithstanding diverse regional planning systems, a �generic� description of local spatial planning can be suggested. At framework level of local planning, the difference between the non-statutory Stadtentwicklungskonzepte and preparatory land use plans as



statutory plan is crucial. Local authorities voluntarily can formulate Stadtentwicklungskonzepte for guiding their decisions on lower tier of local planning. These concepts are used for participatory planning approaches. They can be flexibly adapted to new challenges of local development. Often, they are formulated considering enduring changes of local contexts (e.g. shrinking population in East German Cities, lasting high extent of unoccupied housing stock, and the strategic issue of �urban regeneration�, e.g. Cities of Leipzig and Dresden, see Hutter & Westphal 2003). These concepts formulate the broad spatial pattern of local development. Thereby, channelling new development to low-risk areas could be considered. Local planning authorities have the duty to provide a Flächennutzungsplan (preparatory land use plan) covering the whole area of the municipality that serves as a framework for more detailed land use planning at project level (�Bebauungsplan�). To develop an approved preparatory land use plan can take over ten years. Even after over ten years, some cities do not have a state-approved preparatory land use plan (e.g. the City of Freital within the region of Dresden). Non-statutory plans at the urban district level can be formulated (�Städtebaulicher Rahmenplan�). Only the Bebauungsplan, the land use plan at project level, is legally binding for property owners. It can comprise very detailed regulations, for instance, with regard to site-specific density and open space structure (Hutter et al. 2004). The Bebauungsplan could be used for limiting density in flood-prone areas or avoiding apartments on ground floor. In the UK, as in Germany, local planning authorities are the major actors in the planning system and have responsibility for its day-to-day administration and implementation. Local planning authorities are motivated by central government to develop a Community Strategy that focuses on participatory processes and the building of planning-based local strategic partnerships. Interrelationships between a community strategy and development planning are addressed. As non-statutory plan (except London) spatial development concepts are formulated (e.g. �Going for Growth� strategy of Newcastle upon Tyne, Community and Housing Directorate 2000). Local authorities have the duty to prepare development plans (ISoCaRP 2001, p. 280). As a case study will address flood risk management within the Thames Estuary the following focuses on urban areas within England. �For the major part of the United Kingdom, one tier of local government is responsible both for planning a mandatory development plan combining strategy and detail, and for dealing with planning applications.� (ISoCaRP 2001, p. 286) Unitary development plans consist of two parts. The first addresses strategic policy issues and the second covers detailed policies and broad land use allocations. In London the mayor prepares a Spatial Development Strategy to supplement the strategic elements in the Unitary Development Plans of the London Boroughs. The Planning Policy Guidance 25 (PPG 25): Development and flood risk has described in detail how development plans and related other plans (e.g. community strategy) should be used for pre-flood risk management. Unlike the German local planning system, plans at the project level are rarely provided in the UK (European Commission 1997, pp. 67-69). Development proposals are assessed on a case-by-case basis taking the indicative development plan into account (Rydin 1998, p. 205). In Italy, the fundamental local planning instrument at framework level, the piano regolatore generale, involves a detailed zoning that covers the entire territory under the jurisdiction of the local planning authority. The piano regolatore generale is the dominant approach at framework level. It is the �master plan� for controlling land use change at local level (ISoCaRP 2001, p. 115). The zoning plan is legally binding. It governs the relationship between the local planning administration and property owners until replaced by a new plan. It can comprise land use regulations of extraordinarily minute detail. Currently, more flexible and less detailed planning approaches at framework level are discussed in Italy. At the project level different regulatory instruments exist to provide, for instance, low-cost dwellings and industry estates, initiated by either private or public actors (European Commission, 1997, p. 66-69). Unlike the Germany local planning system, zoning at framework level is detailed and legally binding; unlike the system in England, various types of plans exist at project level.



Why Develop a Strategy to Use Local Spatial Planning for Pre-Flood Risk Management? Local spatial planning for pre-flood risk management is no simple, unitary phenomenon. Different levels of planning should be considered. At each level, various sorts of plans can be deployed. The examples in Table 2 cover only a small part of the totality of possible planning measures at local level (for Germany see Böhm et al. 1999 with regard to flood risk management and Hutter et al. 2004 referring to urban planning). Therefore, deploying local spatial planning for pre-flood risk mitigation requires a strategic approach for developing a consistent combination of different measures over time. Furthermore, up to now, pre-flood risk management is mainly an issue of water authorities; spatial planning is done by specific departments within municipalities and regional bodies in accordance with the very diverse institutional arrangements of spatial planning at the regional level across Europe (European Commission 1997). It can be assumed that these actors already communicate and co-operate to some extent (Hall et al. 2003, Hutter & Schanze 2004, Schanze et al. 2005). Nevertheless, water authorities and local planning authorities basically still operate in different working cultures. Their responsibility is very different. Put simply, the responsibility of water authorities ends where the responsibility of local planning authorities begins (�in front of and behind the dykes�). The dominant orientation of water managers is based on knowledge about water quality and quantity and their determinants. Spatial planners are interested in very different social, economic, and environmental issues as long as they can be related to specific planning instruments. It is likely that using spatial planning for pre-flood risk mitigation will require a higher extent of enduring communication and co-operation. To determine flood risk, water authorities have to consider not only the flood hazard, but, the vulnerability of flood-prone areas. Therefore, they have to take the existing land use as well as new development possibilities and related planning regulations into account. Spatial planning for pre-flood risk management requires extensive analysis with regard to the flood risk of different areas within the whole territory of municipalities and therefore a more detailed understanding how floods develop (precipitation, speed of onset of the flood event, flow velocity, and so forth) and what their impacts are. Hence, both spatial planners and water managers will have to significantly change their knowledge base. This knowledge change is only likely if it is rewarded within water authorities and municipalities, and rewards can be expected if there is some enduring political support for using spatial plans as non-structural measures for risk management. Therefore we ask: What societal context conditions foster enduring high political support to use spatial planning for mitigating flood risk and how can these conditions be enhanced within the given spatial planning system and responsibility of water management? Population growth and shrinkage both have facilitating and inhibiting effects on using spatial planning for pre-flood risk mitigation. From a theoretical point of view, planning is well suited to accomplish the task of controlling and channelling new development. But, under population growth conditions and the related additional new demand for housing, industry estates, and public infrastructure it becomes more difficult to discourage development in flood-prone areas (Brody 2003). On the other hand, the influence of spatial planning on existing development is still rather limited (Burby 1998, p. 11, Müller 2003a). Furthermore, serious attempts to implement pre-flood measures in non-growth areas are less likely than in growth areas because of the limited political attention and resource commitment to these places. What is facilitating is that population shrinkage encourages public as well as private local actors to use flood-prone areas as open spaces and for water retention. Projects investigating the interrelationship of population development conditions and spatial planning are under way (e.g. EU-project LUDA- Large Urban Distressed Areas). The implications for using spatial planning in the context of pre-flood risk management still have to be explored. Therefore, the following question should be considered: What is the influence of population growth or shrinkage on attempts to use spatial planning for risk reduction in flood-prone areas?



2.4 The Process Dimension: Strategic Planning and Learning The content dimension of strategies addresses the question what aims are and should be pursued and what measures of pre-flood risk management are capable of reaching these aims. But, aims and measures can be formulated and implemented in alternative ways. The process of strategy making comprises more variability than is recognized till now within flood risk management research. Usually, a linear model of formulating and implementing strategic alternatives for pre-flood risk management is assumed (explicitly or, more often, implicitly). But, there are alternatives that can contribute to improving the strategy process especially for difficult-to-solve problems as, for example, reducing vulnerability in flood-prone areas through spatial planning. The framework for analysing strategies of pre-flood risk management identifies three main topics with regard to the process dimension: (1) Choosing a linear or adaptive model of formulating and implementing aims and measures; (2) choosing the appropriate mode of planning with regard to the extent of stability of context conditions, and (3) considering the move from the traditional paradigm of �flood protection� to �flood risk management� as complex learning process with different planning horizons, learning types, and levels of societal learning processes. Empirically, as chapter 3 shows, the issue of learning is not in the foreground of analysis. Therefore, the third topic will only be mentioned to give an outlook on further possibilities of understanding strategies for pre-flood risk management.

Linear and Adaptive Model of Strategy Often, in text books of strategy analysis (e.g. Weihrich & Koontz 1992) the strategic process is described for teaching reasons as step-by-step process of (1) analysing the internal and external forces of an organisation, (2) formulating aims and targets on the basis of the fundamental values and beliefs of the decision-makers, (3) defining and evaluating strategic alternatives as well as deciding on a preferred alternative, (4) implementing the chosen alternative. (5) The fifth step comprises the subtasks of controlling and learning in form of new information and knowledge about how to affirm or change the strategy process. There is no problem in using such a step-by-step approach as a heuristic device to understand what different factors and aspects are of importance for strategy analysis. But, nowadays it has become all to clear within the field of strategy research from extensive theoretical discussion (e.g. Schreyögg 1984, Mintzberg 1994, Volberda 1998, Mintzberg et al. 1999, Burgelman 2002, Grant 2005) and empirical findings (Fredrickson & Mitchell 1984, Fredrickson 1984, Grant 2003) that strategy processes do not always follow a simple step-by-step logic to solve complex and dynamic problems. In line with this suggestion, in the field of flood risk management the notion is propagated that decision making (e.g. of water authorities) should be understood as an iterative (e.g. Hall el. 2003, Hooijer et al. 2004) and adaptive process (Schanze et al. 2005, see also Nicholls et al. 2000). To consider real-world conditions of strategy processes, it is a start to consider two process models of strategy making. A process model is a category that refers to organisational decisions and actions. A process model describes how organisations or a subunit decide and act to formulate and implement aims through deploying structural and non-structural measures. Two process models are of special importance (Fredrickson & Mitchell 1984, Chaffee 1985, Volberda 1998, Wiechmann & Hutter 2007): (1) the linear process model and (2) the adaptive model (see Table 3).



Sequential process of planning,programming, and implementation

Top-down strategy making

System of aims, targets and strategicalternatives

Integrated set of strategic,operative and resource plans

Stable

Predictable

Process

Content

Context

Continuous adjustment of societaldecision making and context

Combination of bottom-up initiativesand top-down strategic decisions

System of strategic alternatives, aims and targets

Flexible configurations ofresources

Unstable

Limited predictability

Linear model of strategy Adaptive model of strategy

Table 3: The Linear and the Adaptive Process Model of Strategy

According to the linear model the process of strategy making consists of a well-structured sequence encompassing two main phases: strategy formulation and strategy implementation. Using a linear model of strategy one would describe pre-flood risk management as follows: 1. Analysing the flood hazard and its actual and/or possible damages: Analysis in the linear model is

understood as the analysis of objective relationships which are independent of political and cultural context conditions and societal processes.

2. Formulating aims and targets: On the basis of values and key principles of society, aims and targets for pre-flood risk management are formulated. It is crucial that aims and targets are formulated before action is taken. Hence, in principle, the evaluative basis of mitigating flood risk has to be known.

3. Defining, evaluating, and deciding on strategic alternatives: A decision to implement a certain alternative is only made after comparing and evaluating the effects of possible bundles of measures. No action is taken until a preferable solution has been found (�First decide, than act�).

4. Implementing the chosen strategic alternative: In a narrow sense, implementation can be defined as realising a formally defined programme consisting of more or less specified aims as well as measures to reach the desired effects of the programme (Hucke 1996). Implementation comprises the definition of appropriate organisational structures, staffing, and directing.

5. Controlling and learning: Of course, the linear model of strategy making considers that context conditions can change and expected effects will not occur, but, unexpected will. Characteristically, activities to control, evaluate, and learn are mainly undertaken after the implementation of the strategic alternative is finished.

The linear process model of strategy making has its merits. Aims are formulated on a thorough understanding of the flooding system. The evaluation of a broad range of possible alternatives to reach an aim is undertaken to avoid decisions that refer only to currently already implemented ways of pre-flood risk management. The linear model can encourage decision makers to look for new solutions (Bryson & Roering 1989). It can be seen as a disciplined effort to produce fundamental decisions about how pre-flood risk mitigation should be conducted (see Bryson 1998). The model corresponds with a top-down view on strategy making. Strategic decisions are made mainly by the decision makers



formally responsible for the overall welfare of an organisation or a set of organisations. It is based on the assumption that decision makers are willing to undertake in advance a comprehensive analysis and evaluation of the flood risk system and mitigation alternatives despite the possibility that context conditions can change (e.g. political support for flood risk management can fade), that no easy consistent evaluation of mitigation alternatives in the present is possible (e.g. because of problems in considering how many generations should be taken into account to realise a sustainable flood risk management). To identify the critical elements for improving the current state of management, the linear model assumes a highly stable societal context. The hallmark of the adaptive process model is that strategy does not move forward in a direct way through easily identifiable sequential phases. The process pattern is much more appropriately seen as continuous, iterative and uncertain (Pettigrew & Whipp 1991, p. 27). An adaptive process is characterised by parallel processes of formulating and implementing strategic alternatives. Hence, formulation and implementation are more difficult to differentiate. Decisions for formulating aims and targets, for analysing the internal and external context, and for combining measures are continuously aligned with the changing societal context (Chaffee 1985). Especially political context conditions are taken into account as enabling and inhibiting factors. They do not only affect the implementation of strategic alternatives, but, analysis and evaluation too. Organisational structures not only serve to implement aims and targets, they influence the formulation of strategic alternatives and thereby restrict the organisational potential for solving strategic problems. The linear model works on the assumption that a single decision-maker or an elite of decision-makers (Tonn et al. 2000) can design an explicit �grand� strategy based on a highly top-down, deliberate, analytical process (Volberda 1998, p. 38). Top-down forces are strong. The adaptive model assumes that strategy making requires both strong bottom-up and strong top-down forces (Burgelman 2002). Empirical work has revealed very different roles of managers within such a strategy process. (3) Managers close to stakeholders and operational decisions develop bottom-up initiatives how to improve current management; (2) middle managers are connecting these initiatives with the highly political process at the top level of decision making; (3) decision-makers at the top are responsible for challenging and questioning the status quo, setting performance aims and targets, co-ordinating, evaluating, and legitimising bottom-up initiatives. Empirical evidence shows that the substance of such an adaptive strategy process is significantly influenced by decisions made at lower levels of decision making (Burgelman 2002, Grant 2003). Similar processes of expanding the range of strategy makers and considering the interests of stakeholders can be observed in water management (e.g. House 1999 with regard to citizen participation) and flood management at local and regional level (Fordham 2000, Hutter & Schanze 2004). Strategic spatial planning complements statutory planning (Wiechmann & Hutter 2007). Therefore, it seeks to integrate all stakeholders into the process of discussing notions of long-term spatial development and place making (Healey 2003). Compared with the linear model, the adaptive model is more concerned about developing flexible resources and capabilities for adjusting swiftly to unexpected events and trends (Volberda 1998, p. 43). Adaptive strategy processes can be patterned in a more complex manner. For instance, considering adaptive processes we can distinguish between anticipated implementation during formulation, and actually implementing a formal policy programme for pre-flood risk management. Adaptive processes can lead to the development of strategic projects without the prior intention of implementing a project of strategic relevance for pre-flood risk management. In retrospect, the project is recognised as being of strategic importance (Burgelman 2002). This brief description of the linear and adaptive model underlines the contrast between the two basic models of strategy. Sometimes, a general preference for the adaptive model is expressed. For instance, the �International Manual of Planning Practice� states with regard to the planning system of the UK that �Planning is seen as continuous process. The classical survey/analysis/plan sequence in plan preparation has been abandoned, to be replaced by participatory, objectives achievement approach,



backed up by performance measures, regular monitoring and review. Also, rather than the �predict and provide� method of planning, the emphasis is now on �plan, monitor and manage�. (ISoCaRP 2001, p. 280) However, not necessarily the linear predict-and-provide approach to planning has to be abandoned under all circumstances. The linear and adaptive model of strategy and planning can be understood as complements. The adaptive model argues for a continuous adjustment of societal decision making and context. Hence, under specific conditions it could be appropriate to use the linear model of strategy- making. We already pointed in this direction. The linear model assumes a high degree of stability of the context conditions. Stability can be achieved, for instance, through an established dominant coalition of political and administrative decision-makers at regional and local level in favour of comprehensively analysing the flood risk system and potential damages and implementing strategic alternatives that effectively mitigate flood risk in advance of severe flood events. The question whether the linear or the adaptive model is appropriate for effective strategy making should be answered on the basis of empirical findings. At framework level, local planning often provides general frameworks for guiding land use change and for preparing more detailed regulations on lower tier plans. Framework plans do not determine planning on the lower tier; they help to define the decision situation trough relatively abstract evaluative statements with regard to spatial development as well as social, economic, and environmental aims. Therefore, spatial planning at this level, one could argue, corresponds to the adaptive model of strategy. The linear model with its endeavour to determine all crucial activities in advance, however, seems applicable at the project level (see Table 1). Unfortunately, reality is not as simple as such. It can be argued for just the opposite. Both arguments will be briefly outlined to indicate the background for the first question. The linear model enables decision makers to analyse and assess the current approach to pre-flood risk management at the local level without restricting the analysis right from the beginning to only minor improvements of what has been done till now. The linear model can be an appropriate means against conservatism. But, using the linear model of strategy one runs the danger of finding solutions that cannot be implemented because they lack political support and commitment of private actors. Even if local stakeholders think for themselves that a solution is effective and efficient, they can inhibit its implementation if they were not engaged in the formulation of the solution. From their point of view, procedural justice was sacrificed for swift decision making. For this reason, Burby and associates (2000) argue that integrating all stakeholders into the strategy making process is more important than choosing the �right� specific way of participatory process for mitigating risks through spatial planning. In line with this, local authorities would prefer the adaptive model spatial planning at the framework level and the linear model at the project level. But, is it really obvious that all stakeholders should be integrated at framework level right from the start into the whole framework planning process? The issues of comprehensive spatial planning are complicated, whereas citizens, to put it simply, are mainly interested in planning issues that directly refer to their current individual situation and that can be discussed on the basis of easy-to-understand examples. Powerful private enterprises can pursue strong interests that run counter to a strategic approach to pre-flood risk mitigation (e.g. because this would negatively influence the value of their property in flood-prone areas). Integrating these enterprises into the planning process right from the beginning could hamper a comprehensive analysis of possible locations for additional built development that channels development away from flood-prone areas to areas less at risk. Furthermore, current practices in other local policy fields show that local planning authorities are using the linear model at the framework level, and the adaptive model at the urban district and project level, for instance, for urban regeneration in East German cities (Hutter & Westphal 2003). Against this background it seems useful for empirical inquiry to ask: How are local planning authorities combining the linear and adaptive model for pre-flood risk management with regard to different spatial scales and context conditions?



Propositions for Matching the Process Models to Different Societal Contexts Table 4 compares the linear model and the adaptive process model with the conditions of an effective usage of routines.

Context Routines Linear process model Adaptive process model

Economic growth low to medium High to very high very low to low

Population growth low to medium High to very high -, very low to low

Political support medium to very high High to very high low to medium

Conflict potential very low to low medium to high high to very high

Resources medium to high medium to very high very low to low

Strategic capability very low to low medium to high medium to very high Table 4: Matching Process Models of Strategy Making to Context Conditions (adapted from Tonn et

al. 2000, p. 172)



Strategic Planning

�Think of strategic planning as the organization of hope, as what makes hope reasonable ��

(Bryson 2004, S. 32) Spatial planning for reducing vulnerability in flood-prone areas implies a strategic approach at local level. Analysis and assessment of flood risk and generating flood risk maps require extensive data gathering, skilled staff within municipalities, as well as intensive co-operation between local planning authorities, water authorities, and regional bodies. Integrating citizens into the strategy making enhances public awareness and private engagement (House 1999). Uncertain external context conditions can be considered by formulating scenarios (e.g. identify robust expectations and decisions, defining a set of strategic alternatives to foster flexibility). To learn from experience, framework plans, regulatory measures, and strategic projects have to be connected. However, a strategic approach to spatial planning for risk mitigation demands scarce financial and human resources. Capabilities to develop and deploy these resources have to be built in constant competition with other possible resource uses and competencies for dealing with alternative strategic issues. These other possible uses can be of higher political and administrative weight. They refer to projects that can be interpreted by many stakeholders as �positive� for urban development (e.g. a new school building) and necessary to satisfy the immediate needs of important stakeholders (private enterprises, initiatives of citizens, and so forth). Hence, it should not only be assumed that local planning authorities strategically can decide and act for pre-flood risk management it should be shown how they actually do it and how current practices can be improved. There is some empirical evidence that the required capabilities can be built at the local level (an often cited example from the United States is the strategy of the City of Tulsa, Olshansky & Kartez 1998, Mileti 1999, Meo et al. 2002). But, up to now, the key features and success factors to foster a strategic approach to spatial planning for risk mitigation of many, not only a few, municipalities remain not well understood. Thereby, it is necessary to take the very different context conditions within European catchments into account. Summarising these considerations we ask: How are local planning authorities developing a strategic approach to use spatial planning for pre-flood risk management under very different societal context conditions? Is there a generic approach applicable to very different European contexts? Bryson and Roering summarise their findings from case studies investigating strategic planning in eight local governments in the United States as follows: �� paradoxically, government strategic planning is probably most needed where it is least likely to work. Government strategic planning should appear to work best in units that have effective policymaking boards, strong and supportive process sponsors, superb process champions, good strategic planning teams, enough slack to handle potentially disruptive crisis, experience in coping with major disruptions, and a desire to address what is truly important for the organisation. Any unit with those features probably already uses some sort of �strategic thinking and acting� process.� (1989, p. 606, italics added) Taking this paradox into account, one should focus on those actors that can foster using planning for risk management despite various inhibiting context conditions. Therefore we ask: What actors within municipalities and water authorities are capable of fostering strategic planning to relate spatial planning to issues of pre-flood risk management? Apparently, this question is based on the assumption that it is possible to observe at least attempts of local planning authorities to use spatial planning for pre-flood risk management within catchments (�becoming good cases�). We expect to see anything from improvements to only symbolical alterations and everything in between due to the various conditions of the cases. Empirical research



excludes only local cases where there is presumably no strategic approach at all to mitigate flood risk through spatial planning.

Strategic planning can be defined as �a disciplined effort to produce fundamental decisions and actions that shape and guide what an organisation (or other entity) is, what it does, and why it does it. At its best, strategic planning requires broad-scale yet effective information gathering, clarification of the mission to be pursued and issues to be addressed along the way, development and exploration of strategic alternatives, and an emphasis on the future implications of present decisions.� (Bryson 2004, p. 6, italics not added)

To avoid misunderstandings and endless debates over the net benefits of planning, it is important to make clear wherein the discipline lies. To be disciplined does not necessarily mean to follow a detailed recipe for conducting a strategic planning episode in organizations or regional networks. Empirical research showed that strategic planning can �degenerate� to a mindless bureaucratic ritual that drives out strategic thinking because decision makers follow a prescribed course of action without questioning its appropriateness based on an analysis of current context conditions and future challenges.

To be disciplined means to reflect on the appropriate process for dealing with strategic issues of long-term relevance within specific societal context conditions. Therefore, discipline shows up through � among others � a) recurrent cognitive processes of aligning content with context and b) through creating new categories for catching emerging and uncertain context features, c) through actively searching and welcoming new information and d) paying as much attention to the quality of process as to contents and outcomes (Bryson 2004). The following shows that these core features of strategic planning can be found in planning episodes for long-term flood risk management (e.g., Lonsdale et al. 2005).

In sum, strategic planning as disciplined effort is a stable cognitive process, whereas collective action and formal organizational structures can vary because of changing circumstances. Research on High-Reliability Organizations (HRO) tells us that this is a good recipe for dealing with high complexity and uncertainty (Weick & Sutcliffe 2001).

Apart from these core features of strategic planning, there is great leeway to deploy it for different purposes and different contexts. Hence, to understand planning in practice one has to take a close look at specific cases. For instance, strategic planning can be used to define priorities for allocating a fixed public budget to thousands of flood protection measures after a disastrous flood event like in the Free State of Saxony after the Elbe flood in August 2002. Furthermore, strategic planning can be deployed at regional level for formulating new ideas for long-term flood risk management through scenario analysis. At local level, strategic planning can be deployed for political decision making that exploits the results of regional scenario analysis. These two distinct options facilitate coherent communication about long-term flood risk management at regional and local level (see the Dresden case study). We mentioned before that relating plans and subsequent decisions and actions is crucial. Spatial plans at the framework level of local planning serve as frame of reference; they formulate aims and targets to prioritise projects of spatial development. To illustrate the applicability and implications of strategic plans, strategic projects can be formulated and implemented. Strategic projects are a specific category of projects for spatial development. They are defined as projects that intentionally are tied to strategic plan statements and that depend heavily on co-operation across different levels within organisations and between organisations to achieve their aims. Hurdles during strategic project implementation can be used as learning opportunities to assess the limits of the relevant plan statements and to reformulate them (Figure 8).



