Download - Interactive geodesign tools to support regional adaptation planning · Fig. 1. Board game Settlers of Catan (©Catan GmbH, 2015) Similar to the tasks in the game, the challenges for

208030-L-bw-Eikelboom208030-L-bw-Eikelboom208030-L-bw-Eikelboom208030-L-bw-Eikelboom

Interactive geodesign tools

to support regional adaptation planning

Tessa Eikelboom


Leescommissie: prof. dr. C. Steinitz prof. dr. J.C.J.H. Aerts prof. dr. P. Nijkamp prof. ir. N.D. van Egmond prof. dr. S.C.M. Geertman ‘Interactive geodesign tools to support regional adaptation planning’ PhD thesis, VU University Amsterdam ‘Interactieve beleidsinstrumenten om de ontwikkeling van regionale adaptatie plannen te ondersteunen’ Proefschrift, Vrije Universiteit Amsterdam © Tessa Eikelboom, Amsterdam, April 2015 ISBN: 978-94-6259-687-0 Cover image: Bdenk ontwerpbureau Printed by Ipskamp Drukkers, Enschede

The research on which this thesis is based has been carried out at the Institute for Environmental Studies (IVM) and the faculty of economics and business administrations (FEWEB), VU University Amsterdam. This research has been performed within the framework of Knowledge for Climate programme.


VRIJE UNIVERSITEIT

Interactive geodesign tools to support regional adaptation planning

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan

de Vrije Universiteit Amsterdam,

op gezag van de rector magnificus

prof.dr. F.A. van der Duyn Schouten,

in het openbaar te verdedigen

ten overstaan van de promotiecommissie

van de Faculteit der Aard- en Levenswetenschappen

op woensdag 24 juni 2015 om 9.45 uur

in de aula van de universiteit

De Boelelaan 1105

door

Tessa Eikelboom

geboren te Waalwijk


promotoren: prof. dr. H.J. Scholten prof. dr. P.H. Verburg copromotor: dr. R. Janssen


TABLE OF CONTENTS

CHAPTER 1 Introduction 7

CHAPTER 2 An assessment of geo-information in European regional adaptation strategies 17

CHAPTER 3 Interactive spatial tools for the design of regional adaptation strategies 33

CHAPTER 4 Comparison of geodesign tools to communicate stakeholder values 47

CHAPTER 5 Collaborative use of geodesign tools to support decision making on adaptation to climate change 67

CHAPTER 6 A spatial optimization algorithm for geodesign 87

CHAPTER 7 Spatial analysis of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change, and adaptation 107

CHAPTER 8 Using geodesign to develop a spatial adaption strategy for Friesland. 125

CHAPTER 9 Synthesis and conclusions 139

Summary 145

Samenvatting 151

Bibliography 155

Dankwoord 163

About the author 165


CHAPTER 1

6


Introduction

7

CHAPTER 1 Introduction

1.1 Spatial information and decision making Adaptation is a way to respond to a changing environment, as mitigation measures alone are not sufficient to compensate for the expected climate change (IPCC, 2007). Adapting to climate change involves multiple stakeholders, ranging from individuals, firms and civil society to public bodies and governments at local, regional and national scales, and international agencies. Climate change is only one of multiple factors considered in current decision making. To decide how to incorporate climate adaptation in spatial planning, access to relevant spatial information is needed (Davoudi, 2013). In an increasingly connected environment an immense amount of information is available. Additionally, to ensure transparent and flexible decision making the engagement of stakeholders is increasingly embedded into policy (Reed, 2008). Developing an adaptation plan can be seen as a multiplayer game in which different targets need to be maximized, but it is unknown which changes lead to the best situation for all targets. For example, the board game ‘The Settlers of Catan’ (©Catan GmbH, 2015) is about collecting resources to build settlements and collect 10 victory points (Figure 1). Players can use different strategies to win and are able to trade resources. The allocation of settlements depends on the spatial configuration of the ‘land-use’ tiles in relation to their personal strategy, but also relies on the interaction with other players and their strategy. Each player has to adapt his or her strategy according to the revenues.

Fig. 1. Board game ‘Settlers of Catan’ (©Catan GmbH, 2015)

Similar to the tasks in the game, the challenges for decision makers can be summarized as the need to condense the available data to provide essential information and support collaboration between multiple stakeholders. In the game, no support is provided on the effects of decisions, as this would accelerate the game, and make the game predictable. In practice, however, decision makers can benefit from spatial information and decision support systems. They can be used to evaluate multiple objectives and in this way offer surplus value in plans that meet the needs of stakeholders, and incorporate both scientific and local knowledge (McCall and Dunn, 2012). In contrast to the game, there are methods to support spatial decision making. One of these methods is called ‘geodesign’ and is defined as an iterative design and planning method, whereby an emerging solution is influenced by (scientific) geospatial knowledge derived from geospatial technologies (Lee et


CHAPTER 1

8

al., 2014). Geodesign tools integrate the exploration of ideas with direct evaluation at the same moment. This thesis is entitled: ‘Interactive geodesign tools to support regional adaptation strategies’, where ‘interactive’ refers to tools that can both be used by multiple stakeholders and provide immediate feedback. To interact with spatial information and receive immediate feedback, an interactive communication device was used, the ‘Touch Table’ (Figure 2). Furthermore, the geodesign tools were focused on supporting the development of regional adaptation strategies by offering an interactive platform to the stakeholders. These regional adaptation strategies are plans that consist of feasible measures to shift a region towards a system that is flexible and robust with respect to future climate changes. Integrating geodesign in an interactive way for use in the development of regional adaptation strategies introduces a new methodology in participatory decision making which supports the exchange of information between stakeholders. The objective of the research described in this thesis is to develop and evaluate a new set of geodesign tools that support the development of regional adaptation planning.

Fig. 2. Workshop session

1.2 Spatial planning and climate adaptation Climate change and sustainable development are presenting decision makers with unprecedented challenges, such as interactions at spatial scales, uncertainty, dynamic changes, and contrasting views. These challenges make climate adaptation a complex task from both an information processing and a process point of view. There seems to be little progress towards sustainable development (van Egmond, 2010). As early as in 1987, the Brundtland Report called for changing consumption and production to a


Introduction

9

system that preserves resources for future generations. However, many environmental problems are still preventing the realization of a sustainable society. These problems include the large amounts of waste, water and air pollution, increasing CO2 emissions, loss of biodiversity, plastic waste in our oceans, the unknown long-term effects of chemicals, overgrazing, nutrient pollution, habitat fragmentation, urban heat islands, flooding due to large expanses of paved surfaces in urban areas. Solving these complex issues requires the integration of different disciplines to provide systems where multiple views and objectives are united. Designing for societal and environmental change is not a solitary activity, rather it is a collaborative endeavour. The current threefold crisis (economic, ecological and social) needs simultaneously action, requiring interaction between economists, ecologists and sociologists. Moreover, it is a learning process in which actors from various backgrounds can gain a better understanding of the role of environmental values in a planning problem. Sometimes climate change can be the sole reason for decision making, but more often climate change is only one of the factors that influence planning. Several levels of governance are involved in initiating the development of regional adaptation strategies. Therefore, regions have a significant role to play in adaptation to climate change (Roggema, 2009; Adger et al., 2007). Regional adaptation strategies are plans that describe how to respond to climate change through adjustments in natural or human systems, in order to reduce vulnerability or enhance resilience in response to observed or expected changes in climate and associated extreme weather events. These strategies include anticipatory and reactive, private and public, and autonomous and planned adaptation actions. In practice, adaptations tend to be ongoing processes that preferably are flexible for changing circumstances. The demands of adaptation to climate change vary from place to place. As a result, spatial information plays a key role in the design of adaptation measures, as both the effects of climate change and many adaptation measures have spatial impacts. Spatial planning has an important role in responding to the need to address both the causes of climate change and the impacts of climate change (Wilson and McDaniels, 2007). Also, spatial planning should engage the public and other stakeholders to find consensual approaches to change. Spatial planning is defined as those policy interventions that aim to steer spatial developments in such a way that societal and environmental conditions will improve, whilst also meeting other objectives related to, for example, economic development, water management, and biodiversity conservation. Integration of these often conflicting objectives calls for innovative approaches that provide stakeholders with insight into the planning context and developments. Spatial analysis can provide the required information on past, current, and projected spatial developments, as well as indicate the impact of existing and proposed policy measures. In the context of spatial planning for climate change, plans for new development need to be seen in relation to existing development. Spatial planning studies need to keep in mind that the human world is the result of the combination of religion, science, physical safety, art, spirituality and the environment. Therefore, rationality and inner feelings are both involved when taking decisions. It is important to not just build links between assessment processes and decision making, but also to consider plan implementation and the development processes (Wilson and Piper, 2010).

1.3 From GIS to Geodesign Geo-information-based tools can make spatial information available for stakeholders. Technological developments have changed spatial decision support systems and planning support systems (SDSS and PSS). Initially, land use planners presented their information on large hard copies of maps and used sheets of tracing paper to add stakeholder information to the map (Burrough et al., 1998). In the following years, with the arrival of Geographical Information Systems (GIS), the transparent tracing paper map sheets were replaced by map layers presented within a GIS on a computer screen (Longley et


CHAPTER 1

10

al., 2005), and spatial analyses were facilitated. In practice, however, these tools and systems can fall short in efficiently supporting decision making (Geertman and Stillwell, 2003; Te Brömmelstroet, 2010; Vonk and Geertman, 2008). On the contrary, the advent of methods to host and manage large geo-spatial databases, improvements in computer processing performances and the development of intuitive natural user-interfaces now offer new opportunities for geo-analytical tools to support the design professionals in real time (Dias et al., 2013). Tool development has proceeded towards interactive map interfaces with direct interaction between participants and information (Goosen et al., 2013). Moreover, Dias et al. (2013) felt that the widespread use of the ‘‘Natural’’ user-interface was a decisive development in the acceptability of planning support systems. Recently, several studies have applied natural user-interfaces for direct exploration of geo-information and spatial planning with reported success in the users’ appreciation (Arciniegas et al., 2011; Eikelboom and Janssen, 2013). Technological developments allow for participatory workshops with large screens (Salter et al., 2009) and 3D visualizations (Fisher-Gewirtzman, 2012; Grêt-Regamey et al., 2013). The availability of large touch screens initiated the development of interactive applications to be used for planning and research (Arciniegas and Janssen, 2009; Pelzer et al., 2013). These applications were used in combination with an interactive mapping device, called the ‘Touch Table’. This communication device can be used to organize participatory workshops. One crucial advantage of this type of device is that it can bring stakeholders together to discuss a specific planning problem, supported by one or multiple maps where the stakeholders can interact with the maps themselves. Integrating natural user-interfaces with numerical analysis has resulted in a new approach, called ‘Geodesign’. Using a combination of previous definitions, Lee et al. (2014) defined geodesign as an iterative design and planning method whereby an emerging solution is influenced by (scientific) geospatial knowledge derived from geospatial technologies. In contrast to traditional planning processes, where analysis, design and evaluation are executed in separate steps, geodesign integrates the exploration of ideas with direct evaluation at the same moment (Lee et al., 2014). Geodesign is a collaborative activity where each participant contributes from his/her own profession and identity. It offers methods to connect people from different disciplines in order to tackle complex social and environmental problems. Furthermore, geodesign supports the creation of alternative scenarios that are informed by their geographical context, both science and value-based, providing an important function to the up-front pre-planning required by landscape architects and planners alike (McElvaney, 2012). One of the great advantages of geodesign is the capability to examine the effects of a proposed design across multiple interdependent systems. Steinitz (2012) provides a geodesign framework that shows strategies for thinking about geodesign and is about integrating knowledge from the people of the place, information technologists, and scientist. Impacts resulting from the design can be examined by means of geospatial technology (simulations, modelling, visualization, and the communication of design impacts), and be immediately fed back into the evolution of a design. In the community of geo-information and GIS professions, the term ‘geodesign’ has been well received. Recently, multiple geodesign studies have been published (La Rosa, 2014; McClintock, 2013; Campagna and Matta, 2014). McElvaney (2012) also demonstrates multiple examples of the interactive use of geo-information, including stakeholder mapping and criteria selection.


Introduction

11

1.4 Collaborative planning In an increasingly connected environment an immense amount of information can be collected. More data are shared compared with earlier years, and at the same time citizens are becoming increasingly engaged in producing, altering, and sharing this information. To condense these large amounts of spatial data down to relevant information is a difficult task. In addition, the complexity of planning issues requires that decision making is not a single person’s endeavour, but multiple stakeholders have to be engaged to obtain all relevant information, including environmental scientific data, but also data on economic, social, and other benefits. The engagement of stakeholders from various disciplines implies different aims and priorities. Moreover, the exchange of information between scientists and practitioners can introduce difficulties, and therefore can benefit from support to guide the decision making process. Therefore, the exchange of scientific information to provide valuable knowledge for stakeholders imposes multiple requirements on geodesign tools. As stated by Steinitz (2012), tools should serve as short cuts for handling information. Tools and techniques that are useful are ones that help to either eliminate options or select from among them. Tools must provide essential impact assessment as feedback to evaluate proposals. However, a discrepancy exists between how scientific knowledge is offered and the variety of criteria used in decision making. The scientific impact can be increased if scientists not only calibrate their tools on the basis of scientific evidence, but also tune their methods to problem solving in practical cases (Opdam, 2010). There is a need for insight into which forms of scientific knowledge are most effective, taking into account the demands of the actors and different phases in planning (Beunen and Opdam, 2011). The validity of general scientific relationships is limited at the local level. Important elements that influence the effectiveness of knowledge transfer are interactivity, feedback, spatial evaluation, and visualization. Interactivity and feedback Pouwels et al. (2011) stated that tools should be built on interactions between functions, encourage interaction, allow the incorporation of local knowledge, and generate output in the form of a map that shows where areas of conflict and opportunities are located. Interactivity offers the capacity to interact directly with spatial information. The added value of interactivity has already been recognized in many studies (Andrienko et al., 2002; Bacic et al., 2006; Goosen et al., 2007). The next level of the interactive use of spatial information is the inclusion of real-time feedback. Feedback can be derived from the relevant relationships of a system, from which it is possible to simulate its future state. A common form of simulation (or modelling) is applied in impact assessments that describe the possible consequences of a specific policy. The complexity of interactive models should be based on a fit-for-purpose principle. In some situations complex models are required, but in many cases a simple or simplified model is more suitable. Simple models are perceived less as a ‘black box’, as the underlying relations are easier to explain, but more important are the short response times required for interactive use. Spatial evaluation Spatial multi-criteria analysis (MCA) is a common approach in decision making to make the content of spatial data sets comparable. This valuation method allows for the subsequent ranking of alternative outcomes (see, for example, Van Herwijnen, 1999). MCA often requires spatial aggregation of data, which combines individual values in a single value for a specified location with the use of weights. This loss of data is compensated by the delivery of a comprehensive summary of the original content. Aggregation can be applied to both raster and vector data to generalize the outcomes of analyses of highly detailed data sets to a level at which they are more easily visualized and interpreted.


CHAPTER 1

12

The most straightforward use of spatial MCA is the comparison and ranking of alternatives. Spatial multi-criteria analysis can be used to link policy priorities to rankings but also to provide insight into the spatial distribution of the performance of the alternatives. Spatial multi-criteria methods can also be used in combination with multi-objective linear programming to create optimized alternatives (e.g. Aerts et al., 2003; Janssen et al., 2008; Santé-Riveira et al., 2008). As a special case of design methods, interactive optimization offers solutions to the planner in a number of steps where, after each step, the planner can change the conditions for optimization (e.g. Stewart et al., 2004; Janssen et al., 2008). Visualization The success of interaction with maps depends heavily on how well the information in the map is understood. This requires maps that are easy to understand. The interpretation of maps depends on multiple factors, such as the amount of information covered by the map, level of detail, scale, map visualization and the knowledge level of the map user (Carton, 2007; Janssen et al., 2008). Geo-visualizations are modelled representations of reality, based on spatial information. These visualizations can be difficult to understand for a considerable portion of stakeholders. As Steinitz (2012) stated: ‘It is not self-evident that, when the information is put in a map, this visualization is also understood by the viewer.’ Participatory planning requires communication tools that correspond to the stakeholders’ perception of their environment, and make use of their knowledge and experience. In addition, the spatial information needs to be adjusted to the planning context as the information exchange takes place between diverse combinations of stakeholders and multiple disciplines (Al-Kodmany, 2002). In daily life, multiple maps that use colour coding to communicate spatial information can be found. For instance in weather warning systems, such as forecasts of the risk of avalanches in Tyrol, Austria (Figure 3) (https://lawine.tirol.gv.at/). The classification of danger levels depends on multiple factors, such as weather conditions, snowpack conditions, and steepness. In addition, mountain regions also face other natural hazard risks such as debriflows, flooding, erosion, mud streams, rock falls, mass movements, and landslides. Multi-hazard maps could support land-use planning by visualizing a combination of effects.

Fig. 3. Classification of the risk of avalanches: red is highly dangerous; orange is high risk; yellow is low risk; and green is no risk (www.lawinen.at, February 12th, 2015)


Introduction

13

1.5 Research questions and approach This thesis contributes to improving our understanding of the use of geodesign tools for collaborative planning at a regional scale, and provides tools to support decision making with multiple stakeholders and objectives. The main aim of this research is to develop and evaluate geodesign tools that support the development of regional adaptation strategies. In order to achieve this goal, the following research questions were studied:

1. How is spatial information used in regional adaptation strategies? It is relevant to know how spatial information is used in regional adaptation (Chapter 2).

2. How can adaptation planning tasks be linked to spatial decision support tools? To develop tools that suit adaptation planning, it is necessary to identify which tasks need support (Chapter 3).

3. What is the performance of a new set of geodesign tools? The evaluation of new tools can provide information on the user-performance and -perception of these tools to support specific tasks. The ‘performance’ is defined as whether the user correctly interpreted the information visualized by the tools (Chapter 4).

4. How are these geodesign tools used in collaborative planning? Individual use of the tools does not indicate how the tools function when used simultaneously by multiple stakeholder (Chapter 5).

5. What is the potential for integrating an optimization algorithm in a geodesign tool? The design of different strategies depends on multiple objectives and constraints. An optimization model can support fast generation of alternatives using these constraints (Chapter 6).

6. How can expert knowledge be integrated in a geodesign tool? Geodesign can be used to convert complex scientific knowledge to information that is easy to interpret and valuable for local stakeholders (Chapter 7).

7. What can be learned from the application of the tools in adaptation planning? In order to draw conclusions on a new set of tools extensive experience with the use of such tools in practice is essential (Chapter 8).

The developed geodesign tools were made directly available for practice, in order to evaluate the tools and learn from stakeholders. In total, 13 workshops (from 2010-2014) were organized in which feedback from scientists, regional planners, government officials and local farmers was gathered. This extensive feedback was used for the further development of the tools and to draw conclusions on their usability. An interactive mapping device, called the ‘Touch Table’, was used to support collaborative planning workshops. The ‘Touch Table’ was used in a series of workshops with various stakeholders to generate, assess, and discuss adaptation strategies for peat meadow areas in the Netherlands. The advantage of workshops is that they allow central issues, options and choices to be quickly identified. A workshop requires shared knowledge of the subject, shared assumptions and a shared language (Steinitz, 2012). These conditions were met in the maps that visualize three particular objectives as traffic lights.


CHAPTER 1

14

1.6 Limitations It is important to conduct studies that assess the ‘added value’ of a decision or planning support system (PSS). Added value was defined by Pelzer et al. (2014) as: ‘a positive improvement of planning practice, in comparison to a situation in which no PSS is applied’. In practice, however, there is no such situation in which the same planning process takes place both with and without a support tool. A combination of experiments and application of new tools in planning practice form a useful approach in evaluating the usability of new spatial planning or decision support tools. The developed geodesign tools were made directly available for practice in order to evaluate the tools and learn from stakeholders. Furthermore, evaluation of tools as part of the planning process directly implies that no ‘with and without’ test could be performed. In addition, decision makers had only little time available to participate. As a result, statistical analyses were performed with care as the number of respondents was relatively small. After each workshop, feedback from the workshop was used to improve consecutive workshops. Therefore, each workshop is less comparable to previous workshops. This same approach was used for the participant surveys. After each survey, the questions were changed slightly to improve the quality of the survey. Although the comparability of the surveys was ensured by this methodology, statements were included in each of the surveys to collect user perceptions, thus also gaining insight into the perceived difficulty of the tools. Because the tools were primarily developed for research purposes, limited resources were available for user-friendliness, including, for instance, larger icons. Therefore, user-friendliness was limited, comparisons between tools were primarily focused on how the methodology behind the tools supported the tasks that were given to the users. Before the methodology can be used in new study areas, a structured GIS database is required, as well as information to perform a MCA. This information is often not directly available, as it originates from different sources. Efforts should therefore be made to implement the methodology proposed in this thesis. In addition, the tools were evaluated for peat meadow areas in the Netherlands of around 50 km2. Extrapolation to other or larger regions has not yet been done, but is needed in order to further evaluate the applicability of this methodology to other scales, objectives, or land uses.

1.7 Thesis outline This thesis consists of nine chapters that cover four main steps: 1) identifying the need for geodesign tools; 2) tool development; 3) tool evaluation; and 4) tool application. A flowchart of the thesis outline is provided in Figure 4. Chapter 2 describes the use of geo-information in regional adaptation strategies based on the evaluation matrix from Preston et al. (2011). Chapter 3 illustrates what types of spatial decision support tools can support which tasks of adaptation planning. This chapter shows how the link between tasks, derived from the actual development stages of adaptation, and spatial decision support tools. This link helps to decide what spatial tools are suited to support which stages in the development process of regional adaptation strategies. The practical implication of these links is illustrated for three case studies in the Netherlands. Chapters 4 and 5 describe the systematic development and evaluation of four types of geodesign tools. Chapter 4 compares the effectiveness of geodesign tools by quantifying tool performance. Performance is considered high if a user is able to complete an assignment correctly using the information presented. Four geodesign tools were developed that differ in the use of an aggregation or ranking step. The tools were tested in an online survey to assess their ability to communicate information effectively. Chapter 5 describes the application of the same tools in experimental workshops. Parallel sessions were organized to derive lessons for tool design from observations made when users interacted with the four geodesign


Introduction

15

tools. The quantity and quality of the measures were evaluated to generate insight into the differences in use between planners and researchers for different adaptation planning tasks. Chapter 6 shows how a genetic optimization algorithm can be implemented in a geodesign tool, and demonstrates the ability to generate non-dominated alternatives. This can be single objective alternatives, compromise alternatives, alternatives within set constraints, or stakeholder defined decision variables. The tool can also be used interactively by combining tool output with stakeholder feedback to stepwise generate an alternative in a number of iterations. This study identifies opportunities and limitations of optimization for interactive collaborative spatial planning. Chapter 7 describes how expert knowledge on peat soil subsidence has been implemented in a geodesign tool. Three contrasting study areas were selected for discussions among stakeholders and regional authorities on the current and future challenges, and on the effects of possible adaptation measures. During the workshops, data on current peat soil thickness, land use and water tables were shown at the scale of individual parcels, and the results of soil subsidence calculations were presented for both the current situation and for a situation with climate change. Chapter 8 describes the practical application of a geodesign tool for the development of a future plan for peat meadow areas in Friesland. The Province and Water Board of Friesland have decided to develop a long-term adaptation strategy for the Frisian peat meadow area. A planning process with all the stakeholders involved has already been started to develop this strategy. In a workshop setting, the stakeholders were asked to design spatial plans for the region based on three different strategies. Finally, Chapter 9 summarizes the main findings of this thesis, and discusses the outcomes in the context of the original research aim and of the latest scientific developments.

Figure 4. Thesis outline from the identification of tasks and tools, tool development and evaluation, to tool application.