Impulses,Hurdles,Limits

Aims, Targets,Priorities

Strategic Projects,Strategic Action

Local Planning atFramework Level

Figure 8: Learning to Link Plans at Framework Level with Strategic Projects

Otherwise, a strategic plan can be influenced by elapsed action that not necessarily was guided by the plan. Within formal planning processes it is possible to interpret impulses from outside as relevant for strategy making. Bottom-up initiatives can be integrated into subsequent strategic plans. To take this into account, strategic action is considered as action unguided by plan statements that retrospectively is recognized as experiential basis for formulating plan statements at framework level (for a similar, but not identical definition of the term strategic action see Burgelman 2002). Planning and learning are sometimes seen as antipodes (e.g. Mintzberg 1994, Mintzberg et al. 1999). In fact, planning can be more a way of preserving old views �how things are down and ought to be down around here� than exploring new strategic alternatives. It can enable organisations to gain focused action on a limited set of strategic issues, related aims, and targets. Especially planning procedures within large-sized organisations can become rigid rituals that inhibit strategic thinking (Volberda 1998, Mintzberg et al. 1999). However, planning and learning are not necessarily opposites. Planning as linear programming is appropriate if the relevant context conditions are stable and largely predictable. Local planning at framework level can be fashioned as scenario-based learning which fosters thinking in multiple possible futures and which enables decision makers to think about their �mental models�. Spatial Planning can be understood as process of learning with plans (not despite of plans) for taking into account the diversity of actors relevant for urban and regional development as well as the limited resources of spatial planners to implement �their� aims and targets (Mastop & Faludi 1997, Mastop 2000, Salet & Faludi 2000).



Strategic Planning Under Uncertainty The linear and the adaptive model encompass the formulation and implementation of strategy. Planning modes are solely about formulation at the framework level of planning. Planning at framework level can be defined as a �disciplined effort to produce fundamental decisions and actions that shape and guide what an organisation is, what it does, and why it does it.� (Bryson 1998, p. 2160) Fundamental decisions are not exclusively produced within the formal strategic planning process. They can be made in more informal processes outside planning processes. Hence, it is crucial to define planning modes not with regard to its substance (�fundamental decisions�), but, as a formal procedure (Mintzberg 1994, p. 12) which comprises the analysis, co-ordination, and explicit articulation of fundamental decisions that usually lead to plans. Especially if the predictability of the context is very limited planning at framework level is more about coordinating and articulating basic decisions than about producing these decisions (Grant 2003). Medium- and large-sized organisations establish a strategic planning system. A strategic planning system can be understood as a more or less elaborate set of formal organisational structures (e. g. strategic planning department, process regulations to channel the strategic planning and related control procedure) and processes to coordinate and legitimise strategic decisions. The strategic planning system is closely related to other management systems (information system, human resource system, financial system, Grant 2005). The basic concept of strategic planning was advanced mainly in the late 1960s and the early 1970s (Grant 2005). Stability of external context conditions was assumed (e.g. industry supply structure, demand conditions). The decision problem was to decide in which industries to be in and how to compete in the chosen industries with regard to the strategies of competitors. In the late 1970s it became apparent that only limited predictability of industry structures and of internal conditions (e.g. knowledge assets) is given and that strategic planning has to cope more actively with unexpected, and sometimes disruptive technological and market conditions. Large enterprises as Shell began to understand scenario-based planning not only as a way to consider different possible future developments of external, uncontrollable conditions, but, as mean to make more explicit the mental models of decision-makers to identify flawed assumptions about �the market� and �the consumer�. Nowadays, strategic planning of business organisations is no unitary phenomenon. It is a set of organisational structures, regulations, procedures, and process patterns that can be applied partly to the non-profit and public sector of developed, �formal� societies (Bryson 1998, Bryson 2003). Spatial planning at framework level of local planning can be fashioned in different ways. With regard to the turbulence of context conditions it is useful to distinguish between three effective planning modes (Figure 9):

• The classic mode of programming is appropriate under conditions which are highly predictable;

• Scenario-based planning considers different plausible futures; it is appropriate under conditions of complex and dynamic context conditions that can be predicted with a sophisticated set of routines and information systems;

• Preparedness strategies increase organisational activities for coping with strategic surprise; a minimum of planning devices gives the strategy process a broad, overall direction in which actions and decisions can adaptively emerge.

This conceptual tool of showing the relationship between planning modes and the extent of turbulence is based on one assumption: To maintain an effective planning process, increasing turbulence can be matched with increasing planning activities (e.g. more process regulations to gather information, more meetings, more techniques to analyse data), but, only until a �turning point�. After this point, turbulence is too high, the context becomes largely unpredictable. Planning should be reduced to a



minimum of necessary devices for co-ordinated action. Organisational activities, in contrast, continuously can be expanded.

lowlow

Uncertaintyof context

PreparednessStrategies

Scenario-based planning

Programming

Extent oforganisationalactivities

Extent ofplanningactivities

high

high

high

Figure 9: Matching Strategic Planning Modes to the Uncertainty of Context (Adapted from Volberda

1998, p. 36)

Empirical findings within business strategy research support this assumption for matching planning modes to the turbulence of context. In the 1970s, strategic planning in large business organisations was conducted in a more top-down and formalised fashion aiming at programming the activities of single departments. Increasingly unforeseeable conditions (demand, strategies of competitors, market prices, and so forth) forced the organisations to develop more flexible and communicative strategic planning systems. Strategic planning became a communicative opportunity in large decentralised organisations to discuss assumptions about possible future developments and performance targets, to use scenario planning, and to discuss strategic decisions made relatively autonomous in different subunits of the organisations. Recent empirical findings illustrate this general pattern for the oil industry (Grant 2003, 2005). In what follows, it is shown that the three planning modes can be related to specific problems of pre-flood risk management. (1) Strategic planning as programming is based on an extrapolation of trends within the external and internal context conditions. No fundamental changes of context conditions are accounted for. Organisational activities are directed towards implementing a predefined strategic alternative. The programming mode corresponds with the linear model of strategy making. For instance, after the severe flash flood event in the Weisseritz catchment, a medium-sized tributary of the Elbe river within the Region of Dresden, the water authority is planning to build additional water retention basins. To make an informed decision about the location and capacity of the basins, a decision making process that resembles the programming mode is now taking place. Thereby, the



water authority is communicating with municipalities and regional actors (e.g. the regional planning office) about the conflict potential of different locations (e.g. conflict between nature conservations aims and flood protection). The water authority is combining the structural measure of building additional water retention basins with the non-structural measure of communicating with stakeholders to reach a decision which easily can be implemented. (2) Scenario-based planning considers that various developments of external context conditions are possible. Scenarios are neither predictions nor forecasts. A scenario is defined as a plausible description of how the future may develop, based on a coherent and internally consistent set of assumptions about key relationships and driving forces (e.g., rate of technology changes, prices). Scenarios can serve different purposes. For instance, scenarios can be formulated to identify robust analytical findings with regard to future flood risk and strategic decisions despite very different possible future context conditions (e.g. Foresight project in the UK). They can be defined to match different strategic alternatives to different possible external developments (e.g. Hooijer et al. 2004). Scenario-based planning can be deployed on the basis of a linear or an adaptive model for pre-flood risk management.

(3) Within the context of flood risk management a preparedness strategy aims at ensuring effective responses to the impact of hazards, including timely and effective early warnings and the evacuation of people and property from threatened locations. Preparedness strategies are developed to cope with strategic surprise. Actors are expecting the unexpected (Weick & Sutcliffe 2001). Therefore, planning activities are decreased, whereas organisational activities are increased. Actors relevant for pre-flood risk mitigation focus less on the question �What should we precisely do?�, but, on the questions �What should we be capable of doing and with whom should we communicate?�. Preparedness strategies connect pre-flood risk management and event management. They are of special importance for coping with flash floods. They enable decision-makers and citizens to develop in advance of the flood event the capabilities of effective and swift communication under the condition of a fast developing flood hazard. But, complex and swift communication has its limits. Flash floods can be too fast for effective event communication and action. Therefore, the effectiveness of preparedness strategies should be compared with strategic alternatives for discouraging further development in flood-prone areas (Hall et al. 2003). In case of comparing pre-flood and event mitigation alternatives a broad range of actors has to be taken into account (e.g. insurance companies, developers, and municipalities with regard to pre-flood management, municipalities and emergency institutions for event management).

Implications of Complexity, Uncertainty, and Ambiguity for Strategic Planning

Uncertainty is not the only information- and knowledge problem of FRM. Terminologies in the literature differ, largely dependent on disciplinary origins. We adopt a terminology that is widespread in management- and strategy research (e.g., March 1994, Weick 1995, Volberda 1998, Zack 2000) and that has found application in risk management of technological and natural hazards (e.g., Klinke & Renn 2002). This terminology distinguishes between

Complexity: this information- and knowledge problem is related to the number and heterogeneity of factors relevant to understand causal structures of flood risk and possibilities to reduce it. The shift towards flood risk management which encompasses, for instance, the Sources-Pathways-Receptors-Consequences-Model (SPRC-Model) can be interpreted as attempt to increase the capacity of FRM research to deal with complexity through describing it. The recent emphasis on vulnerability, for instance, increases awareness about factor which could lead to new interventions for reducing damage potentials in flood-prone areas. A system may be complex, yet, predictions are possible due to stable causal patterns of limited dynamic development and due to attempts to gather and interpret information about cause-effect-relationships.

Uncertainty: In contrast to complexity, uncertainty does not necessarily assume a system with a large number of possibilities. A simple example is throwing a dice. Outcomes are uncertain, but only two possibilities are given. Often, however, uncertainty and complexity are intermingled, especially when it comes to reducing flood risk or risk associated with some other natural hazard. Klinke and Renn



argue that uncertainty problems lead to difficult decisions about balancing the costs of overcautious or careless (under- versus overprotection) actions. In their view, dealing with uncertainty and complexity problems require some technical understanding of how the �causal texture� of risk problems functions. Therefore, the strategy process can, under specific circumstance, focus on specific groups of decision makers and stakeholders (see Table 5).

Ambiguity: Ambiguity is the most severe knowledge problem of decision making. In this case, a decision makers lacks an appropriate framework to define what a decision is (in fact, a decision makers does not know that he or she is one) or has to deal with two many conflicting ways of interpreting the same information. Dealing with ambiguity is based on recognizing, interpreting and acting of subtle differences in values, perceptions, and mental models about the causal structure of decision problems. The implications of ambiguity are most significant in work on possibilities for sustainable development. In this report, ambiguity problems are in the background of argumentation.

Knowledge problem

Problem description

Type of conflict

Strategic aims

Process management

Complexity

Low agreement on relevant factors, causal relations and effective means among experts

Cognitive conflict

Reducing damage potentials and limiting the overall level of risk

Technical discourse among agency staff and external experts

Uncertainty

Lack of information and knowledge about future consequences of action possibilities Variability (objective uncertainty) and limited-knowledge uncertainty

Cognitive and evaluative conflict

Balancing costs of under-protection vs. costs of overprotection Increasing diversity, flexibility and preparedness Making robust decisions

Reflective discourse among selected people with minimal technical understanding (agency staff, external experts and eventually stakeholders)

Ambiguity

Low agreement on values, norms and criteria Equivocality (different frames of references) and �true ambiguity� (no frame of reference at all, Weick 1995)

Cognitive, evaluative, and normative conflict

Reaching a consensus about a socially acceptable development path

Broad participation for collaborative learning (agency staff, external experts, stakeholders and representatives of the public)

Table 5: Matching Knowledge Problems and Options for Process Management (Source: Based partly

on Klinke & Renn 2002)

Uncertainty is of special importance for strategic planning. In contrast to managing running floods, strategic planning with its long-term view on FRM focuses on formulating, implementing and controlling strategies for future flood events. Thereby, long-term planning encompasses decision for the mid term (up to 10-20 years) and, in a more explorative mode, for the long term (up to 50-100 years). Of course, given such long planning horizons this raises the question how deal with the fact that some factors of the flood risk system of river catchments, estuaries, and coastal cells are subject to significant uncertainties of external and internal drivers. For instance, flood hazards are sensitive to climate change; the flood vulnerability evolves according to land-use change. Long-term planners need to explore these system�s dynamics and their impacts on future risks. They need to be able to judge how alternative strategies perform under highly uncertain futures. Furthermore, societal context



of the flood risk system is uncertain too. Flood risk management often evolves through incremental change, but dramatic shifts after flood disasters cannot be excluded. Thus, capabilities to deal with specific uncertainties are crucial for effective long-term planning. However, research in natural, engineering and social science on long-term planning in FRM is far from understanding uncertainty in its full antecedents, processes and implications. There are many ways to define uncertainty for FRM (see FLOODsite 2005). In the context of this paper, it is sufficient to distinguish two different meanings of uncertainty (March et al. 1999, Hall 2003, Schanze 2006, Apel et al. 2004):

• Aleatory uncertainty (externally induced uncertainty, objective uncertainty, inherent variability): Some natural and societal processes may be genuinely stochastic. Understanding those processes requires approximating the underlying distributions from which realized events are drawn. �The expected error in estimation can be decreased by aggregating over several events, but history limits the number of comparable events.� (March et al. 1991/1999, p. 146)

• Epistemic uncertainty (internally induced uncertainty, subjective uncertainty, limited knowledge uncertainty): The second kind of uncertainty stems from measurement and interpretation problems. Measurement error, model misspecification, and model system complexity introduce substantial noise into observations and interpretations. With large samples of events, one can tolerate a relatively large amount of noise, aggregating over events to extract a signal. With small samples, however, aggregation is a less powerful procedure.

Implications for dealing with these two kinds of uncertainty are quite different. Epistemic uncertainty calls for better models, increasing model system complexity and dealing with associated uncertainties as well as acquiring more information about future consequences of possible actions. Inherent variability of the outside world calls for specific decision criteria and rules for dealing with an uncertain world in which there is a limit to build better model systems and in which there is a limit to acquire more information in time. In the real world, both kinds of uncertainty are intermingled and decision makers have difficulties seeing and dealing with the fine line between internally and externally induced uncertainties. Consequently, learning to manage uncertainty of flood risk is an important research issue for natural and social scientist alike.

In what follows, complexity and uncertainty will be analysed in some more detail: Firstly, we deal with the question how to think of strategic planning for whole river catchments. From a natural and social viewpoint, river catchments are very complex systems with multiple spatial and temporal scales, multi-level strategy making and complex process patterns and context conditions. What does it mean to speak of strategic planning for river catchments? Secondly, we approach the question of how actors interested in pre-flood risk management can learn how to deal with uncertainty in long-term planning. Expressions like �Lessons learnt� are often used, but seldom explained. What are plausible relationships between learning, uncertainty of flood risk, and strategic / long-term planning?

Strategic Planning to Integrate Strategy Making at Regional and Local level Distinguishing between context, processes, and contents of strategies for pre-flood risk management leaves open the question at what spatial level strategy making takes place and who the strategy maker are. Spatial scales, actors, and planning horizons are crucial for defining the level of integration. This report focuses on small- to medium-sized catchments and the strategies of actors that are relevant for managing flood risk on catchment scale. Hence, transboundary issues as well as European and national policy issues can be understood as context conditions for pre-flood risk mitigation. Especially the implementation of pre-flood risk management requires a close look at regional and local level of strategy making (Parker 2000, Hall et al. 2003, Hooijer et al. 2004, Schanze et al. 2005). Thereby, (1) different planning horizons, (2) levels of societal decision making and related policy issues, and (3) different spatial scales are considered (Figure 10).



Integrating all relevant decisions ondifferent spatial scales (e.g. regional,local, area)

Strategicplanningfor thecatchment

Strategicplanningfor a partof thecatchment

Strategicplanningof a singleorganisation

Example:Key success factors for integrating thestrategic decisions of various actors withregard to a specific area

Examples:� Planning of the water authority� Planning of a municipality

Integrating differentspatial scales and policy issues

Integrating differentplanning horizonsShort

termMedium

termLongterm

Figure 10: Integrating Spatial Scales, Policy Issues, and Planning Horizons

Integrating Different Planning Horizons Strategic planning for pre-flood risk management implies a long-term planning horizon. Planning refers to the development of spatial patterns of the built environment and related capital-intense investments. However, important strategic decisions with a lasting effect on the whole spatial development process can emerge from unforeseen events. They can require an instant response on the basis of a swift analysis. For instance, the post-flood situation following severe flood events serves as �window of opportunity� to rethink and redesign the current strategy for pre-flood risk reduction (Drünkler & Kuhlicke 2004). Hence, short- to long-term planning should be considered.



Strategic Planning of a Single Organisation Strategy research and organisational theory (e.g. Baum 2002, Tonn et al. 2000, Chakravarthy & White 2002) distinguishes between different levels of decision making. Strategies can be formulated and implemented by individuals, groups, organisations, dyads or triads of organisations, and networks of organisations (e.g. the network of political institution that constitute the national political system). Populations of organisations and society are not capable of formulating and implementing strategies. In this report, European and national policy are considered as external context conditions of strategy making. Organisations, dyads of organisations and networks with regard to local and regional level of strategy making are relevant for theoretical analysis. Organisations are in the foreground of inquiry. They are goal-oriented social entities. They formulate aims and targets for self-orientation and for justifying their decisions and resource allocations to important stakeholders (e.g. funding institutions). They distinguish between members and non-members. Membership is determined by various factors. For instance, political institutions, elections, and employment decisions determine who the members of a municipality are and which decisions and procedures they can deploy to reach their aims. Organisations are characterised by a complex bundle of routines to accomplish specific tasks. Especially a large organisation is differentiated into departments and further subunits (e.g. working groups). The terms municipality and local planning authority are interpreted as synonyms. (The terminology is further specified with regard to the institutional setting relevant within the case studies.) Local planning authorities are the main actors of this report. It is indicated when a specific organisational unit within a municipality is meant (e.g. a department for town planning). Local planning authorities refer to a precisely defined territory as geographical part of a catchment. For this reason, usually, more than one municipality is relevant for pre-flood risk management on catchment scale. Horizontal co-operation (co-operation between local authorities and further local actors, e.g. private enterprises) becomes crucial for exploiting win-win solutions. Solutions with positive and negative impacts require more time. In this report, empirical research focuses on strategies of municipalities as single organisations, especially on their current spatial planning practice with regard to pre-flood risk management. Decisions of water authorities and regional bodies are taken into account as far as it is necessary to understand key success factors of planning-based strategies for pre-flood risk mitigation at local level. Unlike municipalities, water authorities are institutionalised at state (e.g. Germany) or national level (e.g. England). Complex formal as well as informal interrelationships between national, regional, and local level exist in most European countries. The relevance of water authorities for pre-flood risk management is obvious. In most European countries they are responsible for river training and operational decisions with regard to flood protection assets on catchment scale. Water authorities are specialised on managing water quality and quantity and related flood risk issues. They have no or only very limited responsibility with regard to development in flood-prone areas beyond a �narrow strip of land very close to river banks, and there have been numerous examples where planning authorities have disregarded the advice of flood agencies on floodplain development �� (Parker 2000, p. 19) Regional bodies (e.g. regional planning office) can be responsible for the whole range of policy issues on catchment scale, but, for the most part with very limited authority for implementation at local level. In dependence of the policy issue involved and the relevant national as well as European regulations, implementation responsibility of regional authorities varies across Europe. To summarise, strategic planning of municipalities, water authorities, and regional bodies can be very diverse with regard to spatial scale and pre-flood risk management. Therefore, integrating the strategies of these actors becomes a crucial challenge for pre-flood risk mitigation. We consider two possibilities: (1) Developing a strategy for a part of the catchment and (2) crafting a strategy for the catchment.



Strategic Planning for a Part of the Catchment A strategy for a part of the catchment is defined as a strategy to mitigate risk within a geographical part of the catchment. Notwithstanding the heterogeneity of single-actor strategies, a consistent pattern of decisions and actions of multiple organisations can be observed. The scale of the part can vary between very limited (�project level�), limited (�area�), and covering more than half of the catchment. The geographical area serves as a common reference for integrating the decisions and actions of various actors. For instance, an enduring political consensus can be observed that a specific area is of high political, cultural, and economic value for regional development (e.g. historical centre of the City of Dresden). For this reason, structural and non-structural measures are implemented to guarantee a safety level above average compared with other areas of the catchment. This example illustrates the assumption that inclusive strategic planning of multiple organisations for a part of the catchment is more likely to emerge than for the whole catchment. The number of relevant organisations is limited and can be defined with regard to the geographical boundaries of the area. Citizens and private enterprises can more easily be included into the strategy process. Complex issues at regional scale of pre-flood risk management are not easy to communicate to lay people. Furthermore, it is plausible to assume that private actors are mainly interested in spatially limited solutions because on this scale consequences of strategic alternatives are more likely to be interpreted as relevant for their current situation. For instance, a real estate owner with assets that are located within an area characterised by brownfields, a deficient open space structure, low economic and population growth potentials can be interested in supporting public attempts to implement measures to preserve more �room for rivers� (see Figure 6). Additional open space improves directly the green structure of the location, indirectly the value of the real estate if the measures are implemented without limiting the development rights of the property owner. But, he will develop only limited interest in increasing the overall open space structure of the whole basin with regard to nature conservation aims (Bizer et al. 1998), and attempts to increase the resilience for coping with floods on catchment scale (Hooijer et al. 2004).

Strategic Planning for the Catchment A strategy to mitigate flood risk on catchment scale is a very complex endeavour. Such a strategy encompasses all relevant spatial levels, all strategies of single organisations, and all issues relevant for pre-flood risk mitigation (e.g. considering interdependencies of strategies to reach �good water quality� and �protection against flooding�). Despite this complexity, an overall pattern exists. Often, the situation of flood risk management practice in catchments is fragmented (Hall et al. 2003). To overcome this state of fragmentation, various challenges have to be met (Schanze et al. 2005). Within the content dimension, different strategic aims and measures of all relevant actors have to be integrated. Various spatial levels have to be taken into account. Process integration aims at combining the different time horizons and process patterns (linear / adaptive model, planning modes, and so forth). Private enterprises can have a much shorter planning horizon than municipalities and water authorities. Even in one municipality time perspectives can differ between a department responsible for fostering economic development and a department responsible for restructuring open spaces. Context conditions are needed that address the double challenge of (1) fostering autonomous action of single organisations to reap the benefits of specialisation and (2) establishing formal and informal institutions for common orientation and constant communication (see Grabowski & Roberts 1997). Developing strategic planning for the catchment implies increasing the complexity of strategy making for pre-flood risk management. Increasing complexity can lead to consistent or fragmented development processes within catchments (see Bressers & Kuks 2004). Therefore, a strategy should not be assumed. It should be demonstrated what, how, and why a consistent combination of aims, measures, and process patterns exists at catchment scale that is continuously aligned to societal contexts.



Strategic Planning and Projects for Strategy Making on Catchment Scale

�Strategy can seem like Mission Impossible: things are so complicated, so interconnected; where do you begin? The answer here: with a single step. Do something, anything. Venture! As you proceed you will learn, and as you learn you will build. Great strategies grow out of little initiatives.� (Mintzberg et al. 2005, p. 9)

To specify how the relevance of projects is assessed with regard to strategy making on catchment scale, six propositions are suggested (see Figure 11):

Integrating all relevant decisions ondifferent spatial scales within acatchment (area, local, regional)

Planningfor thecatchment

Strategicproject

Planning ofsingleorganisations

Key success factors for integrating thestrategic decisions of the water authorityand local planning authority withregard to a specific project withina relatively small area

� Spatial planning of local authorities 1, 2� Strategic planning of the water authority� Spatial planning at the regional level

Integrating differentspatial scales, policy issues

Integrating differentplanning horizons

Strong influenceWeak influence

Shortterm

Mediumterm

Longterm

1

2

3

4

5

6

Figure 11: The Relevance of Projects for Strategy Making: Six Propositions The six proposition are based on the assumption that even �small� successful attempts to decide and act mutually reinforcing with regard to only a part of the catchment are essential for developing a comprehensive strategy for catchments as a whole. This assumption is backed up by theoretical and empirical studies showing that learning for successful strategies under unstable context conditions requires flexible, experience-based approaches (Fredrickson & Mitchell 1984, Sitkin 1996, Volberda 1998, Weick 2001). In the long run, multiple successful experiences to integrate diverse capabilities, knowledge bases and related measures can finally lead to a paradigmatic change of organisations and networks of organisations (Burgelman 2002).