CHAPTER 1

16


An assessment of geo-information in European regional adaptation strategies

17

CHAPTER 2 An assessment of geo-information in European regional adaptation strategies

Abstract Different levels of government initiate the development of strategies to mitigate and adapt to climate change. Climate change is affecting decision making and shaping its priorities as an additional factor in the often already complex domain of environmental governance. Spatial planning has a key role in addressing the causes and impacts of climate change. This study was the first to review 25 European regional adaptation strategies in order to explore relations between their quality and the use of geo-information. The quality of the studies was assessed with use of the evaluation matrix from Preston et al. (2011). The results show a wide variety in the quality of the strategies, and in the type of geo-information included in these strategies. Most of the included strategies turned out to be climate change effect and climate change impact studies, with little information on adaptation options. None of the studied strategies included the appraisal of adaptation options. For many strategies the inclusion of maps increased the quality, however high quality strategies also exist with only a few maps. The engagement of stakeholders and inclusion of decision support tools were found to increase the overall quality of strategies. The expectation is that the need to make trade-offs on where to apply which adaptation option can benefit from the use of spatial information and detailed descriptions of options. Spatial decision support tools have much potential to support the appraisal of options, as the spatial dimension of climate change can serve as an integrating factor for different sectors and multiple stakeholders in adaptation planning. The findings of this study intend to contribute to the ongoing discussion on good practices for adaptation planning, exchange of knowledge and experiences, and results of adaptation practices between countries.

Submitted: Eikelboom T. and R. Lasage (2015) An assessment of geo-information in regional adaptation strategies, Regional environmental change.


CHAPTER 2

18

2.1. Introduction Climate change and climate change adaptation are affecting the domains of policy and decision making (Massey and Huitema, 2013). They are additional factors in the often already complex domain of environmental governance. Climate change adaptation is the process of adjustment to actual or expected climate and its effects (IPCC, 2014). The European Commission (EC) has published a white paper on climate change adaptation (CEC, 2009) and since 2010 the EC has a Directorate General for Climate Action with a unit solely devoted to adaptation (Massey and Huitema, 2013). In 2013 the EC has published the European Adaptation Strategy that requests all EU members to develop a national adaptation strategy (COM 216, 2013). This European Adaptation Strategy main aim is to contribute to a more climate resilient Europe. It states that it is cheaper to take early, planned adaptation action than to pay the price of not adapting. This should be done by prioritizing coherent, flexible and participatory approaches. It also recognizes that due to the specific and wide ranging nature of climate change impacts on the EU territory, adaptation measures need to be taken at all levels, from local to regional to national levels. As there are different requirements for adaptation in different geographical areas and sectors, the involvement of stakeholders is advocated as this is said to ensure that knowledge on and priorities in that specific area will be taken into account (Turner et al., 2003; Schröter et al., 2005). The EC has identified the exchange of knowledge and good practices as an important task for the coming years (COM 216, 2013). Adaptation to climate change is rarely the sole reason for a decision or action. Hence, adaptation measures are often part of broader planning and management, and seldom a response to climate change alone (Adger et al., 2007; Ford et al., 2011). Moreover, Preston et al. (2011) show that weaknesses in adaptation planning are often related to limited consideration of non-climatic factors and suggests the need to consider the broader governance context in adaptation planning. Over the past decade, 19 European countries have developed national adaptation strategies in a response to the expected impacts of climate change (Climate-adapt, 2014). These documents are developed at (supra) national level, and aim to give governments at regional and local level guidance on adaptation planning. The development of more concrete adaptation options, plans and strategies will take place at the sub-national and local level (Roggema, 2009; Adger et al., 2007). At this level the specific conditions, physical, socio-economic and governance, can be included in the plans, making them more sustainable. Therefore regions have a significant role to play in strategic planning and implementation of adaptation to climate change. In this study, regions are subnational, facilitate various functions and cover multiple governmental levels, including municipalities. National levels delegate the identification and implementation of adaptation to their regions. Already several regional adaptation strategies (RAS) have been developed within the EU (Prutsch et al., 2010; Climate-adapt, 2014) and it is expected that more of these strategies will be developed in the coming years in response to the European Adaptation Strategy, following the EU guideline for developing regional adaptation strategies (SWD 134, 2013). Even though the use of spatial planning processes in climate change adaptation has only recently started (Biesbroek et al., 2009; 2010), several different tools have already been developed and applied to support adaptation planning. Building on this experience, a range of climate adaptation guidelines and specific tools are available for the development of adaptation plans (UKCIP, 2005; IISD, 2006; Shaw et al., 2007). Moreover, multiple tools exist that explicitly aim to support the adaptation development process (Ceccato et al., 2011; Gidhagen et al., 2013; Giupponi et al., 2013; Pettit et al., 2013; Wenkel et al., 2013). In addition, multiple decision and planning support systems and geodesign tools (e.g. Geertman and Stillwell, 2009; Dias et al., 2013; Steinitz, 2012) exist that have the potential to contribute to the integration of different sectors and stakeholders (Bacic et al., 2006). These tools can help in



19

evaluating the individual effects of proposed adaptive measures, and in the evaluation of sets of adaptive measures. They also stimulate and support the involvement of stakeholders in the development of these strategies and have been used to consider the different factors in planning and policymaking (Eikelboom and Janssen, 2013). It is not known, however, if and how spatial tools have been applied in developing regional adaptation strategies. Several studies emphasize the importance of spatial planning and geo-information for adapting to climate change (Davoudi, 2013; Nevarra and van der Molen, 2014; Santosh et al., 2014). Adaptation measures often have a spatial consequence, either in design (e.g. peak storage area) or in effect (e.g. zoning of land use functions). Hence, geo-information is very relevant for developing adaptation strategies (Davoudi, 2013; Wilson and Piper, 2010). Regional adaptation strategies can benefit from the use of geo-information as a place-oriented perspective, by integrating different views and objectives based on their spatial configuration. This study is the first to review regional adaptation strategies from a geo-information perspective. The aim of this study is to explore if there is a relation between the quality of regional adaptation strategies and the use of geo-data and decision support tools. For the analysis a comprehensive database was constructed of the most recent regional adaptation strategies from the UK, Germany, Luxemburg, the Netherlands, Belgium and Switzerland, which contains information on different criteria that are used to establish their quality. These are the inclusion of climate change effects, impacts, options and appraisal of options. In addition, the engagements of stakeholders, their use of maps, referring to decision support tools and the relation between these criteria were assessed. These criteria were based on the evaluation framework of Preston et al. (2011) and the adaptation framework of Willows and Connell (2003). To this database additional information was added on other characteristics, like the year of publication, area size and report length. The database is used to explore if there are relations between the quality of a strategy and other factors. The findings of this study intend to contribute to the ongoing discussion on good practices for adaptation planning and exchange of knowledge, as has been identified as an important topic by the EC (COM 216, 2013).


CHAPTER 2

20

2.2. Data and Methods In this study, Smit and Wandel’s (2006) definition of adaptation strategies in changing conditions was applied to the narrower field of climate change adaptation strategies on a regional scale. A regional adaptation strategy is defined as the combination of possible adaptation options that help a region to develop from a current state to one that better manages, adjusts to, or copes with climate change (Eikelboom and Janssen, 2013). As defined by the EC (SWD 134, 2013) adaptation can range from actions that build adaptive capacity to concrete adaptation options. Similarly, strategies range from an incentive for awareness raising to an approach on how to implement explicit adaptation options that are part of an integral plan.

2.2.1 Compilation of the database In order to evaluate the quality of the strategies (including the involvement of stakeholders) and to be able to explore their relation to the use of geo-information and other characteristics, a database was constructed that included information of the strategies: for example, the year it was developed, the total area it covers, etc. For establishing the quality of the strategies, an evaluation matrix was developed using criteria based on the adaptation development stages of Willows and Connell (2003) and the assessment of Preston et al. (2011). The latter has developed an evaluation methodology that combines evaluation theory with adaptation planning guidance as developed by the practitioner community. The six criteria to determine the quality of the adaptation strategies are: (1) articulation of objectives, goals and priorities; (2) assessment of climate change effects; (3) assessment of impacts, vulnerability, and risk, (4) options identification; (5) options appraisal; and (6) stakeholder engagement. These criteria cover the steps 1d to 4c of the EC guideline on developing adaptation strategies (SWD 134, 2013). Each adaptation plan was scored on a three-point scale (0, 1 or 2) on these 6 criteria, and this information was included in the database. Table 1 gives a description of the criteria and how the score was derived. The articulation of objectives, goals and priorities was included in the assessment because it reveals whether it is clear to the authors what the adaptation strategy is about, and whether it is considered how adaptation interacts with other issues in the locality. The assessment of climate change effects and impacts present the available knowledge on climate change, the severity for the region, and is important to identify which sectors are affected, and with which severity. When this information is available on the regional scale it indicates a good quality RAS. If this information is lacking, the quality of the RAS will be lower. In addition, effects and impacts vary spatially and therefore this criterion is an indicator of the presence of spatial information in the reports. The next criterion is whether possible adaptation options to deal with climate change impacts are listed and whether these are described in detail. For the assessment of the relation between the number of spatial options in a strategy and the use of spatial support tools, it was also assessed how many of the listed adaptation options of the strategy can be categorized as spatial or non-spatial. Spatial options are those that either require land change in their implementation, or have spatial effects after implementation. Only detailed descriptions of adaptation options that were considered to be implemented were counted. Options that already had been implemented were not included in the database. For a score of 1 a strategy should include a list of options and for a score of 2 it should also include a list of options including detailed descriptions of the options. The next criterion draws on the assumption that appraisal of the identified adaptation options increases the quality of a strategy. In addition to assessing the adaptation development stages and the way they have been implemented, an assessment of the quality of stakeholder involvement was performed as these are important in the development of good strategies (COM 216, 2013; Schröter et al., 2005). Different stakeholder types can be involved in the development of the strategy such as:



21

government organisations, experts, general public, local authorities, etc. If multiple parties have contributed to the development of the adaptation strategy, it is considered to be of better quality. And if multiple types of stakeholders, besides government organisations, have contributed to the strategy, it is considered to be of even better quality, and hence gets a score of 2 for this criterion. Table 1. Assessment matrix used for the evaluation of regional adaptation strategies

Criteria Score of ‘1’ requires evidence that… Score of ‘2’ requires evidence that…

Articulation of objectives, goals and priorities

The broad purpose of the adaptation strategy has been stated.

The broad purpose of the adaptation strategy has been stated, as well as some discussion of specific objectives, with some consideration of adaptation priorities. There is a clear vision about the goals of the adaptation strategy and how it supports wider goals or targets in that locality, objectives that allow progress towards the goals to be recognized.

Assessment of climate change effects

At least the national effects are included for at least one type of climate change effect ( T, P, and SLR).

Region specific effects are included and illustrated for at least one type of climate change effect.

Assessment of impacts, vulnerability and/or risk

Short description of impacts of climate change. Detailed descriptions, including quantifications, or climate impacts, vulnerabilities or risks.

Options identification Multiple alternative adaptation options are listed, or a few adaptation options are listed including a detailed description.

Multiple alternative adaptation options are considered and detailed descriptions are provided.

Options appraisal An options appraisal has been undertaken. Details of the appraisal framework that has been used to compare and select preferred options.

Stakeholder engagement Consultation and interaction with at least 5 government institutions have taken place during the development of the adaptation strategy.

Consultation and/or interaction with interested parties and/or general public has taken place during the development of the adaptation strategy, additional to government institutions.

For the comparison of the quality of the adaptation strategies with the presence of geo-spatial information, the number and types of maps as well as the use of decision support tools were included in the database. The maps were grouped according to whether they illustrate: (1) climate change effects; (2) climate change impacts; (3) adaptation options; or (4) assessment of adaptation options. Maps of climate change effects visualize primary effects such as the distribution of temperature change. These maps usually cover the total region. Impact maps are those showing secondary or tertiary effects such as urban heat islands, which combine information on the temperature in a city and the locations of vulnerable people (e.g. elderly homes, kinder-gardens, hospitals etc.). Maps that visualize options show where adaptation measures can be implemented. These types of maps can also focus on a specific area within the region. Assessment maps provide information on the consequences of options. The remainder of the maps that could not be classified in these categories were considered as general informative maps that indirectly relate to climate adaptation such as land use maps, topographical maps, or the distribution of inhabitants.

2.2.2 Selection of regional adaptation strategies The report ‘Design of guidelines for the elaboration of Regional Climate Change Adaptations Strategies’ (Ribeiro et al., 2009) was used as a starting database together with the climate-adapt website (climate adapt, 2014). From this database only regional studies that focused on adaptation were selected. Additional strategies were collected from an inventory of secondary information and grey literature. For each region the most recent online available adaptation strategy document was used. The collected adaptation strategies were for practical reasons limited to those that were available in Dutch, English, or German. The list of included adaptation strategies and their characteristics is provided in Table 2. Next to the strategies included in this study, reports were also found that provide detailed information for individual sectors, adaptation plans for smaller areas within the region or other follow up documents.


CHAPTER 2

22

Although containing valuable information for regional adaptation, these reports were not included in the analysis as the purpose of these documents is not to provide a regional adaptation strategy. If strategies explicitly referred to underlying studies, reports and websites on effects, impact and vulnerability, these underlying reports were included in the analysis as if part of the strategy.

2.2.3 Data Analysis The quality score of the strategies was compared with multiple characteristics of the strategies with use of linear regressions and two sided t-tests. Not only total quality, but also the individual criteria were compared. First, the quality was compared with the number and type of maps. Next, quality was compared with the number of spatial and non-spatial adaptation options. In addition, quality was compared with referring to decision support tools. Also, it was checked whether quality differed between publication year, countries, and area size of the regions. Table 2. List of regional adaptation strategies

ID Regional adaptation strategies Year Region Country

Area (km2)

Website

1 Vlaams Adaptatieplan 2013 Vlaanderen BE 13682 www.lne.be/themas/klimaatverandering/klimaattips/klimaattips/wat-doet-de-vlaamse-overheid/vlaams-klimaatbeleidsplan

2 Risiken und Chancen des Klimawandels im Kanton Aargau

2013 Kanton Aargau CH 1404 www.ag.ch/de/bvu/umwelt_natur_landschaft/naturschutz/nachhaltigkeit_2/klimabericht_2010/klimabericht_1.jsp

3 Adaptationsstrategie Klimawandel Kanton Bern Grundlagenbericht

2010 Kanton Bern CH 5959 www.umwelt.nrw.de/klima/klimawandel/publikationen

4 Klimabericht Kanton Graubünden 2012 2012 Kanton Graubünden CH 7105 www.gr.ch/DE/institutionen/verwaltung/ekud/anu/aktuelles/mitteilungen/Seiten/20140902_Klimawandel.aspx

5 Bericht Klimaadaptation 2011 Kanton Schaffhausen CH 298 www.interkantlab.ch/index.php?id=655&L=4

6 Umgang mit dem Klimawandel Klimastrategie des Kantons Uri

2011 Kanton Uri CH 1077 www.ur.ch/de/verwaltung/dienstleistungen/?dienst_id=3655&highlight=klimastrategie

7 Auswirkungen des Klimawandels und mögliche Anpassungsstrategien

2007 Kanton Zürich CH 1729 www.bafu.admin.ch

8 KLARA 2005 Federal State of Baden-Württemberg

DL 35752 www.lubw.baden-wuerttemberg.de

9 Bayerische Klima-Anpassungsstrategie (BayKLAS)

2009 Federal State of Bavaria DL 70552 www.regensburg.de/sixcms/media.php/121/broschuere_bayerische_klimaanpassungsstrategie.pdf

10 Integriertes Klimaschutzmanagement 2007 Federal State of Brandenburg

DL 29479 www.mlul.brandenburg.de/cms/media.php/lbm1.a.3310.de/mk_klima.pdf

11 Bericht zum Stand der Umsetzung der Maßnahmen des Aktionsplans Klima und Energie des Freistaates Sachsen

2008 Federal State of Saxony DL 18416 umwelt.sachsen.de

12 Anpassung an den Klimawandel - Eine Strategie fur Nordrhein-Westfalen

2009 North Rhine Westfalia DL 34088 www.umwelt.nrw.de/klima/klimawandel/publikationen



23

ID Regional adaptation strategies Year Region Country

Area (km2)

Website

13 Regionale Klimaanpassungsstrategie 2011 Oberlausitz-Niederschlesien

DL 4496 www.rpv-oberlausitz-niederschlesien.de/projekte/regionales-energie-und-klimaschutzkonzept-klimaanpassungsstrategie/regionale-klimaanpassungsstrategie/ergebnisse.html

14 Integriertes Regionales Klimaprogramm für die Region Dresden Grundlagen, Zielen und Maßnahmen Klimaanpassungsprogramm

2013 Region of City of Dresden

DL 328 www.regklam.de/klimaanpassungs-programm/

15 Adaptation to climate change- strategies for spatial planning in Luxemburg

2012 Luxemburg LU 2586 www.espon-usespon.eu/library,adaptation-to-climate-change-strategies-for-the-spatial-planning-in-luxembourg-c-change-changing-climate-changing-lives-2012

16 Regionale klimaat AdaptatieStrategie Haaglanden

2014 Haaglanden NL 404 http://haaglanden.nl/regionale-klimaatadaptatie-strategie

17 Developing ADAPTATION to CLIMATE CHANGE in the East of England Living with Climate Change in the East of England Stage 1 Report: Guidance on Spatial Issues

2011 2003

East of England UK 19120 www.sustainabilityeast.org.uk/adaptation http://dev.ukcip.org.uk/wordpress/wp-content/PDFs/EoE_tech.pdf

18 Managing Risks and increasing resilience – the mayor’s climate change adaptation strategy, London

2011 London UK 1569 www.london.gov.uk/priorities/environment/publications/managing-risks-and-increasing-resilience-the-mayor-s-climate

19 A climate change action plan for England’s Northwest 2010-2012

2008 North West England UK 8592 http://enviroeconomynorthwest.com/2012/03/25/northwest-climate-change-action-plan

20 Northern Ireland Climate Change Adaptation Programme A climate change risk assessment for Northern Ireland Preparing for a changing climate in Northern Ireland

2014

2012

2007

Northern Ireland UK 13843 www.doeni.gov.uk/index/protect_the_environment/climate_change/climate_change_adaptation_programme www.doeni.gov.uk/climate_change_risk_assessment_ni_2012.pdf

21 Climate Ready Scotland: Scottish Climate Change Adaptation Programme

2014 Scotland UK 78387 www.scotland.gov.uk/Publications/2014/05/4669/0

22 Climate change mitigation and adaption implementation plan for the draft south east plan

2006 South East England UK 19096 www.espace-project.org/publications/library/climate_change_implementation_plan-300306-v2.PDF

23 The south west climate change action plan 2008-2010 Warming to the idea

2008

2010

South West England UK 23829 www.southwest-ra.gov.uk/media/SWRA/Climate%20Change/Climate_Change_Action_Plan.pdf http://climatesouthwest.org/warming-to-the-idea

24 Climate Change Strategy for Wales 2010 Wales UK 20761 http://cymru.gov.uk/topics/environmentcountryside/climatechange/publications/strategy

25 Yorkshire and Humber regional adaptation study – Weathering the storm

2009 Yorkshire and Humber UK 15420 www.yourclimate.org/pages/regional-adaptation-study


CHAPTER 2

24

2.3. Results

In total 25 regional adaptation strategies have been included in this study. First the results of the analysis of their quality are presented, followed by the use of decision support tools and the use of geo-information. In section 2.3.3 correlations between quality and several characteristics of the strategies were explored.

2.3.1 Quality The quality of the regional adaptation strategies was assessed for six criteria, and the total scores could range from 0 (lowest quality) to 12 (highest quality). Figure 1 shows the scores of the strategies and the score for each criterion separately. The adaptation strategy of Saxony has the lowest quality of 2 and the strategies of Vlaanderen and Northern Ireland have the highest quality, with a score of 10. Most strategies have a good quality score for assessment of effects, impacts and the identification of options, however, options appraisal is missing in all reports. Strategies that score low on assessment of effects in most cases also have a low score for all the other criteria. For seven strategies stakeholder engagement was absent and for four strategies no impact assessment was included in the reports.

Fig. 1. Quality score of each criterion for the regional adaptation strategies



25

Table 3. Scores for individual evaluation criteria applied to adaptation plans. Scores represent the percentages of adaptation plans receiving specific scores (0, 1, or 2) for each criterion

Criteria 0 1 2

1 Articulation of objectives, goals and priorities 0% 76% 24% 2 Assessment of climate change effects 8% 24% 68% 3 Assessment of impacts 16% 20% 64% 4 Options identification 16% 36% 48% 5 Options appraisal 100% 0% 0% 6 Stakeholder engagement 28% 24% 48% Table 3 shows how the quality score varied for the different criteria. The articulation of objectives, goals and priorities ranged from creating a starting point, identifying effects, increasing knowledge and latest findings on consequences of climate change, setting a strategic direction, and primary identification of potential adaptation options. The underlying inducement of developing adaptation strategies was described in the benefits of a proactive approach, because as climate change progresses, the opportunities for successful adaptation diminish and the associated costs increase (e.g. Austria). Some regions also used the strategy to emphasize the need for further risk analyses (e.g. Kanton Uri). Furthermore, the inventory showed that regional adaptation strategies mostly started with information on projected future climate change. The IPCC graph with the expected temperature increase was often found in this part of the strategies. Most strategies continued with describing climate change impacts, which some strategies divided for sectors or themes. Not all strategies clearly distinguished between effects and impacts. Most of the strategies (84%) included adaptation options (figure 2a). The remaining 16% of the strategies did not include any adaptation option. Together the strategies contain a total of 1114 options, of which 20% contained spatially related options. Next, the number of options per strategy ranged from 0 to 149 with an average of 45 options including 9 spatial options. From these, the highest number of spatially related options, 32, was found in Dresden. In addition, the types of options vary widely in scale and content. Most options focused on additional risk assessment and awareness raising. Some strategies focused more on strategic action points and activities to implement climate policy (e.g. London, South east England, north east England). Among these are options as mainstreaming and communication adaptation, building resilience by providing evidence and equipping decision makers with tools and skills, integrating adaptation in regulation and public policy (Wales, Scotland) and awareness raising. Other strategies described options in response to sectorial or local impacts. As an example, Luxemburg described adaptation options for each impact (heat, drought, flood, high precipitation, and wind and storms). The strategies also varied in the emphasis on emission reduction options, which actually is mitigation and not adaptation. Four strategies included maps of options (Haaglanden, North Rhine Westfalia, Dresden, and Luxemburg). These strategies included between 18 and 30 maps in total and the number of options ranged from 17 to 74 of which 4 to 32 were spatial. A map of options explicitly suggests where a certain adaptation measure can be implemented. Maps of options were only found in strategies that also included identification of options and these strategies score above average on quality. Three of the four strategies (Dresden, Oberlausitz-Niederschlesien, Luxemburg, and Kanton Zürich) that contain a high number of spatial adaptation options also contain many maps. Except the strategy of Zürich, which covered 21 spatially related options without including any maps. The two reports that included no maps and no options (East of England, Kanton Uri) score low on quality.


CHAPTER 2

26

The majority of the strategies (72%) described the process of stakeholder involvement. Stakeholder involvement was mostly limited to higher level governmental policy makers and employees of research organizations. In several studies, tools were either used to support the development of adaptations strategies or were a product of the experience of developing an adaptation strategy. Wales mentions different stakeholders: local governments, private sector, third sector and communities, individuals. Scotland refers to tools and outlines stakeholders by defining sector roles and responsibilities. Ireland consulted key stakeholders by telephone consultation and workshops to explore principal areas of risk and potential approaches to adaptation. Several of the strategies are actually in in an early stage of adaptation planning (effect or impact studies) and hence have no high level of involvement of stakeholders.

2.3.2 Use of geo-spatial information The number and type of maps varied between the strategies. The average number of maps per strategy is 21 ranging between 0 and 132. Most strategies included effect maps such as temperature increase or changes in precipitation (Fig. 2). Heat and flood impact maps were common. Eight (32%) of the strategies did not include any maps, and 10 (40%) strategies contained web links that referred to online maps. Many maps did not include basic cartographic elements and are of low quality. The spatial resolution of maps also differs highly between and within strategies. The two strategies that included many maps are Baden-Württemberg and Kanton Zürich. However, these were only maps of climate effects and background information, and they had no maps on adaptation options.

Fig. 2. Percentage of maps for five different categories of adaptation maps

Figure 3 shows an example of which type of maps were typically found in the strategies per stage of the adaptation framework: effects, impacts, options, and assessment. Most maps either visualized climate change effects or provided background information such as climate regions, population density, and topographical orientation (Figure 3a). The strategy for North Rhine Westfalia visualized the change in precipitation with relatively low resolution (Figure 3a) and described potential impacts for different themes: forest, grassland, biodiversity, water management, tourism, health, built-up area, and industry. In addition, a division in regions was made as models show that climate change effects depend on spatial characteristcs (Suderbergland, Eifel, Westfalische Bucht and Westfalische Tiefland, Niederrheinische Bucht und Niederrheinisches Tiefland). The regional adaptation strategy of Haaglanden started with a vulnerability map (Figure 3b) that broadly shows the main impacts of climate change in the region (e.g. heat stress, inundation, flooding, soil subsidence, blue-green algae, and fresh water shortage). In the introduction, only simplified regional maps for heat stress, water management issues,



27

and drought were provided. Figure 3c shows a map of a plan encompassing future adaptation options. This type of map was only found in four strategies. The assessment map (Figure 3d) was taken from a regional adaptation study in the Netherlands (Janssen et al., 2014) as no options appraisal map was found in the studied regional adaptation strategies.

Fig. 3. Each map illustrates a stage of the adaptation cycle: (a) Climate effect map as found in most strategies. The map shows the percentage of change in precipiation in North Rhine Westfalia, (b) Vulnerability map (found in high quality strategies). The maps shows several climate change impacts in Haaglanden, (c) Adaptation options (found in high quality strategies). The map shows a local plan for part of the region Haaglanden, (d) Assessment map (not found in the strategies). The map shows the effect of water drainage on three objectives (from Dutch peat meadow future vision, 2013)

2.3.3 Quality and regional adaptation strategies characteristics In order to identify main drivers, the average quality of the strategies was compared to the score on several criteria, and to several characteristics of the strategies (Fig. 4). The results indicate that the quality of strategies that include stakeholders with governmental institutions and other parties (M=8.2, SD=1.5) is significantly higher than the strategies that do not involve stakeholders at all (M=5, SD=2), t(10)=-3.62, p=0.0046. Also the quality of the strategies referring to tools (M=9.0, SD=1.0) is significantly higher compared to those not referring to tools (M=6.3, SD=2.1), t(13.7)=4.2, p=0.0009. Focussing on the countries of the strategies, the results show that on average the Swiss strategies have a lower score


CHAPTER 2

28

than the German. In addition, the UK strategies (M=8.3, SD=1.2) have a quality significantly higher as German strategies (M=6.0, SD=1.9), t(9.7)=2.8, p=0.019 and Swiss strategies (M=4.5, SD=1.1), t(12.0)=6.5, p<0.0001. The Swiss strategies cover relative small regions and consist of short reports. Most of the German reports had a technical focus. The database was split in three groups on the basis of the year of their publication, distinguishing between before 2010, 2010-2013 and 2014. This division was chosen to see if the more recent strategies had higher score than the older ones. More recent strategies have higher mean quality scores compared to those published prior to 2014 (not significant). For both area size and number of pages the middle category relates to high quality scores. Strategies ranging from 100-200 pages (M=8.6, SD=1.5) have a significant higher score compared to short strategies with less than 100 pages (M=5.9, SD=2.2), t(16.9)=3.2, p=0.005.

Fig. 4. Stratification of the characteristics of regional adaptation strategies towards mean quality (error bars represent standard deviation)

Furthermore, the quality scores were compared with the presence of maps and adaptation options (Fig.5). Figure 5a indicates that a high number of maps were related to strategies of higher quality. However, high quality strategies also exist that have no or only few maps. Figure 5b shows that it is less evident for the number of adaptation options in the strategies. These results did not show any significant correlations. The report of the Federal State of Baden-Württemberg contained a lot of detail on climate change effects, including 31 general informative maps and 33 maps of climate change effects. Saxony was found to have a low quality, but described 84 adaptation options. In contrast, North West England strategy is of high quality but provided only 9 adaptation options, and East of England had no options but still managed a quality score of 8.



29

(a) (b) Fig. 5. Correlation of quality and the number of maps (a) and the number of adaptation options (b)

Five strategies referred to using decision support tools in the process of developing adaptation plans. These strategies all include consultation and interaction with at least five government institutions or interaction with interested parties or general public. Haaglanden included a chapter on tools referring to a climate effect atlas and models to relate climate change to other planning activities. Luxemburg referred to a ‘Climate check’ vulnerability model and the use of workshop sessions. Yorkshire included a paragraph on adaptation tools that referred to the UKCIP Adaptation Wizard and the Yorkshire Forward’s business adaptation Toolkit. Vlaanderen described their ‘climate reflex’ which aims at making decisions that are screened by climate scenarios. Finally, Ireland recommended an insurance industry tool for weather related risks and BACLIAT (Business Areas Climate Assessment Tool) which comprises of a set of workshops for business. Most often (80%) no reports made mention of the use of decision support tools.

2.4. Discussion and conclusions

2.4.1 Status of regional adaptation strategies This study is the first to review 25 European regional adaptation strategies from a geo-information perspective. Main finding is that these regional adaptation strategies vary in their quality and their use of geo-information. The differences in quality score can be mainly explained by the aims of the strategy, their phase in the adaptation process, and the quality of stakeholder involvement. On the basis of this analysis it was concluded that the strategies do not meet the requirements of the used definition of an adaptation strategy (Smit and Wandel, 2006). A regional adaptation strategy is defined as the combination of possible adaptation options that help a region to develop from a current state to one that better manages, adjusts to, or copes with climate change (Eikelboom and Janssen, 2013). Reviewing 25 strategies revealed that strategy aims varied from increasing knowledge and study options to adapt, setting strategic direction in what is required to adapt, identifying the vulnerability of the region, and finding opportunities and assignments. Next, most regional adaptation strategies merely generally described effects (92%), impacts (84%) and options (84%). However, none of the strategies included the appraisal of options. The majority of the strategies (72%) described the process of stakeholder involvement, which contributed significantly to a higher quality strategy. A comparison of adaptation strategies between six large river basins showed that these regions have understanding that climate change is happening, but large differences exist in the implementation level of proposed


CHAPTER 2

30

adaptation measures for which insufficient communication between actors and reluctance of actors to change is an important barrier (Krysanova et al., 2010). This is in line with the added value of stakeholder engagement found in the quality scores. Also the quality of the strategies referring to tools is significantly higher compared to those not referring to tools. About one third of the strategies (32%) did not included maps to support their strategy. From the included maps, most are climate change effect maps (54%), informative maps (34%) and climate impact maps (9%). Some strategies were linked to a website that contained more information such as reports and maps. Online adaptation plans provide the opportunity to be updated based on new developments. The inclusion of maps and adaptation options positively contributed to the quality of a strategy. Furthermore, the regional adaptation strategies propose a lot of options but instructions on how to develop these options into practical actions is lacking. For instance on how to evaluate their effect, how to prioritise them, and how to merge separate options into a cohesive strategy. The options are formulated in a generic way. Implementation of the options as described in the plans will raise questions such as how much and where to apply the measure? This might lead to implementing options that eventually appear to be ineffective, or even to maladaptation. Interactive spatial decision tools can support stakeholders in addressing these questions. None of the included strategies appraises the listed adaptation options. From this it is concluded that the strategies are not in the final stage of the adaptation cycle of Preston et al. (2011), and hence none of the strategies is able to identify which options should be implemented first or where.

2.4.2 Reflections on approach Although the variability in the results from the selection of the German, English and Dutch strategies provided confidence on the representativeness of the sample, future research should look into European regional adaptation strategies which have been published in other languages as those that have been included in this study. Additional to this, it is expected that strategies that will be developed after this study will use the EC guidelines. It would be interesting to repeat this study in several years to assess if these new strategies score better on quality compared to the older ones. Baker et al. (2012) evaluated seven local climate adaptation plans in Southeast Queensland, Australia by developing and applying a quantitative, multi-criteria analysis framework. This was different from the criteria used in this study as Baker et al. (2012) based their criteria on desired adaptation planning outcomes to address key climate threats. For the analysis a 3-point scale was used for each criteria. A larger scale could provide more detailed differentiations, but also requires more classification rules. Also the review was limited to regional strategies, one could consider urban regions as well, but these were not included in the analysis. Recently, Reckien et al. (2014) revealed that from a detailed analysis of 200 large and medium-sized European cities, 72% have no adaptation plan.

2.4.3 Policy implications The high number of adaptation options indicates that a range of possibilities is already available to deal with climate change. A next step is to develop these into applicable and practical options by answering the where and when questions. The adaptation pathways approach supports the ‘when’ question and geodesign tools can support the ‘where’ question while at the same time providing insight into the effects of options for different stakeholder interests, leading to a coherent set of options. Currently, the options and maps are used in a descriptive way the reports. The next stage, the appraisal of options, requires the use of the listed maps and options for making trade-offs to decide where to apply which adaptation measures. This can change the requirements of geo-information and the level of



31

detail of the adaptation options. Spatial decision support tools can be expected to greatly improve the adaptation process (Eikelboom and Janssen, 2013). Most of the reviewed regional adaptation strategies were developed prior to the publication of the European guideline on developing adaptation strategies (SWD 134, 2013) and did not refer to the climate adapt web platform. The European guideline contains a self-check which can be seen as a directory on what to consider when developing an adaptation strategy. Currently, none of the strategies covers the complete check-list. For example, none of the strategies reported identification and understanding of gaps and barriers that hindered adequate response in the past or included assessments of cost-benefits of their options. The European guideline addresses stakeholder involvement as a key element to build the basis for a successful adaptation process. Results show that the engagement of stakeholders, reference to spatial decision support tools, and the use of maps positively influence the quality of regional adaptation strategies. Hence, it is advised that these will be included in the development of new strategies. In addition, these elements should play an important role when incorporating the appraisal of adaptation options in the strategies, which is a critical step before the actual implementation of adaptation measures. To exploit the opportunities and strengths and to reduce the weaknesses and threats within the countries, an exchange of experiences and results of adaptation practices between countries is very useful. This study focused on European strategies, but lessons could also be learned from regional adaptation strategies outside Europe such as Ho Chi Minh City (Lasage et al., 2014). In addition, multiple strategies are available for city regions or municipalities (Mukheibir and Ziervogel, 2007; Hunt and Watkiss, 2011). Countries can learn from innovative strategies, approaches and options to cope with the impacts of climate change in other countries. The results show that the engagement of stakeholders, reference to spatial decision support tools, and the use of maps positively influence the quality of regional adaptation strategies. Hence, it is advised that these will be included in the development of new strategies. In addition, these elements should play an important role when incorporating the appraisal of adaptation options in the strategies, which is a critical step before the actual implementation of adaptation measures.


CHAPTER 2

32


Interactive spatial tools for the design of regional adaptation strategies

33

CHAPTER 3 Interactive spatial tools for the design of regional adaptation strategies

Abstract Regional adaptation strategies are plans that consist of feasible measures to shift a region towards a system that is flexible and robust for future climate changes. They apply to regional impacts of climate change and are imbedded in broader planning. Multiple adaptation frameworks and guidelines exist that describe the development stages of regional adaptation strategies. Spatial information plays a key role in the design of adaptation measures as both the effects of climate change as well as many adaptation measures have spatial impacts. Interactive spatial support tools such as drawing, simulation and evaluation tools can assist the development process. This chapter presents how to link tasks derived from the actual development stages to spatial support tools in an interactive multi-stakeholder context. This link helps to decide what spatial tools are suited to support which stages in the development process of regional adaptation strategies. The practical implication of the link is illustrated for three case study workshops in the Netherlands. The regional planning workshops combine expertise from both scientists and stakeholders with an interactive mapping device. This approach triggered participants to share their expertise and stimulated integration of knowledge. Published as: Eikelboom T. and Janssen R. (2013) Interactive spatial tools for the design of regional adaptation strategies. Journal of Environmental Management, 127:S6-S14.


CHAPTER 3

34

3.1 Introduction Climate change is more and more considered in regional planning. Mitigation and adaptation are two ways of responding to climate change, where adaptation is judged to be inevitable as the effects of mitigation measures such as emission reduction take several decades to become apparent. The IPCC defines adaptation as adjustment in natural or human systems in response to actual or expected climatic stimuli or their effects, which moderates harm or exploits beneficial opportunities. Various types of adaptation can be distinguished, including anticipatory and reactive adaptation, private and public adaptation, and autonomous and planned adaptation (IPCC, 2001). Adaptation measures can deal with the effects of climate change on a much shorter time scale (Füssel, 2007). Adaptation strategies can be developed on different scales from the international level to the level of households and individuals where each level has its own quality. The scale of autonomous adaptation by individuals and/or businesses is expected to be insufficient to deal with climate change impacts (CEC, 2009). Higher levels such as the national or European scale can monitor adaptation actions and promote action by legislation (CEC, 2009; Adger et al., 2005). National plans need downscaling to smaller regions because the regional scale is the appropriate scale to respond in terms of spatial measures (Roggema, 2009; Adger et al., 2007). The increasing interest in adaptation strategies is reflected in the development of adaptation frameworks and guidelines (e.g., Dessai et al., 2005; Willows and Connell, 2003; Bruin et al., 2009). These adaptation frameworks illustrate the different stages in the adaptation process. However, practical tasks and spatial tools to support the development of regional adaptation strategies are lacking. This study focuses on how to support planned adaptation on a regional scale. Stakeholders with conflicting goals and objectives are involved in regional planning. One major challenge in regional planning with multiple stakeholders is to initiate discussion whilst at the same time to create consensus. The importance of participation in regional planning is recognized in earlier research (Jankowski, 2009; Reed, 2008). The involvement of participation is taking place at a much earlier stage in the planning process, while previously participation mainly took place in the final decision phases (Geertman, 2006). Interaction can be defined as two-way communication between stakeholders and spatial information. Other studies have agree that the degree of interactivity of a tool is an important aspect that can positively influence regional planning (Andrienko et al., 2002; Bacic et al., 2006; Goosen et al., 2007). Interaction encourages cooperation between stakeholders and offers the possibility for the user to independently perform map operations. As interaction promotes learning by doing, it increases the understanding of underlying mechanisms, which subsequently leads to increased credibility and acceptance (Andrienko et al., 2002). Spatial information plays a key role in designing adaptation measures as both the effects of climate change and adaptation measures contain a spatial component. More data have become available in digital form as Geographical Information Systems (GIS) are increasingly used by other disciplines (Scholten et al., 2009). Spatial support systems that attempt to include the complexity of a system tend to lose transparency and are not used in practice (Vonk et al., 2005). The limited usage of planning support systems can be explained by the lack of awareness and lack of experience of stakeholders as well as instrument quality (Vonk et al., 2006). Therefore, spatial support tools are recommended to be developed in cooperation with stakeholders (Vonk et al., 2007). The complexity of support tools can be further reduced through transparency such as the explanation of limitations and by emphasizing the benefit of their usage for the planning process. Carsjens and Ligtenberg (2007) and Goosen et al. (2007) underline the need for interactive and participatory tools that can quickly and indicatively show the environmental impact of spatial plans for each individual stakeholder as well as the combined impact. The aim of this chapter is to show what type of spatial support tools can be used to support different stages of the development process of regional adaptation strategies with stakeholders.



35

More specifically, the following questions are addressed: What are regional adaptation strategies? What are the different stages and tasks in the adaptation process? How can stakeholders be supported to develop regional adaptation strategies?

First, the characteristics of adaptation on a regional scale are derived from the wide range of definitions of adaptation strategies (section 2). Second, adaptation assessment frameworks and guidelines are compared (section 3). A literature review was carried out to define regional adaptation strategies and to summarize adaptation frameworks. The review was limited to adaptation assessment frameworks where the emphasis was on the process of formulating practical adaptation options. The adaptation frameworks and guidelines illustrate the different tasks in the assessment stages in the adaptation process and serve as a basis to link spatial tools that can support the adaptation process. This chapter only describes tasks that can be fulfilled by stakeholders in an interactive multi-actor setting. Tools are coupled to the adaptation process to support both the spatial aspect and interaction. In this chapter, ‘tools’ refer to spatial and interactive instruments used by stakeholders during regional planning. The practical implication of the connection between spatial support tools and adaptation frameworks is illustrated in the context of regional planning workshops in the Netherlands (section 4). A workshop approach was applied that combines expertise from both scientists and practitioners with interactive spatial support tools. An interactive mapping device (‘Touch Table’) was used to support the workshops.

3.1.1 Method A literature review is used to derive tasks for each adaptation stage. Based on these tasks, specific types of interactive spatial tools were developed and applied in local planning workshops. For each workshop a ‘pre-workshop’ and ‘post-workshop’ questionnaire was filled out by the participants. A total of 49 stakeholders (from the water board, province, nature and environmental organizations, and farmers and farmer represents), evaluated their experiences in the questionnaires. The questionnaires were used to evaluate five main themes. A list of statements revealed changes in expectations, knowledge and opinion. Next, both scaled and open questions were used to evaluate the different maps, the workshop approach and the use of a ‘Touch Table’. The questionnaires contained both similar questions to make them comparable between workshops, as well as specific questions tailored to the study area. The tools applied in the workshops were especially developed for the workshop regions within the ArcGIS environment and used the Touch table as their interface. The simulation and evaluation tools were developed using CommunityViz Scenario 360TM, an ArcGIS extension for interactive spatial planning. All workshops were directly related to existing planning processes, which enabled the participation of local stakeholders. The scale of the different study areas ranges from a few to a maximum of about 50 km2. The influence of the stakeholders on tool development lays the purpose of the workshop process and the characteristics of the study area such as main objectives.

3.1.2 Regional adaptation strategies and the role of stakeholders There is a wide range of definitions of climate adaptation (Smit et al., 1999; IPCC, 2007), but their descriptions lack details to serve for the regional scale. This thesis applies Smit and Wandel’s (2006) broader definition of adaptation strategies in changing conditions to the narrower field of climate effects on a regional scale; a regional adaptation strategy is defined as the combination of possible measures that help develop from a current state of a region to one that better manages, adjusts to or copes with climate change. The development of an adaptation strategy is an iterative, continuous learning process (Niang-Diop and Bosch, 2004). At the regional scale, adaptation strategies can provide direction in dealing with specific effects of climate change across a larger area of land than an individual municipality. Regional planning comprises


CHAPTER 3

36

multiple land use activities, infrastructure and settlements. This means multiple local stakeholders are involved in regional adaptation planning. A climate change adaptation strategy may include a mix of policies and measures, selected to reduce vulnerability. As defined by the IPCC, ‘Vulnerability is the degree to which a system is susceptible to, or unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate of climate variation to which a system is exposed, its sensitivity, and its adaptive capacity’ (IPCC, 2001). Adaptation measures can be either autonomous or planned (anticipatory or reactive). The appropriateness of an adaptation measure depends on multiple factors such as vulnerability, costs, effectiveness, flexibility and stakeholder perception. This study concentrated on adaptation assessment, ‘The practice of identifying options to adapt to climate change and evaluating them in terms of criteria such as availability, benefits, costs, effectiveness, efficiency, and feasibility’ (IPCC, 2001). spatial measures such as land use change and changes in water management were identified for the study areas in this chapter. Measures such as insurance, acceptance and evacuation were not included. The development of an adaptation strategy requires the presence of adaptive capacity. ‘Adaptive capacity is the ability or potential of a system to respond to climate variability and climate change successfully’ (Adger et al., 2007; IPCC, 2007). The physical capacity of a region can be influenced over time by changes in, for example, land use and water management. In practice, a high adaptability will not automatically translate into successful adaptation, because there are different types of barriers to adaptation such as technological, financial, and social barriers. Climate change is in practice only a minor issue in regional planning. Climate change can provide a sole reason for making a decision, but more often climate change is only one of the factors that influence planning. Adaptation measures are often part of broader planning and management activities, and seldom a response to climate change alone (Adger et al., 2007). Soil subsidence in fen meadow areas, for example, also occurs under current climate conditions. Adaptation planning often combines climate change with secondary factors such as a motivation for adaptation (Ford et al., 2011). A recent evaluation study by Preston et al. (2011) shows that weaknesses of adaptation strategies are often related to limited consideration of non-climatic factors.

3.2 Assessment frameworks for the formulation of regional adaptation strategies The adaptation process is described in several adaptation frameworks and guidelines (e.g. (Willows and Connell, 2003; Prutsch et al., 2010). Exploration of these frameworks is needed to select a basis that helps to determine practical tasks and suitable support tools. Smit et al. (2000) summarize the adaptation process in four main questions: ‘adaptation to what?; who or what adapts?; how does adaptation occur?; and how good is the adaptation?’ The European Commission prepared a guideline on how to include climate change into management plans. This guideline on the formulation and implementation of regional adaptation strategies exists of four steps: (1) prepare the ground; (2) assess vulnerability within the region; (3) set strategic direction; and (4) plan and implement concrete adaptation measures (Ribeiro et al., 2009). The purpose of the adaptation assessment defines the use of scenarios (Dessai et al., 2005). Broad strategies or policies only use scenarios indicatively, whereas more detailed scenarios are needed to make specific operational decisions. Dessai et al. (2005) distinguishes the various adaptation frameworks based on the role of climate scenarios in three major approaches: the IPCC approach, risk approaches, and human development approaches. The Adaptation Policy Framework (APF) (Burton et al., 2002) is a cross-cutting approach and is composed of five basic steps, where engaging stakeholders and enhancing adaptive capacity are crosscutting components: (1) defining project scope; (2) assessing current vulnerability; (3) characterizing future climate risks; (4) developing an adaptation strategy; and



37

(5) continuing the adaptation process. This framework builds on the IPCC approach but also contains risk-based approaches (3) and human development approaches (2, 4 and 5). The decision-making framework of Willows and Connell (2003) complements the APF with more detail and covers the steps of the frameworks discussed above. Within this framework, climate scenarios are just tools to help assess the risk of climate change and its influence on the decision-making process. This framework is composed of eight stages (Fig. 1): (1) identify problem and objectives; (2) establish decision-making criteria; (3) assess risk; (4) identify options; (5) appraise options; (6) make decision; (7) implement decision; and (8) monitor. According to the IPCC, adaptation assessment is the practice of identifying options to adapt to climate change and evaluating them in terms of criteria such as availability, benefits, costs, effectiveness, efficiency, and feasibility (IPCC, 2001). For this thesis, the widely used (Ribeiro et al., 2009; Brown et al., 2011) decision-making framework of Willows and Connell is selected as it describes the actual development of adaptation strategies in three detailed stages: assess risk; identify options; and appraise options. Another advantage of this framework is that it shows adaptation as an iterative process. The framework is circular and contains feedback and iteration to refine the problem, objectives and decision-making criteria. Iteration is important to achieve robust decisions (Willows and Connell, 2003). This framework is used in the next sections to link adaptation tasks and spatial tools.