These small, but strategic relevant experiences have to be described in more detail. Therefore, the case studies analyse the dynamic and context conditions of strategic projects. It is expected that data gathering with regard to spatial planning and strategic projects will concentrate on propositions 1 and 3. Hence, considerations and conclusions referring to catchment-wide strategy making are primarily based on theoretical considerations (e.g. Benz & Fürst 2002) and the interpretation of empirical findings produced elsewhere (e.g. Hutter & Schanze 2004, Evans et al. 2002, Bressers & Kuks 2004, Nissling-Näf & Kuks 2004). Proposition 1: In the short term, the current approaches to planning within single organisations have a strong influence on strategic projects. Especially large organisations like the City of Dresden need strong and enduring external and internal influences to change strategy (�organisational inertia�, see, for example, Burgelman 2002). Therefore, project-based management aims at maintaining communication and dialog between organisations with different responsibility, working cultures and political conditions of strategy making. Strategic projects for combining structural measures and spatial planning can be seen as a specific issue that focuses exchange of information and discussion. Proposition 2: In the short term, successful strategic projects have only a weak influence on the regional strategy process. This proposition is based on the observation that flood risk management within catchments is fragmented rather than integrated (Hall et al. 2003, Schanze et al. 2005). Successful projects can be interpreted as a first step towards integrated flood risk management, without a guarantee that these steps will lead to an enduring strategic patterns for pre-flood risk management at catchment scale. Proposition 3: In the medium term, the current planning approach of single organisations and strategic projects can be mutually reinforcing. In the medium term, integrated strategic decisions of single actors (e.g. local authorities and water authorities) can have a mutually reinforcing effect towards an integrated pre-flood risk mitigation strategy within a specific part of the catchment. This kind of positive feedback loop between the strategic plans and projects to reduce vulnerability can be expected if the area to be protected is of enduring political relevance (e.g. area that comprises the well-known historical centre of Dresden). Proposition 4: In the medium term, many successful strategic projects and the regional strategy process can be mutually reinforcing. Recurrent success of integrating structural and non-structural measures at project level can be interpreted as opportunity for enhancing the overall societal support for pre-flood risk management within catchments. Proposition 5: In the long run, many successful strategic projects have a strong influence on the planning approach of organisations engaged in these projects. Recurrent successes in integrating structural and non-structural measures are expected to have a lasting effect on those local planning authorities and water authority that are engaged in the projects. Despite being embedded in different working cultures and responsibilities, the authorities consider the catchment-wide effects of their planning decisions. Proposition 6: In the long run, strategies of single organisations, strategic projects, and regional strategies can be mutually reinforcing. The strategy process finally leads to a complex learning pattern considering different types of learning, different planning modes, different models of formulating and implementing strategic alternatives with regard to the different contexts of pre-flood risk management within catchments.



Learning for Pre-Flood Risk Management

�Intelligent action is driven both by one´s understanding of the world and adaptive responses to prior experiences. The former is limited by one´s representation or mental model of the world, while the latter is constrained by the limited number of experiences that one may have relative to the vast set of possible actions that one may take. As a result, cognitive and experiential processes are complementary. Cognitive search is broad in that it considers a wide array of alternatives simultaneously, but misspecified in that these alternatives are evaluated on the basis of an incomplete mental model of the world. In contrast, experiential search is narrow because it enables actors to explore only a small set of alternatives at a given moment but lets them test these alternatives on the basis of the actual environment rather than a mere representation of the environment. Models of bounded rationality should incorporate both forms of intelligence.� (Gavetti & Levinthal 2000, p. 135)

Usually, flood disaster trigger hopes that crisis motivates people to learn something new to avoid catastrophes in the future. Scientists write reports entitled �Hochwasservorsorge in Deutschland. Lernen aus der Katastrophe 2002 / Flood Prevention in Germany. Learning from the Disaster 2002� (DKKV 2003). After a disastrous flood, boards of inquiry are convened to allocate responsibility and to prevent further disasters of the same kind. They publicize reports entitled, for instance, �Learning to live with rivers� (ICE 2001). In this context, learning usually refers to the notion of acquiring and using something new: a new way to think of floods as physical processes with increasing variability and flood disasters as largely man-made interruptions of daily life; a new way to deploy the full spectrum of physical measures and policy instruments for improving flood prevention, for instance, through spatial planning to assess and regulate new development on flood plains (Penning-Rowsell 2001). Learning something new is seen as something positive. Learning always improves FRM. The question is if decision makers learn. The question is not how they learn. In contrast, this report treats learning as a broad concept for different forms of learning. Learning can facilitate long-term planning. Learning also can make life difficult for decision makers to deal with uncertainty of flood risk. The notion of learning comes from psychology (see Weick 1996). A traditional definition of learning sounds like this: Learning is �to become able to respond to task-demand or an environmental pressure in a different way as a result of earlier response to the same task (practice) or as a result of other intervening relevant experience �. The sign of learning is not a shift of response or performance as a consequence of change in stimulus-situation or in motivation, but rather a shift in performance when the stimulus-situation and the motivation are essentially the same.� (English & English 1958, p. 289 quoted in Weick 1996, p. 116) Learning is easily observable when context conditions stay the same and performance improves due to recurrent task-relevant experiences of decision makers. In case of repetitious task requirements and a decision maker with the willingness and the resources to learn, learning behaviour will be observed. This definition depicts learning as a �conservative� process through which the learner accumulates knowledge about a given task and context. This definition fits only partially with what has been mentioned at the beginning of this section. It sees learning less as a cognitive process where a new way of thinking is developed. It is more a way of acquiring some predefined behaviour, for instance, learning to effectively use a typewriter for writing letters. To take into account the attempts of decision makers to make some fundamental changes in thinking and acting, the literature on individual and organizational learning has adopted a broader



understanding (Easterby-Smith & Lyles 2003). This understanding elucidates learning as a cognitive process for questioning basic assumptions about the appropriateness of values, aims, tasks, and rules in specific contexts. There are different ways of dealing with the vast literature on learning. We adopt the approach of James March, not least of all, because he has extensively thought about linking learning, risk taking, and strategy making (March 1994, 1999). His distinction between exploration and exploitation is of utmost importance for the arguments of this report: �Organizations divide attention and other resources between two broad kinds of activities. They engage in exploration � the pursuit of new knowledge, of things that might come to be known. And they engage in exploitation � the use and development of things already known.� (March & Levinthal 1999, p. 210) The distinction between exploration and exploitation is not identical with, but resembles others, for instance, collaborative learning and conflict management. Conflict management can be adopted if the problem at stake is moderately known, but, different interests of the involved actors require a mediated process to come to a solution, often in form of a compromise. Collaborative learning is a specific mode of social learning. It addresses an issue that is widely acknowledged to be neither well-understood nor easily addressed. The process is likely to be long and iterative (Tonn et al. 2000, p. 171, Hutter & Schanze 2004). In case of collaborative learning even the problem is not yet well-defined. Strategy has to deal not only with conflicting interests, but, with fundamental uncertainty about the problem to be solved. Under these conditions a long-term learning process is necessary for solving �the problem�. To manage this process and to orientate people, it is useful to identify broad categories of different phases and their characteristics (see the comparative longitudinal research work of Van de Ven et al. 1999 on administrative and technical innovations). Researchers that base their arguments on the distinction between exploration and exploitation have a much narrower scope of study (first and second order effects) than people interested in conflict management and collaborative learning (first to third order effects, see Table 6).

First Order Effects Second Order Effects Third Order Effects • Social Capital: Trust,

Relationships • Intellectual Capital:

Mutual Understanding, Shared Problem Frames, Agreed Upon Data

• Political Capital: Ability to Work Together for Agreed Ends

• High Quality Agreements • Innovative Strategies

• New Partnerships • Coordination and Joint

Action • Joint Learning Extends

into the Community • Implementation of

Agreements • Changes in Practice • Changes in Perceptions

• New Collaborations • More Co-evolution, Less

Destructive Conflict • Results on the Ground:

Adaptation of Cities, Regions, Resources, Services

• New Institutions • New Norms and

Heuristics • New Discourses

Table 6: First, second, and third order effects of strategic planning (Source: Adapted from Innes &

Booher 1999, p. 419



The distinction between exploitation and exploration is a generic one. It should be carefully applied to the specific problem and context conditions at hand. We do this by elucidating exploration and exploitation by using a three-dimensional approach (Figure 12):

Extens

iveRan

ge of s

trateg

ic

altern

ative

s

Limite

d

Exploitation

Near

RationalModel

Exploration

Org

aniz

ing

plan

ning

Ong

oing

Pro

ject

Consequences intime and space Distant

Figure 12: Eploitation, Exploration and Rational Decision Making (Adapted from Gavetti & Levinthal 2000, p. 116) Organizing planning: Case studies on long-term planning in form of scenario-based strategy games show (Lonsdale et al. 2005) that decision makers from different institutions (e.g., water management, spatial planning, emergency services) and different spatial levels (e.g., national, regional, local) think differently about the future dependent on whether they are engaged in ongoing decision processes within their focal organisations (�on-line�) or whether they are involved in activities �outside� their daily work (�off-line�, Gavetti & Levinthal 2000). Organizing long-term planning as a project broadly covers all resources and activities to plan for the future in a team with representatives from different institutions, fixed schedule and time horizon for project realisation, and limited resources, but leeway to think differently (compared to business as usual) and to adopt a fresh perspective on current planning practice. Range of strategic alternatives: Nowadays, it is common to think of bundles of measures and policy instruments as strategic alternatives to realise a given safety standard. Theoretically, it is possible to take a wide range of possible strategic alternatives into account. In practice, history, site-specific socioeconomic conditions and physical processes of the flood hazard will strongly influence the range of relevant alternatives. Furthermore, the appropriate range of strategic alternatives does not only depend on external context conditions, but on the purpose of long-term planning (van der Heijden et al. 2002). For instance, in case of a worst-case scenario planning exercise in form of a strategy game, it can be useful and reasonable to restrict the set of possible actions to three simple strategic alternatives (Lonsdale et al. 2005). This is so because the purpose of the exercise is less to decide between simple alternatives and more to understand decision difficulties that arise while attempting to understand how decision makers could and would decide in distant futures. Consequences in time and space: A central concern of intelligence is the relation between the exploration of new possibilities and the exploitation of old certainties. �Exploration includes things



captured by such terms as search, variation, risk taking, experimentation, play, flexibility, discovery, and innovation. Exploitation includes such things as refinement, choice, production, efficiency, selection, implementation, and execution.� (March 1994, p. 237) Exploitation is driven by experience. Exploitation, therefore, generates clearer, earlier, and closer feedback than exploration. It corrects itself sooner and yields more positive returns in the near term. �As a result, the primary challenge to sustaining an optimal mix of exploration and exploitation is the tendency of rapid learners and successful organizations to reduce the resources allocated to exploration.� (March & Levinthal 1999, p. 213) Against this background, the learning approach of this report is as follows: • Learning and context: Learning is context sensitive because it a) refines existing knowledge and

affirms existing beliefs or b) leads to fundamentally new knowledge and beliefs about causal and evaluative relationships. Whatever the case, affirmation or significant change, learning refers to the context of starting conditions of the learning process. For instance, assuming that attempts to combine structural and non-structural measures for pre-flood risk management in the UK and The Netherlands are more developed than in most other European countries (e.g. Germany), using spatial planning for risk mitigation can be analysed against a different background of already effectively implemented strategic alternatives . For this reason, aspiration levels between European countries will probably differ. Hence, it is crucial to take the societal context of different European countries into account.

• Analysing interfaces between exploration, exploitation, and rational decision making: Experiential learning (exploitation), a broad cognitive search and understanding (exploration), and rational decision making in line with a �logic of consequences� (March 1994) are complementary (see the quote at the beginning). However, research up to now has either focused on (1) explaining how experience influences beliefs and expectations (e.g., Olson et al. 1996), (2) explaining how exploitation and uncommitted resources lead to exploration and innovation (e.g., Van de Ven et al. 1999), and (3) arguing that only extensive analysis of possible futures and strategic alternatives in off-line projects can be qualified as intelligent decision making (e.g., Hastie & Dawes 2001, Merz & Emmermann 2006). Interfaces between exploitation, exploration, and rational decision making have been neglected (Gavetti & Levinthal 2000). This report welcomes the desire to explore these interfaces. However, its focus is more on how exploration influences exploitation and rational decision making than vice versa.

• Exploring distant consequences of strategic alternatives through a project for knowledge creation: Speaking of learning how to deal with uncertainty of flood risk in long-term planning resembles saying that learning can take on the form of project-based exploration how a limited set of strategic alternatives performs in distant futures. Both expressions underline thinking about multiple futures. The former makes explicit the content of thinking (�uncertainty of flood risk�) and the latter how that thinking is organized in planning (�off-line project�) and what it leads to (�new knowledge�).

• Recommendations for intelligent decision making: The report is based on the assumption that balancing exploration and exploitation as well as rational decision making is an important step towards intelligent decision making. Decision makers can increase the odds of intelligence through following recommendations how to make decisions (following Bryson 2004, see March 1994 for a pure outcome notion of intelligence).

• Distinguishing between exploration and exploitation in practice: The distinction between exploitation and exploration is based on analytical thinking. In the �real world�, in most cases a lot of very detailed context-specific and process knowledge is necessary to understand how to entangle exploratory and exploitation activities. Case studies, therefore, are an appropriate way to deal with these complexities (Yin 2003, see examples below).

In the past, people engaged in long-term planning have lost some of their credits because of their disposition to overestimate the possibilities of formulating plans that can and should be implemented in the future. Speaking explicitly about uncertainty in long-term planning should contribute to



regaining some of these lost credit points. The following example from flood risk management research and practice show that this is not only theoretically possible, but under way in practice. Understanding learning as complex process that encompasses exploitation as well as exploration is important for dealing with uncertainty of flood risk in long-term planning. Two examples show why: • The London example puts exploratory activities into the foreground. The example illustrates

some important aspects how to explore uncertainty and its implications for long-term strategy making. It shows that the willingness and ability to anticipate uncertainty in decision making can be more important than coming to definite conclusions how to adopt the findings of such exploration. The example refers to an off-line project (�strategy game�) deploying a worst-case scenario about rapid sea level rise relevant for the Thames estuary (see Lonsdale et al. 2005).

• The Dresden example takes a closer look at how exploration and exploitation influence each other over time in the context of a recent flood disaster. Flood disasters disrupt daily life. People experience high uncertainty. Through disasters, the �crisis-reform thesis� tells us (Boin & t´ Hart 2003), decision makers can learn about changed implications for dealing with uncertain futures rather than about �how to predict or control similar occurrences in the future� (March et al. 1999, p. 140). In Dresden in the short term after the Weisseritz flash flood disaster 2002, this kind of deep learning for dealing with uncertain futures did not materialize. Current decisions for the long-term focus on protecting against similar occurrences in the future through technical measures. Against the background of institutional constraints and post-disaster politics, this raises the question how to sustain exploration in times when discussing highly uncertain futures (e.g., in form of a worst-case scenario) is less than welcome.

If one takes both examples together, this makes clear that learning how to deal with uncertainty of flood risk in long-term planning has a double meaning: (1) It is about taking a long-term view and designing distant futures, often through scenario planning. (2) It is about a long-term effort to balance exploration and exploitation within a complex learning process prone to interruptions, irrelevance for ongoing decision making, and (post-disaster) politics.



Changing the Way How We Think about Change Change can take many forms. It can be planned or unplanned, incremental or radical, recurrent or unprecedented. Changes are observable only over time. The basic concept of change involves three aspects: difference, at two or more temporal moments, between states of a unit or system. Research on organizational change, learning, and innovation (Poole & Van de Ven 2004, Easterby-Smith & Lyles 2003) produced rich findings depicting change processes at different levels of strategy making (individual, group, organization, network of organizations, and population of organizations). In this report, the interplay of episodic and continuous change at group and organizational level is in the foreground of analysis.

Episodic and Continuous Change The distinction between episodic and continuous change is of growing importance for understanding and managing change (Weick & Quinn 1999, Tushman & Smith 2002, Poole 2004). The difference seems to apply to all levels of strategy making (Gersick 1991). Table 7 compares episodic with continuous change. Characteristic

Episodic Change

Continuous Change

Metaphor of organization

Organizations are inertia-prone and change is infrequent, discontinuous, and intentional.

Organizations are emergent and self-organizing and change is constant, evolving, and cumulative.

Analytic framework

Perspective: Macro, distant Emphasis: Short-run adaptation Key concepts: Inertia, deep structure, interrelated parts, triggering, replacement and substitution, discontinuity, revolution Change is an occasional interruption or divergence from equilibrium. It is externally driven. It is seen as a failure of the organization to adapt to a changing environment.

Perspective: Micro, close Emphasis: Long-run adaptability Key concepts: Recurrent interactions, response repertoires, emergent patterns, improvisation, translation, learning Change is a pattern of endless modifications in work processes and social practice. It is driven by organizational instability and alert reactions to daily contingencies. Numerous small accommodations cumulate and multiply.

Intervention theory

Intentional change: �Unfreeze, change, freeze�. Change is inertial, linear, progressive, and requires outside intervention.

Redirection of existing tendencies: �Freeze, reframe, unfreeze�. Change is cyclical, without an end state.

Role of change agent

Prime mover who creates change by finding points of leverage in organization. Change agent changes meaning systems, structures, power balance, control system, constellation of actors / management team.

Sense maker who redirects and shapes change. Change agent recognizes, makes salient, and reframes current patterns. Change agent unblocks improvisation, translation, and learning.

Table 7: Episodic and continuous change (Source: Adapted from Weick and Quinn (1999), p. 366)



Weick and Quinn (1999) characterize episodic and continuous change in terms of its tempo, defined as �characteristic rate, rhythm, or pattern of work or activity� (Random House Dictionary, cited in Weick & Quinn 1999, p. 365). Episodic change is conceived to be �infrequent, discontinuous and intentional� (p. 365), whereas continuous change is viewed as �ongoing, evolving and cumulative� (p. 375). The two forms of change are associated with different metaphors of organization, analytical frameworks, theories of intervention, and roles attributed to change agents. The distinction between episodic and continuous change is correlated with several others, including strategic and incremental change (Poole 2004, p. 5).

Strategic and Incremental Change In this report, the terms strategic and incremental change refer to the outcome of change processes (see Boyne et al. 2000 for further meanings of incremental and incrementalism). Incremental change (Burgelman 2002) happens within a well-established frame of strategic orientation that the dominant coalition of the organization holds for gathering, analysing, interpreting internal and external data and for action (Daft & Weick 1984). Tushman and Romanelli (1985) propose that a frame encompasses strategic aims and key products, organizational and power structures, as well as reward and control systems. Incremental change is not necessarily small. It is crucial to recognize that incremental change is well-known to the organization which can hold true for small and large resource commitments (Burgelman 2002). In contrast, strategic change is a change of the frame itself and therefore accompanied by perceived uncertainty of organizational members with regard to the implications of change for delivering key products, realizing important aims, and holding on to established organizational and power structures.

Towards a Complex Process Theory of Strategic Change Based on the two dimensions of (1) basic change process and (2) realized outcome of change, the distinction between strategic reorientation and renewal becomes clear (see Table 8).

Basic change process

Episodic change

Continuous change

Strategic change

Reorientation

Renewal

Outcome of the change process

Incremental change Incrementalism as theory

Organizational everyday life

Table 8: Four types of organizational change processes and their outcomes Strategic Reorientation: Tushman and Romanelli (1985, p. 179) define strategic reorientations of organizations as the result of swift and complementary changes in core products and markets, organizational and power structure, as well as reward and control systems. They argue that effective strategic orientations are often accompanied by turnover in the composition of the top management team and significant changes of corporate culture. Leaders do not only engage in symbolic activities. They make �hard� and concrete strategic decisions about how to fashion a new organization. Change can be identified by referring to new symbols, new people, new products, new structures, and new performance measures. The role of charismatic leaders is to change inertia-prone organizations that cannot renew themselves without outside intervention. Tushman and colleagues have a pessimistic view on deploying continuous processes for strategic change. Strategic Renewal: Robert Burgelman (1991, 2002) posits that strategic change not necessarily happens in form of an episode of reorientation. Based on a longitudinal single case study, he shows



that continuous change processes can lead to strategic renewal. Referring to evolutionary theory, he defines strategy as theory of the organization about the reasons of past and current successes (2002, p. 11). Strategy as a collectively shared cognitive structure explains success and retains explicit knowledge about external conditions (e.g., important stakeholders). Strategy is a precondition to avoid dissipating scarce resources over a set of activities that lack coherence (�strategic neglect�). Strategy as success theory is embedded in distinctive organizational competences and the �character� of a corporation. In line with theories of episodic change, Burgelman assumes that organizations are inertia-prone, but this does not necessarily mean that change happens only through episodic change. Leaders can shape cultural conditions (e.g., rules for agreeing in case of diverse perspectives and interests) and structural mechanisms (e.g., criteria for resource allocation) to continuously balance the competing demands of following a given strategic orientation on the one hand and experimenting with new initiatives on the other. Strategic Change as Complex Process: Whether in form of reorientation or renewal, strategic change can be understood as the outcome of a complex organizational process that encompass multiple organizing modes at multiple levels of strategy making (e.g., Machado & Burns 1998, Pettigrew 1990). To understand such processes, Van de Ven and Poole distinguish between four process theories of organizational development and change: teleological, dialectical, life-cycle (e.g., institutional) and evolutionary change (Poole & Van de Ven 2004, see Figure 13).

EVOLUTION

LIFE CYCLE TELEOLOGY

DIALECTIC

Pluralism (Diversity)ConfrontationConflict

Population scarcityEnvironmental selectionCompetition

Purposeful enactmentSocial ConstructionConsensus

Immanent programRegulationCompliant adaptation

Stage 4 (Terminate)

Stage 1(Startup)

Stage 2(Grow)

Stage 3(Harvest)

Dissatisfaction

Search /Interact

Set / EnvisionGoals

ImplementGoals

Variation Selection Retention

MultipleEntities

Unit ofchange

SingleEntity

Prescribed ConstructiveMode of change

Conflict SynthesisThesis

Antithesis

Figure 13: Process Theories of Organizational Development and Change (Source: Van de Ven &

Poole 1995, p. 520) Complex theories of change combine multiple types of process theories. The propositions of Tushman and Romanelli (1985, 1994) and Burgelman (1991, 2002) both can be interpreted as complex theories for explaining strategic change (Hutter 2006). Strategic reorientations are due to a teleological change processes driven by new leaders. Subsequent periods of convergence and stability within a given strategic orientation are dependent on evolutionary forces and product life cycles. In contrast, explaining strategic renewal requires the deployment of all four types of process theories.



2.5 The Context Dimension: External and Internal Constraints Some elements of societal context relevant for pre-flood risk management are listed in Figure 1. The list is not exhaustive and it is doubtful if such a list could ever be defined. The context of pre-flood risk management is not a fixed entity, but follows from which management problem is in the foreground of inquiry. The context comprises variables that cannot be altered within the chosen time perspective. This is not to say that there is no dynamic relationship between context, content, and process. Context reduces the complexity of possible decisions and actions to a manageable amount. Hence, context enables actors to make informed decisions because many relevant variables can be treated as given (e.g. responsibility of water authorities in Germany for defining floodplains (�gesetzliches Überschwemmungsgebiet� nach § 32 WHG) and responsibility of spatial planning authorities to formulate development plans for areas behind structural measures as dykes etc.). Context enables and controls human agency. Human agency and societal processes can change context, e.g. the context of integrated water basin management (Bressers & Kuks 2004). However, it is very demanding to study the dynamic relationship between context and process (Chakravarthy & White 2002). A pragmatic approach to consider context seems appropriate. In this report, external context conditions are understood as boundary conditions for analysing the content and processes of pre-flood risk management at local level. From the point of view of strategy making within small- to medium sized catchments, the external context comprises political, economic and social conditions (e.g. real estate market prices, policy guidelines of federal and state government, national spatial planning system). Internal conditions influence urban and regional development from within. • Political factors determine the extent and duration of support for pre-flood risk management and

the way of dealing with (potential) conflicts between public and private interests. For instance, it can be assumed that discouraging further development of settlements on floodplains systematically through spatial planning requires an enduring high extent of political support.

• However powerful politicians may be, embedding political decisions and actions in local and regional cultural conditions should not be underestimated. Strategies to channel growing demands for housing to areas unlikely to be flooded is more likely when politicians and officials responsible for land-use planning are supported by citizens and intermediate organizations (e.g., voluntary initiatives to increase citizen involvement) that follow the �instructions� of risk conscious people (e.g., Merz & Emmermann 2006). In this case, local and regional actors decide to take the responsibility for pre-flood risk management without referring primarily to the external political and administrative context. They believe that the adverse impacts of flooding are not an external event due to a hazard, but consequences of prior decisions to invest in flood-prone areas (Mileti 1999).

• Internal financial resources have an important impact on how pre-flood risk management is developed. A high extent of resources currently not needed to accomplish predefined tasks (�organisational slack�) facilitates experimentation with new structural and non-structural solutions because strategic projects can be implemented without significantly changing existing priorities. However, fiscal resources deeply depend on the structure of the national fiscal system producing specific forms of fiscal stress (see Table 9).

These are only simple examples how the societal context of strategies for pre-flood risk management can be considered. A more detailed understanding of context and the interplay of context, process, and contents requires a closer look at specific cases of using local planning for pre-flood risk reduction. Therefore, case studies are conducted (see chapter 3). Across Europe, context conditions of pre-flood risk management, strategic planning, and learning are highly diverse. Even within one case study, it is not possible to analyse all potentially relevant conditions. Case studies focus on specific external and internal context conditions. In a way, it is the aim of case studies to determine which context conditions are of special significance to understand



case-specific decisions and actions. Table 9 shows that the Task-13 case studies focus on questions how context conditions arising from multi-level governance, a recent flood disaster, initiatives for area-based integration of functional policies and local cultures are in the foreground of analysis. This does not mean that the others are negligible. Further studies should shed more light on how economic growth policies, new legislation, and fiscal stress could influence building strategies for pre-flood risk management. Furthermore, Table 9 makes clear that Task 13 focuses on driving forces and constraints towards flood risk management that are ground predominantly in internal factors of local and water management authorities. Context

Driving Forces for Change

Inertial Forces (Forces for not-changing)

Economic Competition

In the long-run, competition favours regions with a high safety standard and a well-developed risk culture

In the short run, competition presses economic and political actors to avoid costs for increasing the safety standard and for developing a risk culture

New laws and Guidelines

New laws and guidelines at national (e.g., PPG 25 in the UK) and European level (e.g. Draft of the Floods Directive) foster a more risk-based approach for dealing with floods

New laws and guidelines are not fully supported by existing laws, guidelines and informal rules

Fiscal stress

Fiscal stress favours efficient management solutions and joint financing among public authorities and between public and private actors

Fiscal stress forces politicians and officials to sacrifice safety for pursuing economic growth (�safety follows growth, but not vice versa�)

Multilevel governance*

New forms of multilevel governance support a catchment-based risk governance approach

Old forms of multilevel governance (e.g., �functional alliances�) can be a hindrance for developing a catchment-based risk governance approach

Local response to a flood disaster*

The shock of a flood disaster triggers a local response consistent with a flood risk management approach; the response is articulated by leading politicians and officials

The shock of a flood disaster triggers a local response consistent with the pre-disaster flood protection approach; the response is articulated by leading politicians and officials

Flood risk and risk culture*

Pre-flood risk management is part of a wider risk culture that explores the unexpected and that wants to learn from successes and failures

Protecting against floods and searching for �absolute safety� is part of a wider local culture that prefers the expected, routines, and learning from successes

Area-based policy integration*

Local politicians and officials search for new modes of area-based policy integration to supplement or supplant the functional organisation of the welfare state

Politicians and officials have only limited capabilities and resources to establish new modes of area-based policy integration

Socio-cultural local groups

Socio-cultural groups want to integrate local knowledge and decisions about environmental quality, the quality of life of places, and also about natural hazards

Socio-cultural local groups focus on environmental quality and the quality of places, but not on rare and uncertain natural hazards and their consequences

Table 9: Context Conditions for Pre-Flood Risk Management in European Member States � A List of Potential

Forces for Analysing Specific Cases (Based Albrechts et al. 2003 and other sources)



3. Case Studies 3.1 Aim, Approach, and Research Design The aim is to use three case studies to show the issues involved in managing flood risk in large-urban areas where there are pre-existing flood defence structures, but an emerging new approach that moves toward flood risk management rather than flood protection. After various recent flood disasters in European Member States, there is a wide recognition of the centrality of flood risk management and the need for a pro-active approach to strategy development. However, there may be considerable inertia within societal decision processes due to high investments in existing physical structures and planning decisions. This leads to the overall research questions of the case studies:

What are specific driving forces to change from flood protection to strategies for pre-flood risk management in European Member States? What are inertial forces? How should we manage driving and inertial forces of change?

In what follows, case study findings from three European Member States are presented: Germany, Italy, and the United Kingdom (England). The studies show important, but not all challenges and opportunities for moving from flood protection towards risk management. They focus on deploying physical measures, spatial planning and communicative instruments at local level in receptor areas. Hence, the following is a contribution to pre-flood risk management, not the whole story. The case study findings should be complemented by studies that place strategy making at regional level in the foreground of analysis.

Why Case Studies? Compared to research strategies like surveys and experimental designs, the advantage of case studies is that they allow a rich account of the complexities involved in situations where an old approach (�flood protection�) is still important and a new approach (�flood risk management�) gaining currency. In this case, it is difficult to reduce the object of research to a limited set of variables that can be easily controlled (rationale for experiments) or easily distinguished from the wider context of investigation (rationale for a survey). Furthermore, case studies are appropriate when questions how to manage the process are important. Based on the case study findings and the framework presented in Chapter 2, Chapter 4 will formulate recommendations to local planning agencies and water authorities for developing strategies for pre-flood risk management.

Multilevel Approach To facilitate the comparison of findings from different case studies, cases with similar abstract strategic content (aims, targets, measures) were chosen: Case studies analyse if, and if yes, how physical measures are combined with spatial planning and communicative instruments (informal spatial planning in particular) at local level to reduce vulnerability and to increase preparedness in flood-prone areas. To analyse the change from flood protection to flood risk management not only in abstract terms (see Chapter 2), but also by considering examples where a change is likely, each case study analyses specific sites. Hence, we faced the challenge of combining abstract terms and arguments with very context-specific information. Based on various research team meetings and e-mail communication, a multilevel methodological approach was chosen to face this challenge (see Table 10):



Level Features Functions

Level 1: Research questions and general concepts

• General • Non experiential • (1) Similar abstract strategic

content, (2) different processes, and (3) very different contexts

• Input to empirical analysis

• Help define research design • Facilitate comparison of case

studies (e.g., common approach for defining high-level strategic alternatives)

Level 2: (1) Conceptual questions and (2) strategic issues

• Specific • Substantive • Suggestive • Formulation (1) at the beginning

and (2) the end of empirical analysis to interpret results

• (1) At the beginning: Specify

research questions • (2) At the end: Suggest common

strategic issues in terms of recommendations to practitioners how to develop strategies for pre-flood risk management in Europe

Level 3: Case-specific questions

• Particular • Experiential • Embedded in narratives • (1) Different concrete strategic

content, (2) different processes, and (3) very different contexts

• Test the practical relevance of

conceptual questions • Illustrate strategic issues

Table 10: A Multilevel Approach to Conducting the Case Studies (Based partly on Burgelman 2002)

Research questions helped to define the design of the comparative case study approach (see Hutter 2005). Research questions are general and abstract. They have to be complemented by conceptual questions / strategic issues and case-specific questions. Conceptual questions are an input to empirical analysis. After having completed empirical analysis, interpreting the analytical results leads to the definition of common strategic issues in terms of recommendations to developing strategies for pre-flood risk management in flood-prone areas. Hence, like conceptual questions strategic issues are located at Level 2. Unlike conceptual questions, they are emerging during and are fixed at the end of the empirical research process. Case-specific questions directly apply to only one specific case study.

Research Design Empirical studies analysing spatial planning for pre-flood risk management at local level within European catchments are rare. For instance, the work of Böhm et al. (2002) is restricted to regional planning and interregional relationships in large river basins (e.g. the Rhine river basin). Spatial planning of municipalities is not in the foreground of analysis. Studies that share with this report the focus on process, local level, and different societal contexts are to be found in the field of event management research (e.g. Parker & Handmer 1998, Rosenthal & t´Hardt 1998). Studies referring to integrated water management can be instructive (Bressers & Kuks 2004). In a research situation where well-known examples with a similar research design are lacking, we should look for general methodological considerations that are applicable to a wide range of research problems for defining the case study design of this report. The case study design is defined in line with the widely acknowledged methodological considerations put forward by Robert Yin (1993, 2003). Four features of the design are salient:



• Embedded design: Defining the case as Local spatial planning at framework level to reduce vulnerability in flood-prone areas and a subunit as Strategic project to combine structural and non-structural measures, especially spatial planning (no holistic case study design),

• Collecting mainly retrospective data (limited real-time analysis, no longitudinal research design),

• Multiple-case study design: Comparing cases with similar contents, different processes and very different contexts,

• Describing the cases and subunits of analysis (no explanatory or exploratory study).

Embedded Design: Defining the Case and the Subunit Case studies can be conducted as holistic or embedded designs. The advantage of a holistic design is that research focuses on the whole case and tries to avoid getting lost in the details. But, holistic designs can be too general. In the case of an embedded design a two-level analysis is conducted. Context, case, and subunit(s) have to be defined. An embedded design has the advantage of allowing (1) cross subunits analysis within a case and (2) cross-case analysis in case of a comparative study. Embedded designs risk the danger of getting lost of the case because the subunits are easier to study. This report follows an embedded design. Table 11 compares the case and subunit.

Case study Contents Process Questions

(1) Case: Local spatial planning at framework level to reduce vulnerability

• Spatial plans of local authority • Strategic plans of water authorities • Spatial plans at the regional level

• Linear or adaptive process? • Strategic planning: How, Why?

Who uses it? • What are the implications for

learning?

(2) Subunit: Strategic project to combine structural and non-structural measures, especially spatial planning

• Project within an area close to

rivers (often within a municipality) • Plan or concept with a limited set

of aims and targets formulated for the specific area

• Relevance of physical measures, formal and informal spatial planning

• Is the project explicitly related to

statements in strategic plans? • How is the project related to

statements in strategic plans (linear/adaptive)?

• How do the water authority and the local planning authority co-operate?

Table 11: Comparing the Case and the Subunit of a Case Study



Defining the case In general, a case is defined as a complex phenomenon that cannot be easily isolated from its context (Yin 2003). In this report, a case is defined as Local spatial planning at framework level to reduce vulnerability in flood-prone areas. This case definition primarily refers to decisions and actions of local planning authorities. However, spatial planning within catchments is a highly communicative phenomenon that comprises contributions of multiple local planning authorities, representatives from water authorities, and regional bodies. Therefore, in this report, formal planning processes and related documents as well as informal co-operative practices are taken into account. It is assumed that planning activities of municipalities and water authorities (as well as regional bodies) are most important for pre-flood risk management at local level. This notion holds true for cases of pre-flood risk management within catchments that currently show at least significant attempts / initiatives to deploy spatial planning for risk mitigation and for considering the overall relationships between water management and spatial planning (�becoming good cases�). Analysing how flood risk is managed within catchments without such attempts (�still bad case�) would probably lead to the normative conclusion that spatial planning is needed for pre-flood risk reduction, but, it would be rather difficult to find empirical information on this topic. In case of a well-established planning-based strategy to reduce flood risk on catchment scale (�already good case�), it would be necessary to consider actors from sector policies (e.g. public infrastructure policy, housing policy). Because spatial planning within catchments will be analysed with regard to two strategic projects, the plans and concepts of two local planning authorities (local authority 1 and 2) will be investigated in some detail.

Defining the subunit In this report, a subunit is defined as Strategic project to combine structural and non-structural measures, especially spatial planning. It is closely related to the case. Spatially, a subunit refers to a geographical area which is a relatively small part of the catchment. Temporally, a project has a short- to medium term planning horizon. Because of the limited spatial scale a well-defined set of decisions of local planning authorities and water authorities is relevant. Local planning authorities and water authorities intentionally tie strategic projects to plan statements at the framework level of spatial planning. Often, but not necessarily, explicit plan statements can be found that articulate the features and expected effects of a strategic project. They depend heavily on co-operation between local planning authorities and water authorities to achieve their aims. Dealing with hurdles during strategic project implementation can be used as learning opportunity to assess the limits of the relevant plan statements. Grounded in experience, plan statements can be reformulated. In particular, three criteria for defining subunits are important: (1) Reducing vulnerability at the project level should be a critical element of the project. (2) There is a preference for flood-prone areas where public ownership and private assets can be found. For identifying the ownership structure of a strategic project documents and interviews are appropriate methods. (3) It is not necessary to identify successfully implemented strategic projects to reduce vulnerability in flood-prone areas. We expect to see anything from major improvements to only symbolical alterations and everything in between, due to the various conditions of the subunits (see Bressers & Kuks 2004 for a similar approach).



Collecting Mainly Retrospective Data Defining the case and subunit is crucial for data collection and analysis. Table 12 illustrates that data collection emphasises the level of case analysis. This allocation of data gathering resources assures that the analysis of the case does not get lost while studying the more concrete phenomenon at project level.

Case study Content, Process, Context Data collection

(1) Case

• Strategic planning of local authority • Strategic planning of water authority • Spatial planning at the regional level

• Documents 1, 2; Interviews 1-3 • Documents 3, 4; Interviews: 4-6 • Documents 5, 6; Interviews 7-9

(2) Subunit

• Strategic project

• Documents 7, 8; Interviews: 10-12

Case or Subunit

• Flexible Allocation • Documents 9-n; Interviews 12-n

Table 12: Allocation of Data Collection Resources

To identify interesting strategic projects, planning documents are analysed distinguishing between two types of documents: a) project-specific documents intentionally formulated for reducing vulnerability in flood-prone areas through structural and non-structural measures; b) spatial planning documents that serve a wide spectrum of different purposes (e.g., land use plan at the framework level of local planning to foster the reuse of brownfields and to develop the green structure of an urban district). Changing from the paradigm of flood protection through deploying structural measures to flood risk management is a complex and long-term endeavour (Sayers et al. 2002, Hall et al. 2003, Schanze et al. 2005). Therefore, it could be promising to undertake real-time analysis of strategic processes within different societal contexts. Studies conducted by Pettigrew & Whipp (1991) and Van de Ven et al. (1999) show that longitudinal comparative case studies with extensive field observation enable us to analyse how complex process patterns are becoming reality in practice. Scientific generalisations can be combined with a detailed understanding of linear and adaptive processes of managing in the real world. However, if research resources are too limited for a comparative longitudinal case analysis, it seems appropriate to �identify interesting strategic dynamics and work backwards to study the precursor conditions and processes that stimulated, guided, channelled, directed and limited subsequent actions and decisions.� (Chakravarthy & White 2002, p. 201, italics added) The case studies of this report follow a retrospective comparative approach to strategic dynamics. Hence, interesting strategic dynamics have to be defined to analyse the precursor conditions and processes that stimulated and channelled decisions and actions that led to these outcomes. Thereby, a step-by-step approach seems to be useful: Firstly, document text analysis indicates interesting strategy dynamics. Secondly, the preliminary interview analysis can enhance the findings from text analysis. Thirdly, interview analysis considers data on context conditions and processes. Fourthly, the findings are compared with different and similar findings in the literature (Eisenhardt 1995, p. 67). The step-by-step approach is slightly different with regard to the level of case and subunit analysis: To define relevant strategic projects, specific planning documents as well as information collected at the level of case analysis can be useful. Hence, case information should be collected to some extent before analysing specific decisions and actions at the subunit / strategic-project level (successive approach to data collection).



The Replication Logic: Comparing Cases and Subunits Compared to single-case investigations, multiple-case study designs are preferable. Multiple-case studies enable richer comparisons. This enhances the investigation of differences and similarities between cases. Multiple case studies should be interpreted in analogy to conducting multiple experiments (Yin 2003, pp. 46 ff.). Strategic projects will not be analysed across cases because the contexts of the cases substantially differ. Conclusions aiming at showing similarities between strategic projects risk the danger of fostering a project-based approach to strategy making which is clearly not appropriate to understand strategies for pre-flood risk management as a multidimensional phenomenon. Therefore, projects are strictly analysed within the overall case study. Analysis is mainly concerned about the overall question how to develop a strategic approach for using spatial planning at the local level to mitigate flood risk within catchments. Therefore, the details of specific strategic projects illustrate and connect the strategy analysis with concrete information on current practices of pre-flood risk management. Nevertheless, the analysis remains relatively abstract to define critical strategic decision situations of local planning authorities with regard to their specific societal context which is heavily influenced by regional and national policies. Of course, this does not exclude that similarities across European catchments can be found (e.g. influence of EU-policies on local decision making, see Bressers & Kuks 2004).

Describing the Cases and Subunits Case studies can be conducted to realise exploratory, descriptive, and explanatory purposes. This report is based on a descriptive purpose. The reasons for this are the following: The case studies address different flood types (flash floods, plain floods, tidal flooding). Furthermore, because of very different context conditions between the German, English, and Italian cases, it is expected that cases, subunits, and contexts differ (�theoretical replication logic�, see Yin 2003). Differences in process patterns can be attributed to differences in flood types and to differences in contexts (economic and social conditions, political support for flood risk management, and so forth). Up to now, no overarching theory to explain flood risk management strategies with regard to different flood types and very different context conditions has emerged as a basis for formulating hypotheses which can be tested through surveys and case studies. Hence, from a pragmatic point of view, it seems reasonable to conduct descriptive case studies which contribute to a step-by-step approach for developing a descriptive as well as causal understanding of pre-flood risk management within river catchments. A descriptive study wants to describe a phenomenon with regard to theoretically relevant features. Hence, description is based on theory. Its aim is not to explain a phenomenon referring to propositions and related hypotheses claiming to link dependent and independent variables, but to describe the phenomenon systematically. Within this research context, theory facilitates decisions what to describe and how to organise and interpret the description with regard to the questions of the study and the relevant body of scientific knowledge and policy issues. Therefore, even if the case study design is descriptive, it is useful to formulate and use propositions for aligning process patterns effectively to the context of local flood risk management. Unlike hypotheses (see Van de Ven 1999), the propositions are not �tested� in a strict sense through conducting the case studies. They focus data gathering through defining interesting relationships between variables of content, context and process.



Dimension

Comparing the cases

Content (�What?�)

Similar strategic content: • Similar Aims: Reduce vulnerability (exposure) in flood-prone areas • Similar plans: Spatial plans of local authorities, strategic plans of

water authorities, and plans of regional bodies • Similar issue for co-operation: Strategic project to combine structural

and non-structural measures; emphasis on spatial planning • But, different types of floods (flash flood, plain floods, tidal flooding)

Process (�How, Who?�)

Different process patterns to formulate and implement strategic alternatives with regard to the context conditions of the cases (�situational approach�):

• Linear process model (strict sequence of analysis, evaluation of alternatives, implementation, monitoring, and so forth); adaptive process model (�top-down and bottom up�, �learning-by-doing�)

• Strategic planning modes: Programming, scenario-based planning, preparedness strategy

• Connecting spatial plans at framework level with strategic projects • Learners and different learning processes (exploitation, exploration)

Context (�Why, When?�)

Very different natural and societal context: • Different types of catchments • Different history of flood events • Different economic development (different economic growth rates and

economic structures) • Different size of cities (metropolitan city, city, small and medium sized

city) • Different population development (shrinkage, stagnation, growth) • Different external political and administrative context (e.g. spatial

planning system in Germany, England, and Italy; different responsibility of water authorities)

• Different institutional history and culture of municipalities / metropolitan cities

• Different internal political context on local level (e. g. different perception and usage of EU policies on local level, political support for pre-flood risk reduction through spatial planning)

• Different local potential for conflict between public and private interests

• Different resources for the formulation and implementation of area-based initiatives to combine structural measures and land use plans (financial resources, staff, and so forth)

• Different strategic capabilities of local planning authorities to manage low probability / high consequence events (e.g. flash flood with high negative impacts on people and properties)

• Different capabilities of local planning authorities to combine structural and non-structural measures (e.g. capability of constant and integrative communication between different departments of municipalities, capabilities to change knowledge-bases)

• Different capabilities of citizens to cope effectively and timely with unexpected events / floods (�preparedness�)

Table 13 Comparing the Content, Processes, and Contexts of the Cases



Assessing the Validity and Reliability of the Results Table 14 summarises how the validity and reliability of the case study findings are being assessed.

Criteria Case Study Tactic Phase of research

Construct validity

- Use multiple sources of evidence - Establish chain of evidence

Data collection

Internal validity

- Explicitly define case and subunit - Use models

Data analysis

External validity

- Use replication logic in multiple-case studies - Use typology to describe context

Research design

Reliability

- Use case study protocol - Develop case study database

Data collection

Table 14: Assessing the Validity and Reliability of the Results (adapted from Yin 2003)

The criteria construct validity is about the concern of establishing correct operational measures for the concepts being studied. Valid case studies assure that theoretical categories and propositions are transferred into concepts which are specific enough to be related directly to the data of the case (Burgelmann 2002). Two points should be made with regard to the criteria of construct validity: (1). It is important, to define conditions that hold true for related planning activities at framework level of local spatial planning, strategic planning of water authorities, and regional planning. For instance, planning statements within a Städtebaulicher Rahmenplan (development plan at urban district level) referring to plans and concepts of the water authority can be interpreted as indicators of related planning decisions. (2) At project level, a valid concept referring to the notion of a strategic project to combine structural and non-structural measures, especially spatial planning is needed. Multiple sources will be used to qualify a project as a strategic project (document analysis, interviews). There is some empirical evidence (Wiechmann 2003, Brake 2000) that especially big cities (like Dresden, Leipzig) and capital cities (like London, Wien, Berlin) use the term �strategic project� within their planning documents. Successive data collection (from case level analysis to analysis of subunits) will enable us to use data from documents and interviews at framework level for identifying interesting attempts to combine structural and non-structural measures at project level. The criteria internal validity usually refers to the task of avoiding false conclusions about causal relationships between at least two variables. Our case study design follows, as mentioned above, a descriptive purpose. Hence, the question of internal validity is of relevance for a different reason. To conduct an internal valid analysis can be interpreted more widely as the task of generating correct conclusions from empirical data. This task is relevant for exploratory, descriptive, and explanatory case study designs. With regard to this criterion, two points should be stated: Firstly, for internal validity an explicit attempt to link the analysis of cases and subunits is crucial (see above). Secondly, qualitative models are used for describing different process patterns of strategy (see chapter 2). External validity deals with the problem of generating general knowledge from case studies. The mode of generalising from the analysis of data is different between survey research and case studies. Case studies rely on analytical generalization (Yin 2003, p. 37). In the case of a descriptive case study analysis, generalization aims at generating conceptual frameworks for describing specific process patterns within context. The description (e.g. in form of a process model) should be tested by replicating the findings in a second or more case studies. The results might be accepted as providing strong support for using the model as an appropriate description of strategies. The objective of using the criterion of reliability is to be sure that if a later investigator followed the same procedures as described by an earlier investigator and conducted the same case study all over again, the later investigator should arrive at the same findings and conclusions. Appropriate tactics to



deal with this concern are case study protocols or developing a specific database for the case analysis. Especially protocols will be developed to facilitate the gathering and analysis of data across the very different context features of the European cases and subunits. Protocols are developed for analysing (1) planning documents at framework level and (2) documents related to strategic projects.

Comparing the Research Design with Alternatives A comparison with the design of other empirical studies highlights some features of the comparative case analysis outlined in this report: Firstly, in this report, strategies are not analysed on the basis of longitudinal case studies which incorporate extensive real-time observations (�catching reality in flight�, Pettigrew 1991). Longitudinal case studies are an appropriate design to study multidimensional (context, process, context) and complex multilevel strategy making (local, regional, national, and so forth) from the perspectives of multiple actors (Pettigrew 1991, Pettigrew 1995, Van de Ven et al. 1999). Different path dependencies and time horizons are compared. In this report, empirical research considers two levels (strategy of a single organisation and strategic projects / strategy for a part of the catchment) without extensive real-time analysis. To take catchment-wide effects of strategy making into account, theoretical arguments and empirical findings produced elsewhere are considered. Secondly, analysing learning for flood risk management processes requires at least a comparative static design (Rosenthal & t´Hardt 1998). Because learning processes are complex mixtures of different learning forms (e.g, exploration, exploitation, see above) which cannot easily be separated from their societal contexts. As a consequence, extensive analysis of qualitative data is necessary. A survey design (e.g. Brody 2003) can only be seen as a first step into this direction. In this report, the case study analysis cannot refer to empirical information generated at different time points. Empirical research focuses on one situation in time. Hence, we will draw only modest inferences to illustrate the implications of the case studies to learn for pre-flood risk mitigation. For instance, we are investigating who the actors are that are presumably learning from flood disasters as low-probability, high-consequence events without analysing in detail how the learning processes unfold in practice.



3.2 Case Study Results: An Overview To cover a wide spectrum of natural and societal context conditions, three cases were selected:

• Case 1: Strategy Making within the Thames estuary • Case 2: Strategy Making within the Weisseritz river catchment • Case 3: Strategy Making within the Adige river catchment

Case 1: Strategy Making within the Thames Estuary: The Thames Barrier was completed in 1982 after 30 years of debate, design and construction. An Environment Agency �TE2100� project is now under way to explore what needs to be done next, in the context of climate change, economic change in the Estuary, and the fact that the Barrier will be past its design life in another, say, 60 years. The Floodsite case study is designed to benefit from that TE2100 project and evaluate the strategies for flood risk management in the Estuary, and to identify the measures that could be used in this respect and the criteria that we need to use to make the necessary choices. We need first to simplify the issues, to gain understanding, and then add complexity in the option comparison. The following are high level tactics that should be at the extremes of what will be implemented:

• Do Nothing (for instance, "walk away"), • Continue present practice, • Integrated flood defence management, • Integrated flood risk management.