Fig. 1. Adaptation framework of Willows and Connell (2003)


CHAPTER 3

38

3.2.1 Stages and tasks in development of regional adaptation strategies The focus is on the three actual development stages (stage 3-5 in Fig. 1), as the goal is to link interactive spatial tools to practical tasks to support stakeholders with the design of regional adaptation strategies. The three stages can be divided into multiple practical tasks, such as the exploration of available information; validation and improvement of information; exploration of current and future vulnerability and risks; exploration of adaptation options; prioritization of options; and evaluation of adaptation options. These tasks can be grouped into six task categories: analysis, validation, exploration, design, evaluation and negotiation (based on Carton (2007) and Bruin et al. (2009). The task categories can be coupled to the selected stages (Table 1). As shown by the arrows, the order of action can be in multiple directions. The purpose of stage 3 is to determine the risks of climate change effects. This is accomplished by two main tasks: analysis and validation. Analysis is the quantification of effects from either climate change scenarios or the application of adaptation measures. Validation is the verification of available information based on the expertise and knowledge of stakeholders. During the identification of options (Stage 4), the purpose is to find adaptation measures; this involves exploration and design. Exploration is the identification of problems and opportunities from the maps. Design refers to the actual development of the spatial configuration of adaptation measures. The appraisal of options in Stage 5 results in a ranking of options to formulate a preferred strategy, based on the input of earlier stages. In this stage, options are evaluated and negotiated. Negotiation is the discussion between stakeholders concerning the outcomes. Evaluation is the comparison of results to decide on further action. Iterations of the risk assessment, options identification and appraisal loop result in strategies.

3.2.2 Link between adaptation tasks and interactive spatial support tools Tools in this thesis refer to spatial instruments that can be used interactively by stakeholders during planning workshops. A geographical information system (GIS) often serves as a platform for interactive spatial tools. Different type of tools can be used to support the tasks, as described in the previous section. This chapter divides spatial support tools into three categories:

Drawing tools Simulation tools Evaluation tools

The tools are named after their function in supporting adaptation tasks. The tools mentioned in the chapter were all developed by the authors and tailored to the study areas. All tools were developed within a GIS environment and used a Touch table as their interface. The simulation tool was developed using the ArcGIS model builder combined with Visual Basics for the user interface. The evaluation tool was developed using CommunityViz Scenario 360TM, an ArcGIS extension for interactive spatial planning. Drawing tools give stakeholders the opportunity to add comments to maps and to highlight specific features. Lines or polygons can be sketched on a map to indicate various opinions or knowledge given by stakeholders, such as good/bad, high/low, wet/dry, and agree/disagree. These maps can provide insight into possibilities for adaptation measures. Areas with high current and future vulnerability can be highlighted to identify bottlenecks. Annotation maps can serve as background information in the next stages. Simulation tools are instruments that can indicatively provide insight into the impact of changes in various spatial parameters. A simulation tool exists of interactive maps that allow adjustments to the current situation and provide feedback based on these changes. This tool can be applied during the



39

exploration stage or for the design of adaptation measures. This type of tools allows changes made by stakeholders to improve or correct an alternative. Evaluation tools can help to value, compare and rank strategies. For the evaluation of the current situation or an alternative strategy, it is helpful to assign values to its effects on the objectives for the study area. Value maps can be combined to evaluate total effects for different land uses and objectives. Multi-criteria analysis (MCA) can, for example, help to identify trade-offs, combine multiple stakeholders and prioritize measures. The combination of MCA and geographical information systems (GIS) has been applied successfully in previous studies (e.g. Arciniegas et al., 2011; Arciniegas and Janssen, 2012, Alexander et al., 2012). The value maps can serve as input to combine multiple objectives such as the identification of ‘best and worst’ parcels for land use change. This can quickly and indicatively show the environmental impact of spatial plans for each individual stakeholder, as well as the combined impact. Other spatial support tools can be either categorized as one of the three types of tools mentioned in this section, or can apply to a task category from section 3.1. The final step is to cluster the above tools based on their descriptions to the identified tasks from section 3.1 in one overview. The resulting matrix shows which interactive spatial tools are suited for what stages in the development process of regional adaptation strategies (Table 1). The exact realization and implementation of each tool can differ based on the variety of options within a tool or due to the characteristics of a case study. Table 1. Matrix that couples tasks, tools and the stages in the adaptation framework for spatial adaptation

Stage Task Tool Stage 3 Assess risk

Analysis Evaluation Validation Drawing, Simulation

Stage 4 Identify options

Exploration Drawing Design Drawing, Simulation

Stage 5 Appraise options

Evaluation Evaluation Negotiation Drawing


CHAPTER 3

40

3.3 Application of tools in stakeholder workshops This section describes the application of interactive spatial support tools for three case studies in the Netherlands. For each case study, a planning workshop took place. Table 2 illustrates the tasks and tools coupled to the various stages of the planning process in the three case studies. Multiple stakeholders, such as the local water board, the province, farmers’ and nature conservation organizations, as well as individual farmers were involved in the workshops. In the first workshops on the Dutch Wadden Sea Island ‘Texel’, drawing tools were applied to validate model results and to explore problems and opportunities for adaptation measures. For the second workshop in ‘Zevenblokken’, a simulation tool was available to design options for water management. In the third workshop in ‘Friesland’, an evaluation tool was used to compare different water management alternatives. All workshops were organized in cooperation with local authorities. Though, the workshop programs and tools were developed by scientists, the content and focus of the workshops was set by the local stakeholders based on their progress in the planning process. This means the tools were adapted to the stakeholder preferences. Figure 2 shows the distribution of different type of stakeholders over the different areas. This figure excludes the researchers involved in the organization of the workshop sessions. Table 2. Derivation of tasks and tools from the planning process

Fig. 2. Distribution of participants over stakeholder types

6

2 3 2 3

2

1 23

1

4

1

3

2

14

0

2

4

6

8

10

12

14

16

Part

icipa

nts

(n)

Type of stakeholder

Texel

Drenthe

Friesland

Workshop Stage in the planning process Task Tool Dutch Wadden Sea Island ‘Texel’ 3. Assess risk

4. Identify options Validation Exploration

Drawing tool Drawing tool

Fen meadow area ‘Zevenblokken’ 4. Identify options Design Simulation tool Fen meadow area ‘Friesland’ 5. Appraise options Evaluation Evaluation tool



41

3.3.1 Application of a drawing tool On the Dutch Wadden Sea Island ‘Texel’, the aim of the stakeholder workshop was to make a first inventory of potential effects of climate change on agriculture. Analyses of the effects of climate change on salinity and water level were carried out beforehand, using a hydrological model (Witjes, 2011). The resulting maps were used as input for the workshop. Drawing tools were used to support the validation and exploration of the maps. Three types of drawing tools were applied in three follow up sessions. The first drawing tool was used to indicate parcel ownership (Fig. 3a). This introduces the participants with the use of the digital drawing tools, but it also gives insight in the distribution and representation of stakeholders in the area. Second, the farmers were asked to indicate areas that they estimated to be too wet, dry or problematic for their farming activities based on the water level map of the current situation (Fig. 3b). Thirdly, the drawing tool was used to propose adaptation measures (Fig. 3c).

Fig. 3. Drawing results: a) Ownership parcels, b) Validation of model results, and c) Identification of measures

The participants had no difficulty in identifying their own parcels based on the aerial photograph, and they drew the outlines of their property with high accuracy. The validation task reveals that participants’ expertise is very valuable to evaluate model results. The participants were only prepared to comment on the model results for their own parcels. They felt unconfident to extrapolate their expertise to similar adjacent parcels. At the same time, some participants also indicated water levels that would lead to an unprofitable situation. In practice, the identification of problems resulted in a preference for drawing adaptation options. Some of the adaptation options are largely innovative, but other measures were more realistic and practical. One of the suggested measures, which is shown at the eastern edge of the island, is a fresh surface water basin that is protected from salt water by a surrounding dike to provide for sufficient fresh water. An option that is more connected to current infrastructure is to turn the direction of effluent water flow in the opposite direction, and is indicated by the arrows. In addition, the development of fresh water bodies below the dunes was suggested. Other measures that were mentioned but that were not indicated on the map are changes in soil cultivation, changes in crop types and changes in the timing of harvesting.


CHAPTER 3

42

3.3.2 Application of a simulation tool In ‘Zevenblokken’, the objective of the workshop was to design options for water management. This means that the task was to design a future spatial configuration that includes adaptation options. The workshop started with a drawing tool to search for problems and possible measures. Two types of background maps were used for the identification. The first map showed the water surplus (Fig. 4a) and the second map was a value map for agriculture (Fig. 4b). The value map is classified in three classes, where a score below 75% means unprofitable, a score between 76 and 90 is acceptable and a score between 91 and 100 is optimal for agriculture. After the identification of possible measures by drawing, a simulation tool was provided to design options. The simulation tool consists of a control panel with three types of measures to choose from: 1) changing the water level of the ditches; 2) changing the surface height; and 3) applying drainage (Fig. 4c). Stakeholders defined the extent of one or more measures by using the control panel. Next, the stakeholders collectively select one or more spatial units to apply the defined type of measure and the start the calculation. All measures directly affect the mean water level of the selected water units. In turn, changes in the water level directly lead to changes in the value map for agriculture.

Fig. 4. Simulation tool: a) Water level, b) Value map for agriculture, and c) Control panel

Before the actual design of options, the participants first tested the simulation tool. They applied large changes for only one single measure and one single water level unit to investigate the effect of the different measures. As an example, first the water level in the ditches of one water level unit was decreased by half a meter. For another water level unit, drainage was applied 150 cm below the surface. Based on these first results, a more serious run of the tool was performed. A farmer who owned several parcels in the area controlled the tool the most, and continued applying measures until his parcels reached optimal scores. The questionnaires show that the simulation tool was preferred in a format where changes are based on adaptation measures instead of directly changing the height of the water level without describing the associated measures. The mostly applied measure was changing the water level in the ditches. A majority of 63% found the proposed measures from which a choice could be made to be sufficient, whereas 37% requested more possibilities to choose from. The simulation tool meets with the transparency request as the underlying mechanisms were explained and kept simple.



43

The simulation tool was applied on water level sections, which is the most appropriate spatial unit according to 89% of the users. Next to the changes, the simulation tool also calculates the absolute maps of the water level and value map. This was necessary according to 44% of the respondents and was only needed as background maps for another 44% of the participants. The application of changes can result in multiple adjusted maps. It is important to discuss which type of output is preferred by the stakeholders for the workshop purpose.

3.3.3 Application of an evaluation tool In ‘Friesland’, the aim of the workshop was to compare water management alternatives. Due to historical land consolidation, the number of water level sections is very high. In addition, for the purpose of agriculture, the water levels in this fen meadow area are kept low, which causes soil subsidence. Together, these two aspects lead to high maintenance costs for the water board. The aim of the workshop was to reduce the number of water level sections whilst having the minimum interference with the objectives of the stakeholders. The alternatives are based on merging water level sections from different perspectives. An evaluation tool helps the comparison by supporting evaluation and negotiation tasks. The evaluation tool was based on the valuation of multiple objectives: nature, agriculture, greenhouse gases, soil subsidence and landscape. The resulting value maps were used to identify differences. Each objective was scaled from 0-100 based on contributing criteria such as land use and water level. Figure 5b gives an overview of the scores for each of the objectives in each alternative. The differences in values for agriculture between the current and alternative configuration could also be observed from a transparency overlay of both layers. The difference map was used for the identification of parcels where the agricultural value decreased. The causes of the decrease where observed by replacing the value map with a map of water level changes between the current and alternative configuration. It was observed that although the value decreased for some parcels, the score still remained acceptable for farming. For other parcels the scores increased in the alternative configuration, which suggests an improvement for agriculture. The difference maps show both the increase and decrease of the score for agriculture. This indicates positive and negative effects of the alternatives. They support negotiation as they provide input to decide about the next steps in the development process. A drawing tool (same as 4.1) was used to search for areas with opportunities and bottlenecks for implementation of the alternatives (Fig. 5a).

(a) (b)

Fig. 5. Difference in agricultural value between current and alternative (red circles are bottlenecks)

88 87

31

4250 53

0

25

50

75

100

Valu

e

Objective per scenario (current and alternative)

Agriculture Current

Agriculture Alternative

Soil Subsidence Current

Soil Subsidence Alternative

Nature Current

Nature Alternative


CHAPTER 3

44

The value and difference maps initiated a lively discussion among the participants. None of the alternatives was fully accepted, but together they served as an inspiration for making changes to the alternatives. The alternatives revealed missing information such as the costs and effectivness of measures. On the one hand, the results of the scoring unexpectedly show an increase in the score and only minor decreases for agricultural activities. On the other hand, the benefits of implementation of merging water level units are mainly for the water board; negative side effects were not fully covered in this first inventory of alternatives.

3.3.4 Evaluation of the workshop approach In the workshops, the input of the participants was an important driver for the adaptation process and contributed to a broader support of the resulting plans. The use of support tools by the participants was intuitive, which created confidence in the participants about the reliability of the tools and on the transparency of scientific data. Two-thirds of the participants claimed that their understanding of the effects of climate change increased due to the workshop. Map operations with the highest added value in the Touch Table were found to be in- and out zooming, transparency overlay with background maps, drawing, and horizontally or vertically swiping of one layer on top of another background layer. User-friendliness and the accessibility of information were found to be the most important points for improvement. After the workshops, the confidence in the added value of the tools in combination with the Touch Table increased compared to when they were asked to value the same statement before the workshop. Also, after the workshops more people disagreed with the Touch Table being just a new gadget. In particular, the combination of available information, stakeholders and scientists was found a useful source for the majority of the participants and helped to analyse the problem. The workshops did fulfil the expectations (and sometimes even exceeded expectations) of the participants (86%). Although some of the respondents (55%) had little or no experience with the use of spatial support tools with an interactive mapping device, most of them (86%) found it fairly easy to use the Touch Table. Almost all participants (98%) would recommend the workshop concept to others, and 78% agree that regional discussions can be better supported with interactive tools instead of paper maps. A comparison of the observations of the different case studies revealed that the content of an adaptation strategy greatly depends on regional characteristics. The regional problems differed between the fen meadow areas and ‘Texel’; in the case of fen meadow areas, soil subsidence is the main process behind the local problems, whereas on Texel salt intrusion and recycling of fresh water plays a major role. The comparison of stakeholder workshops in fen meadow areas with the stakeholder workshops in Texel demonstrates the relatively low level of importance of the effects of climate change for fen meadow areas as soil subsidence also occurs under current conditions. For each case study, a specific compilation of tools was applied tailored to the progress in the adaptation process for each study area.



45

3.4 Discussion and conclusions This study focused on the specific domain of interactive spatial tools to support the development of regional adaptation strategies. Spatial information plays a key role in the design of adaptation measures as both climate change and many adaptation measures have spatial impacts. Regional adaptation strategies are plans that consist of feasible measures to shift a region towards a system that is flexible and robust for future changes. They apply to regional impacts of climate change and are imbedded in broader planning. The selected adaptation framework of Willows and Connell (2003) consists of three looped development stages: assess risks, identify options and appraise options. Each of the three stages contains two task categories: risk assessment includes analysis and validation; identification of options consists of exploration and design. Finally, appraise options consists of evaluation and negotiation as tasks needed to make improvements for the next round. As all these tasks require the use of spatial information and as many people find it difficult to use spatial information effectively interactive spatial tools are used to support these tasks ((Andrienko et al., 2007; Carton and Thissen, 2009)). Several map based methods are available to support multiple stakeholders in their use of spatial information to design spatial plans. However, there is limited empirical evidence on the effectiveness of different types of methods to support specific tasks ((Arciniegas et al., 2012); (Ozimec et al., 2010)). Geertman and Stillwell (2009) state that the number of successful applications of geo-technology by planning practitioners to support their activities are far from commonplace. Uran and Janssen (2003) identify the mismatch between the decision problem of the end-users and the answers produced by the system as the main factor for this lack of success: the technology-driven systems produce the correct answer to the wrong question at the wrong moment. For practical reasons this study was limited to interactive spatial tools. In practice there are many other spatial and non-spatial tools, such as visualization methods, physical and statistical models, and cost-benefit analyses (e.g. Rinner et al., 2008; Appleton and Lovett, 2005). This study started with the identification of tasks linked to the various stages of the planning process and tried to find interactive spatial tools that were suited to support these tasks. The tools were developed especially for the use with a ‘Touch Table’ as all tools need a communication platform. Depending on the tool, it is also possible to apply the underlying methodology of the tools in a setting without a Touch Table. Though, the perception and performance with the tool could differ as the Touch Table as a communication platform was positively received by all participants. Three types of tools were tested in three separate workshops: 1. drawing tools, 2. simulation tools and 3. evaluation tools. Drawing tools were tested on the Dutch Wadden Sea Island ‘Texel’. The tools were used to validate modeling results, to explore problems and to identify opportunities for adaptation measures. The use of the tools was successful as they got participants involved and provoked a large number of comments and suggestions. The aerial photograph was used for reference but participants were also very interested in more technical background maps related to water levels and water quality. A simulation tools was tested in the ‘Zevenblokken’ case study. This tool was used to design water management strategies. The tool allowed the participants to interactively change model input and immediately see the results. As a first step participants used the tool to test if the model produced plausible results. After passing this test the tool was used to interactively design a water management plan. The possibility to interactively test the model combined with short response times is essential for successful use of the tool. An evaluation tool was used in the ‘Friesland’ case study to compare different water management strategies. Central to this tool was the use of value maps. These maps proved useful in communicating relative qualities of the plans and also triggered specific questions related to underlying information. The value maps were considered more useful than the aggregated value scores for the whole area.


CHAPTER 3

46

In all three case studies, it proved necessary to tailor the spatial support tools to the regional context. The time and effort this takes may be a limitation to the use of these tools in practice. All three workshops showed that interactive participation promotes stakeholder involvement and encouraged knowledge exchange and acceptance of workshop products. It is important to allow the participants to play around and test the tools before the real work starts. The number of different tools should be limited and the tools should be relatively simple and transparent. For the interactive application of tools, the calculation time has to be limited to seconds. Similar conditions for effective design of adaptation strategies were found by Füssel (2007). The results of this study agree with the suggestion of Opdam (2010) that communication between science and society is valuable for planning. The suggestion for future research is to continue the collaboration between science and society in both the development and evaluation of interactive spatial support tools.


Comparison of geodesign tools to communicate stakeholder values

47

CHAPTER 4 Comparison of geodesign tools to communicate stakeholder values

Abstract Geodesign tools are increasingly used in collaborative planning. An important element in these tools is the communication of stakeholder values. As there are many ways to present these values it is important to know how these tools should be designed to communicate these values effectively. The objective of this study is to analyse how the design of the tool influences its effectiveness. To do this stakeholder values were included in four different geodesign tools, using different ways of ranking and aggregation. The communication performances of these tools were evaluated in an online survey to assess their ability to communicate information effectively. The survey assessed how complexity influence user performance. Performance was considered high if a user is able to complete an assignment correctly using the information presented. Knowledge on tool performance is important for selecting the right tool use and for tool design. The survey showed that tools should be as simple as possible. Adding ranking and aggregation steps makes the tools more difficult to understand and reduces performance. However, an increase in the amount of information to be processed by the user also has a negative effect on performance. Ranking and aggregation steps may be needed to limit this amount. This calls for careful tailoring of the tool to the task to be performed. For all tools it was found maybe the most important characteristic of the tools is that they allow for trial and error as this increases the opportunity for experimentation and learning by doing. Published as: Eikelboom T. and Janssen R. (2015) Comparison of geodesign tools to communicate stakeholder values, Group Decision and Negotiation.


CHAPTER 4

48

4.1 Introduction In spatial planning maps can be used to combine stakeholder values with different types of spatial information. Maps can also serve different functions in the planning process such as reaching agreements, exchanging information, and setting objectives. However, the role of map representations is not always well understood. The influence of maps depends both on the quality and presentation of the information as on the processing capabilities of the decision-maker (Dühr, 2007). Maps integrated in geodesign tools are used to support stakeholders in collaborative planning. Geodesign tools combine geography with design by providing stakeholders with tools that support the evaluation of design alternatives against the impacts of those designs (Flaxman, 2010). Little research has been undertaken on the communicative function of map graphics in planning (Dühr, 2007). Researchers in the field still remark on a lack of extensive testing and quantitative evaluation of spatial planning and decision support tools (Vonk et al. 2005; Geertman and Stillwell 2004; Geertman and Toppen 2013). Only a few studies have explicitly tested tool effectiveness (e.g. Inman et al., 2011; Arciniegas et al., 2012). It is not self-evident that when information is put in a map, it is also understood by the viewer (Steinitz 2012). Multiple attributes are mostly combined in a suitability map. However, a suitability map of a single objective shows the spatial differentiation of the performance of this objective but does not present the values of other objectives. Furthermore, a suitability map derived from combining multiple objectives only shows the total suitability and does not give any detail about the aggregated objectives. Maps that present a combination of multiple attributes are often complex. Janssen and Uran (2003) for example showed that participants overestimated their ability to use this type of maps. Geodesign tools intend to increase the effectiveness of spatial planning. However, effectiveness is a broad concept that can include many aspects. Previous studies have discussed various aspects of effectiveness (Nyerges, et al. 2006; Salter, et al. 2008). Effectiveness has been associated with the usability of a system in the context of human-computer interaction (Sidlar and Rinner, 2009; Meng and Malczewski, 2009). Jonsson, et al (2011) characterizes effectiveness as making sure that the right things are done and that they are done right. Budic (1994) considers effectiveness as operational effectiveness and decision-making effectiveness. The former concerns improvements in quality and quantity of data, whereas the latter is about the facilitation of planning-related decision making. Goodhue and Thompson (1995) distinguished effectiveness as the extent to which instruments enable stakeholders to carry out the intended tasks and the fit of the instruments to the capabilities and demands of the stakeholders. Gudmundsson (2011) states that, besides measuring effectiveness to assess instrumental use, a tool can also have a more conceptual role where use involves general enlightenment. Use of information can be described as receiving information, reading information or understanding information. Use can also be described as the amount of influence of the information on decision-making in terms of contribution or actions. This study focused on visualizing the spatial pattern of multiple stakeholder values simultaneously. A comparison was made between four types of geodesign tools to communicate these values. The tools were tested in an online survey to assess their ability to communicate information effectively. The potential of interactive geodesign tools to contribute to decision processes is more and more recognized (Steinitz, 2012; Dias et al., 2013). Not many studies, however, address the effectiveness of these tools. An unique element of our study is that it directly links effectiveness to task performance and therefore explicitly includes the interactive element of the toll in the evaluation. The tools designed for this study vary in the way information on values is processed and presented. The tools differ in the use of an aggregation or ranking step (Fig. 1). Aggregation means that the values are weighted and summed in a total value. Aggregated values support stakeholders by combining multiple sources in a single attribute. Aggregation prevents a stakeholder from having to combine objectives



49

themselves. Ranking means that the order of the value is used to select values. The result is information that only shows the best and worst objective values. The variations resulted in four geodesign tools: 1) objective value tool, 2) relative objective value tool, 3) total value tool and 4) stakeholder value tool. The objective value tool just presents the objective value and does not require any aggregation or ranking. The relative objective value tool shows how each parcel performs compared to all other parcels and therefore requires ranking. The total value tool aggregates the objective values into an overall value. Finally, the stakeholder value tool uses both aggregation and ranking in order to visualize which exchange is best for each stakeholder.

Fig. 1. Four geodesign tools to present stakeholder values

A pre-test was conducted to finalize the graphic design of the tool. Next, the tools were evaluated in an online survey. Section 2 first describes the current literature on the use of geovisualization for stakeholder value mapping and then the results of the pre-test. The methodology of the survey is described in Sect. 3. Section 4 presents the four geodesign tools that were developed to present stakeholder information. Section 5 shows the results of the survey comparing tool performance of the four tools. Finally, Sect. 6 provides conclusions on the usefulness of these tools to support spatial planning.


CHAPTER 4

50

4.2 Visualization of stakeholder values in geodesign tools Geovisualizations can be used to present the spatial distribution of stakeholder values in decision support and planning tools. A stakeholder value is the score of an objective that is found important by a stakeholder and can consist of multiple objectives of equal or unequal importance. Maps are useful for spatial communication (Arciniegas et al., 2011; Carton, 2007) and can even be more effective when incorporated in a decision support tool. In literature, multiple decision and planning support tools integrated in a geographical information system (GIS) can be found (e.g. Geertman and Stillwell, 2009; Batty, 2008). Maps are evolving in more exploratory and interactive tools that serve as an interface (Kraak, 2004). Recent literature also emphasized the need for a multi-objective view to cartographic design (Xiao and Armstrong, 2012).