The strategy for the Thames Estuary is contextualised within the metropolitan city region of London with its above EU-average socioeconomic growth conditions. The aim is to use the case study to show the issues involved in managing flood risk in a large urban area where there is pre-existing flood defence infrastructure but a new philosophy that moves thinking towards flood risk management rather than flood defence. Case 2: strategy making within the Weisseritz river catchment: The Weisseritz river is a tributary of the Elbe river within the city region of Dresden. Due to extremely high precipitation in August 2002 a flash flood occurred with severe social, economic, and ecological impacts. The flood event was embedded in the slow-rise flood of the Elbe river. In reaction to this flood event various structural as well as non-structural measures were deployed for recovery and improving the pre-flood strategy for the Weisseritz river. How these measures are related can be studied under conditions of moderately growth at framework level and at district and project level with regard to the aim of reusing brownfields under conditions of difficult economic restructuring and population shrinkage (see EU-project LUDA � Large Urban Distressed Areas � Case Study Dresden). Case 3: Strategy making within the Adige river catchment: The case study �Adige� investigates the characteristics of the current flood risk management framework in an environment suffering from both flash flooding and slow-rise flooding. A specific effort will be placed on the feasibility analysis of using modern hydrometeorological flash flood detection technology and communication technology to increase social resilience at a regional level. To summarise, the case studies foster a rich description of different strategies for reducing vulnerability in flood-prone areas through planning. The cases cover different flood types (flash flood, plain floods, tidal flood), different spatial planning systems, and very different economic and population development conditions. It can be presumed that the internal context conditions will vary significantly between the three cases. It is expected that cross-case analysis reveals similar contents, different process patterns, and very different context conditions (see Table 13). Interpretation will connect these empirical findings with theoretical considerations how to develop effective pre-flood strategies, thereby, making sense of the findings within a larger context of scientific knowledge and policy issues.



3.3 The Thames Estuary Study This study sought to examine the specific opportunities and challenges faced by decision-makers involved in strategic planning in a highly complex physical and social environment, subject to increasing pressures from both climatic and social change. The aim of the case study was to highlight issues involved in managing flood risk in a large urban area where there is pre-existing flood defence infrastructure but a new philosophy that moves towards flood risk management rather than flood defence. The study was comprised of an extensive desktop review of the current IFRM project within the Thames Estuary (UK Environment Agency Thames Estuary 2100 Project) and in-depth interviews with key regional stakeholders with interest and input to the strategy development process. In accordance with the overall approach to Task 13 case-studies, the strategy process for flood risk management in London was analysed according to the content, process and context of the decision-making process.

Context The River Thames (see figure 14) is multi-faceted, functioning on many essential levels: the river and its corridor act, for example, as an artery for communication, and a resource for commerce, industry, commodities, housing, biodiversity, recreation, drainage and water supply. It is a functional tidal system and continuing improvements in water quality together with the intrinsically rich diversity and continuity of habitat has meant that the Thames now supports one of the widest varieties of fish, wildfowl and invertebrates of any Estuary in Europe. The lands adjoining the Thames comprise a number of valuable habitats including: fresh and salt water marsh, grazing marsh, mud-flats, shingle beach and inter-tidal vegetation. The hinterland includes areas of open marshland and extensive Brownfield areas.

Figure 14: The River Thames and London

Along the Thames Estuary there are a number of important environmental sites designated under the EC Habitats and Birds Directives. There are also Sites of Special Scientific Interest within the floodplain area, many of which are water dependent and situated both riverward and landward of the flood defences. As well as the environmental designations, there are also three world heritage sites, Scheduled Ancient Monuments, Areas of Archaeological Importance, Conservation Areas and buildings of national importance within the flood risk area. In addition, there are also a number of hazardous industrial sites (COMAH sites) and waste disposal sites that may need to be protected form flooding to prevent pollution problems. Strategic planning for flood risk management occurs within the context of a wide range of EU directives, national Planning Policy Guidelines (PPGs), Planning Policy Statements (PPSs) and sub-regional polices and plans. The policy context which surrounds planning in London and Thames Estuary region is outlined in Appendix 1. The Thames Estuary floodplain is extensively developed. London produces 17% of the UK's GDP and the City of London is one of the world's major financial centres: it is the banking centre of the world, and Europe�s main business centre. The London foreign exchange market is the largest in the world,



with an average daily turnover of $504 billion, more than the New York and Tokyo exchanges combined and the economy generates $365 billion annually. London is the largest city in the European Union with an estimated population on 1 January 2005 of 7,421,328 and a metropolitan area population of over 1.25 million. Future planned development to the east of London as part of the Government�s �Thames Gateway� regeneration scheme will add 120,000 homes to the region. Major floods, both from the Thames and its tributaries, have long been a part of London life. The construction of the tidal defences along the Thames Estuary can be dated to the 12th century. The �law of the marsh�, established in Essex in 1280, required every man to contribute to the upkeep of defences in proportion to his benefits and land rights. A parliamentary authority was established in 1427 and reinforcement of the Act in the 16th century established the principles of land drainage and tidal defences for the next 300 years (Peter Brett Associates, 2002). A series of increasingly damaging high tides between 1874 and 1877 resulted in the passing of the Metropolis Management (Thames River Prevention of Floods) Amendment Act in 1879, which led to the raising and maintenance of defences to prescribed levels above Ordnance Datum. The Act was further amended to raise these statutory levels in subsequent years when high tides exceeded the defence levels. In January 1928, a surge tide combined with high freshwater flowing the Thames caused flooding in which 14 people drowned in central London. Further rising of the defences followed and these 1930 defence level remains today upstream of the Thames Barrier. These 1930 levels protected London during the catastrophic tidal surge tide of 1953, but over 300 people died on the East coast of England, as much as 24,000 houses were flooded, up to 400 destroyed and over 32,000 people were evacuated from their homes (Johnson et al., 2004). On Canvey Island alone, 59 people died. Research into the causes of tidal surges and a method for controlling surge tides in London ultimately led to passing of the Thames Barrier Act in 1972 and the construction of the Thames Barrier and associated defences (Gilbert and Horner, 1984). The Thames Barrier was operated for the first time in 1983 and between 1983 and May 2003 it was closed 87 times to protect London from flooding. The tidal defence systems comprises the Thames Barrier, seven other major flood barriers owned and operated by the Environment Agency, over 400 moveable defences and 487km of tidal walls and embankments. The current standard of protection provided is generally 1:1000 and the current design standard has an allowance for sea level rise to the year 2030. However beyond this time, if no improvements are undertaken the standards of defence will begin to fall below the design standard with increasingly serious consequences for the Thames Estuary and London. In accordance, the Thames Estuary 2100 (TE2100) Project has been set up by the UK Environment Agency to develop an integrated approach to flood risk management: the final strategic mangement option discussed above. The aim of the project to develop a Flood Risk Management Plan for the Thames Estuary to the year 2100 that will provide sustainable long-term flood defence solutions, maximise the reduction of flood risk and provide best value for money � now and in the future. Because of the complexity of the study and consultation programme required to develop the Plan, a phased approach was developed to guide the progress of strategy development (Figure 15, next page). The case study examined phase 1 and the recently completed phase 2 of the strategy processs: the development of early conceptual options for flood risk management (FRM). It focused on an extensive desk-top study, complimented by a series of in-depth interviews with key stakeholders across the study area. Using the characteristics of an adaptive process described above, as benchmarks on progress towards Integrated Flood Risk Management, the report focuses on evidences and key limitations as experienced by stakeholders in achieving an adaptive strategy planning process.



PHASE MILESTONE

Figure 15: Flood Risk Management Plan Development and Review Process

This report examines phase 1 and the recently completed phase 2 of the strategy processs: the development of early conceptual optoins for flood risk management. It is important to note that the project is on-going and evolving. The project aim is to have a completed flood risk management plan by 2010.

Content A series of strategic options and associated policy considerations exist for flood risk management in the Thames Estuary region, with varying implications for resources within the floodplain (Environment Agency, 2004c). These options can be summarised as:

• Do nothing (i.e. �walk away�): The risk of flooding will increase rapidly over the next 30-40 years with the flood risk area becoming more frequently inundated.

• Continue present practice (i.e. �do minimal�): This results in a gradual reduction in standard of flood protection, with potential for damages as above but over a longer time frame. Reactive management carried out in an uncoordinated manner will also give increased risk of flooding to the protected areas.

• Integrated Flood Defence Management: An integrated approach to managing the defences will be maintained to a high standard of protection. This approach is reactive to changes in the estuary and only addresses the defences, which will need to be more substantial in time to address the environmental/social impacts. Two sub-options exist within this strategy approach: (1) Enhancement of flood risk management standards within the existing option structure: reconfigure the barrier and enhance the downstream embankments to provide a higher standard of protection within the existing flood defence philosophy and the type of assets that this has created and (2) Enhancement of flood risk management standards within a new option structure: provide and maintain the same standard of flood protection in some other way ("throttle" in the Estuary; downstream barrier to complement the existing one etc)

• Integrated Flood Risk Management: Proactive approach based on a good understanding of the process drivers would not only sustain an appropriate standard of protection but also ensure that all options for managing flood risk i.e. engineering and spatial planning are maximized. This approach will include integrated but robust development control, which will reduce risk but also have the potential to lower costs. All stakeholders are actively involved in the decision-making process which raises awareness and gives workable solutions.

Phase 1 � Scoping and Strategy Development Strategy Envelope

Phase 2 � 4 Developing and implementing a robust Plan, which is acceptable to government and stakeholders Phase 2 Technical, environmental and social studies and the formulation of Early Conceptual Options Phase 3 Options Appraisal and Recommendations � the first refinement of the High Level Options Phase 4 Second iteration of High Level Options � finalising and implementing the Flood Risk Management Plan

Early Conceptual Options

High Level Option 1, High Level Option 2 etc

Estuary Flood Risk Management Plan

STA

KEH

OLD

ER



Process Four primary characteristics were considered within this case-study as indicative of a strategic planning process for an adaptive response to managing flood risk:

• Nurturing diversity in decision-making: governance: practices that build resilience and a social network with trust and respect in the decision-making process.

• Combining the range of existing knowledge systems into the decision-making process (e.g. engineering, physical modelling, social science): allows the integration of wide range of system behaviours and functionalities into the strategy development process.

• Embracing uncertainty and change: the management process depends on institutional learning, incorporating previous crises and management strategies resemble risk spreading and insurance building within society.

• Creating opportunity for self-organisation: adaptive management and planning continuously tests, learns and modifies its activities and understanding for coping with change and uncertainty.

Governance: The flood risk management process model (Figure 15) clearly recognises the role of stakeholder participation. The reality is that, to date, this aspect has been very limited in its scope. The importance of clear communication and consistent goals in the strategy process across the decision-makers emerged as a key issue from the interviews, with questions raised as to the current effectiveness of this process. However, there was a mixed response within the interview group to the lack of public participation: a response which reflects a challenge to the development of an effective governing framework for the strategy process. From the opinions expressed by the respondents, it would seem that there are key organisations with input to FRM that do not consider the reality of public participation a necessity or, at the very least, one which should relate across the range of content within a strategy framework. This constrasts with the view of members of the public, as articulated for example through the Thames Estuary Partnership, which is one of interest and desire to be involved in the wider strategic planning process. Key steps have been taken by the current FRM project in defining the needs and opportunities towards implementing policy, so that the decision-making process has moved towards �good� governance. However, there are important questions which remain to be addressed. An example of such a question is the relationship of communities and individuals with interest and concern regarding FRM, to the organisations with the power and the resources to determine the nature of the strategy process. Ensuring all stakeholders are engaged in a decision-making process is critical to understanding the range of conflicts and choices for a management process: as well as promoting the commitment of individuals or groups of stakeholders to any management plans which emerge from the process of decision-making. Improving the current FRM strategy process in this direction would promote our understanding of the nature of vulnerability with the area and would equip decision-makers to develop better options for these needs. Knowledge integration: Interviewees linked knowledge integration with stakeholder engagement, so it could be presumed that views on governance also relate to combining knowledge systems in the decision-making process. An interesting point emerging from the interviews is that the respondents gave a distinct internal-organisational context to knowledge integration: influencing or commenting on regional-scale integration of the range of knowledge on flood risk was not considered by individual stakeholders to be within their organisation�s remit. It would seem there is an assumption among stakeholders that different aspects of �the science� are being effectively integrated by the Environment Agency as scientific-leaders of the flood risk management strategy. The literature evidence from the first two phrases of the project suggests that this has not yet happening to any significant level.



Increasing our knowledge of the integrated system i.e. relationships and linkages between the spatial and temporal responses of the physical and socio-economic environment is fundamental to understanding the true impact of external forcing on the system and so to effective management. This research suggests that more effective moves towards integration must be made within the current strategy process for the London region. However, the interviews also reflect a dissociation of stakeholders with the existing knowledge basis of the project and highlight a distinction between �science� and �governance� in the strategy development process. Linking integrated scientific knowledge with an integrating approach to stakeholder engagement is critical to understanding the differences in the functional scales at which the science of the system and at which the decision-making power operates, and ensuring effective communication and decision-making links across these scales. This is important to creating an adaptive strategy process. Embracing uncertainty and change: The current FRM plan development process has moved towards facilitating uncertainty, including mechanisms in the decision-making process through which strategic alternatives for management can be tested against a range of potential physical and socio-economic futures for London and the Thames Estuary region. However, whilst this positive reflection can be made of Environment Agency leadership of the management planning process, there is concern among stakeholders that primary decision makers setting the context of flood risk management are not giving sufficient regard to the range of possible futures for the region. Such discussions on uncertainty were related by interviewees to the UK Government development proposal �The Thames Gateway�, with the concern that the Government approach to proceeding with a very fixed socio-economic future in mind for the region, is constraining the effectiveness of the strategic planning process. The research indicates a distinction between local-scale and broader-scaled organisations on ability, or in the perceived need, to incorporate uncertainty into decision-making. At the local level incorporating uncertainty into decision-making, would seem to be considered by some stakeholders as irrelevant or impossible. At the regional scale, and particularly for organisations with objectives which relate to a relatively longer-term, uncertainty is recognised an issue that needs to be understood and managed more effectively. nother point emerging from the interviews is that greater clarity of both the impacts of uncertainty (particularly at the general public level) and the limitations on current decision-making (particularly at the policy level) is required to prevent uncertainty being used as an issue to block �unfavourable� alternatives for flood risk management. One of the most difficult tasks in formulating an adaptive strategic planning process may be the challenge of developing relationships, both within and between, the social, physical and institutional environments that allow a region to manage uncertain futures and adapt in an integrated manner to drivers of change. Opportunities for self-organisation: As with the theme of uncertainty, there is some difference in the Environment Agency�s anticipated approach to self-organisation through a re-iterative development and review process, and the reality among other stakeholders of modifying behavioural responses and preferences on management options given change in the drivers of flood risk within the system. The research suggests that there is generally an absence of a learning process among stakeholders at the local scale and that the ability of regional organisations to adopt an iterative approach to strategy development is very much constrained by limited resources. Perceptions of the need to be involved in a longer-term strategy process also varied across the stakeholders interviewed, reflecting to some extent the scale of organisation and its role or relationship to FRM. Given that the current strategy process is managed by the Environment Agency, it may be argued that this leadership is sufficient to ensure that management options in an emerging plan have reflected a process of testing and learning. However, this lack of self-organisation ability across the wide stakeholder basis of the region would have implications beyond the life of the TE2100. This would relate in particular to the ability to develop a long-term adaptive strategy process, in which there is a continuous cycle of learning and testing. A series of other interesting and important themes emerged during the course of the interviews conducted within this research. The lack of dedicated investment for implementing FRM options



which may be identified through the strategy process was considered to be one of the greatest challenges to the future success of FRM. The need for more focus on the building industry to encourage more adaptable housing was highlighted as a means for achieving relative simple but effective reduction in the vulnerability within the Thames Estuary floodplain. The challenge represented by the lack of compensation mechanisms for home owners in high risk flood areas was also discussed. On a positive note, a general theme emerged of �stepping up�, or more stakeholders � including key regional organisations and local communities � seeking to take responsibility in a decision-making process. The London case study is an example of a management process in which a series of important steps have been taken towards the integrated management of a highly complex physical and social environment. However, it is also clear that the current strategy process still has significant limitations if assessed from the perspective of adaptive management. A primary conclusion emerging from this work is that translating the vision of an adaptive strategy process into the reality of flood risk management remains a challenging task. It requires a thorough understanding of the physical, economic and social processes and responses � and the flood defence assets � which characterise the system: and an institutional and organisational landscape which facilitates self-organisation, learning and careful planning.



3.4 The Weisseritz River Study It is evident that the capacity of local government to develop a strategy for pre-flood risk management depends both on the type of flood to be managed and the historic pattern of strategy-making in a multilevel governance context. The case study �Weisseritz river / Dresden� focuses on the challenges involved in managing the risk of flash floods in a small left tributary of the Elbe river in Dresden, Germany. The study refers to the historic pattern of strategy-making to explain why the flash flood disaster in August 2002 triggered only a limited change of the local management strategy from flood protection to a more risk-based approach. The study shows that moving towards pre-flood risk management requires specific combinative capabilities of spatial planners and water managers. Within the City of Dresden, these capabilities were in short supply despite the long-standing local tradition of combining physical measures and spatial planning for managing the flood risk of the Elbe river. Hence, the case study focuses on a local response to a flood disaster and the problem of area-based policy integration. The study involved officials from local authorities (spatial planning, flood risk management, environmental protection, nature conservation, project development) and state authorities (predominantly water management and officials responsible for engineering work for pre-flood risk management). The researcher conducted interviews based on a thorough literature review (planning documents, newspapers, reports, scientific literature) and extensive participant observation (e.g., to capture how citizens and business organizations perceived and interpreted the flash flood disaster in August 2002 over a period of 3 years). Furthermore, case study materials served as study material in a master course developed at the Technical University of Dresden (FLOODmaster).

Context After German unification in 1990, the City of Dresden was suddenly confronted with a highly competitive global market. Industries that once employed every citizen under the socialist regime found themselves dramatically downsizing their workforces in the free market context. In addition, the shift from manufacturing to a service economy only aggravated fast-rising unemployment. As a result, people � especially younger generations � moved to western Germany. In the 1990�s the city lost approx. 10% of its population, the housing vacancy rate jumped up to 14 %. The city responded to urban shrinkage by working diligently to attract new high-tech industries, earning it the moniker �Silicon Saxony.� Since 1999 Dresden has reversed its population decline and even boasts some small growth due to increases in immigration and birth rate. Currently, the performance of Dresden in demographic and economic terms is superior to all other cities in Eastern Germany with the exception of Berlin. However, some areas in Dresden are characterized by enduring low economic and social development. In Germany, these areas are called �benachteiligte Stadtgebiete / distressed areas�. The Weisseritz river forms part of one of these distressed areas in Dresden, the Weisseritz area. During the middle ages, the Weißeritz area was located outside the fortress walls of Dresden. The area was shaped by the Weißeritz river. In the 19th century, the city of Dresden expanded out to the area and landscape features diminished. The river bed of the Weisseritz was relocated outside the city centre in the year 1893. Adjacent residential areas for workers were characterised by urban structures of high density, but relatively attractive open spaces for leisure. Today the Weißeritz area has an impact on the urban region as a whole due to its radial location within the city and its large size. It includes an inner city part, the former industrial core, residential areas from the 19th/20th century, attractive landscape close to the city border in the south of Dresden as well as massive road and rail infrastructure. Road and rail infrastructure and the river Weißeritz are dominant elements within the area constituting mainly barriers for daily life and business. In sum, the Weisseritz river is part of a heterogeneous area with economic and social development below average. Figure 16 gives an overview over real (not planned) land-use of the Weisseritz area and illustrates the heterogeneous structure of the area. One could argue that environmental, economic, and social problems accumulate in the Weisseritz area. In summary it can be stated that the area suffers due its heterogeneous structure from an accumulation of environmental, economic and social problems in addition to those of urban structure.



Figure 16: Real Land-use within the Weisseritz River Area (Source: Müller et al. 2006)

In August 2002, people in the Weißeritz area experienced a hitherto unknown flood disaster in approximately one from both the rivers Weißeritz and Elbe. A few days later, The Elbe rose to the dramatic level of 9.21 meters. Beside the inner city part the whole Weißeritz area, especially the Friedrichstadt was hardly affected, when the Weißeritz used its old riverbed (replaced in the 19th century) as a second discharge channel again. Many thousand people along both rivers had to be evacuated and situated in emergency accommodations. The power of water destroyed streets, bridges, buildings and public places in a few days with a total loss of ca. 250 million Euro in Dresden. The catastrophe had enormous consequences for the redevelopment process of the Weißeritz area in the short and long term. Immediately after the flood several projects were stopped as the priority was set to clear the damage and activate the infrastructure again. In a long term view, existing plans and concepts had to be questioned to take flood protection issues into account. As one interviewee said, politicians �discovered� their responsibility for the neglected distressed areas in Dresden like the Weisseritz area.



Figure 17: Inundated Areas by the Weisseritz River in August 2002 (blue), Statutory Flood Plain in

Accordance with the German �Wasserhaushaltsgesetz (WHG)� Delineating the 100-year flood (red).



Content Before the fact, nobody wants to learn from a disaster. Afterwards, we are urged to do so. Learners prefer to pursue a strategy of extracting knowledge from non-rare events with small losses. Unfortunately, this strategy cannot be applied under all circumstances. This is the case in Dresden. In August 2002, people in Dresden, Eastern Germany, experienced two rare flood events. Afterwards, citizens and state authorities urged local government to learn deeply from this severe interruption of everyday life. Based on a case study, the paper reports what and how local government learned from this disaster. It highlights the importance of learning how to deal with uncertainty of flood risk in long-term planning. The 2002-disaster triggered two different response processes from which one is converging and consensus-oriented and the other diverging because of multiple cognitions and conflicts. The case study shows a multi-level response of local government to a flood disaster. It elucidates that learning from rare events through managing diversity means two things: (1) recognizing and exploiting different processes and associated contents and (2) managing diversity within one process. In Dresden, local government has its difficulties to manage this complex response due partly to the cultural bias for consensus-oriented policy solutions. In what follows, case study findings illustrate this interpretation. In August 2002, Dresden was threatened by two rare flood events within only a few days. Local officials classify the flash flood event of the river Weisseritz in 2002 as a 1-in-500 year flood. It was characterised by an extremely short warning time and very high inundation dynamics. What is striking is that many local actors were completely surprised by the event. This can be explained partly through history. Before 2002, the last major flood happened in 1958. Furthermore, until August 2002 most local actors focused on the river Elbe with its slow-rising flood waves and possible lead times up to several days. At the end of 2005, there is a consensus within local government that flood protection assets in Dresden should be capable of controlling peak discharges of the river Weisseritz up to the levels corresponding with the August 2002 event (�design flood�). To realize this safety standard, the Free State of Saxony and local government in Dresden agreed on a specific division of responsibility: • Under certain circumstances the Free State is willing to provide flood protection assets for a 1-

in-200 year flood. This is the case with the river Weisseritz in Dresden: • To realise a safety level for a 1-in-500 year flood, local government is willing to finance the

additional planning and construction costs of physical measures. This implies a significant change of the local flood protection strategy. Before the event, hydraulic structures were designed to protect against a 1-in-100 year flood only.

Reaching consensus within local government was more difficult when it comes to discussion about flood events with a peak discharge that exceeds the one in August 2002 and / or with unexpected inundation dynamics. This is the problem of reaching a consensus about how to manage residual risk. Two groups are observable: • Officials responsible for environmental protection argue for an open discussion about residual

risk. They argue for a strategy that takes the full spectrum of possible flood events and their consequences into account. This group is supported by some spatial planners. These planners suggest that physical measures to protect against a 1-in-500 year flood could be combined with instruments to realise a green space corridor along the course of the river.

• At present, it is difficult to justify a green space corridor for coping with residual risk. A majority of politicians and officials wants to restrict discussion to realising the safety standard of a 1-in-500 year flood through building new hydraulic structures. They want to avoid restrictive implications for urban development in flood-prone areas, especially nearby the historical centre.



To sum up, consistent with the notion that shocks trigger activities to develop new approaches and management solutions, responding in a passive way to the flash flood disaster of the Weisseritz river in August 2002 was clearly out of question. What is less clear is to what extent the disaster triggered a significant change from flood protection to pre-flood risk management. At an abstract level of strategy-making and especially shortly after the event, leading politicians (e.g., the mayor) and officials (e.g., from the Office for Environmental Protection) claimed that absolute safety against flash floods of the Weisseritz river is impossible and managing residual risk inevitable. At the more concrete level of making specific decisions for managing flood risk, local government focused on physical measures for flood protection: In December 2005, the City Council decided to provide financial resources for changing flood defence structures designed for a 1-in-100 year to a 1-500-year flood safety level. The Free State of Saxony agreed to co-finance these investments to some extent. Even if these planned investments are realised, there will remain a residual risk. Yet, despite the abstract call for pre-flood risk management put forward after the flood disaster, leading local politicians and officials now want to restrict discussion and decision-making to decreasing the probability of flooding through physical measures. At present, in local government there seems to be a majority of politicians and officials with a preference for enhancing safety through physical measures (Alternative III, see Table 15), whereas only a minority argues for pre-flood risk management at the Weisseritz river. This shows that a more risk-based strategy (Elbe river) and a protection-oriented approach (Weisseritz river) can co-exist in one municipality. Approach

Alternative

Strategic Content

Alternative VI

Reduce probability of flooding as well as vulnerability and increase preparedness through physical measures, informal spatial planning and communicative instruments

Pre-Flood Risk Management

Alternative V

Reduce probability of flooding as well as vulnerability through physical measures and formal spatial planning

Alternative IV

Reduce probability of flooding within a new options structure � physical measures

Flood Protection (= Flood Defence)

Alternative III

Reduce probability of flooding within the existing options structure � physical measures

Alternative II

Do minimal � maintain existing assets

Passive approach

Alternative I

Do nothing � walk away

Table 15: High-Level Strategic Alternatives of Flood Protection and Flood Risk Management

The notion is well-known that learning from flood disasters as low-probability, high-consequence events is fostered if the immediate time after their occurrence is used as �window of opportunity� to change from a reactive approach to a pro-active, strategic approach. But, the available time after an event is limited. Memories of the event fade and after a while flood risk issues are in the background of local policy making. Therefore, using a �window of opportunity� already requires actors with political and strategic capabilities as well as responsibility to establish a strategic approach to spatial planning for risk mitigation. Obviously, using planning for pre-flood risk reduction is a mutually reinforcing feedback circle. The Weisseritz river case study shows that local authorities that lack political support and strategic capabilities have difficulties using severe floods as window of



opportunity for deploying spatial planning. Subsequently, it becomes even more difficult to develop a strategic approach to risk management.

Process Strategic planning as disciplined cognitive effort can be deployed at different spatial levels. This report focuses on regional and local level within the physical context of the small- to medium-sized catchment of the Weisseritz river and its specific institutional context. Two options out of the wide range of possible options to deploy strategic planning for long-term FRM are considered (see Table 16).

Strategic planning at�

Regional level

Local level

Focus and overall thrust

- External context and content - Formulating new ideas

- Internal context and process - Adopting new and good ideas

What? - Knowledge problem - Strategic issues

- Complexity, uncertainty - Testing existing protection concepts - Wide range of strategic alternatives - Low-probability flood events

- Ambiguity, uncertainty - Priorities of urban development - Local planning culture

How? - Decision mode - Change mode

- Organizing a �study group� - Scenario analysis: Multiple futures - Episodic change

- Pluralistic leadership - Continuous change

Who? - Actors

- Officials from different levels and with different institutional positions - External experts (content / process)

- Local officials - Local politicians - External experts (content / process)

Where? - Social setting(s)

- New forum

- Existing forums and arenas - New forums and arenas

Main challenges?

- Legitimizing a planning episode - Mobilizing officials - Embedding the results - Building a new forum

- Initiating strategic planning - Maintaining the process - Connecting forums and arenas - Formulating implementation plans

Table 16: Deploying Strategic Planning at Different Spatial Levels of FRM � Strategy Making in the

Weisseritz River Catchment

Both options claim to supplement rather than to supplant existing formal concepts and measures for long-term FRM at Länder and state level (see the overview in Samuels et al. 2006). Furthermore, they complement formal spatial planning. They are based heavily on the motivation of (some) local officials and politicians to search new solutions for long-term FRM. Thus, strategic planning in this report presumes some experience of local politicians and officials with flood events as disasters (e.g., the disastrous flood events of August 2002 in Dresden). Furthermore, it presumes sufficient resources and capabilities to comply with state rules and to develop own strategic plans for long-term FRM. It is more likely to find these resources and capabilities in large cities than in small ones. Focus and overall thrust: External context (e.g., climate change, economic development in flood-prone areas) and content are in the foreground of this regional planning option. Its overall thrust is to develop new ideas for long-term FRM. The term �new� has a subjective meaning here. �New� means �new� for the people with responsibility for FRM in a specific case. Hence, deploying strategic



planning requires careful analysis of existing flood risk management concepts as well as important structural measures and policy instruments to determine possible innovations. Content (�What?�): To legitimize investments (mainly measured in time), strategic planning should clearly follow a focused theme-driven approach. Possible contents can vary widely (Galer & van der Heijden 2001, p. 862), for instance, • Testing existing long-term flood protection concepts based on scenario-analysis to consider a

wide range of future external context conditions and to assess the robustness of existing concepts,

• Defining and assessing strategic alternatives for long-term FRM, thereby considering strategic alternatives that range from pure flood protection alternatives to alternatives that emphasize spatial planning as policy instrument to reduce vulnerability in flood-prone areas (for more details with a slightly different terminology see Klijn et al., this issue),

• Discussing existing capabilities of long-term FRM to deal with disruptive external events (see Lonsdale et al. 2005 for a good example with regard to rapid sea-level rise for the Thames Estuary). Officials and external experts are attempting to �expect the unexpected� (Weick & Sutcliffe 2001).

To legitimize regional strategic planning for long-term FRM, it should have one thematic focus from the beginning. The recommendation is that a �study group� is organized that can provide a list of thematic options and a proposal which issue is most important for context-specific innovation (�strategic issue�). The alternative approach would be broad informal co-operation at catchment level to deal with multiple themes that can encompass predefined as well as emerging issues. In this case, a minimum of organizational structure is needed to deal with a continuous and complex co-operation process (see the example of �Weißeritz-Regio� within the region of Dresden, Hutter & Schanze 2004). Process (�How?): Following a theme-driven approach to generate new ideas corresponds with an analysis-centred process that tries to incorporate some elements of creative thinking. Strategy research as well as research on long-term FRM shows that scenario analysis can foster such a process. Thereby, the specific method of defining strategic alternatives and scenarios is less important than following an approach that motivates officials and external experts to construct explicit knowledge about possible futures. Thereby, scenario-based planning takes on the form of a project. It has some clear boundaries in terms of participants, duration, work programme, and required resources. Hence, regional strategic planning is more a planning episode than a continuous process of strategy making (van der Heijden et al. 2002, pp. 229-275). Context (�Who? Where?�): Innovation research shows (e.g., Van de Ven et al. 1999) that people engaged in the invention and implementation of new ideas need to balance the need for homogeneity and heterogeneity, especially in the exploration period. In contrast, implementation can be a more narrow process of collective action �to make it happen�. This report argues that regional strategic planning for long-term FRM should try to combine homogeneity in cognitive orientation with heterogeneity in institutional affiliation (e.g., water management, spatial planning at local and regional level, economic development). The former ensures that discussion can focus on difficult content questions that require a minimum in technical understanding. The latter ensures that results of the scenario planning episode can be communicated within different institutional contexts. Hence, theme-driven selection of participants (Klinke & Renn 2002) should focus on officials (broadly covering agency staff and officials in a narrow sense) and external experts (e.g., consultants) for content and process (e.g., moderation of meetings). Main challenges: Discipline is required to combine a limited range of actors with a theme-driven analysis of possible flood events, their consequences and strategic alternatives in such a way that results of relevance for ongoing decision processes at local level are gained. In this regional case, strategic planning aims at creating a new forum for long-term FRM that inspires existing arenas and forums at local level. �Forums are the basic settings we humans use to create shared meaning through



dialogue and discussion.� (Bryson 2004, S. 310, see also Bryson & Crosby 1993). This new forum is detached from existing social settings not by mistake, but to give significance to strategic issues that are forgotten or in the background of current FRM practices at local level. The danger of a new regional forum is that it is too detached from existing settings to have a significant effect on local strategy making. Strategic planning at local level: Unfortunately, new ideas are often seen as bad ideas when their implications are detailed in existing local forums and arenas for political decision making. In line with this, the focus on flood protection is due partly to limited capabilities of planners to manage the strategy process and partly to limitations of strategic alternatives to the chosen alternative of physical measures. Already before the flood event in August 2002, politicians and spatial planners could not agree on priorities and a more process-oriented approach for managing spatial development within the areas adjacent to the Weisseritz river. At this time, flood risk was not an issue for spatial planning. After the event, planners were not able to integrate the different planning processes for the Weisseritz area (planning for flood protection, an open space corridor, for nature conservation, built development, economic development, and so forth). The linear (sequential and non-participatory) process pattern of these activities did not match with societal context (low investment pressure, medium political support for spatial planning, high conflict potential within local government as well as between public and private actors). Therefore, local strategic planning for long-term FRM should opt for focusing on internal context and process, not to guarantee innovation success, but to increase its odds. Pluralistic leadership is of special importance here (Van de Ven et al. 1999, pp. 112-124). It balances the contributions of idea champions of long-term FRM and the arguments of sceptics through strong institutional leadership. Thereby, it is difficult, but crucial to include sceptics at an early stage of discussion (see Hooijer et al. 2004).

Formation ofInnovation Unit

DevelopmentalPeriod

Implementationof Innovation

Critic

InstitutionalLeader

Mentor/Sponsor

Number ofEvents

Involved

Few

ManyEntrepreneur

Figure 18: Relatively High Involvement of Critics in the Formation Period of Innovations Increases

the Odds of Innovation Success (Source: Van de Ven et al. 1999, p. 116).



3.5 The Adige River Study The case study �Adige� investigates the characteristics of the current flood risk management framework in an environment suffering from both flash flooding and slow-rise flooding.

Context The study area is the Adige river basin closed at Trento (Figure 19), with a contributing area of 9.763 km2. The Adige river is the second longest river of Italy, 409 km long, rising in the Tyrolean Alps, N Italy, and with a total area of 11.954 km2. It flows south, past Bolzano, Trent, and Verona, to the Po valley where it turns east to flow into the Adriatic Sea, south of Venice. The research is focused on the mountainous part of the basin, which includes two distinct administrative units: Provincia Autonoma di Bolzano (7100 km2) and Provincia Autonoma di Trento (2663 km2). Altitude ranges from 180 m a.s.l. up to 4000 m a.s.l.. The average altitude is 1735 m a.s.l.; almost 28% of the catchment area lies above 2000 m a.s.l.. Glacial areas amount to 150 km2. The catchment is located south of the inner alpine province (Fliri, 1975): it ranges between a dry climate (with a mean annual precipitation around 600 mm/yr, due to the dual sheltering effect of the range to both the north and the south) and a wet climate (around 2500 mm/yr) along the venetian borders. The mean areal precipitation is 873 mm/yr. The southern range experiences showery precipitation with thunderstorm and hail, particularly in summer and autumn (Bacchi et al., 1996). Precipitation quantiles exhibit large spatial variations: 100-year return time, 24-hour duration precipitation ranges between 100 and 200 mm, whereas the corresponding 1-hour duration precipitation ranges between 30 and 60 mm (Borga et al., 2005). Two flood types could be distinguished in the Adige river basin: flash floods affecting smaller catchment areas in the Adige tributaries and slowly rising riverine floods along the Adige River. Floods of short duration with a relatively high peak discharge (Untied Nations Educational, Scientific, and Cultural Organization, International Glossary of Hydrology, http://www.cig.ensmp.fr/hubert/glu/aglo.htm, accessed 1 August 2006), in this case caused by very intense rainfall, were defined as ��flash floods�� in this study. Cases of riverine floods are those occurred over the whole basin in 1882, 1965, 1966, 1998 and 2002. The city of Trento, as well as several towns in the basin, was flooded and heavily hitted during the 1966 event. Flash flood events are generally combined with important geomorphic phenomena (debris flows and landslides). For example, a flash flood event occurred near Fortezza on 15 August 1998 and triggered a debris flow which killed 5 people on the motorway between Bolzano and Vipiteno. Other important flash flood events hitted Termeno (Rio Inferno, June 1987), Cortaccia (26 June 2001) and Vandoies (17 July 2002). These floods involve some tens km2 and may trigger important geomorphic phenomena.

Figure 19: The Adige River Basin closed at Trento



Around Trento, the landscape is characterized by Alpine geomorphology and strong relief energy. Elevation ranges between 200m and 2000m a.s.l.. During the ice ages the valley has been completely glaciated which is evident in the overall morphology and deposits that are found. The Adige river is the key-feature of the valley and it has created a large alluvial plain. Before the beginnings of the 19th century, the river changed its curse very frequently, making the flood plain unfit for any permanent settlement. Then in the nineteenth century, the river was straightened and forced into a fixed bed. Flood levees were built or improved based on the largest flood on record (at that time), which occurred in 1882. The main reason for these interventions was the reclamation of the entire valley bed, which presently hosts various types of productive activities. Old riverbeds and meander-curves are still visible in the terrain, sometimes accentuated by land-use patterns. The main valley of the Adige river basin (running in the north to south direction) hosts truly important infrastructural ways. It is one of the main routes linking the Mediterranean to North Europe. A highway and a railroad lead to the Brenner pass and connect thus Verona (Italy) with Munich (Germany). In many cases, where the valley is narrower, the floodplain is almost entirely occupied by normal routes, highway and railway. Two tributaries join the Adige just upstream Trento, the Noce and the Avisio. These rivers have created large alluvial fans. The Alluvial fan of the Noce � also know as the Piana Rotaliana � has its apex near the snout of the gorge of La Rocchetta at an altitude of 238 metres above sea level (a.s.l.) and an elevation of 205 m. (a.s.l.) at its former confluence with the Adige, giving it an average gradient of 5 m/km. The second tributary, the Avisio River, enters the Adige valley near Lavis. The elevation of its fan at that point is 237 m. (a.s.l.) 3 kilometres downstream, at the confluence the altitude is 195 m. (a.s.l.), giving the Avisio an average gradient of 14 m/km. The Adige at Trento has an average discharge of 221 m3/s; during floods the peak discharge can rise to over 1000 m3/s. In 1966 a maximum discharge of around 2300 m3/s was measured at Trento. The basin of the Noce has a surface area of 1390 km2 and the average discharge at Mezzolombardo is 30 to 40 m3/s. The surface area of the Avisio is 930 km2 and the average discharge is around 10 m3/s. The Adige river and its tributaries are an important source for hydropower. There are 27 reservoirs of varying size spread over the basin, with storage capacity of almost 540 106 m3. This is about 8 % of the average annual runoff. The hydroelectric exploitation of the water course brings about significant further modifications of the original state. Such alterations are due to both the reduction in the water discharge in the diverted reaches and to the type of management of the hydroelectric plants. The production of hydropower energy �peak time� results in the intermittent release of water by the plants. The effects are daily variations of discharge in a ratio of 1:10 to 1:40. Two areas are considered to exemplify the different facets of flood risk management in the region: (1) The city of Trento, on the Adige, which exemplifies the problem related to plain flood risk management in a moderately large urban concentration (110.000 inhabitants) on the floodplain. (2) The village of Vermiglio (Cortina), in the upper Noce valley, which exemplifies the problems of flash flood management in a small village in a mountainous area. Trento is a moderately large urban concentration (110.000 inhabitants) placed on the floodplain of the Adige river. It is characterised by heterogeneous socioeconomic structure and land owner structure, with fastly increasing population at district level and high investment pressure. Generally speaking, the social perception of flood risk is rather low in comparison with other sites in the basin. The flat topography of the Adige River Valley results in a very large flooded area when the river leaves the main channel, as occurred during the 1966 major flood. The relatively low slope of the river channel and the consequent slow water velocities facilitate advanced warning of flooding conditions (at least in comparison with the warning lead time expected in the river tributaries). However, local areas surrounding Trento may also be hitted by flash flood or by debris flows, as witnessed by the event occurred during the 2000 flood on Romagnano.



Vermiglio is a village (1,856 inhabitants, Census 2001) in the upper part of the Noce Torrent valley. The municipality surface area is 103.7 sq km and Vermiglio is located at 1,261 m a.s.l. According to the last 40 years Census data, it is noticeable that after a period of heavy reduction in the number of people living in Vermiglio, in the 1980s the number of inhabitants began to increase. Fieldwork was conducted in the hamlet of Cortina (367 inhabitants), because it is the most dangerous part of the village, which takes its name from the stream Rio Cortina. In year 2000 four different events between 14th and 17th of November caused the deposition of 1,500 m3 of sediment in different locations inside the village of Cortina. Check dams previously built along the channel were not capable to store the sediment load carried out by the flood. A new check dam was realized after this event. On November, 14, 2002 a new event occurred, which destroyed the new check dam and all other works along the channel. The sequence of the two events generated a high awareness of flood hazard and risk in the area.

Content The evolution of the flood risk management policy is described in this Section in the broader context of the river basin management policy, both in Italy and in the two Provinces. A comprehensive reform of the whole water management system has been occurring in the last decade of 20th century. Among its most important elements, there is the Water Resources and Soil Conservation Act (L. 183/1989), aimed at the creation of a system of planning in the water domain inspired to the concept of Integrated River Basin Management (IRBM). According to the law, each watershed should have been provided a �River Basin Master Plan (RBP)� intended as a superior-ranking system of guidelines for urban development, land use, water resources use and water quality. In this period a more general innovation concerns the completion of the process of decentralization starting in the 70s. Constitutional reforms passed in the last 90s empower fully Regions in a wide number of matters, including water policy, with the definitive retirement of Central State; the latter will only maintain general supervisory powers and frame legislation. L. 183/89 (Water Resources and Soil Conservation Act) is designed to secure coordination of the different sectorial policies for the water cycle, soil conservation, water pollution abatement and protection operations, use and management of water resources for the purposes of rational economic and social development, and protection of the associated environmental aspects. To achieve these objectives, the law assigns duties to all the competent authorities dealing with land use and water management, namely central government, the regional, provincial and municipal administrations, the mountain community administration, the reclamation and irrigation consortia, including the mountain watershed consortia. The real institutional innovation is represented by the formation of the watershed authorities, which have as their geographical reference framework the river basin or watershed, a division that supersedes the earlier regional and sub-regional administration boundaries. The authorities can be separated into three different levels of representation: watershed of national, interregional and regional concern. The operative instrument for attaining the objectives identified by the law is the watershed plan, which serves as an area plan for the sector. It represents the means of collecting information, and defining the legislative and technical-operational aspects to be used in the preparation of the programmes and regulations for the conservation, protection and improvement of the soil and for correct water use, according to the physical and environmental features of the geographical area concerned. Under the new laws on soil conservation Italy has been divided into:

• 6 watersheds of national level; • 18 watersheds of interregional level; • Watersheds of regional level; • Experimental watershed of the River Serchio.



Table 17 describes the current Water Resources Management Institution Framework.

Table 17: Water Resources Management Institutional Framework in Italy This picture exhibits some degrees of variation across the Country. In particular, Regions have some autonomy in deciding what powers and competences retain to themselves, and what others delegate to local authorities. The elaboration of the RBP in the frame of the law 183/89 has resulted in a very complex process. Currently, no RBP has been completed and approved. In 1993 the law n°493 introduced the concept of River Basin �Stralcio Plan� (RBSP). This allows to limit the area of the RBP development only to critical sectors of the river basin. The objective of the law is to allow the development of the RBP at least for those areas where the flood and landslide risk is considered to be higher than elsewhere. In 1998, following the catastrophic flash flood occurred nearby Naples (147 killed people in Sarno and Quindici), a significant improvement in the flood risk management legislation was introduced by the law n°267/98 which includes three main elements: • A first order identification of the high-risk areas, classified in four classes (from R1 �

corresponding to low risk to R4 � corresponding to high), in a River Basin Extraordinary Plans (RBEP), elaborated by the RBAs. Risk was identified has a combination of hazard and vulnerability.

• A program of urgent measures to mitigate the flood and landslide risk in R4 areas. Structural works where funded in the R4 areas. In order to verify the functionality and the environmental quality of these measures, a specific monitoring was adopted by ANPA (Italian Agency for the Environmental Protection).

• The development of a wide range flood monitoring, forecasting and warning system. Although flood and landslides continue to hit Italy, there is no doubt that the law n° 267/98 marked the shift from the traditional safety mentality or promise of protection to a risk culture. Such a risk culture is based on a comprehensive analysis of the flood risk and an appraisal of potential risk reducing measures. A key element in this frame is an open dialogue about risk and risk-reducing options, involving all stakeholders. An important element in decreasing the vulnerability of local communities



is the development of flood forecasts, warning and response systems. For example, during the huge flood occurred in Piemonte and Valle d�Aosta in October 2000, the use of flood forecasts and warning as a mean to govern the operations of the Civil Protection units was largely appreciated. Emergencies after a catastrophic flood: the role of Civil Protection Civil defence in Italy is organised as a coordinated resources system where national, regional, provincial and local authorities work together with local and public institutions, the scientific community, private institutions and organisations, voluntary organisations, and professional associations. Each of these authorities, community institutions and organisations has developed its own part of the National Civil Defence Service and makes its own contribution to the achievement of the aims of civil defence. This involves a series of practical measures in the various sectors involved in civil defence and, as a consequence, it makes necessary to place responsibility for managing and coordinating to a central unit. In the event of exceptional cases of this type, the Council of Ministers can make use of a special ordinance, prior to a declaration of state of emergency (of limited duration and scope), with the consequent assumption of shared responsibility at the highest level of government. Emergency powers can also be used in order to prevent dangerous situations or serious damage to persons or property. Emergency management activity is based either on emergency plans drawn up on a national level by the Department for Civil Defence and on a local level by the Prefecture, or else on national emergency programmes. The special organisation that is required to ensure the central management and coordination of emergency assistance is the Operations Committee of Civil Defence. When the state of emergency is proclaimed, the prefects of the provinces affected will take all necessary measures to guarantee relief; they will implement the dispositions of the provincial emergency plans on behalf of the President of the Council of Ministers or the Minister Charged with Coordinating Civil Protection. In the study region, the Civil Protection Service has a long tradition of efficiency. For the case of Trento, the flood risk management strategy can be described as follows: The limit discharge for overflowing of the Adige at Trento is around 2100 m3/s, with a return time of 30-40 years. The local plan for flood risk management provides an indication to increase the expected return time of the overflowing discharge to 80-100 years (with a discharge around 2400-2500 m3/s). This will be accomplished by: i) increasing the embankment elevation; ii) building a number of polders on the Adige floodplain; iii) retrofitting two hydropower reservoirs for flood peak alleviation; iv) establishment of a regional flood forecasting and warning system. At the same time, a wide spectrum of non-structural measures has been taken into account. To enable the regional implementation of the plan, according to the current legislation, flood hazard and flood risk areas have been mapped. Flood risk map have been prepared, taking into account flood vulnerability and exposure.

Process More effective use of regulatory instruments: New flood risk maps across the region could cause a drop in housing values in vulnerable areas. People are unwilling to accept restrictions to economic activities and to use of private property based on risk maps drawn by the experts. Main criticisms are the following: • Maps are based on uncertain knowledge of real processes: this is made evident when flood-

related damages are located in areas not considered to be hazardous; • Maps are too rigid, i.e. the maps are not revised and updated on the basis of better knowledge

gained during past flood events; • Restrictions are rigid. They are not modified even when expensive protective works for hazard

mitigation are put in place. This means that regulatory instruments (understood here essentially as land use regulations) may be effective when i) their uncertainty is quantified; ii) they are revised according to new events; iii) they allow to take into account the mitigation exherted by protection structures. From a land-use planning perspective, land developers could also be directly involved in the discussions. Regulation implies



communication. A particularly important point is the concept of residual risk, which brings into the debate a term of ambiguous communication. Generally, people don�t understand / accept the idea of the residual risk, particularly when they have to pay taxes for protection works or when they have to accept land use regulation to mitigate flood vulnerability. The confidence that risk is being properly controlled ( a risk under control) is not sufficient to make it acceptable: i.e.: when people is told to be �safer, but still at risk�, they are surprised. This means that more effective use of regulations implies better communication and better education. Integration of planned physical measures and policy instruments: The implementation of structural protection measures generally diminishes people�s awareness of flood risk. Dikes, levees, barriers, etc. generate a false sense of total safety, whereas these can only actually increase (at the best) the return time of damaging events. Policy instruments should take into account of new flood protection structures; if they fail to do this, they loose part of their trustable information content. Physical measures and policy instruments need to be combined through careful communication, since physical measures have an implied symbolic power (see above). Services in the region starts to maintain symbols of past flood disaster as potential icons. Feasibility of using modern hydrometeorological integrated flood and flash flood forecasting and warning to decrease vulnerability: The experience with flood forecasts is generally positive. This experience is limited to riverine plain floods. Experience with convective events and flash flood outlook is generally much more limited. The feedback is very positive for the fire brigade of Bolzano, which makes use of radar-based observations for localization of high intensity storms and their ground effects. A more general discussion was held on the feedback between efficiency of Civil Protection Service and flood risk management. It has been observed that, as the civil protection services are very efficient, people tend to rely on them entirely, delegating responsibility for safety. As a consequence they loose self protection skills which are based on local knowledge and may contribute to reduce damage and at the same time they will be less prone to adopt safety oriented behaviours. There is a need to be aware of the ambiguous relation between trust in expert agencies and personal responsibility (involvement versus delegation). A form of normalization bias (Mileti and O�Brien, 1992) has also been observed. This describes how people infer from an ability to cope successfully with (objectively) minor impacts a capability to deal with any future larger occurrence. This bias can result in people overestimating their perceived preparedness and/or underestimating the risk attributed to hazard. Efficiency of the Civil Protection Service can generate a form of normalization bias. A counterpart in this is that episodes with less capability to counteract from Civil Protection are less tolerated, as well as episodes where interventions from Civil Protection generate damages to private properties. Ways to enhance awareness to flood risk and preparedness of the population: A preliminary factor to improve awareness to flood risk is the knowledge of the flood awareness patterns among and inside the various communities in the region. Perception from the experts is that the areas where awareness of flood risk is lowest are the major urban areas and their surroundings, as well as communities which were not affected by flood since long time. In general, flood memory is considered as a major ingredient in building flood risk awareness. In major urban areas this may be lost due to major immigration fluxes of population and to coexistence of other perceived factors of risk (environmental pollution, car accidents, and so forth). In rural areas, if no flooding occurs for a long time the awareness of flood risks decreases. Particularly sensitive points are the locations on the flood risk system where flood risk analysis from experts does not match with flood risk perception from local residents. Major ways to modify flood risk awareness are i) organization of exercitations for flood evacuations; ii) interaction with local resident network an stakeholder to integrate them into the flood hazard analysis process.