4.2.1 Mapping stakeholder values The interests of each stakeholder can be presented in the form of maps. The information in these maps has to be combined. Instead of offering all available information to the planners, spatial evaluation methods can help decision makers to structure and simplify the decision problem (Herwijnen 1999). Two ways to aggregate information can be distinguished 1) approaches that start with individual problem solving followed by aggregation of the solution maps, and 2) approaches that start with the aggregation of stakeholder values which will then be processed in a multi criteria analysis (Boroushaki and Malczewski, 2010; Herwijnen, 1999). Depending on the decision issue, values need to be ranked. In multi criteria analysis ranking is often used (Belton and Stewart, 2002). The need for testing the effectiveness of decision support has been recognized for a long time (Densham 1991; Crossland et al., 1995). Only a few researchers have explicitly studied the effectiveness of visualizations of the spatial distribution of stakeholder values in spatial planning and decision support tools (e.g. Inman et al., 2011; Arciniegas et al., 2012). The review of Te Brömmelstroet (2012) showed that different types of evaluation criteria are applied and concluded that a systematic analysis of performance is missing. Inman et al. (2011) described the application of a quantitative approach to evaluate environmental decision support systems with small groups of stakeholders in two case studies. The objective of these case studies was to facilitate the participatory decision-making process in water management projects. Stakeholders’ perceptions of effectiveness were elicited and compared using statistical analysis. The results of the two case studies suggested that stakeholders’ backgrounds influences their perceptions of effectiveness. The experiments of Arciniegas et al. (2012) show that using a set of collaborative spatial decision support tools, it was found that the cognitive effort related to the volume and format of information is a critical issue in spatial decision support. Usefulness, clarity and impact were the dimensions on which effectiveness was evaluated. Ozimec et al. (2010) evaluated multiple types of symbols and tasks that differed in the level of complexity. The evaluation was based on decision accuracy which was measured by performance, decision efficiency which was measured by duration, and decision confidence and ease of task which were derived from ratings. The results of this study show that the type of symbolization strongly influences decision performance. The findings indicated that graduated circles are appropriate symbolizations for use on thematic maps and that their successful utilization seems to be virtually independent of personal characteristics, such as spatial ability and map experience. There are also studies that explicitly test the different uses of symbols in maps. Dong et al. (2012) measured deviation and response time to assess the quality of dynamic symbols. The results show that size is more efficient and more effective than colour for dynamic maps. Garlandini and Fabrikant (2009) and Fuchs et al. (2009) used eye tracking to study the effectiveness of maps. Garlandini and Fabrikant (2009) propose an empirical, perception-based evaluation approach for assessing effectiveness and efficiency of longstanding cartographic design principles. The visual variable size was found to be the



51

fastest and the most accurate to detect change if it was flashed on and off on the map. The form style and use of cartographic visualizations in spatial planning differ between nations and even between regions (Dühr, 2004). This is mainly determined by the functions of plans in the planning system. This means that above findings cannot directly be used for the communication of stakeholder values in a small scale study area. However, the studies advocate that symbols are useful for map design. In this study the evaluation of the tools is limited to testing communicative tool performance and task functionality. Criteria to measure the impact of the tools on the decision making process, such as user confidence and satisfaction, were not studied. The evaluation of the cartographic design was not the main focus of this study and was limited to the pre-test described below.

4.2.2 Pre-test An empirical pre-test was used to find perceived preferences of map presentations. A site was selected from a previously studied area (Janssen et al., 2014; Brouns et al., 2014). The key issue in the region is the trade-off between the prevention of soil subsidence and the conservation of agricultural production. A small study area was selected with three types of land use: intensive grassland, extensive grassland and nature. The map includes 13 parcels that were numbered. Three main objectives were identified: (1) maximize agricultural production; (2) minimize soil subsidence; and (3) maximize natural value. The objective values depend on both land use and water level. A high water level results in high objective values for soil subsidence and nature, but in low values for agriculture. A high value for soil subsidence means low subsidence rates, high values for nature means high quality and low values for agriculture means low productivity. Different symbolizations, such as squares, patterns and bar charts were used to develop the semiology of the maps of Figure 2 and 3 (Bertin, 1983; Slocum et al., 2009).

Fig. 2. Stakeholder value maps: a) traffic light boxes, b) bar charts

Figure 2 shows the current land uses with the stakeholder values for three objectives in each parcel. In Figure 2a, boxes are used to show, for each of the objectives, if their performance is good (green), intermediate (yellow) or bad (red). Nature has a low value for all parcels. For intensive and extensive grasslands the value for agriculture is on average high, except for parcel number 3 and 4. In Figure 2b the value of the objectives is reflected by the height of the bars. These reveal that for the extensive


CHAPTER 4

52

grasslands, values for soil subsidence are close to the values for agriculture. From the bar charts the variation in values for nature can also be derived, whereas in the boxes nature was all classified as low. In the pre-test, a laboratory experiment was set up with 41 students and researchers. The respondents were asked whether, when given the task to change the land use pattern, they would choose the traffic light or bar chart presentation (Fig. 2). Respondents who favoured the bar charts mentioned the importance of being able to see the actual height of the objective values. Although the boxes provide less information, they were preferred by 63% of the respondents. This map was found easier to read and better suitable for the identification of parcels that need change. The visualization of stakeholder interests with boxes was used for visualization of the geodesign tools that were tested in section 4 of this study. Figure 3 uses the same information to present the land use changes that result in an increase in the total value of the plan. A weighting and aggregation of the three main objectives was applied to determine if a land use change would increase total value of the plan. The map presented in Figure 3a combines colours and symbols. The colour of each parcel presents the current land use. The colour of the boxes indicates the land use that would result in the highest increase of total value. If the current land use is also the best land use the parcel, it is left blank. In Figure 3b, primary colours are used for the original land use and secondary colours to indicate the preferred transitions. The original land use is visualized by the colour of the border of the parcel and the preferred change by the colour of the parcel itself. Finally in Figure 3c, the colour of the parcel itself indicates the current land use and the texture indicates in which direction the land use change is favourable. Similarly, the respondents were asked to select one of the three tools to change the land use.

Fig. 3. Maps indicating preferred land use change: a) symbols b) colours c) pattern

The symbols map (Fig. 3a) was preferred by 83% of the respondents over the map using borders with primary and secondary colours (Fig. 3b). The main reasons were that respondents 1) preferred the more intuitive colours, 2) preferred the use of the symbols on top of current map layer, and finally, 3) that the limited number of legend classes was easier to see. This was expressed by one of the respondents as: “it's quicker to compare two sets of different variables than to go back and forth between the legend and the map with the extensive colour use.” Respondents who were in favour of the colours presentation named, as an advantage, that the legend identifies all possible changes, but also mentioned that they did not like the use of the coloured outlines. In comparing the colours map (Figure 3b) with the patterns map (Fig. 3c) 61% of the respondents voted for the patterns map. One of the respondents remarked: “I like the mixing colours because the colour showing the change combines the



53

colours of the old and new land use, so you can pick out patterns easier. The patterns are not so easy to read but can be recognized easily from the legend.” It is interesting to see how stakeholder values and environmental indicators are visualized in existing geodesign tools. There have only been few studies where multiple objectives were presented in a single map layer (Alexander et al., 2012; Arciniegas et al., 2011). A more widely used approach is to combine multiple criteria in a single map or indicator (Jankowski et al., 2001). The planning support system ‘UrbanSim’ presents single maps of costs, number of residents (Waddell, 2002). ‘Urban strategy’ uses indicator maps of for example noise (Borst, 2010). The online planning support system ‘What if’ uses suitability mapping (Pettit et al., 2013). A suitability map advises positively and negatively about the suitability of locations for a specific change. On the map, regions are marked as ‘suitable’, less suitable or unsuitable (Carton, 2007). Carton and Thissen (2009) show that different frames of stakeholders result in different preferences regarding the suitability maps.

4.3 Method Four geodesign tools were developed based on a pre-test and a literature review. These four geodesign tools were evaluated in an online survey. This section first describes the structure of the survey, next the questions and assignments are explained and finally some practical issues are discussed. The aim of the survey was to test tool performance. Tool performance was divided into communicative performance and task performance. Communicative performance is defined as the ability to deliver information from the map to the user. Task performance is defined as tool functionality and describes how well a tool supports a specific task (Vonk and Ligtenberg, 2010).

4.3.1 The survey The survey consisted of 40 multiple-choice questions and was designed to take about 30 min to complete. This method of data collection was selected to (1) expose the tools to a large number of students and researchers, (2) ask in depth questions about the tools, and (3) test the tools with independent respondents. Students and researchers from Faculty of Earth Sciences from the VU University were contacted to complete the online survey. The survey consist of four categories of questions: (1) respondent characteristics, (2) communicative performance, (3) task functionality, and (4) user perception. Each question is accompanied with a map including a title, legend and map description (Fig. 4). Tool performance was assessed for each tool and for each dimension. The dimensions are map patterns, map relations, map change, tool selection and tool application. This means that a total of 20 questions was used to determine overall tool performance. The remaining questions constitute of respondent characteristics, a ranking question, perception on tool difficulty and an open question for respondents to leave comments. Communicative performance was evaluated in three dimensions of map interpretation: (1) map pattern, (2) map relation and (3) map change. The first two referred to static performance. Map pattern refers to the spatial pattern of the information. Map relation referred to how the various map layers lead to the map pattern. Map change referred to dynamic performance and referred to the extent that a change in map pattern was understood. The assessment of dynamic performance provided insight into the ability to use the tools in an interactive setting with dynamic attributes. Interactive maps provide opportunities for including spatio-temporal changes and allow user interaction with spatial data (McCall and Dunn, 2012). Task performance was evaluated by (4) tool selection and (5) tool application. The survey started with a short explanation of the stakeholder objectives and the relationships between the objectives and the physical conditions. The parcels were numbered to ask about tool information


CHAPTER 4

54

and characteristics of specific parcels. Tool performance was assessed with multiple choice questions of four options from which only one answer was correct. The answers to the multiple choice questions were labelled as correct (1) and incorrect (0) and were statistically compared to find significant differences in the performance of the respondents for the four tools with paired t-tests. All question had to be completed but each question contained a ‘do not know’ option to prevent gambling. The survey was developed with SurveyMonkey® (www.surveymonkey.com, last accessed February 2013), which is an online survey tool. It provides online questionnaire software to design, collect and analyse data. The final questionnaire was pre-tested to check if the questions were understood and to test the length of the survey. Access to the survey was distributed by e-mail.

4.3.2 Assignments First characteristics of the respondents were collected. Experience levels were scored on a 5-point scale ranging from very low (-2) to very high experience (+2). Experience with maps was divided into experience with maps in general, experience with land use maps and experience with GIS. Experience with maps was used to divide the respondents into ‘experts’ and ‘non experts’. Non-experts are those with experience level up to average on a 5-point Likert scale. Experts are those respondents that classified their experience level with maps as high or very high.

Fig. 4. Example of survey questions: a) map pattern, b) map change

Static communicative performance was assessed in map patterns and map relations. In terms of map patterns, the respondents were, for instance, asked to answer: ‘What is the nature score for parcels 10-13?’ (Fig. 4a). Next, a question was asked to find out if the respondent understood the underlying relations with a question such as ‘Why have parcels 3 and 4 got a low value for agriculture?’ The dynamic communicative performance was evaluated to determine whether a change in the map was understood after changes were made. Respondents were asked to name how the new map originated from the original map. For example: For which parcels resulted the land use change in an improved value for the agriculture objective (Fig. 4b). The last part of the survey evaluated how the respondents



55

linked the tools to specific tasks. Task functionality was evaluated in four questions about tool selection and four questions about tool application. Four tasks were formulated that were expected to be best supported with one of the tools. Figure 5 shows a question for one of the tasks. The respondents were asked to select the tool that was most appropriate to perform the task. Next, the respondents were asked to complete the task. If the respondents choose the associated tool and were able to complete the assignment, it was regarded to be a plus for task functionality. The order of the task assignments was changed randomly to prevent a learning effect. Finally, the respondents were asked to indicate their opinion towards the individual tools on a 5-point scale ranging from very difficult (- 2), to very easy (+2).

Fig. 5. Example of survey question to test task functionality

4.4 Geodesign tools This section describes the planning tasks that were formulated and the tools that were developed to support these tasks. First the tasks that were formulated are explained, followed by a detailed description of each of the tools that were assumed to support these tasks. Four geodesign tools were developed to present multiple stakeholder values differently. Section 5 describes the survey results.

4.4.1 Stakeholder tasks As the tools were developed based on stakeholder tasks, this section first describes the planning tasks that were formulated. A stakeholder task is the assignment that has to be accomplished during a planning stage. Spatial planning and decision making consist of multiple planning stages. Each stage is assumed to contain multiple stakeholder tasks (Eikelboom and Janssen, 2013). This study evaluated the influence of aggregation and ranking in presenting stakeholder values in geodesign tools. The variations resulted in the following stakeholder tasks: (1) assess the spatial pattern of the objective values, (2) identify bottlenecks, (3) find compromises and (4) discover trade-offs to support negotiation (Fig. 6).


CHAPTER 4

56

Fig. 6. Stakeholder tasks

The first task was the assessment of the spatial pattern of the objective values. Spatial planning typically starts with an exploration of the state of an area. In terms of stakeholder objectives, this means that each stakeholder searches for high and low scoring areas. Stakeholders that have interests in multiple objectives also search for areas with acceptable or alternative values for multiple objectives. For this task it was necessary to have information about each objective simultaneously for each spatial decision unit. The second task was the identification of bottlenecks. Bottlenecks are situations where change is needed. Respondents had to select areas that are sub-optimal or problematic. In case of multiple problematic areas and a limited budget, information on priorities is needed. Stakeholders need to know which regions have the lowest performance. Time and energy can also be saved when parcels that are close to optimal are excluded or neglected. The identification of outliers in the regions is a task that calls for ranking of the objective values. The first two tasks require separate presentation of each objective. The thirds task was the search for the best compromise for all stakeholders. This results in a direct advice on what to change in the interest of all stakeholders. This task is supported by a tool that shows the best compromise by combining the stakeholders in a predefined manner. For the fourth task respondents were asked to find parcels that are candidates for negotiation with other stakeholders. This meant that the task was to find information that supports the identification of desirable exchanges of land use. To support this task information is needed on which measure leads to the highest value for each of the stakeholders. The last two tasks required integration of stakeholder objectives in aggregated values. The tasks were operationalized in assignments such as ‘create extensive grassland when extensive grassland is the best land use’ or ‘raise the water level for parcels when the objective value soil has a 20% worst value’.

4.4.2 Tools The support these tasks four geodesign tools were made available: (1) objective value tool, (2) relative objective value tool, (3) total value tool, and (4) stakeholder value tool. Each tool was designed for one of the specific tasks. The four stakeholder tasks were: (1) Assess the spatial pattern of the objective values, (2) identify bottlenecks, (3) find the best compromise, and (4) discover trade-offs to support negotiation. Parcels were used as the spatial unit for evaluation with land use of water level as background. Within each parcel values were presented in one, two or three boxes. The tools were designed for dynamic use as the effects of a change on the value of objectives was shown immediately. The tools were constructed with Community Viz software version 4.3 (http://placeways.com/communityviz, last accessed December 2014). An overview of the tools is given in Figure 7.



57

Fig. 7. Visualization of four geodesign tools with the current land use as a background layer: a) objective value tool, b) total value tool tool, c) relative objective value, and d) stakeholder value tool


CHAPTER 4

58

The first tool is named the ‘objective value tool’. The value of an objective depends on land use and water level and each objective responds differently. When more than one objective is shown, a comparison between objectives can be made at the same time that land use or water level is changed. The objective values vary between 0 and 10 and are represented in three classes: worst (0-7), average (7,8) and best (9,10) ( Fig. 7a). Consequently, three red boxes indicate a low value for all objectives. Second, the ‘total value tool’ visualizes the best option for the stakeholders as one group and is a consensus driven approach. This total value is derived from weighting the stakeholder objectives for each land use. The total value tool has a different map lay out as it only shows one box instead of three (Fig. 7b). The background colour of each parcel represents the current land use. The colour in the box in the middle of a parcel shows the land use type that results in the highest total value. If the current land use is the same as the best land use no box is shown. The following weights were used to calculate total values (Table 1): Table 1. Objective weights

Total Agriculture Soil Nature 1.00 0.50 0.25 0.25

The ‘relative objective value tool’ shows a percentage of the best- and worst- scoring objective values. The tool can be seen as a reduced version of the first tool as it only shows relative values on the map. The aim is to have less information on the map so that selection of areas of interest will become easier or faster. For each objective the current values are ranked. Using this ranking the highest and lowest ranking parcels are identified. In this study the 20% highest and lowest were presented (Fig. 7c). The relative value of each objective is presented in three classes where red represents the lowest 20%, green the best 20% and white all intermediate parcels. The relative objective values depend on both land use and water level. After improving a parcel by changing either land use or water level, a new ranking decides which parcels have again the 20% highest and lowest value at that moment in time. This can be used to search for possible bottlenecks to identify areas in need of improvements. The amount of information to be processed by stakeholders is reduced as only the top and bottom 20% are presented. The final tool is the ‘stakeholder value tool’, which is linked to land use. The map shows which change is preferable for which stakeholder. Table 2 shows three stakeholders. Intensive farmers are assumed to be only interested in agriculture. Extensive farmers are assumed to also have an interest in soil and nature, while the stakeholder responsible for nature is assumed to have an interest in soil and nature only. By weighting the objectives the stakeholder values can be calculated. This tool shows potential values for each stakeholder: the potential value if the land use is changed to the preferred land use of the stakeholder. The map shows the best and worst 20% of the parcels for each stakeholder. The three boxes now represent the three land uses (Fig. 7d). The left box represents intensive grassland, extensive grassland in the middle and right nature. A threshold is specified say 20%, indicating the best and worst 20%. A box is red or green if it is with the worst or best 20%; otherwise it is white. These values are independent of the land use of a parcel. The land use type that leads to a high value is green and the land use type that has a low value is red. As a response users can change land use if the current land use is presented with a red box or if another land use type is presented with a green box. Table 2 shows how the objectives are linked to the stakeholders.



59

Fig. 8. Visualization of four geodesign tools with the current water level as a background layer: a) objective value tool, b) total value tool tool, c) relative objective value, and d) stakeholder value tool.


CHAPTER 4

60

Table 2. Stakeholders weights

Stakeholders Agriculture Soil Nature Intensive agriculture 1 0 0 Extensive agriculture 0.5 0.25 0.25 Nature 0 0.25 0.75

In summary, the tools differ in the amount of information, the number of calculation steps to process the underlying information, and the degree of ambiguity. The first three tools show information about three objective values with each three legend classes and provide information that leaves room for discussion. The last tool only shows a single value and directly suggests a change. This tool is more prescriptive to the decisions to be made. The tools differ from each other as the second tool includes an additional ranking. The third tool is even more complex as weighting was included, and the aggregation of the last tool results in less information on the map. The values are presented as boxes which leave the background map visible. Linking the boxes with different maps can provide insight into the influence of different factors on objective values. From Figure 7a, for example, it does not become clear why parcels 3 and 4 have a low value for agriculture. In Figure 8 the tools are shown on top of the current water level. Figure 8a reveals that parcels 3 and 4 have a high water level which decreases the value of agriculture.

4.5 Results This section presents the results of the survey. Four geodesign tools were developed to present stakeholder objectives. The tools vary in the way the values of these objectives are presented. Tool performance was divided in communicative performance and task performance. Communicative performance is defined as the ability to deliver information from the map to the user. Task performance is defined as tool functionality and describes how well a tool supports a specific task. The results of the survey are described in four steps: (1) respondents’ characteristics, (2) communicative performance, (3) task performance, and (4) overall performance.

4.5.1 Respondents The online survey was completed by 49 of the 78 respondents (completion rate of 63%). The respondents that finished questions about the first tool completed the survey. The respondents that dropped out, had already stopped after the first substantive question. The respondents were students (63%) and researchers (27%). They were experienced or very experienced with maps (51%), land use maps (35%) and GIS (22%). The average duration of completing the survey was 43 minutes, though it was not registered whether the respondents had small breaks between questions. The questions in the survey were found to be difficult by 39% of the respondents (e.g. those that checked difficult or very difficult on a five point Likert scale).

4.5.2 Communicative performance A distinction was made between static and dynamic performance. Static performance was assessed in two dimensions. First, the respondents were asked how they interpreted the map patterns. Secondly, they were asked about the underlying relations. Dynamic performance was tested by changing the land use in all four tools followed by questions about changes in the maps. For static performance, 22 respondents (45%) answered all questions correctly. From the 27 (55%) that had made mistakes, there were 15 respondents who made multiple mistakes. The 22% that were wrong on the relative objective tool all picked the same wrong answer. They interpreted the map as if it was the absolute objective tool.



61

For dynamic performance, 9 respondents (18%) were completely correct and 31 (63%) made only 1 or 2 mistakes. The scores of communicative performances for all respondents are shown in table 3. Table 3. Communicative performance rates for different geodesign tools. Performance is expressed as the percentage of correct answers (n=49). * indicates significant different performance compared to the other tools (p < 0.05)

Map type Performance dimension

Obj. value tool

Rel. obj. value tool

Total value tool

Stakeholder value tool

Static Static Dynamic

Map patterns Map relations Map change

96%

96%

96%*

76%*

80%*

59%

100%

67%*

61%

94%

92%

53%

The objective value tool includes no additional calculation steps. As expected this tool scored high on all categories. A score of 96% implies that two respondents (4%) gave the wrong answer. The tools that include a ranking step, the Relative objective value tool and the Stakeholder value tool, have lower rates. This is especially the case for dynamic performance. Map patterns were found most difficult to understand using the relative objective value tool as compared to the objective value tool, the total value tool, and the stakeholder value tool. Understanding the underlying relationships was easier for the objective value tool compared to the relative, and the total value tool. This suggests that aggregation as well as ranking decreases the ability to understand the relations that formed the map. The dynamic performance of the total value tool is low partly because of the relative high percentage that was indicated as ‘do not know’ and listed as wrong (12%), compared to 0-6% for all the other questions. The relative value tool and the stakeholder value tool use a percentage to calculate the best and worst parcels. This percentage is dependent on the planning task, for example, 20%, when the assignment is to allocate extensive grasslands in 20% of the area. It was tested whether the respondents understood the effects of a change in this percentage. The respondents were presented with a map based on different percentage and were asked whether they thought that the percentage was increased, decreased, unchanged or whether they had no idea. Only 51% of the respondents correctly understood in which direction the percentage had changed, 14% of the respondents had no idea and another 33% had the direction of change wrong. From this it can be concluded that tools presenting individual performance of spatial units are easier to understand compared to tools based on ranking of these units. The objective value tool and the total value tool were best understood based on dynamic performance.

4.5.3 Task performance Task performance was evaluated by two questions. First, respondents were asked to select the tool they found most suitable for a specific task. In a follow up question respondents were asked to apply the selected tool to complete the task. The results are shown in table 4. Table 4. Task performance rates for different geodesign tools. Performance is expressed as the percentage of correct answers (n=49). * indicates significant different performance compared to the other tools (p < 0.05)

Performance dimension Objective value tool

Rel. obj. value tool

Total value tool

Stakeholder value tool

Tool selection Tool application (/Total)

80%

71% 92%

86% 84%

73% 67%* 65%

Tool application (/Correct tool) 90% 93% 97% 97%


CHAPTER 4

62

Tool application (/Total) performance represents the percentage of the respondents that selected the correct tool and also applied it correctly. Therefore, it is never higher than the tool selection score. The last row, tool application (/Correct tool) shows the success rate as a percentage of the respondents that selected the correct tool. It was not possible to derive the correct answer with the wrong tool. Table 4 shows that in general respondents picked the right tool for the assigned task with the exception of the stakeholder value tool which was only selected by 67% of the respondents for the associated task. Although the objective value tool scored best on communicative performance it scores lowest on corrected tool application. This can possibly be explained by the amount of information that needs to be processed to perform the task. Although this tool is the easiest to understand it requires the most information to be processed. The tools that involve ranking, the relative objective value tool and stakeholder value tool, only present the best and worst parcels and therefore there is less information to process. In addition, the stakeholder value tool aggregates the information for each stakeholder. This could explain the high performance of this tool. From the respondents 47% selected the correct tool for all tasks, 33% made one mistake and 20% went wrong on multiple tools.