4. Recommendations Changing from the paradigm of flood protection to flood risk management will require a lasting learning process at local, regional, national, and European level (Schanze et al. 2005). It is plausible to assume that, at present, the position of European countries within this learning process is different. Hence, societal context becomes crucial for understanding and improving strategies for pre-flood risk mitigation (Hutter 2006). Societal context has a strong influence on attempts to use planning for pre-flood risk reduction, especially in case of spatial planning. Spatial planning in general and land use planning in particular have even been described as dependent variables of societal context conditions, e.g. institutional arrangements regulating the property rights of landowners and power structures within the national political systems. However, there is evidence within the flood risk management research community and related policy fields that leeway to exploit strategic choices inherent in spatial planning is being recognized and used more intensely than before for pre-flood risk management. Enhancing this evidence through case studies in three European countries (Germany, UK, Italy) is the task of coming reports. Thereby, it is unlikely to find one dominant cause for current practices of spatial planning. It is more likely that robust improvements will come from changing many different components over time (Pettigrew 1991, Pettigrew et al. 2003). For instance, improvements can be achieved through integrating stakeholders into decision making at project level, using scenarios to identify robust assumptions about the development of flood risk, connecting planning at framework level with strategic projects that combine structural and non-structural measures for pre-flood risk mitigation. Grounded in a multidimensional understanding of strategy, the case studies in European catchments provide knowledge about how strategic planning at local level currently is used for pre-flood risk management, what societal context conditions should be taken into account, and who important actors within municipalities and water authorities are for improving the practice of mitigating flood risk. Informed in previous sections by strategy research, it is certainly inappropriate to use this knowledge for prescribing a simple, linear model how to make better strategies in the real world. As results of practical relevance we should expect (1) a set of conceptual tools to understand strategy making for pre-flood risk management by deploying spatial planning at local level, (2) important questions that address true strategic issues and related answers, (3) and knowledge about forces influencing the strategy process without determining it. This set of recommendations will give local planning authorities enough leeway to apply it to their specific context conditions. Burby and associates argue that local governments �must take care in carrying out hazard reduction planning, minding both the political and technical details. Some of the lessons from local experience show that communities must be both visionary and pragmatic. They need to be far-sighted in gathering credible data, preparing maps, building consensus through planning, and paying attention to development management well before pressures built to use hazard areas more intensively. They also must be practical in using site-specific approaches, integrating hazard mitigation into their normal development review procedures, taking advantage of post-disaster windows of opportunity, and being prepared to purchase property if necessary.� (Burby et al. 2000, p. 19f.) Like practice, research on strategies for pre-flood risk management has to be visionary and pragmatic. Significant improvements in understanding pre-flood strategies will come from a comprehensive analysis which takes into account all natural and societal factors of flood risk management on catchment scale (Schanze et al. 2005). At the same time, improvements depend on more pragmatic approaches that analyse specific management problems that are, by definition, of limited extent. These problems can be analysed empirically. The focus of this report is the specific relationship of pre-flood risk management and strategic planning at the local level, especially with regard to spatial planning. Six recommendations shall be used to draw together what emerges from recurrent themes and findings:



Adaptive Model of Strategy is the Rule, Effective Linear Processes are Exceptions

(1) In the �real world�, adaptive process management prevails and linear processes are exceptions: More often than not, claims for a rational approach to pre-flood risk management are described as a very orderly process of open dialog and decision making that fits into well-defined components and tasks for accomplishment. Even if such description is used to give an overview, to point to iterations, to make connections clear, the background assumption is that adaptive processes are (or should be) more the exception than the rule. The conclusion in this report is somehow different. In the �real world�, we should expect adaptive processes as rule, and linear processes as exceptions. Consequently, the burden of proof should be on the side of proponents of linear processes which instruct flood risk managers to go step-by-step to a better solution. Such solutions are certainly possible, if they are on the political agenda with high priority, if they are supported by powerful politicians, perhaps because of media attention, if an extensive analysis of strategic alternatives, possible flood events and consequences is possible. Considering adaptive processes as rule and linear ones as exceptions urges decision makers to judge from the beginning if high investments in data-heavy linear processes are worthwhile.

Strategic Planning for Pre-Flood Risk Management

(2) Use strategic planning as an umbrella concept for content- and process-oriented approaches to long-term FRM: Long-term FRM is based partly on defining and assessing strategic alternatives with regard to a wide range of criteria (sustainability, robustness, flexibility, see Klijn et al. this issue). Effective scenario analysis is simple, coherent, maximizes differences between scenarios and strategic alternatives and can have visionary power. This design-oriented and scenario-based approach to exploring strategic alternatives is an effective way to formulate new ideas for long-term FRM. However, more often than not, the relevance of new ideas for local decision making is in question. Strategic planning at regional level helps to initiate and conduct a scenario planning episode (e.g., a �strategy game�, see Lonsdale et al. 2005) that has the chance to be of significance for local officials and indirectly for politicians too (e.g., through mobilizing officials that are sufficiently homogenous in their understanding of FRM and related to different institutional positions). Strategic planning serves as an umbrella concept that encompasses content-oriented (exploring strategic alternatives) and process-oriented approaches (mobilizing a limited range of actors, legitimizing the planning episode, and so forth, see Penning-Rowsell & Winchester 1992).

(3) Use strategic planning to facilitate travelling across spatial levels: Strategic planning can be applied to FRM at different spatial levels. This is the main reason why FRM can benefit from using strategic planning. The rationale for this claim is as follows: FRM as �holistic and continuous societal analysis, assessment and reduction of flood risk � (Schanze 2006, p. 4) is the overall challenge of integrating � among others � technical expertise and political decision making in various policy fields and at different spatial scales (site, local, regional, state, and so forth). European Member States are moving in this direction � with different starting points and speed. However, it is evident that integration should be combined with �division of labour� between policy fields, levels and so forth. Strategic planning is one way of focusing on a specific spatial level, a limited range of actors, selected issues and strategic alternatives and important forums as well as arenas for decision making without forgetting that planning results have to be embedded in an overall strategy for FRM at catchment level. Because the term �strategic planning� can be used flexibly without loosing its meaning it serves as a bridge between different spatial levels and policy fields. Hopefully, in the long run, this will contribute to develop full blown strategies for long-term FRM.



Learning to Deal with Uncertainty of Flood Risk in Long-Term Planning To deal with uncertainty of flood risk in the context of long-term planning, research mostly points to the content and context dimensions of FRM (e.g., strategic alternatives, scenarios of physical and socioeconomic development). In contrast, this report highlights strategic choices that reside within the process dimension. Three conclusions show that content, context, and process are tightly interwoven for effective strategy development. These conclusions take on the form of recommendations to local flood risk managers supported by state authorities. (4) Think of dealing with uncertainty of flood risk in long-term planning as a social process prone to interruption, irrelevance for ongoing decision making, and post-disaster politics. Long-term planning, as strategic planning (Bryson & Roering 1989/2000, Bryson 2004), is prone to be interrupted because of decision makers shifting their attention to pressing problems of the day. This is so especially in the case when politicians and citizens are included into the planning process. Usually, elected politicians and citizens have much more on their agenda than long-term planning in flood risk management. Thus, either dealing with uncertainty in long-term planning mainly takes place between agency staff, officials from different authorities, and external experts (e.g., consultants and scientists, see Klinke & Renn 2002) or additional resources for managing an inclusive process are allocated to long-term planning (e.g., additional resources for process champions and facilitators measured, for instance, in attention of politicians and financial resources for an inclusive long-term planning project). However, good process management and sufficient resources are necessary, not sufficient conditions for dealing with uncertainty in long-term planning. Two further conclusions show that local flood risk managers should pay ample attention to questions of justifying planning and to matching process management with context conditions. (5) Justify both learning for exploring uncertain futures and learning for exploiting a certain past! Make connections explicit! In principle, flood risk managers cannot choose between learning from the past and learning for the future. They have to do both (which does not exclude that single projects emphasize the past or the future). Being aware of this necessity will motivate managers to justify each in relation to the other. Managers who do this deliberately increase the odds of learning how to deal with uncertainty of flood risk in long-term planning. An example that illustrates this conclusion can be found in the ongoing discussion about formulating a new long-term strategy for the Thames Estuary. Numerous contributors to this discussion mention uncertainty as a challenge for designing a new long-term strategy for the Thames estuary (e.g., uncertainty about the specific impacts of global climate change on regional sea-level rise and inundation dynamics). Some researchers and practitioners expand their imagination about uncertain distant futures in FRM through conducting strategy games (e.g., Lonsdale et al. 2005). However, exploring uncertain futures takes place against the background of common knowledge about the past, for instance, knowledge about the decades it took to decide on and build the Thames Barrier as a main flood protection scheme within the estuary. Considering the time needed for building the Thames Barrier serves as a factual statement for justifying the need to think early about a new long-term strategy (e.g., Lavery & Donovan 2005). Learning about the past and exploring possible futures are connected in justifying current attempts to define a new strategy for the long run. Flood risk managers should identify such connections and should make them explicit to facilitate communication about uncertainty of flood risk and long-term planning. (6) Use forums for exploring uncertain futures of flood risk in long-term planning! Thereby, it is crucial to find a fit between historical context and process options and to not overestimate the possibilities of open dialog with politicians and citizens! Arenas are social settings where decisions are made and implemented. Forums are made for discussion as well as �nice talk�, if discussion is without consequences for decision making. However, local flood risk managers should not underestimate the power of forums. Indeed, planners usually gain leverage through their ability to organize forums in which no one decision maker can dominate. �The possibilities are thereby enhanced for the emergence of a collective interest or vision that transcends narrow partisan interests. Once this collective vision emerges, it can have a profound impact on subsequent decisionmaking in arenas, or on conflict



management in courts� (Bryson & Crosby 1993, p. 191). Existing arenas and forums deeply reflect historical context of long-term planning in FRM. Case studies show (Hutter et al. 2007) that considering historical context is a nontrivial task. For instance, some people that experienced a recent flood disaster are willing to search for fundamentally new ways to understand flood risk in its full variability. They see comprehensive understanding as a precondition for reducing flood risk to an acceptable level. Some other people that suffered from the �same� disaster focus their attention on avoiding in the future what they encountered as a deeply disturbing outside influence in the past. The Dresden example shows that the latter is not only a matter of people, but of powerful politicians too. To consider such complex context conditions, flood risk managers should be capable of doing two things: (1) they should avoid interpreting open dialog with all potentially relevant actors as the �one best way� to rational decision making (e.g., Merz & Emmermann 2006). (2) They should deploy a wide range of process options to use and design forums for communication about uncertainty of flood risk (e.g., choosing an actor- or theme-driven process, organizing long-term planning as project or ongoing process). For instance, this report proposes that, in a context where open communication about uncertainty of flood risk is less than welcome among a majority of politicians and citizens, local flood risk managers interested in discussing uncertainty of flood risk (e.g., through formulating worst-case scenarios) should seek project-based discussion with like-minded people from different organizations (e.g., water management at state level, scientific organizations). This is a forum with open discussion among selected people that are willing to learn how to deal with high uncertainty of flood risk even after a disaster. Results of discussion can spread step-by-step into the ongoing decision making process at local level. Only in the long run we can tell if this leads towards a risk culture that supports open communication about the whole range of uncertainty issues within long-term planning. In sum, learning how to deal with uncertainty of flood risk in long-term planning always takes place in a historical context. This will influence the content (e.g., referring to the Thames Barrier for justifying long-term planning) as well as the process of learning how to effectively communicate about and decide between strategic alternatives for uncertain futures (e.g., project-based discussion in Dresden). Managers that exploit strategic leeway within the process dimension of long-term FRM should be more successful than managers that take this dimension for granted. Uncertainty is one of these cross-cutting themes that foster significant progress in very different scientific disciplines (e.g., hydroinformatics, economics, management and strategy research). In contrast, the search for certainty and order has been described as one of the overarching dispositions of human nature. This report is based on the notion that learning to deal with uncertainty implies letting go a preference for framing issues as either/or-problems. Effective strategy makers acknowledge uncertainty and certainty; content, process, and context; forums and arenas; as well as exploration and exploitation. It is hard for people to find a simple solution for these complexities. As a consequence, effective strategy development itself remains an uncertain endeavour.



5. References 1. ALBRECHTS L, HEALEY P, KUNZMANN K R (2003), Strategic Spatial Planning and

Regional Governance in Europe, APA Journal, Vol. 69, No. 2, pp. 113-129.

2. APEL H, THIEKEN A H, MERZ B, BLÖSCHL G (2004), Flood Risk Assessment and Associated Uncertainty, Natural Hazards and Earth System Science, Vol. 4, 295-308.

3. BARTUNEK J M (1993), The Multiple Cognitions and Conflicts Associated with Second Order Organizational Change. Murnighan, J. K. (ed.): Social Psychology in Organizations. Advances in Theory and Research, Englewood Cliffs, Prentice Hall, 322-349.

4. BAUM J A C (ed.) (2002), Companion to Organizations, Oxford/UK, Basil Blackwell.

5. BENZ A & D FUERST (2002), Policy Learning in Regional Networks, European Urban and Regional Studies, Vol. 9, No. 1, pp. 21-35.

6. BIZER K, EWRINGMANN D, BERGMANN E, DOSCH F, EINIG K, HUTTER G (1998), Mögliche Maßnahmen und Instrumente einer Steuerung der Verkehrs- und Siedlungsflächennutzung, Berlin, Springer.

7. BOEHM H R, HEILAND P, DAPP K, HAUPTER B (2002), Spatial Planning and Supporting Instruments for Preventive Flood Management. Final report of IRMA-SPONGE Project 5, Darmstadt University of Technology.

8. BOEHM H R, HEILAND P, DAPP K, MENGEL A (1999), Anforderungen des vorsorgenden Hochwasserschutzes and Raumordnung, Landes-/Regionalplanung, Stadtplanung und die Umweltfachplanungen � Empfehlungen für die Weiterentwicklung, Berlin, Umweltbundesamt.

9. BOIN A, T´HART P (2003), Public Leadership in Times of Crisis: Mission Impossible? Public Administration Review, Vol. 63, No. 5, 544-553.

10. BOYNE G, ASHWORTH R, POWELL M (2000), Testing the Limits of Incrementalism: An Empirical Analysis of Expenditure Decisions by English Local Authorities, 1981-1996, Public Administration, Vol. 78, No. 1, pp. 51-73.

11. BRAKE K (2000), Strategische Entwicklungskonzepte für Großstädte � mehr als nur eine Renaissance der �Stadtentwicklungspläne�?, Archiv für Kommunalwissenschaften (AfK), No. 2, pp. 269 � 288.

12. BRAMLEY M & BOWKER (2002), Improving Local Flood Protection to Property, Proceedings of ICE, pp. 49-54.

13. BRESSERS H & KUKS S (2004), Integrated Governance and Water Basin Management. Conditions for Regime Change and Sustainability, Dordrecht, Kluwer.

14. BRODY S D (2003), Are We Learning to Make Better Plans? A Longitudinal Analysis of Plan Quality Associated with Natural Hazards, Journal of Planning Education and Research, Vol. 23, pp. 191-201.

15. BRYSON J M (2004), Strategic Planning for Public and Nonprofit Organizations. A Guide to Strengthening and Sustaining Organizational Achievement, Jossey-Bass, San Francisco.

16. BRYSON J M (1998), Strategic Planning, Shafritz J (ed.), International Encyclopedia of Public Policy and Administration, New York, Holt and Co., pp. 2160-2169.

17. BRYSON J M, CROSBY B C (1993), Policy Planning and the Design and Use of Forums, Arenas, and Courts, Environment and Planning B: Planning and Design, Vol. 20, 175-194.

18. BRYSON J M & ROERING W D (1989/2000), Mobilizing Innovation Efforts: The Case of Government Strategic Planning, Van de Ven A, Angle, H, Poole M S (eds.), Research on the Management of Innovation, Reprinted 2000, pp. 583-610.



19. BURBY R J (1999), Unleashing the Power of Planning to Create Disaster-Resistant Communities, APA Journal, pp. 247-257.

20. BURBY R J (ed.) (1998), Cooperating with Nature. Confronting Natural Hazards with Land-Use Planning for Sustainable Communities, Washington, D.C., Joseph Henry Press.

21. BURBY R J, DEYLE R E, GODSCHALK D R, OLSHANSKY R B (2000), Creating Hazard Resilient Communities through Land-Use Planning, Natural Hazards Review, Vol. 1, No. 2, pp. 99-106.

22. BURBY R J, NELSON A C, PARKER D & HANDMER J (2000), Urban Containment Policy and Disaster: Is there a Connection? Paper Presentation at the 2000 Conference of the Association of Collegiate School of Planning November 2-5, 2000, Atlanta.

23. BURGELMAN R A (1991), Intraorganizational Ecology of Strategy Making and Organizational Adaptation: Theory and Field Research, Organization Science, Vol. 2, No. 3, pp. 239-262.

24. BURGELMAN R (2002), Strategy is Destiny. How Strategy-Making Shapes a Company´s Future, New York.

25. CARROLL J S, RUDOLPH J W & HATAKENAKA S (2003), Learning from Organizational Experience, Easterby-Smith M & Lyles M (eds.), Handbook of Organizational Learning and Knowledge Management, Malden/USA, Basil Blackwell, pp. 575-600.

26. CHAFFEE E E (1985), Three Models of Strategy, Academy of Management Review, Vol. 10, No. 1, pp. 89-98.

27. CHAKRAVARTHY B S & LORANGE P (1991), Managing the Strategy Process. A Framework for a Multibusiness Firm, Englewood Cliffs/New Jersey, Prentice Hall.

28. CHAKRAVARTHY B S & WHITE R E (2002), Strategy Process: Forming, Implementing and Changing Strategies, Pettigrew A, Thomas H, Whittington R (eds.), Handbook of Strategy and Management, London, Sage, pp. 182-205.

29. CHANDLER A D (1962), Strategy and Structure. Chapters in the History of the Industrial Enterprise, Cambridge/Massachusetts, The MIT Press.

30. COMFORT L K, SUNGU Y, JOHNSON D & DUNN M (2001), Complex Systems in Crisis: Anticipation and Resilience in Dynamic Environments, Journal of Contingencies and Crisis Management, Vol. 9(3), pp. 144-158.

31. COMMUNITY AND HOUSING DIRECTORATE, CIVIC CENTRE, NEWCASTLE UPON TYNE (2000), Going for Growth. A Green Paper. A Citywide Vision for NEWCASTLE 2020, Newcastle.

32. DAFT R L, WEICK K E (1984/2001), Toward a Model of Organizations as Interpretation Systems. Reprinted in: Weick, Karl E. (2001): Making Sense of the Organization. Oxford, Blackwell, 241-258.

33. DAWSON R J, HALL J W, BATES P D, NICHOLLS R J (2005), Quantified Analysis of the Probability of Flooding in the Thames Estuary under Imaginable Worst-case Sea Level Rise Scenarios, Water Resources Development, Vol. 21, No. 4, 577-591.

34. DE BRUIJN K (2004), Resilience and Flood Risk Management, Water Policy, Vol. 6, pp. 53-66.

35. DE BRUIJN, K M & KLIJN F (2001), Resilient flood risk mangement strategies, Guifen, L & Wenxue L, (eds.), Proceedings of the IAHR congress, September 16-21 2001, Beijing China, Tsinghua University press, pp 450-457.

36. DE GEUS A P (1996), Planning as Learning, Starkey K (ed.), How Organizations Learn, London, International Thompson Business Press, pp. 92-99.



37. DKKV � DEUTSCHES KOMITEE FÜR KATASTROPHENVORSORGE (2003), Hochwasservorsorge in Deutschland. Lernen aus der Katastrophen 2002 im Elbegebiet, Bonn, DKKV.

38. EA � ENVIRONMENTAL AGENCY (2003), Strategy for Flood Risk Management (2003/4 � 2007/8). Version 1.2, London, EA.

39. EASTERBY-SMITH M & LYLES M (HRSG.) (2003), Handbook of Organizational Learning and Knowledge Management, Malden/USA, Basil Blackwell.

40. EUROPEAN COMMISSION (2001), Einheit Europas, Solidarität der Völker, Vielfalt der Regionen. Zweiter Bericht über den wirtschaftlichen und sozialen Zusammenhalt. Luxemburg.

41. EUROPEAN COMMISSION (1997), The EU Compendium of Spatial Planning Systems and Policies, Luxembourg, Office for Official Publications of the European Commission.

42. EUROPEAN COMMISSION (2000), The EU Compendium of Spatial Planning Systems and Policies. Italy, Luxembourg, Office for Official Publications of the European Commission.

43. EUROPEAN COMMISSION (2000), The EU Compendium of Spatial Planning Systems and Policies. United Kingdom, Luxembourg, Office for Official Publications of the European Commission.

44. EC � EUROPEAN COMMUNITIES (2000a), The Urban Audit. Towards the Benchmarking of Quality of Life in 58 European Cities. Volume I: The Yearbook. Overview. Comparative Section. Luxembourg.

45. EC � EUROPEAN COMMUNITIES (2000b), The Urban Audit. Towards the Benchmarking of Quality of Life in 58 European Cities. Volume II: The Yearbook. Summary Results for Each City. Luxembourg.

46. EISENHARDT K M (1995), Building Theories from Case Study Research, Huber G P & Van de Ven A H (eds.), Longitudinal Field Research Methods. Studying Processes of Organizational Change, Thousand Oaks, Sage, pp. 65-90.

47. EVANS E, RAMSBOTTOM D M, WICKS J M, PACKMAN J C, PENNING-ROWSELL E (2002), Catchment Flood Management Plans and the Modelling and Decision Support Framework, Proceedings of ICE, pp. 43-48.

48. FLEMING G (2002), Learning to Live with Rivers � the ICE´s Report to Government, Proceedings of ICE, pp. 15-21.

49. FLOODSITE (2005), FLOODSite Report T32-04-01. Language of Risk � Discussion Document. Wallingford, FLOODsite.

50. FORDHAM M (2000), Participatory Planning for Flood Mitigation: Models and Approaches, Parker D J (ed.), Floods. Volume II, London, Routledge, pp. 66-79.

51. FREDRICKSON J W (1984), The Comprehensiveness of Strategic Decision Processes: Extension, Observations, Future Directions, Academy of Management Journal, Vol. 27, No. 3, pp. 445-466.

52. FREDRICKSON J W & MITCHELL T R (1984), Strategic Decision Processes: Comprehensiveness and Performance in an Industry with an Unstable Environment, Academy of Management Journal, Vol. 27, No. 2, pp. 399-423.

53. GALER G S, VAN DER HEIJDEN K (2001), Scenarios and Their Contribution to Organizational Learning: From Practice to Theory, Dierkes M., Berthoin Antal A., Child J., Nonaka I. (eds.) Handbook of Organizational Learning and Knowledge, Oxford University Press, Oxford, 849-864.

54. GAVETTI G, LEVINTHAL D (2000), Looking Forward and Looking Backward: Cognitive and Experiential Search, Administrative Science Quarterly, Vol. 45, 113-137.



55. GERSICK C (1991), Revolutionary Change Theories: A Multilevel Exploration of the Punctuated Equilibrium Paradigm, Academy of Management Review, Vol. 16, No. 1, 10-36.

56. GHEMAWAT P (1991), Commitment. The Dynamic of Strategy, New York, The Free Press.

57. GODSCHALK D R (2003), Urban Hazard Mitigation: Creating Resilient Cities, Natural Hazards Review, Vol. Z, pp. 136-143.

58. GRABOWSKI M & ROBERTS K (1997), Risk Mitigation in Large-Scale Systems: Lessons from High Reliability Organizations, California Management Review, Vol. 39(4), pp. 152-162.

59. GRANT R M (2005), Contemporary Strategy Analysis, Malden/USA, Blackwell.

60. GRANT R M (2003), Strategic Planning in a Turbulent Environment: Evidence from the Oil Majors, Strategic Management Journal, Vol. 24, pp. 491-517.

61. GREIVING S (1999), Hochwasserschutz in der räumlichen Planung, Raumforschung und Raumordnung (RuR), No. 1, pp. 25-34.

62. GRUNTFEST E & RIPPS A (2000), Flash Floods: Warning and mitigation efforts and prospects, Parker D J (ed.), Floods, Volume I, Routledge, pp. 377-390.

63. GUNNE-JONES A (2003), Land-Use Planning: How Effective is it in Reducing Vulnerability to Natural Hazards? Unpublished paper.