4.5.4 Overall performance The previous sections showed differences between performance rates for both communicative and task performance. The first relates to the dimensions (1) map patterns, (2) map relations, (3) map change, and the second results from (4) tool selection and (5) tool application. For each tool five questions were asked. If everyone was correct on all questions the total performance is 100%. The influence of each dimension on the overall performance is shown in Figure 9.

Fig. 9. Overall performances of the four tools and the contribution of each dimension



63

The summation of all performance questions provides information on the total performances for each of the tools. The tools differ in the level of aggregation and the use of ranking. The results in figure 10 show that the objective value tool has highest percentage of correctness and the stakeholder value tool has the lowest percentage. The stakeholder value tool includes ranking and aggregation and was therefore also expected to be the most complex.

Fig. 10. Influence of ranking and aggregation on tool performance (expressed as the percentage of correct answers)

The bars of Figure 11 show the dimension performances for each of the respondents ranked from low to high performance for the two levels of expertise. The respondents 1-25 are non-experts and 26- 49 are experts. Those with low performance rates have no correct answers for dynamic performance (map change) and task performance (tool application and tool selection). Performance scores range from 30 to 100%. Almost all people managed to answer more than half of the questions correctly but only two respondents reached a score of 100%.


CHAPTER 4

64

Fig. 11. Respondents’ performances for the five survey dimensions (n=49)

The overall performance is also influenced by the level of experience of a user. Comparing the total performance of the non-experts (M=14.8, SD=3.01) with the experts (M=17.0, SD=2.02) showed a significant difference t(23)=2.01, p=0.004.The performance rates are higher for experts compared to non-experts for all survey dimensions. In addition, the variance in the performance of the non-experts is higher (Fig. 12).

Fig. 12. Box plot of the overall performance of non-experts and experts (n=49)



65

At the end of the survey, respondents were asked to assess the difficulty of the tools using a five point scale from very easy to very difficult. The results show that the objective value tool was perceived easier (M=2.86, SD=0.91) than the relative objective value tool (M=3.22, SD=0.90), t(49) = 1.98, p < 0.05. No correlation with completion time and no correlation with experience could be found. From the performance scores it could be concluded that none of the tools was found to be too difficult (performances >72%). However, the static dimensions of communicative performance showed that the inclusion of ranking has a negative influence on the interpretation of map patterns and understanding underlying mechanisms (map relations) although the differences remain small. The assignment on the understanding of map change suggests the tool that included both ranking and aggregation was less suitable for interactive use. The analyses of the performances based on respondents characteristics indicated that the tools were better understood by users with some experience with maps, though were not too difficult for non-experts. In general, the tools without ranking were perceived easier compared to those including ranking or aggregation. The tools described were only used to present a maximum of three objectives. This is in accordance with the findings of Arciniegas et al. (2012) who concluded in an empirical analysis of the effectiveness of map presentations of stakeholder values that no more than three objectives should be presented simultaneously. Recently, Pelzer et al. (2014) qualitatively evaluated the perceived added value of planning support systems (PSS) by frequent users of a touch table device. The results of this study show that the practitioners found improved collaboration and communication as main advantage of the tools. The tools were specifically designed for interactive use. This required short calculation times and the need for explanation should be limited.


CHAPTER 4

66

4.6 Conclusions Geodesign tools are used to increase the effectiveness of spatial planning. However, effectiveness is a broad concept that can include many aspects. Only few studies explicitly tested the effectiveness of geodesign (e.g. Inman et al., 2011; Arciniegas et al., 2012). This study measured the effectiveness of geodesign tools as tool performance. Performance was considered high if a user is able to complete an assignment correctly using the information presented. Knowledge on tool performance is important for future tool use and tool design as it can provide arguments for selecting a specific type of tools or to design a new tool.

4.6.1 Ranking and aggregation The tools differed in the use of an aggregation or ranking step. Aggregation means that the values are weighted and summed into a total value. Aggregated values support stakeholders by combining multiple sources in a single attribute. Aggregation prevents a stakeholder from having to combine objectives themselves. Ranking means that the order of the value is used to select values. The result is information that only shows best and worst objective values. The tool that performed best on communicate performance was the objective value tool. This was the most simple tool without an additional aggregation or ranking step. Adding a ranking step lowered performance, especially dynamic performance. This was in line with the results on perceived difficulty of the tools. Performance on map patterns was found most difficult for the relative objective value tool. Understanding the underlying relations was easier for the objective value tool compared to the relative and total value tool. This suggests that aggregation as well as ranking decreased the ability to understand the relations that formed the map. Although the objective value tool scored best on communicative performance it scored lowest on tool application. This could possibly be explained by the amount of information that needed to be processed to perform the task. Although this tool was the easiest to understand it required the most information to be processed. The tools that involve ranking, the relative objective value tool and stakeholder value tool, only presented the best and worst parcels and therefore there was much less information to process. In addition, the stakeholder value tool aggregated the information for each stakeholder. The summation of all performance questions provided information on the total performance for each of the tools. The results showed that the objective value tool performed best and the stakeholder value tool had the lowest performance. The stakeholder value tool included ranking and aggregation and was therefore also expected to be the most complex. The overall performance is also influenced by the level of experience of a user. The average performance rates for dynamic and task performance were higher for experts compared to non-experts.

4.6.2 In conclusion Tools should be as simple as possible. Adding ranking and aggregation steps makes the tools more difficult to understand. On the other hand tools should also limit the amount of information to be processed by the user of the tool. This may well call for including ranking and aggregation steps. This stresses the importance of tailoring methods to tasks. Further research is needed to experiment with the tools in a workshop setting (Eikelboom and Janssen, 2015), to test the tools in practice (see for example Janssen et al. 2014) and to test the tools in different contexts (see for example Alexander et al., 2012). But maybe the most important characteristic of the tools is that they allow for trial and error. Steinitz (2012) emphasizes that this is very important as it increases the opportunity for experimentation and learning by doing.


Collaborative use of geodesign tools to support decision making on adaptation to climate change

67

CHAPTER 5 Collaborative use of geodesign tools to support decision making on adaptation to climate change

Abstract Spatial planners around the world need to make adaptation plans. Climate adaptation planning requires combining spatial information with stakeholder values. This study demonstrates the potential of geodesign tools as a mean to integrate spatial analysis with stakeholder participation in adaptation planning. The tools are interactive and provide dynamic feedback on stakeholder objectives in response to the application of spatial measures. Different rationalities formed by underlying internalized values influence the reasoning of decision making. Four tools were developed, each tailored to different rationalities varying between a collective or individual viewpoint and analytical or political arguments. The tools were evaluated in an experiment with four groups of participants that were set around an interactive mapping device: the Touch Table. To study how local decision making on adaptation can be supported, this study focuses on a specific case study in the Netherlands. In this case study, multiple different stakeholders need to make spatial decisions on land use and water management planning in response to climate change. The collaborative use of four geodesign tools was evaluated in an interactive experiment. The results show that the geodesign tools were able to integrate the engagement of stakeholders and assessment of measures. The experiment showed that decision-making on adaptation to climate change can benefit from the use of geodesign tools as long as the tool is carefully matched to the rationality that applies to the adaptation issue. Although the tools were tested to support the design of adaptation plans in a Dutch setting, the tools could be used for regional adaptation planning in other countries such as the development of RAS (Regional Adaptation Strategies) as required by the European Union or on a national scale to support developing NAPAs (National adaptation plans of action) as initiated by the United Nations Framework Convention on Climate Change (UNFCCC) for Least Developed Countries. Published as: Eikelboom T. and Janssen R. (2015) Collaborative use of geodesign tool to support decision making on climate adaptation, Mitigation and Adaptation Strategies for Global Change.


CHAPTER 5

68

5.1 Introduction Spatial planners around the world need to make decisions on adaptation to climate change. This could be adaptation to droughts, floods, pests and diseases, increasing forest fires. Dealing with these problems asks for local action. As a result, adaptation takes place at multiple spatial scales, in urban regions, mountainous regions, in both temperate as well as tropical climate zones. Recent examples are: increasing flood resilience in New York (Aerts et al., 2014), community-based water storage in semi-arid areas such as Ethiopia to adapt to climate change and mitigate household water shortages (Lasage et al., 2013), flood adaptation strategies for coastal cities such as Ho Chi Min City, Vietnam (Lasage et al., 2014), and developing climate resilience for alpine tourism (Wyss et al., 2014). In practice, climate adaptation is not the main goal of regional planning (Adger et al., 2007; Ford et al., 2011). The development of adaptation plans is a complex task from both an information processing as a process point of view. Especially, because the consequences of climate change are uncertain, multiple, complex, and controversial. Adaptive spatial planning is essentially a game of mutual gain (van Buuren et al., 2013). The climate adaptation process concerns a large number of stakeholders with different backgrounds and skills for processing the information. Stakeholders are those who influence a decision, as well as those affected by it (Hemmati and Enayati, 2002). The involvement of stakeholders in the planning process is increasing. Participatory approaches in environmental knowledge production are commonly propagated for their potential to enhance legitimacy and performance of decision-making processes (Hage et al., 2010). However, the involvement of stakeholders is generally costly and time consuming (McIntosh et al., 2008). Careful preparation work on planning activities is required for successful participation. Furthermore, participation must be organized as an explorative process to create operational collaboration (Celino, 2011). Collaborative planning is an interactive process of consensus building using stakeholder and public involvement. Margerum (2002) stressed out the importance of a pragmatic approach and a skilled facilitator. Another barrier for collaborative planning is the availability of data and methods to develop, assess, and select measures (Moser and Ekstrom, 2010). Improvements of the current organization of the spatial planning system are desirable to enable the realization of climate adaptation (van Buuren et al., 2013) as adaptation plans are largely underdeveloped (Preston et al., 2011). Improved climate-related decision making requires the acknowledgement that information may be scientifically relevant without being decision-relevant. The tools and scientific information that scientists consider as simple and useful are not always perceived that way by practitioners (Beunen et al., 2011; Kirchhoff et al., 2013). In addition, different actors perceive the usefulness of scientific information differently. Decision support characterized by one-way communication and a focus on products as opposed to process has been demonstrated to be ineffective (Weaver et al., 2013). Spatial information plays a key role in the design of adaptation strategies as climate change has spatial impacts (e.g. Wilson, 2006). In addition, the adaptation strategies themselves are spatial as they involve the spatial allocation of measures. The spatial information includes multiple layers of information ranging from detailed technical information to more general and sometimes qualitative information on development paths for a region. A geographical information system (GIS) is an indispensable tool for planners to design and visualize the effects of their decisions. The capabilities of a GIS can help to educate stakeholders about potential impacts of certain decisions on their objectives (Schatz et al., 2013). Multiple tools already exist that support the adaptation development process such as decision support systems (DSS) and planning support systems (PSS) (Geertman et al., 2013). Three recent examples are LandCaRe DSS, a vulnerability assessment model (Giupponi et al., 2013); a decision support system for urban climate change adaptation named SUDPLAN (Gidhagen et al., 2013); and a DSS for identifying and



69

exploring the potential of adaptations strategies to cope with flood risk (Ceccato et al., 2011). In addition, more general planning tools exist that do not explicitly refer to climate change, but which can be suitable for adaptation such as the ‘Online What If’ planning support system (Pettit et al., 2013). Wenkel et al. (2013) found from stakeholder workshops that for decision support systems in climate change science, tools should interactively communicate state-of-the-art knowledge, provide easy-to-use regional climate information and enable simulations of adaptation options. Geodesign tools combine geography with design by providing stakeholders with tools that support evaluation of design alternatives (Flaxman, 2010). The integration between spatial data, collaboration and decision-making in geodesign tools poses several challenges. Firstly, there is the social dimension of the problem as the technology must be perceived useful to be accepted. Secondly, emotional factors such as satisfaction and commitment to the process can play a role. Thirdly, there is the organizational dimension of the problem (Antunes et al., 2013). The behaviour of stakeholders can differ based on underlying internalized values that form the foundation of a rationality or scheme of reasoning. Different rationalities can result in different spatial designs based on the same empirical observations. Carton (2007) described two types of classifications to distinguish between rationalities; the viewpoint of actors and the type of reasoning. The division of viewpoint is between collective rationality, where the stakeholders have the willingness to engage based on a common ground, and individual rationality, which is about strategic behaviour of stakeholders to strive for their own benefit. The collective viewpoint builds on the concept of communicative rationality and the individual approach focuses on the behaviour of individual actors in safeguarding their values. The classification of rationality by the nature of arguments is in analytical and political reasoning. The reasoning behind decision making can be more analytical, where the focus is on objectives, or political which relies on the creation of benefits from stakeholder perspective. Political distinguishes itself by focusing on stakeholders separately, whereas the collective view combines stakeholders. Developing usable tools requires systematic research to better understand the science–practice interface. Experimentation with new scientific tools in practice helps to observe how practitioners respond and how the tools affect the social process (Opdam, 2013).

Fig. 1. Classification of four geodesign tools by rationality (based on Carton, 2007)

In this study, a case study was used to show how geodesign tools that are tailored to different rationalities to support the development of local adaptation plans. These tools support the identification of adaptation measures by providing feedback on different objectives and stakeholders interactively. The tools were designed based on different rationalities as it was expected that the collaborative use of tools differs between adaptation issues and involved stakeholders.


CHAPTER 5

70

Four geodesign tools were developed according to the classification of rationalities (Fig 1). The tools are named the ‘objective value tool’, the ‘relative objective value tool’, the ‘stakeholder value tool’ and the ‘total value tool’. Each tool is expected to suit specific adaptation issues based on the type of stakeholders that are involved, the adaptation development stage of the planning process, and the regional characteristics of the area and problem itself. The analytical and individual tool is expected to be useful for adaptation issues that involve few or collaborative stakeholders and more quantifiable issues such as crop change, whereas the collective- and political-based tools provide information from stakeholder perspectives. The tools were evaluated in an experimental setting to study how the tools and the associated rationalities influence the decisions of the planners and researchers in designing land use and water management changes in response to climate change.

5.2 Material and methods To study how regional decision making on adaptation can be supported, this study focuses on a specific case study in the Netherlands. In this case study multiple stakeholders need to make spatial decisions on land use and water management planning. The collaborative use of four geodesign tools was evaluated in an interactive experiment with two groups of researchers and two groups of planners. The participants were asked to use each tool to design spatial measures, in the form of land use and water management changes, to improve the value of one objective while at the same time minimize the decrease in the other two objectives. The designs were compared by the amount of measures initiated by the tool, the correspondence of the measures with tool information and the change in objective values. In addition, the communication and behaviour of the participants was observed and the perception of the participants was assessed in a survey containing 70 statements. The next sections illustrate the format of the experiment, provide tool descriptions, and describe how the results were compared.

5.2.1 Experiment Two experimental sessions were organized to evaluate the geodesign tools in a controlled setting. The advantages of an experiment in a controlled setting compared with a planning process are (1) time taken to introduce new tools to stakeholders, (2) the opportunity to ask them for feedback, and (3) allows for comparative studies by applying multiple tools in a row for the same task. The tools were integrated in an interactive decision support system. An interactive mapping device, Microsoft Surface 2.0 Touch Table, was used as the communicative platform, similar to previous studies (Alexander et al. 2012; Arciniegas et al., 2012; Janssen et al. 2014). The advantages of a Touch Table, such as learning by doing, availability of a geo-spatial database, and intuitive control, have been described in several studies (Arciniegas et al., 2011; Eikelboom and Janssen, 2013; Pelzer et al., 2013). Two simultaneous groups worked on separate devices to compare results. The first experiment was organized for researchers and the second experiment was organized for regional planners. The researchers were associated to environmental sciences related to peat meadow areas. The participants of the planners session were from Dutch provincial authorities and water boards involved in planning in peat meadow regions. Both groups are involved in peat meadow area but not in this particular study area. In total there were 14 participants. Before and after the experiment the participants filled in a survey. The opinion of the participants was assessed by 70 statements using a 5-point Likert scale. The categories of the survey were (1) personal characteristics, (2) decision context, (3) role of climate change, (4) information needs, (5) experience, (6) experiment feedback, and (7) tool feedback. The study area was a peat meadow area of about 50 km2 in the northern part of the Netherlands. The Province and Water board of Friesland have decided to develop a long-term adaptation strategy for the peat meadow areas of the province. Primary activities in this region are highly productive dairy farming,



71

nature conservation, recreation and housing. The region is currently mainly used for commercial dairy farming but is also important for its high natural, cultural and historical values. Important problems in the region are soil subsidence causing damage to buildings and infrastructure, deterioration of landscape values, inefficient water management, poor water quality, and the changing perspectives for dairy farming (Janssen et al., 2013). More details on application of the tools in peat meadow areas in the Netherlands can be found in Brouns et al. (2014) and Janssen et al. (2014).

Fig. 2. Location of peat meadow study area

Measures needed to be taken to address the problems of soil subsidence without too much damage to agriculture. Spatial measures such as land use change and changes in water management were found to be relevant adaptation measures. Agriculture, soil subsidence and nature were identified by the stakeholders as the three main objectives and their value was dependent on water level and land use. The effects of climate change were accounted for in the water level and in soil subsidence and a change in these variables resulted in a change of the objective values. Climate change was incorporated by using the W+ scenario from the Royal Netherlands Meteorological Institute (KNMI) and predicts a temperature increase of 2°Celsius and a modified atmospheric circulation, resulting in drier summers (Hurk et al., 2006). The change in subsidence rates was assumed to increase at a rate of 1.5 (Brouns et al., 2014).


CHAPTER 5

72

The tools were constructed with Community Viz software version 4.3 (CommunityViz, 2014). Figure 3 shows the graphical user interface, which was used by the participants during the experiment. The interface enabled real-time analysis of applied measures, but did not include sophisticated physical, social or economic models. Instead, the tools used an expert-based multi-criteria analysis to determine the effects of measures on multiple objectives. In addition, the values were aggregated to the level of parcels. The experimental sessions contained four rounds of 1 h each. The same assignment was used for each group and each tool. The assignment was to improve the value of one objective (soil subsidence) while at the same time minimize the decrease in the other two objectives (agriculture and nature). To achieve this the groups could change the land use or water management of the region. The changes in land use should be done in such a way that the totals for each land use stayed the same. In each round, the participants applied measures on the level of parcels. This resulted in a set of 16 maps with different spatial designs. During each round, participants experimented with applying measures as they were also able to undo measures or to adjust the measures based on tool feedback. The amount of potential designs was high as the possible measures could vary between three types of land uses and ten classes of water level change. The designs can be dominated by a single participant. On the other hand, the suggestions of a single person could as well be easily turned back by the other participants. For each group, the geodesign tools were used by the same group of participants to ensure comparability between tools.

Fig. 3. Graphical User Interface of a geodesign tool with the (1) map library, (2) bar charts of land use surface area and average objective values, (3) navigation toolbar and (4) design toolbar for applying measures. The map shows the start situation. The traffic lights represent the values of three objectives where red means low, white medium and green is high



73

5.2.2 Tool description Four geodesign tools were developed tailored to different rationalities. This means the tools were divided into individual or collective viewpoint and analytical or political reasoning. Political reasoning differs from the collective viewpoint as political treats each stakeholder separately, whereas collective concerns the interests of multiple (Carton, 2007). The individual tools strive for strategic decisions. The analytical tools present objective values. The collective tools combined stakeholder values using weighting factors to provide a common ground for stakeholders by suggesting locations for action. The political tools stimulate negotiation by showing preferred changes and can be used to make trade-off decision. Each tool fits one unique class of combination from the division of rationality in individual and collective reasoning, and the split in arguments for analytical or political tool use. The tools are named (1) the objective value tool, (2) the relative objective value tool, (3) the stakeholder value tool, and (4) the total value tool. The objective values were derived from expert-based look-up tables that used land use and water level as input parameters. The objective value tool was designed to support a decision-making process that is focused on analysing objectives at the individual level. For each parcel the absolute objective values are shown. These values change when water level or land use are changed. Each objective is presented as a traffic light symbol of three classes where red is low, white is average, and green is a high value. In Figure 4a, the left parcel is scoring high for the objective soil and low for agriculture and nature. The right parcel is scoring high for agriculture but low for the other two objectives. The difference between the objective values is caused by the difference in land use and water level. Conversely, the relative objective value tool aimed to support collective planning. The relative objective tool shows relative values instead of absolute values. The tool indicates the status of the parcel in relation to the values of the other parcels. In this way, relative differences can be observed even if the whole area has low values for an objective. The tool can be used to easily find parcels with high and low combinations which can be subject to change. This means that traffic lights are green if the objective value is high compared to other parcels or red if the value is low compared to other parcels. If the height of a value is average then the traffic light is white. A fixed border of the percentage of best and worst objective values is set. In this example the threshold was set at 20%. The 20% best are given a green light, the 20% worst are given a red light, and the 60% in between are white. When the objective value of a parcel changes due to water level of land use change, the ranking is updated and the distribution of red and green traffic lights changes. Although the objective value tool indicates that the agricultural value of the intensive grassland of the left parcel in Figure 4a is low, this value is still relatively high compared to other parcels in the area based on Figure 4b. Similarly, the agricultural value of the extensive grassland is relatively high. As planning involves multiple stakeholders, a tool that combines objectives by weighting according to individual preferences is the stakeholder value tool. This tool contains the potential objective values for each land use type. The left traffic light is intensive grassland, extensive grassland in the middle and the right traffic light represents nature. Weights for intensive grasslands were set 1.0 for agriculture, 0.0 for soil and 0.0 for nature. Weights for extensive grasslands were set 0.5 for agriculture, 0.25 for soil and 0.25 for nature. Weights for nature were set 0.0 for agriculture, 0.25 for soil and 0.75 for nature. The weights were set a priori based on expert knowledge. The participants were informed about the reasoning behind the weights and were given the opportunity to change the weights during the experiment, but they did not use this opportunity


CHAPTER 5

74

The traffic lights show whether a land use type potentially has a relative high value, a relative low value or in between. The traffic lights are static and independent of the current land use. The traffic lights that represent a land use type with a high potential total value is green. This tool stimulated participants to search for favourable swaps from red to green. If for example, the current land use is intensive grassland and this light is red, but the light of extensive grassland is green, a land use change from intensive to extensive grassland is favourable. Figure 4c shows for the left parcel that intensive grassland is the best scoring land use according to the current land use. On the other hand, the right parcel is extensive grassland but extensive grassland is classified as being one of the worst land uses for this parcel. The total value tool used weighting factors, defined by stakeholders, to combine objectives in collective values. It has a different map lay out as it only presents one single traffic light for each parcel instead of three (Fig. 4d). The background colour of each parcel again shows current land use. The traffic light shows the land use that results in the largest total value. If the current land use is the same as the best land use no traffic light is shown. Figure 4d shows that for the intensive grassland a change to extensive grassland would lead to the highest total value. The properties of the tools correspond to the demands of Pouwels et al. (2011) who purposed that tools should be built on interactions between functions, encourage interaction, allow incorporation of local knowledge, and generate output in the form of a map that shows where conflict areas and opportunities are located. More detailed descriptions of the tools can be found in (Eikelboom and Janssen, 2015).

Fig. 4. Visualization of four geodesign tools for two parcels that show: (a) absolute objective values, (b) relative objective values, (c) relative scoring of each land use based on stakeholder weights, (d) the land use change that is best for all stakeholders

5.2.3 Analysis of results The influence of geodesign tools on the use of tool information were analysed from the designs on the maps and from observations of the communication in the groups. The designs were evaluated by, the type of measures, the spatial distribution, the number of changes, the number of changes that correspond with tool information and the change in objective values. The last three were assessed numerically. The correspondence to tool information was defined as the number of parcels that were changed that fitted the information provided by the colour of the traffic light in that parcel. The percentages were expressed in relation to the total number of parcels in the area. A parcel that was changed in multiple steps is not counted separately. Only the final change map compared to the original map was used to calculate the number of changed parcels compared the total number of parcels.

Tool 1 – Objective value tool

Tool 2 – Relative objective value tool

Tool 3 – Stakeholder value tool

Tool 4 – Total value tool

(a) (b) (c) (d)



75

The objective values were expressed as absolute means for the whole area. The number of measures applied by the participants was used as an indicator for the functioning of the tool, the stimulating effect, and the perceived usefulness. The assumption was that a more difficult tool would use more time to be interpreted which would cause less time to be spent on allocating measures. In addition, the number of changes made with the tool provides an indication for the extent to which the tool stimulates change. The communicative potential of the tools was derived from the correspondence of the measures with tool information as that indicates whether the information as indicated by the tool was used for the decision. Furthermore, differences between the groups in using tool information were derived from observations and recordings. The responses to statements of the survey were averaged to discover statements with high impact and to compare the responses of planners and researchers.

5.3 Results and discussion First the quantitative results were provided as the percentage of changed parcels, the percentage of changes in accordance with tool information, and the mean objective values. Next, the results of comparing the sessions of the four different groups for the objective value tool were described on the basis of the four designs. Third, the observations of the experiment were described for each type of tool including an example of measures applied with each tool. Finally, the feedback of the participants was described.

5.3.1 Comparison of the spatial designs The amount of measures from each group and for the geodesign tools is shown in Table 1. The use of the objective value tool by the first group of researchers resulted in the most changes (66%). The other groups changed between 16 and 19% of the parcels with the objective value tool. The relative objective tool induced less changes for the second group of researchers and the second group of planners. Again the first group of researchers applied most changes. Also for the stakeholder value tool, the first group of the researchers applied most changes, though the amount of changes was only 7%, which corresponds to 25 parcels. The mean percentages demonstrate that the objective value tool induced most changes and the stakeholder value tool the fewest. Both tools were classified as individual, which in view of rationalities indicated that more changes resulted from the analytical tools compared to the political tools. The observations showed that participants made less use of the tools when they felt the tool was less useful. This finding corresponds to the theory of the Technological Acceptance Model (TAM) (Davis, 1985; Dias and Beinat, 2009). Table 1. The amount of measures from each group and for the geodesign tools expressed as the percentage of the area

Table 2 shows that the measures designed with support of the objective value tool and the total value tool were more in correspondence with tool information compared to the other tools. The rates of the objective value tool are above 70% for all four groups. The correspondence of the relative objective value tool is poorest (57%). This suggests that many changes designed with support of this tool were not based on tool information. The correspondence of the relative objective value tool is lower than the

Geodesign tool Researchers 1 Researchers 2 Planners 1 Planners 2 Mean 1) Objective value tool 66% 16% 18% 19% 30% 2) Relative objective value tool 21% 8% 18% 2% 12% 3) Stakeholder value tool 7% 2% 5% 2% 4% 4) Total value tool 10% 19% 6% 6% 10%


CHAPTER 5

76

objective value tool and the total value tool. In general, suggestions from the total value tool were followed which is reflected in a mean performance rate of 91% correctness. Table 2. Correspondence rate of measures with tool information from each group and for the geodesign tools

Linking the results does not show a relation between the amount of changes and the correspondence of the changes. As an example, ‘Researchers 1’ applied many changes according to tool information with the objective value tool as they changed 66% (Table 1) and of these parcels 81% (Table 2) were changed according to the tool information. Another finding is that despite changing a high number of parcels simultaneously, the correctness of the changes is still high. There are also examples with only a low percentage of change but with a high performance as well as examples of few changes with low performance. Both tables show a high variability between groups and tools. Furthermore, the average value for each objective was calculated as the area weighted sum of the values for each parcel. The values suggest that it was fairly easy to keep the value for agriculture above 0.80. In the current situation agriculture has a mean value of 0.85, soil subsidence is 0.49 and nature has a value of 0.08. The objective value tool supported the first group of planners to accomplish the highest score for soil of 0.61 and a score of 0.25 for nature as they combined extensive agriculture with high water levels. The same group again produced the highest value for soil (0.54) with use of the relative objective value tool. The political tools, stakeholder value tool and total value tool, did not or only minor result in water level changes, though the water level highly influences objective values. Consequently, no increases in objective values were achieved with the stakeholder value tool.

5.3.2 Comparison between groups The comparison of the maps supported with the objective value tool revealed different designs for each group of participants without consistent differentiation between researchers and planners. Figure 5 shows the original situation that was used to start the design session. The first group of researchers introduced a large area of extensive grassland (Fig. 6) and decided also to raise water levels on multiple parcels. In addition, the participants decided to move the maize parcels along the right side of a river as it was observed from the soil map that these were clay soils. Next, the second group of researchers noted that nature was fragmented and decided to cluster nature and moved it northeast. This group also focused on land use changes such as moving maize. The group also added different small groups of extensive grassland. Subsequently, the first group of planners introduced a buffer of extensive grasslands in the east as this area is adjacent to a large nature area that is located outside the study area (Fig. 7). However, they did not combine this measure with raising the water level. This group also focused on buffers of extensive grasslands and combined this with changes in water level. Furthermore, only a few nature parcels were removed. The group shifted a lot of parcels from scoring high on agriculture and low on soil and nature into low for agriculture and high on soil and nature. Finally, the second group of planners also introduced a large area of extensive grasslands and two smaller clusters. The group increased the water level for some deep wells based on the elevation map. No changes were made for maize or nature. Similarities in the designs can be observed for the first group of researchers and the second group of planners as well as for the second group of researchers and the first group of planners.

Tool Researchers 1 Researchers 2 Planners 1 Planners 2 Mean 1) Objective value tool 81% 70% 91% 91% 83% 2) Relative objective value tool 49% 60% 68% 50% 57% 3) Stakeholder value tool 73% 100% 78% 50% 75% 4) Total value tool 94% 99% 100% 72% 91%



77

Explanations for the different designs between groups and tools were found in variations of the observations. The first difference was observed in the time spent on tool interpretation. Some groups used more time to discuss what could be observed from the tool. Secondly, the duration of negotiating the allocation of measures varied. In some groups less negotiation preceded the design of measures. Related to this, the second observed difference was that some groups applied large clusters of measures, whereas others changed individual patches. The different designs changed the objective values as observed from the traffic lights. The objective values also differ due to changes in water level, which cannot directly be observed from the figures in this chapter but the water level maps were available in the application used during the experiment. Despite the design of different maps, the types of discussions were similar for the four groups. The same logic towards deciding on the type of measures was restated such as moving of the maize parcels, create higher water levels for intensive grasslands or combine the high water levels with extensive grassland, and cluster nature areas together. Only the spatial allocation of these measures differed for the four groups.

Fig. 5. Original land use map with objective values of start situation


CHAPTER 5

78

Fig. 6. Spatial design of the first group of researchers. The traffic lights show the updated objective values

Fig. 7. Spatial design of the first group of planners. The traffic lights show the updated objective values



79

5.3.3 Comparison between tools The previous section described differences between the participants for a single tool. Similar to the results of the objective value tool, the other tools also showed similarities in the type of applied measures. Each tool supported the group decision process differently varying between collective, individual, analytical or political rationality (Fig. 8). A review of how performance of planning support systems is reported (Te Brömmelstroet 2012) showed that most studies hypothesize about which dimensions were improved but did not report on measuring this increase. The designs in this study were judged quantitatively by the amount of changes and the correspondence of the changes to tool information. The resulting percentages suggested that some tools encouraged less measures. In addition, the performance of the measures indicated that some tools induced more changes that were not in correspondence with tool information. The performance of the measures with the tool tailored to collective and analytical rationality was poorest. This was either caused by the difficulty of tool information or by the use of expert knowledge to decide on measures. This section describes the typical observations for each tool.

Fig. 8. Percentage of the applied measures in correspondence with tool information

The following examples of measures illustrate the use of each tool (Fig. 9):

1. Objective value tool-raising the water level improved the objective agriculture and soil. 2. Relative objective value tool-the change of extensive to intensive grasslands was unfavourable. 3. Stakeholder value tool-allocation of nature to suited parcels with some connecting parcels. 4. Total value tool-nature was changed to intensive grassland according to tool information.


CHAPTER 5

80

Fig. 9. Examples of measures applied during the experiment with the original situation on the left and the changes on the right for each tool



81

The objective value tool corresponded to individual and analytical rationality. The use of the objective value tool resulted in changing large clusters of parcels where the focus was on the objective value of soil. Only little time was spent on discussing tool interpretation and negotiating the allocation of measures. In the example of Figure 9, the participants decided to raise the water level for parcels with three red lights. Comparison of the new situation (Fig. 9b) with the original (Fig. 9a) reveals that the higher water levels were beneficial for both agriculture and soil, but the objective value of nature was unchanged. Apparently, the original water level was too dry for agriculture, soil and nature. The new situation was still too dry for nature. The best and worst values obtained with the relative objective value tool were suited for more collective behaviour but was analytical, because it describes three separate objectives. Application of the relative objective tool resulted in changes for single parcels with a focus on relative low agricultural values. Much time was spent on discussing tool interpretation and negotiating the allocation of measures. The examples for the second tool showed that tool information was not always the main driver for changes. The participants changed some extensive grassland into intensive grasslands. The traffic lights indicated a relative high value for the three objectives. This means no suggestion for applying land use change is initiated by the tool. However, the participants decided to investigate whether the objective values would decrease when changing land use to intensive grassland. This is an example of experimental use of the tool. After changing to intensive grasslands, the traffic lights indicated that for the most left parcels the relative objective value for agriculture was not within the best anymore. The change of the traffic lights due to the application of measures was found to be confusing. The tool calculated a new subset of best and worst objective values. Next, participants changed two maize parcels to nature. One of the maize parcels has a green traffic light for the objective nature and the other two maize parcels are within the worst scoring parcels for agriculture (Fig. 9c). It is therefore reasonable to change these parcels to another land use type such as nature as performed by the participants (Fig. 9d). Also a single maize parcel was changed to extensive grassland. The maize parcel in the lower left corner has a relative low value for the three objectives. For such a parcel, the tool suggests that any change could be beneficial. Changing land use to extensive grassland improved the relative score for nature. From the observations of the conversations, it was found that this tool was less intuitive and resulted in more discussion on the response of the tool. This is reflected in more random changes to discover how the traffic lights respond in comparison to their expectations. The stakeholder value tool presents the interests of individual stakeholders and couples to political use. The use of the stakeholder value tool mainly led to single changes, but these changes concentrated on parcels with a combination of red and green traffic lights. In the study area only red, green, green parcels existed. The other parcels had either three blank, three red or three green traffic lights. Again, a lot of time was spent on discussing tool interpretation and on negotiation of allocating measures. The traffic lights suggested that intensive grassland was not the best scoring land use. In contrast, for some of these parcels the right traffic light indicated that nature is a potential high scoring land use. The participants used this explicit information to decide to change the parcels from intensive grasslands to nature (Fig. 9f). The total value tool used a summation, which implies a collective context, and used weighting of the objective values, which refers to political reasoning. The application of the tool showed that little time was spent on discussing tool interpretation, whereas more time was spent on negotiation of allocating measures. The designs were strategic and in clusters. The participants used the tool to look for sets of parcels where one suggests a change from A to B and the other suggests the reverse. Once the sets were finished no further measures were taken as the assignment was to preserve the amount of hectares per land use. In exchanging land use, the participants mainly designed clusters of changes. In addition to changes suggested by the tools, participants also made their own changes based on local knowledge or to test the response of the tool. This is reflected in the results of Figure 9h. Apart from the tool the


CHAPTER 5

82

participants also decided to add extensive grasslands. However, the tool showed that the total value was highest for intensive grasslands. For both analytical tools the participants mainly based their decision on a single traffic light. The relative objective value tool, collective and analytical, and the stakeholder value tool, individual and political, were subject to much more discussion about the interpretation of the tools which led to less changes. This was especially observed for the second group of planners.

5.3.4 Feedback on the tools from the participants Before and after the experiment the participants filled in a survey. The survey included 70 statements assessing (1) personal characteristics, (2) decision context, (3) role of climate change, (4) information needs, (5) experience (6) experiment feedback, and (7) tool feedback. A selection of the five strongest statements was made (both positive and negative deviations). These five statements were: (1) the tools support a first identification of measures, (2) the effects of changes visible in bar charts is useful, (3) usage of indicative values when no exact values are known is useful, (4) the tools supports finding measures to reduce soil subsidence, and (5) exchanging land use was easy. The decision context and role of climate change were studied to gain information on the institutional settings of the study area. Statements on the decision context revealed that the complexity of the decision process of soil subsidence does not so much relate to contradictory objectives, stakeholder disagreements, politics, the long time horizon or history. On the contrary, the statements towards climate change indicated the need to reduce uncertainties, create consensus on the occurrence of effects, and identify what types of adaptation measures exist, the consequences, their timing and location. In terms of information needs, the participants prefer to receive information about breaking points instead of trends. Currently prevailingly scenarios were used to illustrate the effects of climate change, however information is requested on what situation can no longer be maintained. Moreover, they like to see the differences compared to the current situation. The participants highly indicate their need on information on consequences of climate change (64%), possible measures (79%), effectiveness of measures (71%), and cost-benefits of measures (71%). To deal with the problems in this study, multi-criteria and cost/benefit approaches were preferred above design or expert based approaches. Although, the planners were less experienced with the use of a Touch Table compared to the researchers, the four groups indicated that the tools were easy to use. The average rate of the experiment on a scale from 1 to 10 was 7.5. The participants stated they would recommend the methodological concept to others. Furthermore, the experiment contributed positively to extending their knowledge on the possibilities of interactive spatial support tools (mean rate of 3.93 on a 5-point scale). Finally, participants were asked to provide feedback on the tools. The evaluation of the four different tools showed a preference for the objective value tool (78.6%) and the total value tool (35.7%) compared to 7.1% for both the relative objective value tool as well as the stakeholder value tool. As stated by one of the participants: “The analytical tools show the effects of measures on single objectives, whereas the other tools force integration.” Participants’ comments on the tools also provided suggestions for tool requirements, such as the need for a professional operator to control the use of a tool during a session. Furthermore, it was suggested to provide more explanations on how the traffic lights change differently for each type of tool. Also, information was requested about the stages of the planning process in which the tools can best be applied. In addition, consensus was needed between stakeholders on the weighting of values that was used in some of the tools. Some mentioned drawbacks of the tools were the time effort for applying the tools on new study areas, the fact that decision making also depends on aspects that were not included in the tools such as politics and finances and that the results were highly dependent on the formulas and models used in the tools.



83

Comments on the tools also provided suggestions for tool improvements. Currently the valuation does not incorporate spatial adjacencies of similar land uses in adjacent parcels. Suggestions on the appearance of the tools were to fade out parcels that are not subject to change and combine the tools with 3D visualizations and photos. Ideas on the content were to only show priority areas, to include a filter to select parcels with similar characteristics at once, and to include scenarios that show the influence of a single measure for the whole area such as increasing summer water levels. Furthermore, the statements were compared between planners and researchers. Table 4 shows the statements that had different responses for each group. Interestingly, planners agreed more on the statement that the maps in the touch table highly influenced their decisions. They also were more positive about that applying changes on the level of parcels was useful and that the tools supported the tasks. Table 4. Different responses to statements between researchers and planners (5-point scale)

Statements Mean researchers

Mean planners

Experience with touch table 3.10 1.50 Information availability of the touch table 2.70 3.75 Better use a touch table instead of paper maps 4.00 3.00 The maps in the touch table highly influenced the decisions 3.10 5.00 The information in the touch table was synoptic 3.30 4.25 The applied changes on the level of parcels was useful 3.30 4.50 The tools fit the spatial level of the problems 2.56 1.75 The tool support the tasks 3.44 4.25 Climate change is a large problem for the region 4.00 3.25

The selection of a tool should depend on the type of rationality that fits the adaptation problem. This is influenced by the area specific characteristics of the adaptation issue and the tasks involved in designing a plan for the region. These tasks vary during the process as each adaptation development stage has different tasks (Eikelboom and Janssen, 2013). The collective tools suit a communicative approach, whereas the tools based on individual rationality focused on the behaviour of individual actors in safeguarding their values. Next, there is the difference between the need for an analytical approach of the effectiveness of measures or a political approach that focuses on how stakeholders prioritize measures. As an example, a climate change driven decision for changing crop type due to, for instance drought or diseases, by a single farmer can be defined as an individual and analytical decision. The objective value tool can support this decision by providing insight into the influence of crop changes on single objectives. On the contrary, a regional dike reinforcement project is much more a collective endeavour. In this case, the relative objective value tool can provide support by visualizing the best and worst performing areas. The allocation of flood prone areas is certainly a political example of adaptation as the measure solves a larger scale problem at the expense of local interests. The stakeholder value tool can provide support by visualizing changes for each stakeholder separately as it uses weighting factors according to stakeholder preferences. In a planning situation where there is agreement on the relative importance of the different objectives, the total value tool is a tool that gives fast and specific suggestions on optimal land use changes.


CHAPTER 5

84

5.4 Conclusions Decision support to develop viable climate change adaptation strategies encompasses a wide range of measures and issues. Assessments of climate change impacts vary widely, depending on the situation (e.g. a natural resource/production system such as agriculture, or an economic activity such as investment in infrastructure development); time frame (e.g. near-term consistent with annual crop planning, or longer timeframe comparable to the design lifetime of road transport system); region and area (e.g. a trans boundary watershed or a single site); and purpose of the assessments (e.g. to raise awareness of climate change, or to inform the technical design of large/expensive infrastructure). Furthermore, successful adaptation not only depends on governments, but also on the active and sustained engagement of stakeholders including national, regional, multilateral and international organizations, the public and private sectors, civil society and other relevant stakeholders (United Nations Framework Convention on Climate Change, http://unfccc.int/adaptation/items/7006.php last accessed on January 10, 2015). This study described the application of four geodesign tools to support collaborative adaptation planning. The spatial designs that resulted from the use of the tools can contribute to the establishment of a climate resilience society as these local plans can serve as ingredients for adaptation to similar issues at larger scales or in other regions. Both planners and researchers considered the tools developed in this study useful at the scoping stage of an adaptation planning process. The learning by doing aspect of the tools was reflected upon as very effective. The tools were easy to use (Eikelboom and Janssen, 2015) and their application positively contributed to extending the knowledge of the participants. The results indicated that the choice for a tool influences the decision-making process as each tool yielded different designs of adaptation measures. The discussions revealed that for each tool similar logic was used by each group to decide on measures and that the participants tend to cluster changes. Little time was spent on tool interpretation for tools tailored to individual analytical and collective political rationalities, whereas much time was spent on discussing what could be interpreted from the other tools. The application of the individual analytical tool induced many changes while only little time was spent on negotiation of measures. Less changes were made using the remaining tools where more time was spent on negotiating the allocation of measures. Moreover, the individual analytical tool was preferred by both researchers and planners and was also found to induce many measures in correspondence with tool information. Therefore we argue that careful selection of methods and tools supports the development of adaptation plans and rationality can be used to choose between different geodesign tools. If the rationality behind the decision process is unclear, the analytical and individual approach would be best as the interpretation and use of this tool was found quick and easy. Improvements of geodesign tools as well as the users learning process must be seen as an interactive and iterative process. The communication between science and policy can benefit from further tool development by improving user friendliness such as the integration of urban strategy and phoenix (Dias et al. 2013), the inclusion of downscaled climate scenarios, adding information on costs, inclusion of filters to reduce the amount of information in the maps, to add scenarios of measures and improving modelling and visualization techniques to further tailor tools to specific planning tasks. Further improvements are needed in terms of the availability of the tools to a wider audience (e.g. web tool), and making the tool flexible for different scales and users. Next, the participants emphasized the need for a professional operator and indicated that additional explanation is needed on the exact interpretation of the traffic lights for each type of tool. Furthermore, the tools need constant updates on the latest findings and best available data. To conclude, the development and application of this type of tools is a process rather than a product (Wenkel et al. 2013).



85

Although the tools were tested to support the design of adaptation plans in a Dutch setting, the tools could be used for regional adaptation planning in other countries such as the development of RAS Regional Adaptation Strategies(RAS) as required by the European Union or on a national scale to support developing NAPAs (National adaptation plans of action) as initiated by the United Nations Framework Convention on Climate Change (UNFCCC) for Least Developed Countries.


CHAPTER 5

86


A spatial optimization algorithm for geodesign

87

CHAPTER 6 A spatial optimization algorithm for geodesign Abstract This chapter describes a genetic algorithm that can be used to generate land use plans that maximize both additive and spatial objectives in a vector-based GIS environment. To test the usefulness of the algorithm it was integrated in a geodesign tool and applied to a planning process in a peat meadow area in the Netherlands. The objective of this chapter is to demonstrate the potential and limitations of a genetic optimization algorithm to support collaborative land use planning workshops. The chapter shows how the algorithm can be used to: (1) generate a set of alternatives to start a decision process, (2) identify similarities and differences between collaborative planning results and results from optimization, and (3) as an interactive tool using feedback from stakeholders. It proved possible to generate a relevant set of non-dominated solutions to begin the planning process. Comparing results from the optimizer with stakeholder results demonstrated that both approaches generated plans with similar values for the objectives but with large differences.in the maps that were produced. Integrating the optimizer in a geodesign tool demonstrated how the optimizer can complement stakeholder input if it is used as an interactive geodesign tool. Collaborative planning is based on the assumption that the stakeholders have knowledge that is not, or even cannot be represented in a formal model. The challenge is to combine optimization with stakeholder input in such a way that both approaches complement each other to get the best of both worlds. Submitted: Eikelboom T., Janssen R., Stewart T. J. (2015) A spatial optimization algorithm for geodesign, Landscape and Urban planning.