64. HALL J W (2003), Handling Uncertainty in the Hydroinformatic Process, Journal of Hydroinformatics, 215-232.

65. HALL J W, MEADOWCROFT I C, SAYERS P B, BRAMLEY M E (2003), Integrated Flood Risk Management in England and Wales, Natural Hazards Review, Vol. 4, No. 3, pp. 126-135.

66. HANDMER J, DOVERS (1996), A Typology of Resilience: Rethinking Institutions for Sustainable Development, Organization & Environment, 482-511.

67. HANDMER J, HENSON R, SNEERINGER P, KONIENCZNY R, MADEJ P (2001), Warning Systems for Flash Floods: Research Needs, Opportunities and Trends, Gruntfest E & Handmer J (eds.), Coping with Flash Floods, pp. 77-89.

68. HANDMER J, PENNING-ROWSELL E & TAPSELL S (1999), Flooding in a Warmer World: The View from Europe, Downing T E, Olsthoorn A A, & Tol R S J (eds.), Climate, Change and Risk, London, Routledge, pp. 125-161.

69. HASTIE R, DAWES R M (2001), Rational Choice in an Uncertain World. The Psychology of Judgement and Decision Making, Sage, Thousand Oaks.

70. HEALEY P (2003), Collaborative Planning in Perspective, Planning Theory, Vol. 2, No. 2, pp. 101-123.

71. HOOIJER A, KLIJN F, KWADIJK J & PEDROLI B (eds.) (2002), IRMA-SPONGE, Towards Sustainable Flood Risk Management in the Rhine and Meuse River Basins, Main results of the research project, Delft.

72. HOOIJER A, KLIJN F, PEDROLI B, VAN OS A (2004), Towards Sustainable Flood Risk Management in the Rhine and Meuse River Basins: Synopsis of the Findings of IRMA-SPONGE, River Research and Applications, Vol. 20, pp. 343-357.

73. HOUSE M A (1999), Citizen Participation in Water Management, Wat. Sci. Tech., Vol. 40, No. 10, pp. 125-130.

74. HREBINIAK L G & JOYCE W F (1985), Organizational Adaptation: Strategic Choice and Environmental Determinism, Administrative Science Quarterly, Vol. 30, pp. 336-349.



75. HUCKE J (1996), Die Implementation umweltpolitischer Programme � Strukturen und Erfolgsbedingungen, Zeitschrift für angewandte Umweltforschung (ZAU), Sonderheft 8, pp. 59-72.

76. HUTTER G (2006), Strategies for Flood Risk Management � A Process Perspective, Schanze J., Zeman E., Marsalek J. (eds.) Flood Risk Management � Hazards, Vulnerability and Mitigation Measures, Springer, Berlin, 229-246.

77. HUTTER G (2004), Strategies of Local Governments to Increase the Preparedness of Households Facing Flash Floods, Geller, Walter et al. (eds.), 11th Magdeburg Seminar on Waters in Central and Eastern Europe: Assessment, Protection, Management. Proceedings of the international conference 18-22 October 2004 at the UFZ. Leipzig (UFZ), pp. 87-88.

78. HUTTER G, SCHANZE J (2004), Potenziale kooperativen Lernens für das Hochwasserrisikomanagement � am Beispiel der Vorsorge gegenüber Sturzfluten im Flussgebiet der Weißeritz [Co-operative Learning for Flood Risk Management � The Case of Prevention within the Catchment of the Weißeritz River], Felgentreff C., Glade Th. (eds.) Von der Analyse natürlicher Prozesse zur gesellschaftlichen Praxis, Universitätsverlag Potsdam, Potsdam, 63-87.

79. HUTTER G & WESTPHAL C (2003), Wettbewerbsstrategien von Kommunen zur Weiterentwicklung des Wohnungsbestands � am Beispiel von Leipzig und Münster, Hutter G, Iwanow I, Müller B (eds.), Demographischer Wandel und Strategien der Bestandsentwicklung in Städten und Regionen, IÖR-Schriften Bd. 41, Dresden, pp. 127-148.

80. HUTTER G, WESTPHAL C, SIEDENTOP S, JANSSEN G, MÜLLER B, VORMANN M, EWRINGMANN D (2004), Handlungsansätze zur Berücksichtigung der Umwelt-, Aufenthalts- und Lebensqualität im Rahmen der Innenentwicklung von Städten und Gemeinden � Fallstudien, Berlin, Umweltbundesamt.

81. INNES J E, BOOHER D E (1999), Consensus Building and Complex Adaptive Systems. A Framework for Evaluating Collaborative Planning, APA Journal, Vol. 65, 412-423.

82. INSTITUTION OF CIVIL ENGINEERS (ICE) (2001), Learning to Live with Rivers. Final Report of the Institution of Civil Engineer´ Presidential Commission to Review the Technical Aspects of Flood Risk Management in England and Wales, ICE, London.

83. ISDR � INTERNATIONAL STRATEGY FOR DISASTER REDUCTION (2004), Terminology: Basic terms of disaster risk reduction, http://www.unisdr.org/eng/library/lib-terminology-eng%20home.htm (17.06.04)

84. JANSSEN G (2005), Hochwasserschutz, Akademie für Raumforschung und Landesplanung (ARL), Handwörterbuch der Raumordnung, Hannover, ARL, in press.

85. JOHNSON C L, TUNSTALL S M, PENNING-ROWSELL E C (2005), Floods as Catalysts for Policy Change: Historical Lessons from England and Wales, Water Resources Development, Vol. 21, No. 4, 561-575.

86. KLIJN F, DE BRUIJN K, LIGTVOET W (2007), Exploration of Strategic Alternatives for Long-Term Preventive Flood Risk Management: How to Deal with an Uncertain Future? Schanze J (ed.), Flood Risk Management Research � From Extreme Events to Citizens Involvement. Proceedings of European Symposium on Flood Risk Management Research (EFRM 2007), Dresden, IOER, 112-121.

87. KLINKE A, RENN O (2002), A New Approach to Risk Evaluation and Management: Risk-Based, Precaution-Based, and Discourse-Based Strategies, Risk Analysis, Vol. 22, No. 6, 1071-1094

88. KISSLING I & KUKS S (2004), The Evolution of National Water Regimes in Europe. Transitions in Water Rights and Water Policies, Dordrecht, Kluwer.



89. KUHLICKE C & DRÜNKLER D (2004), Vorsorge durch Raumplanung? Das Problem der Prävention in zeitlicher und raumplanerischer Hinsicht: Eine Lösungsskizze, Raumforschung und Raumordnung (RuR), No.3, pp. 169-176.

90. KUNDZEWICZ Z W (2002), Non-structural Flood Protection and Sustainability, Water International, Vol. 27, No. 1, pp. 3-13.

91. KUNDZEWICZ Z W & SAMUELS P G (1997), Real-time Flood Forecasting and Warning. Conclusions from Workshop and Expert Meeting, Proceedings of Second RIBAMOD Expert Meeting, Published by DG XII, European Commission.

92. LAVERY S, DONOVAN B (2005), Flood Risk Management in the Thames Estuary Looking Ahead 100 Years, Philosophical Transactions of the Royal Society, 1355-1474.

93. LAWA � LÄNDERARBEITSGEMEINSCHAFT WASSER (2004), Instrumente und Handlungsempfehlungen zur Umsetzung der Leitlinien für einen zukunftsweisenden Hochwasserschutz, Düsseldorf, Ministerium für Umwelt und Naturschutz, Landwirtschaft und Verbraucherschutz des Landes Nordrhein-Westfalen (NRW).

94. LfUG � SÄCHSISCHES LANDESAMT FÜR UMWELT UND GEOLOGIE (2004), Ereignisanalyse. Hochwasser August 2002 in den Osterzgebirgsflüssen, Dresden (LfUG).

95. LONSDALE K G, DOWNING T E, NICHOLLS R J, PARKER D, VAFEIDIS A T, DAWSON R, HALL J (2005), Plausible Responses to the Threat of Rapid Sea-Level Rise for the Thames Estuary, Hamburg University and Centre for Marine and Atmospheric Science: Working Paper FNU-77, Hamburg.

96. MACHADO N, BURNS T R (1998), Complex Social Organization: Multiple Organizing Modes, Structural Incongruence, and Mechanisms of Integration, Public Administration, Vol. 76, Summer, 355-386.

97. MARCH J G (1994), A Primer on Decision Making. How Decisions Happen. New York, The Free Press.

98. MARCH J G (1999), The Pursuit of Organizational Intelligence, Malden, Blackwell.

99. MARCH J G, SPROULL L S, TAMUZ M (1996), Learning from Samples of One or Fewer, Cohen M D & Sproull L S (eds.), Organizational Learning, Thousand Oaks, Sage, pp. 1-19.

100. MASTOP H, FALUDI A (1997), Evaluation of Strategic Plans: the Performance Principle, Environment and Planning B: Planning and Design, Vol. 24, pp. 815-832.

101. MCDANIELS T L, GREGORY R (2004), Learning as an Objective within a Structured Risk Management Decision Process, Environmental Science & Technology, Vol. 38, No. 7, 1921-1926.

102. MEO M, ZIEBRO B & PATTON A (2004), Tulsa Turnaround: From Disaster to Sustainability, Natural Hazards Review, Vol. 5, No. 1, pp. 1-9.

103. MERZ B, EMMERMANN R (2006), Zum Umgang mit Naturgefahren in Deutschland: Vom Reagieren zum Risikomanagement [Dealing with Natural Hazards in Germany � From Reaction towards Risk Management], GAIA, Vol. 15, No. 6, 265-274.

104. MILETI D S (1999), Disasters by Design. A Reassessment of Natural Hazard in the United States, Washington D.C, Joseph Henry Press.

105. MINTZBERG H, AHLSTRAND B & LAMPEL J (2005), Strategy Bites Back. It is a lot More, and Less, than You ever Imagined�, Harlow, Prentice Hall/Financial Times.

106. MINTZBERG H, AHLSTRAND B & LAMPEL J (1999), Strategy Safari. Eine Reise durch die Wildnis des Strategischen Managements (Strategy Safari: a guided tour through the wilds of strategic management), Wien, Ueberreuter.



107. MUELLER B (2003a), Regionalentwicklung unter Schrumpfungsbedingungen. Herausforderungen für die Raumplanung in Deutschland, Raumforschung und Raumordnung (RuR), Vol. 61, No. 1/2, pp. 28-42.

108. MUELLER B (2003b), Regionale Kooperation bei der Planung und beim Management des Wasserrückhalts in Flusseinzugsgebieten, Sächsisches Staatsministerium für Umwelt und Landwirtschaft (SMUL) (ed.), Hochwasserkonferenz. Tagungsband. Dresden, pp. 69-73.

109. NICHOLLS R J, JOHNSON C, GREEN C, SHACKLEY S (2000), An Adaptive management framework for Climate Change, Proceedings of SURVAS Expert Workshop on European Vulnerability and Adaptation to impacts of Accelerated Sea-Level Rise (ASLR), Hamburg, Germany, 19th�21th June 2000.

110. OLSHANSKY R B & KARTEZ J D (1998), Managing Land Use to Build Resilience, Burby R J (ed.), Cooperating with Nature. Confronting Natural Hazards with Land-Use Planning for Sustainable Communities, Washington, D.C., Joseph Henry Press, pp. 167-201.

111. PAHL-WOSTL C, MÖLTGEN J, SENDZIMIR J, KABAT P (2005), New methods for adaptive water management under uncertainty - The NeWater project. Paper accepted for the EWRA 2005 conference.

112. PARKER D J (2000) (ed.), Floods. Volumes I and II. London, Routledge.

113. PARKER D J (2000), Introduction to Floods and Flood Management, Parker D J (ed.), Floods. Volume I, London, Routledge, pp. 3-39.

114. PARKER D J & HANDMER J W (1998), The Role of Unofficial Flood Warning Systems, Journal of Contingencies and Crisis Management, Vol. 6, pp. 45-60.

115. PENNING-ROWSELL E (2001), Flooding and Planning: Conflict and Confusion, Town & Country Planning, April, pp. 108-110.

116. PENNING-ROWSELL E & FORDHAM M (eds.) (1994), Floods Across Europe. Flood Hazard Assessment, Modelling and Management, London, Middlesex University Press.

117. PENNING-ROWSELL E & PEERBOLTE B (1994), Concepts, Policies and Research, Penning-Rowsell E & Fordham M (eds.) (1994), Floods Across Europe. Flood Hazard Assessment, Modelling and Management, London, Middlesex University Press, pp. 1-17.

118. PENNING-ROWSELL E, WINCHESTER P (1992), Scenario Construction for Risk Communication in Emergency Planning: Six �Golden Rules�, Parker D, Handmer J (eds.) Hazard Management and Emergency Planning. Perspectives on Britain, Earthscan, London, 203-217.

119. PETTIGREW, A (1997), What is a Processual Analysis?, Scandinavian Journal of Management, Vol. 13, No. 4, 227-248.

120. PETTIGREW A, THOMAS H, WHITTINGTON R (eds.) (2002), Handbook of Strategy and Management, London, Sage.

121. PETTIGREW A & WHIPP R (1991), Managing Change for Competitive Success, Oxford/UK, Blackwell.

122. PETTIGREW A, WHITTINGTON R, MELIN L, SÁNCHEZ-RUNDE C J, VAN DEN BOSCH F J, RUIGROK W, NUMAGAMI T (eds.) (2003), Innovative Forms of Organizing. International Perspectives, London, Sage.

123. PLATE E J (1999), Flood Risk Management: a Strategy to Cope with Floods, Bronstert A, Ghaz A, Hladny J, Kundzewicz Z W & Menzel L (eds), Proceedings of the European Meeting on the Oder Flood 1997, Ribamod concerted action, European Communities, Office for Official Publications of the European Communities, Luxemburg, pp. 115-128.



124. POOLE M S (2004), Central Issues in the Study of Change and Innovation, Poole, M S, Van de Ven A (eds.) (2004): Handbook of Organizational Change and Innovation, New York, Oxford University Press, 3-31.

125. POOLE M S, VAN DE VEN A (eds.) (2004), Handbook of Organizational Change and Innovation, New York, Oxford University Press.

126. REINKE M (2005), Urban Planning and SEA, Schmidt M, Joao E, Albrecht E (eds.), Implementing Strategic Environmental Assessment, Berlin, Springer, pp. 571-584.

127. ROSENTHAL U & T´HART P (eds.) (1998), Flood Response and Crisis Management in Western Europe. A Comparative Analysis, Berlin, Springer.

128. ROYAL GEOGRAPHICAL SOCIETY (2001), Summary Statement 9. The Floods � What have they taught us? Outcome of a dinner discussion held May 2001, London, Royal Geographical Society.

129. RYDIN Y (1998), Urban and Environmental Planning in the UK, Houndmills, MacMillan.

130. SALET W & FALUDI A (eds.) (2000), The Revival of Strategic Spatial Planning, Amsterdam.

131. SAMUELS P, KLIJN F, DIJKMAN J (2006), An Analysis of the Current Practice of Policies on River Flood Risk Management in Different Countries, Irrigation and Drainage, Vol. 55, S141-S150.

132. SANCHEZ R & HEENE A (2004), The New Strategic Management. Organization, Competition, and Competence, New York, John Wiley.

133. SAYERS P B, HALL J W, MEADOWCROFT I C (2002), Towards Risk-based Flood Hazard Management in the UK, Proceedings of ICE, pp. 36-42.

134. SCHANZE J (2002), Nach der Elbeflut � die gesellschaftliche Risikovorsorge bedarf einer transdisziplinären Hochwasserforschung, GAIA, Vol. 11, No. 4, pp. 247-254.

135. SCHANZE J (2006), Flood Risk Management � A Basic Framework, Schanze J., Zeman E., Marsalek J. (eds.) Flood Risk Management � Hazards, Vulnerability and Mitigation Measures, Springer, Berlin, 1-20.

136. SCHANZE J, SAYERS P, KLIJN F, HUTTER G, MESSNER F (2005), Flood Risk Management Approach of FLOODsite. Background paper derived from the Dresden Meeting on September 13th and 14th, 2004 (Dresden Paper), Dresden, FLOODsite.

137. SEO M-G, PUTNAM L L, BARTUNEK J M (2004), Dualities and Tensions of Planned Organizational Change. Poole M S, Van de Ven A (eds.), Handbook of Organizational Change and Innovation. Oxford/UK, Oxford University Press, 73-107.

138. SITKIN S B (1996), Learning Through Failure: The Strategy of Small Losses, Cohen M D & Sproull L S (eds.), Organizational Learning, Thousand Oaks, Sage, pp. 541-577.

139. THOMPSON M, ELLIS R, WILDAVSKY A (1990), Cultural Theory, Boulder, Westview Press.

140. TONN B, ENGLISH, M, TRAVIS C (2000), A Framework for Understanding and Improving Environmental Decision Making, Journal of Environmental Planning and Management, Vol. 43, No. 2, pp. 163-183.

141. TUSHMAN M L, ROMANELLI E (1985), Organizational Evolution: A Metamorphosis Model of Convergence and Reorientation, Research in Organizational Behavior, Vol. 7, 171-222.

142. TUSHMAN M L, SMITH W (2002), Organizational Technology, Baum J A C (ed.): The Blackwell Companion to Organizations. Oxford/UK, Blackwell, 386-414.



143. VAN DE VEN A H (2000), Professional Science for a Professional School. Action Science and Normal Science, Beer, M. & Nohria, N. (eds.), Breaking the Code of Change, Harvard Business School Press, Boston, M.A., pp. 393-413.

144. VAN DE VEN A H, POLLEY D E, GARUD R & VENKATARAMAN S (1999), The Innovation Journey, Oxford/UK, Oxford University Press.

145. VAN DE VEN A H, POOLE M S (1988), Paradoxical Requirements for a Theory of Organizational Change, Quinn R E, Cameron K S (eds.), Paradox and Transformation. Toward a Theory of Change in Organization and Management, Cambridge/Mass, Ballinger, pp. 19-63.

146. VAN DE VEN A, POOLE M S (1995), Explaining Development and Change in Organizations, Academy of Management Review, Vol. 20, No. 3, 510-540.

147. VAN DER HEIJDEN K, BRADFIELD R, BURT G, CAIRNS G, WRIGHT G (2002), The Sixth Sense. Accelerating Organizational Learning with Scenarios, John Wiley, Chichester.

148. VOLBERDA H (1998), Building the Flexible Firm. How to Remain Competitive, Oxford, Oxford University Press.

149. WEICK K E (1982), Management of Organizational Change Among Loosely Coupled Elements. Reprinted in: Weick K E (2001): Making Sense of the Organization. Oxford/UK, Blackwell, 380-403.

150. WEICK K E (1987), Substitutes for Strategy, Reprinted in: Weick K E (2001), Making Sense of the Organization, Oxford/UK, Basil Blackwell, 345-355.

151. WEICK K E (1996), The Non-Traditional Quality of Organizational Learning, Cohen M D, Sproull L S (eds.), Organizational Learning, Thousand Oaks, Sage, pp. 116-124.

152. WEICK K E, QUINN R E (1999), Organizational Change and Development, Annual Review of Psychology, Vol. 50, 361-386.

153. WEICK K E & SUTCLIFFE K (2001), Managing the Unexpected. Assuring High Performance in an Age of Complexity, San Francisco, Jossey-Bass.

154. WEIHRICH R & KOONTZ H (1992), Management, New York, McGraw-Hill.

155. WHIPP R (2001), Strategy: Organizational, Smelser N J, Baltes P B (eds.), International Encyclopedia of the Social & Behavioral Sciences. Volume 22, Amsterdam, pp. 15151-15154.

156. WHITE I, HOWE J (2002), Flooding and the Role of Planning in England and Wales: A Critical Review, Journal of Environmental Planning and Management, Vol. 45, No. 5, pp. 735-745.

157. WIECHMANN T (2003), Neuere Erfahrungen mit Strategischer Planung in europäischen Großstädten. Recherche im Auftrag der Stadt Dresden, Stadtplanungsamt. Dresden, unpublished paper.

158. WIECHMANN T (1998), Vom Plan zum Diskurs? Anforderungsprofil, Aufgabenspektrum und Organisation regionaler Planung in Deutschland, Baden-Baden, Nomos.

159. WIECHMANN T & HUTTER G (2007), Die Planung des Unplanbaren � Was kann die Raumplanung von der Strategieforschung lernen?, Dangschat J, Breitfuss A, Frey O, Hamedinger A (eds.), Strategieorientierte Planung im kooperativen Staat, Wiesbaden, VS � Verlag für Sozialwissenschaften, in preparation.

160. WILDAVSKY A (1991), Searching for Safety, New Brunswick/USA, Transaction Publishers.

161. YIN R (1993), Applications of Case Study Research, Thousand Oaks, Sage.

162. YIN R (2003), Case Study Research, Thousand Oaks, Sage.



6. Task-Specific Glossary The glossary overlaps with and complements the FLOODsite document �Language of Risk (LoR)�. Terms that are defined in the LoR are indicated by the sign*. Adaptive process model of strategy � parallel processes of analysing, evaluating, deciding, implementing, controlling, and learning for deploying measures. Collaborative learning � the process of co-operation between various actors to address an issue that is widely acknowledged to be neither well-understood nor easily addressed. The process is likely to be long and iterative (Tonn et al. 2000, p. 171, Hutter & Schanze 2004). Collaborative learning is a specific mode of social learning. *Exposure � refers to people, assets and activities threatened or potentially threatened by a hazard (e.g. flooding). *Flood hazard map � map with the predicted or documented extent of flooding, with or without an indication of the flood probability. *Flood risk zoning � delineation of areas with different possibilities and limitations for investments, based on flood hazard maps. *Floodplain � part of alluvial plain that would be naturally flooded in the absence of engineered interventions. Flood-prone area � area which is potentially threatened by floods. *Hazard � A physical event, phenomenon or human activity with the potential to result in harm. A hazard does not necessarily lead to harm. Learning � learning is defined as process of drawing inferences from experiences of self or others. Learning leads to a knowledge change of individuals, groups, organizations, and networks about cause-effect and evaluative relationships. Linear process model of strategy � strict sequence of analysing, evaluating, deciding, implementing, controlling, and learning with regard to deploying measures. *Risk reduction � The reduction of the likelihood of harm, by either reduction in the probability of a flood occurring or a reduction of vulnerability of the receptors. *Scale � Difference in spatial extent or over time or in magnitude; critical determinant of vulnerability, resilience etc. *Scenario � A plausible description of how the future may develop, based on a coherent and internally consistent set of assumptions about key relationships and driving forces (e.g., rate of technology changes, prices). Scenarios are neither predictions nor forecasts. The results of scenarios (unlike forecasts) depend on the boundary conditions of the scenario. *Social learning � process through which the stakeholders learn from each other how to better manage the system in question. Spatial development � changes in the distribution of activities in space and the linkages between them through the conversion of land and property uses.



*Spatial planning � public policy and actions intended to influence the distribution of activities in space and the linkages between them. It will operate at EU, national and local levels and embraces land use planning and regional policy. Strategic alternative � combination of structural and/or non-structural measures. Strategic decisions � subset of organisational decisions that influence the overall welfare of an organisation. In contrast to tactical decisions, strategic decisions cannot be easily undone. They have a strong influence on the allocation of resources, responsibility, power and career advancements of employees. Strategic decisions happen inside and outside the strategic planning process. Planning mode � process patterns that determines how future conditions of strategy making are taken into account. A distinction is drawn between three modes: Programming, scenario-based planning, preparedness strategies. Strategic project � a project that is intentionally formulated and implemented with regard to strategic aims and targets. Its implementation heavily relies upon informal as well as formal co-operation between local planning authorities and water authorities within catchments. Strategic planning � process of coordinating and legitimising strategic decisions within organisations and between organisations (dyad of two organisations, networks). Strategic planning is more or less formalised. Within medium- and large sized organisations strategic planning is often institutionalised as strategic planning system. Strategic planning system � set of formal organisational structures (e. g. strategic planning department, process regulations to channel the strategic planning process) and processes to coordinate and legitimise strategic decisions within medium- and large-sized organisations. The strategic planning system is closely related to other management systems (information system, human resource system, financial system). Strategic plan � document as the result of strategic planning processes. The primary function of strategic plans is to reduce variety within and between organisations and to serve as a frame of reference for orientation. Strategic plans can comprise analytic statements about the present situation and its causes, expectations about future development (possibilities, e.g. formulated as scenarios), and statements about aims, specific targets, and measures. Strategic spatial planning � social process through which actors in diverse institutional relations and positions come together to design plan-making processes and develop contents for the management of spatial change. Strategic spatial planning with its emphasis on informal co-operation is distinct from strategic planning as formal procedure to produce, co-ordinate, and legitimise fundamental decisions. *Susceptibility � the propensity of a particular receptor to experience harm. *Vulnerability � is defined as the potential of a system to be harmed. This potential can be defined as the product of susceptibility and value.

Strategies for Pre-Flood Risk Management - FLOODsite

Documents

Transcript of Strategies for Pre-Flood Risk Management - FLOODsite