CHAPTER 6

88

6.1 Introduction Combining expert knowledge with local knowledge in collaborative workshops is becoming common in land use planning. In the past, land use planners presented their information on large hard copies of maps and used sheets of tracing paper to add stakeholder information to the map (Burrough et al., 1998). The arrival of Geographical Information Systems (GIS) replaced the transparent maps by map layers presented within a GIS on a computer screen (Longley et al., 2005). This has now advanced further towards interactive map interfaces with direct interaction between stakeholders and underlying data. Parallel to this process has been a movement to combine the sketching approach, common in landscape architecture, with numerical analysis available in GIS (Bishop, 2013). This combination has recently been labelled ‘Geodesign’ and is defined as follows: “Geodesign is a design and planning method which tightly couples the creation of design proposals with impact simulations informed by geographic contexts, systems thinking and digital technology” (Steinitz, 2012, p.12). Combining the creation of design proposals with numerical analysis requires interactive approaches and opens the way to quantitative decision support techniques such as multi-criteria analysis and optimization. This may well lead to a revival of the use of optimization in the planning process as optimization systematically searches through the space of management options (Seppelt et al., 2013). This chapter presents a genetic algorithm for spatial optimization to be used in collaborative planning workshops. A current planning process to develop a long-term adaptation strategy for the peat meadow area of the province of Friesland is used as an illustration of the potential use of the algorithm. Results of the optimization are compared with the results achieved by stakeholders in collaborative workshops (see also Janssen et al 2014). The utility of optimization models to generate planning alternatives and to facilitate their evaluation and elaboration was already recognized in 1979 by Brill et al. and is still considered a promising method for generating alternative land-use designs, for further consideration in spatial decision-making (Ligmann-Zielinska et al., 2008). Using these algorithms creates analytical and computational problems. Geographical information systems make use of two types of data: attribute data and topology (Longley et al., 2005). An attribute table includes both the decision variables, e.g. types of land use, together with any attribute values which need to be included in the objective functions. Topologies are typically arranged using one of two models: vector or raster. Entities in vector format are represented by strings of coordinates (points). Two points can be connected to form a line segment, while sequences of lines can be connected end to end, returning to the starting point to form a polygon (parcel or area). Attribute data are stored for each polygon (which can be of varying sizes). Data in a raster model are stored in a two-dimensional matrix of uniformly sized cells on a regular grid. Depending on the model used, each grid cell is assumed to have homogeneous properties. By their nature raster data are substantially easier to include in mathematical representations for purposes of optimization. As a result most GIS-based applications of multi objective optimization use raster-based data as their input (Klein et al., 2013; Karakostas and Economou, 2013; Porta et al., 2013; Cao et al., 2011; Ko and Chang, 2012). Using a raster-based representation of a planning region Janssen et al. (2008) showed that it was possible to formulate a spatial planning problem in mathematical terms and apply multi objective optimization to generate optimal solutions interactively. Unfortunately, they also had to conclude that using a grid size which would realistically describe the planning region leads to unrealistically long computation times, as a result of the large number of decision variables. This



89

prevented use in a fully interactive setting where short response times are essential. Although it is clear that a vector representation leads to a more efficient representation of the problem, it is also clear that the switch from raster to vector creates new complications. In a raster each grid cell has the same shape and size, borders on exactly four other grid cells and has four borders of equal length. In a vector format each polygon can have any shape and size, and have any number of borders of various lengths. Janssen et al. (2008) differentiate between two types of objectives, namely: simple additive objectives, which associate costs and/or benefits with the allocation of any particular land use to a specific cell, then sum these cost and/or benefits across all cells; and spatial objectives which indicate the extent to which the different land uses are contiguous (i.e., the extent to which activities are or are not fragmented across the region). The shift from a grid to vector-based representation does not create great difficulties for the additive objectives. The differences in area size of the decision units can easily be accommodated using area size as a weighting factor in calculating overall performance. The shift from raster to vector is not so easily implemented for the spatial objectives, as we shall discuss in the next section. This chapter describes a genetic algorithm that can be used to generate land use plans that maximize both additive and spatial objectives in a vector-based GIS environment. The algorithm is a follow up to a similar algorithm developed for raster environments (Stewart et al., 2004; Janssen et al., 2008). A full mathematical description and numerical testing of the vector-based genetic algorithm can be found in Stewart and Janssen (Stewart and Janssen, 2014). The present chapter focuses on application of the algorithm. To test the usefulness of the algorithm it was integrated in a geodesign tool and applied to a planning process in a peat meadow area in the Netherlands (see also Janssen et al., 2014). The objective of this chapter is to demonstrate the potential and limitations of a genetic optimization algorithm to support collaborative land use planning workshops. The following questions will be addressed:

Can the algorithm be used to generate a relevant set of alternatives to be used to start the decision process? This could be alternatives linked to objectives of specific groups of stakeholders, but could also be intermediate alternatives or even the full set of non-dominated alternatives.

What are the similarities and differences between collaborative planning results and the results achieved by the algorithm and can these differences be explained?

How can stakeholder input be combined with results from the algorithm in an interactive design process to get the best of both worlds?

Section 2 provides a brief outline of the model developed in Stewart and Janssen (2014), describing the formal problem formulation in mathematical terms, the (generalized goal programming model and the numerical solution to this model by means of a genetic algorithm approach. Section 3 addresses the integration of the algorithm within the decision support system and describes the graphical interface between algorithm and user. Sections 4, 5 and 6 present results. Section 4 shows how the algorithm can be used to generate a set of non-dominated alternatives and to start the collaborative workshop. Section 5 focusses on the differences between interactive and computational design, where plans generated by the stakeholders in collaborative workshops are compared with plans generated by the algorithm. Section 6 combines stakeholder input with model results and describes the use of the algorithm in an iterative design process. Finally, sections 7 and 8 provide discussion and conclusions.


CHAPTER 6

90

6.2 The optimization model This section describes the structure of the Proximity Optimized Land Use Model. To suit interactive decision making, the optimization model has to permit land use changes, changes in restriction on total area for each land use, changes in objective weights, and the possibility to influence the importance of additive and spatial objectives. The primary decision problem is that of selecting a land use (out of a list of K possibilities) for each of N parcels of land defined in a GIS. It is convenient to represent the problem in terms of NK binary variables pk where 1=pk if and only if land use k is allocated to

parcel p ( otherwise 0=pk ). Each parcel is allocated one and only one land use. This implies the algebraic constraint:

Nppk

K

k,1,2,=for 1=

1=

(1)

Two further sets of constraints may be imposed to limit the choice of pk :

Bounds on total area allocated to a given land use: Let pa be the area of parcel p . The total area

allocated to land use k is thus given by pkpN

pa

1=. It is possible that lower and upper bounds on the

area allocated to land use k , say LkA and U

kA respectively, may need to be imposed, requiring then that choice of decision variables need to be subject to constraints of the form:

KkAaA U

kpkp

N

p

Lk 1,2,=for

1= (2)

Restrictions on specific combinations: In some cases, a parcel p may be unsuited to land use k , in which case the restriction 0=pk is imposed for this combination. Alternatively, the user may wish to

fix the land use allocated to parcel p to choice ; this may be achieved by specifying this parcel to be unsuitable for all land uses k .

6.2.1 Additive Objectives Any optimization study requires specification of the objectives to be optimized. In most cases of interest, there exist multiple objectives that need to be aggregated in some way. The easiest objective form is additive: for each of M concerns (often referred to as criteria), some form of value can be realized by the allocation of land use k to parcel p . Let pjkv be the value realized in terms of criterion j if land use k is allocated to parcel p . ‘Additive’ in this context, means that the overall value achieved

in terms of criterion j is obtained as the area-weighted average of the values in each parcel, i.e. jV defined by:

Mja

avV

p

N

p

ppkpjk

K

k

N

pj ,1,2,=for =

1=

1=1=

(3)

6.2.2 Spatial Objectives



91

A less directly quantitative type of objectives is to ensure that the spatial extent of each land use should be sufficiently compact to allow proper management. Various compactness measures have been suggested (e.g.,Aerts et al., 2003; Porta et al., 2013; Vanegas et al., 2011; Santé-Riveira et al., 2008). In Stewart et al. (2004) a series of cluster-based measures was derived from qualitative expressions of preference for different layout patterns, but these were based on a regular grid representation of the region. This cluster-based approach raised many implementation problems when extended to a more general vector-based GIS structure. With small adaptations, the model presented below is that of the proximity-based approach described initially in Stewart and Janssen (2014), in which it was demonstrated that the resulting land use clusters were more-or-less indistinguishable from those obtained when optimizing the cluster-based measures directly. The proximity model however is much more rapidly and easily implemented than the clustering approach. The proximity spatial objective is based on the centroid-to-centroid distances between pairs of land parcels (Fig. 1). Let ),( pp yx be the coordinates of the centroid of parcel p . The distance between any pair of parcels p and q was defined by the Euclidean distance between their centroids:

22 )()(= qpqppq yyxxd . Then, for each parcel p , all other parcels were ranked in order of

increasing distance from p , and select the nearest n parcels as proximate to parcel p . Denote the set of parcels proximate to p by p , and define p as the maximum distance pqd for all pq . In

effect, all parcels proximate to p have centroids within a circle of radius p around the centroid of parcel p .

Fig. 1. Illustration of proximity between four polygons (Stewart and Janssen, 2014)

The aim is then, for each type of land use, to maximize the degree of proximity of those parcels which are allocated this land use. See Stewart and Janssen (2014) for a formal definition of this measure of proximity.

6.2.3 Aggregation of Multiple Objectives


CHAPTER 6

92

In effect, there is a total of KM objectives. For ease of further explanation, the relevant achievement measures to be maximized are defined by iz for KMi ,1,2,= , where jj Vz = for

Mj ,1,= and kMz represents the proximity measure defined by Stewart and Janssen (2014) for

land uses Kk ,1,= . In this study, the objectives are all comparably scaled between 0 and 1. Aggregation of the objectives by weighted summation , i.e. to optimize a function of the form

iiKM

izwZ

1== , is a flawed approach in general, and can generate highly misleading results (Miettinen,

2008), p11 and p40. In brief, there are two reasons for this assertion: In non-convex problems (the spatial objectives will not in general be expected to be convex), there may exist efficient (potentially optimal solutions) which are not reachable by any weighted summation; Even for convex problems, the weighted sum approach implies constant preference tradeoffs at all performance levels, a property which is seldom true in general (for example, decreasing marginal values are typical). This study followed the interactive reference point approach as described in Stewart and Janssen (2014), but with a modification to what they term the scalarizing function to produce a smoother function which was found numerically convenient. For each objective, first an estimated ``ideal'' level, iI , was established. For the spatial objectives, a tight enough clustering will make the proximity objective close to 1, so that it suffices to set 1=kMI . For the additive objectives, the constraints and individual parcel

values may make 1=jV far from achievable. In order to obtain a more realistic estimate, the pjkv for the given objective j of all allowable combinations of p and k were sorted, and then select from the top (taking only the best suitable land use for each parcel), until all parcels are allocated. The resulting value of jV for this allocation is an estimate of jI . Then, for each objective, a ``reference point'' (or ``aspiration level'') was selected less than the estimated ideal. Let iA represent the aspiration level chosen in this way. An aggregated measure of goal achievement is then defined by the following scalarizing function to be minimized:

4

1==

ii

iii

KM

i AIzIWS

(4) Each of the bracketed terms in (4) has a functional form as illustrated in Fig. 2, which shows the relevant term for two possible values of the aspiration level when the underlying achievement level is scaled so that the ideal=1. Comparing the two curves it is clear how the term with a lower aspiration level is less than that for the higher aspiration at the same achievement level. However, when achievement for the objective with higher aspiration level is high enough, further increases in achievement may contribute less to the improvement in S than increases in achievement for the objective with lesser aspiration but at a currently lower current level of achievement. More specifically, the following features may be observed: 1. When the value of iz exceeds the aspiration level iA , further gains have very small marginal value, so that minimization of S avoids the problem which occurs with weighted summation in which extreme solutions often occur (very good on some objectives, while very poor on others). 2. On the other hand, the existence of positive marginal values, however small, ensures that the resulting solution is efficient (Pareto optimal).



93

3. As values of iz decrease below iA , an increasingly severe penalty is placed on further reductions, tending to generate balanced solutions. Clearly, a shift in the aspiration level modifies the point at which the higher penalties come into play. It is possible also to affect the solutions by changing the weights iW , which affect the magnitude of the penalties applied at the same relative underachievement of aspirations. Experience has however shown that the numerical results obtained are more sensitive to adaptation of the aspiration levels than to changes in the weights (Supplementary information).

Fig. 2. Scalarizing function terms for two possible values of the aspiration level relative to the ideal, namely 0.85 (corresponding to an objective with high aspiration) and 0.65 (moderate to low aspiration)

6.3 Implementation in a geodesign tool The optimization routine was implemented in a geodesign tool. The model parameters were linked to interactive buttons. The tool was constructed with CommunityViz 4.3 Scenario 360, which is an extension to ArcGIS (CommunityViz, 2015). A previous version of the geodesign tool was successfully applied to support decision making in various planning activities (Janssen et al., 2014). Tool performance was studied in various evaluation studies (Eikelboom and Janssen, 2015). Hommerts, a small area in The Netherlands is used to demonstrate the use of the tool (Fig. 3). This area is part of a planning process to develop long term strategies for the peat meadow areas of the Dutch province of Friesland (Janssen et al., 2014). The size of the area is about 50 km2 and is mainly used for commercial dairy farming, but is also important for its high natural, cultural and historical values. Important problems in the region are soil subsidence causing damage to buildings and infrastructure, deterioration of landscape values, and inefficient water management (Brouns et al., 2014). Options to change the area are land use changes, introduction of new crops and changes in water management.


CHAPTER 6

94

Fig. 3. Dutch peat meadow area ‘Hommerts’



95

The geodesign tool builds on the input of M * K value maps that represent the value of objective k for land use m. As an example, Fig 4. shows the value maps for three objectives for land use intensive grassland. In the application the value maps are combined in the visualization of traffic light boxes that provide information on three stakeholder objectives simultaneously. The traffic light boxes show the status of three objectives where red means a low value and green a high value. When users apply changes that influence the values of the objectives, the colours of the boxes change accordingly. This visualization can be used to evaluate the impact of changes directly and the advantage of the boxes is that they can be placed on top of different maps and that they visualize three objectives simultaneously. The graphical user interface shows (Fig. 5) allows the decision makers to design spatial measures and receive real-time feedback on the effects. For this study, the tool was extended with buttons to change the optimization constraints. Users can interact with the interface in four ways: A) change land use, B) apply water management measures, C) set optimization constraints D) start optimization. After each change the map is dynamically updated.

(a) (b) (c) Fig. 4. Value maps for land use intensive grassland for objectives agriculture (a), soil (b) and nature (c)


CHAPTER 6

96

Fig. 5. Graphical user interface of the geodesign tool that allows four types of interactions: A) land use changes, B) water management changes, C) set optimization constraints, D) start optimization



97

6.4 Non-dominated alternatives to start decision process

It is useful to start with a set of alternatives that provide an overview of the range of possible solutions. One possibility to do this is to generate a set of alternatives where each objective is maximized separately (single objective maxima) and combine this set with compromise alternatives. In this study the optimization algorithm was used to generate alternatives of single objective maxima for Mobjectives and one intermediate alternative based on equal weights. The objective weights iW and

additive objective aspiration levels iA were varied to generate single objective maxima for agriculture, soil and nature, and an intermediate alternative. The spatial aspiration level was set at its lowest value (0.4) because inclusion of spatial objectives decreases the importance of satisfying the additive objectives. Five different types of land uses were available: intensive and extensive grasslands, nature and wet crops hemp and algae. To provide an unrestricted decision space for the allocated amount of each land use no area constraints were applied. As expected, the generation of the single objective maxima with the optimization routine resulted in alternatives that each scored highest on the maximized objective. Fig. 6 shows that for each objective a different composition of land uses is preferred, also reflected in the allocated area for each land use (Table 1b). The intermediate alternative (Fig. 6d) results in intermediate values for all objectives. The alternative maximized for agriculture and the intermediate alternative have higher proximity scores compared with those alternatives maximized for soil and nature. This can also be observed from the maps (Fig.6). The result is a set of non-dominated solutions to begin the planning process. The range of alternatives allows users to explore the decision space. In addition, these alternatives can support discussion as stakeholders often know exactly what they do not want instead of having a precise idea of their preferred solution. Proposals of other stakeholders can have a similar role, but the use of an optimization model can accelerate this negotiation process by providing the solution space systematically.

Fig. 6. Optimization results of Maximized for agriculture (a), Maximized for soil (b), Maximized for nature (c), Intermediate (d)


CHAPTER 6

98

Table 1. Non-dominated alternatives: a) Objective weights, aspiration level and values, b) land use, minimal and maximum area, allocated area, aspiration level, proximity score

(a) Additive objectives (b) Spatial objectives

Alte

rnat

ives

Obj

ectiv

es

Wei

ghts

Aspi

ratio

n le

vel

Valu

es

Alte

rnat

ives

Land

use

Allo

cate

d ar

ea (h

a)

Aspi

ratio

n le

vel

Prox

imity

scor

e

Initial Agriculture Soil Nature

- - 0.80 0.24 0.05

Initial Intensive grassland 2168 - 1.00

Maximized for agriculture

Agriculture Soil Nature

0.8 0.1 0.1

0.95 0.50 0.50

0.78 0.33 0.13

Maximized for agriculture

Extensive grassland Intensive grassland Nature Wet crop hemp Wet crop algae

1146 1013 0 71 28

0.85 0.85 0.40 0.40 0.40

0.86 0.82 0 0.08 0

Maximized for soil Agriculture Soil Nature

0.1 0.8 0.1

0.50 0.95 0.50

0.45 0.48 0.33

Maximized for soil Extensive grassland Intensive grassland Nature Wet crop hemp Wet crop algae

548 95 696 576 254

0.40 0.40 0.85 0.85 0.85

0.54 0.05 0.65 0.73 0.62

Maximized for nature


0.1 0.1 0.8

0.50 0.50 0.95

0.32 0.48 0.40

Maximized for nature


0 80 649 686 753

0.40 0.40 0.85 0.40 0.85

0 0.18 0.66 0.57 0.59

Intermediate


0.33 0.33 0.33

0.95 0.95 0.95

0.61 0.38 0.18

Intermediate


652 639 475 240 162

0.85 0.85 0.85 0.85 0.85

0.84 0.81 0.75 0.75 0.74



99

6.5 Interactive or computational design? This section summarizes the differences and similarities between interactive and computational designs. The optimization routine was used to generate alternatives similar to the strategies that were developed by stakeholders during a planning process in Friesland, the Netherlands (Janssen et al. 2014). For both approaches, the value maps representing the performance of the three objectives for each land use drove the decision making process. However, in addition to the value maps stakeholders also included local knowledge and information that was only indirectly related to the objectives such as land ownership, recent investments and water management infrastructure. To make the information manageable for the stakeholders the three objectives were presented to the stakeholders in three legend classes, whereas the optimization routine used the absolute values to allocate measures. In the workshop stakeholders were asked to design a land use and water management plan for the following three policy strategies:

Business as usual: low impact technical measures only, no changes in land use; Parallel tracks: create buffers to separate conflicting functions such as agriculture, housing and

nature; New Horizons: introduction of new crops, large changes in land use and water management.

Each strategy implies different assumptions which were translated to qualitative objective weights. In the optimization constraints, these qualitative weights were translated to a numerical weight. The assignment for the first strategy ‘Business as usual’ was to apply low impact technical measures without changing land use. As a relatively simple measure the stakeholders decided to change the water level for intensive grasslands (Fig. 7a). To do this they grouped parcels according to existing physical boundaries and selected parcels with one or more red traffic lights. In the optimization, stakeholder results were used to set the constraints. (Table 2b). As this strategy aims to improve agriculture, the weight and aspiration levels were set high for both the additive objectives (Table 2a) and the spatial objectives (Table 2b). Table 2 shows that both stakeholder and optimizer changed about 600 ha. Both did this in such a way that high proximity scores were achieved. Both approaches resulted in similar values for the objectives. However, Fig 6 shows large differences in the spatial allocation of the measures: the stakeholders created four clusters of measures and the optimization routine grouped the changes in a single cluster (Fig. 7b). Testing the sensitivity of this result by reducing the spatial aspiration level in the optimization resulted in higher values of the objectives but lower proximity scores. Furthermore, removing the area constraints led to 1500 ha with a change in water level and an increased in value for all three objectives. This alternative could serve as a reference to show stakeholders that allocating a larger area would increase the values.


CHAPTER 6

100

(a) (b) Fig. 7. Strategy 1 ‘Business as usual’: stakeholder plan (a) and optimization result (b)



101

Table 2. Strategy 1 ‘Business as usual’: a) Objective weights, aspiration level and values, b) land use, minimal and maximum area, allocated area, aspiration level, proximity score


For the second strategy, ‘Parallel tracks’, stakeholders were asked to apply measures to separate different functions in the area. Houses in the area suffer from soil subsidence resulting from too low water levels and so stakeholders decided to allocate extensive grasslands with high water levels along the main road; this separates the houses from intensive grasslands with low water levels (Fig. 8a). In the optimizer this was incorporated as a land use restriction such that only along the road extensive grasslands could be allocated. Contrary to the stakeholders the optimizer allocated the extensive grasslands to the south (Fig. 8b). These parcels were found to contribute to higher values for all three objectives (Table 3a). The stakeholders did not perform a comparative analysis and therefore were not aware of this variance. Other, unknown, reasons may have determined their decision to limit extensive grassland to the northern part of the area. The alternative generated by the optimizer could serve as a reference alternative for the stakeholders.

Busin

ess a

s usu

al

Obj

ectiv

es

Wei

ghts

Aspi

ratio

n le

vel

Valu

es

Land

use

Min

are

a (h

a)

Max

are

a (h

a)

Allo

cate

d Ar

ea (h

a)

Aspi

ratio

n le

vel

Prox

imity

scor

e

Initial


- - 0.80 0.24 0.05

Intensive grassland 0 2168 2168 - 1

Stakeholders


+++ ++ +

- 0.78 0.35 0.12

No change Change in water level

1500 0

2168 2168

1598 571

- -

0.91 0.82

Optimization Agriculture Soil Nature

0.5 0.4 0.1

0.95 0.85 0.65

0.76 0.37 0.14

No change Change in water level

1500 500

1650 650

1515 654

0.85 0.85

0.95 0.88


CHAPTER 6

102

(a) (b) Fig. 8. Strategy 2 ‘Parallel tracks’: stakeholder plan (a) and optimization result (b)

Table 3. Strategy 2 ‘Parallel tracks’ a) Objective weights, aspiration level and values, b) land use, minimal and maximum area, allocated area, aspiration level, proximity score


Para

llel t

rack

s

Obj

ectiv

es

Wei

ghts

Aspi

ratio

n le

vel

Valu

es

Land

use

Min

are

a (h

a)

Max

are

a (h

a)

Allo

cate

d Ar

ea (h

a)

Aspi

ratio

n le

vel

Prox

imity

scor

e

Stakeholders Agriculture Soil Nature

++ ++ ++

- 0.78 0.26 0.06

Extensive grassland Intensive grassland

- - 284 1908

- 0.56 0.95


0.33 0.33 0.33

0.50 0.50 0.50

0.79 0.28 0.10

Extensive grassland Intensive grassland

227 1526

341 2289

354 1772

0.85 0.40

0.77 0.93



103

For the strategy ‘New horizons’, stakeholders were asked to make radical changes to the area by allocating new crops by changing water management and land use. Stakeholders decided to allocate substantial areas of reed and wet crops. They tried to limit the changes to only a few landowners and did this in areas with a high number of red traffic lights for agriculture (Fig. 9a). In the optimizer the areas allocated were included as area constraints. The optimization alternative resulted in similar objective values (Table 4a) and proximity scores (Table 4b) as the stakeholder plan. However, the maps show large differences (Fig. 9). It is clear that more than one solution exists. As the values of both plans are similar the result from the optimizer may serve as a point of reference for the stakeholders.

(a) (b) Fig. 9. Strategy 3 ‘New horizons’: stakeholder plan (a) and optimization result (b)


CHAPTER 6

104

Table 4. Strategy 3 ‘New horizons’: a) Objective weights, aspiration level and values, b) land use, minimal and maximum area, allocated area, aspiration level, proximity value


New

hor

izons

Obj

ectiv

es

Wei

ghts

Aspi

ratio

n le

vel

Valu

es

Land

use

Min

are

a (h

a)

Max

are

a (h

a)

Allo

cate

d Ar

ea (h

a)

Aspi

ratio

n le

vel

Prox

imity

scor

e

Stakeholders Agriculture Soil Nature

++ +++ +

0.82 0.74 0.62

Intensive grassland Reed Wet crop hemp Wet crop algae

- - 1440 373 143 212

0 +++ +++ +++

0.88 0.84 0.75 0.72


0.3 0.6 0.1

0.85 0.95 0.500

0.83 0.77 0.60

Intensive grassland Reed Wet crop hemp Wet crop algae

0 0 0 0

1500 500 200 200

1432 403 134 199

0.40 0..85 0.85 0.854

0.90 0.89 0.70 0.68

6.6 Iterative design This section demonstrates how the optimizer can be combined with user feedback to generate plans in an iterative way. In each step the optimizer presents a plan to the user. The user evaluates the plan and based on his/her preferences and local knowledge sets specifications and restrictions for the next round. The principle of this approach is shown in the example presented in Fig 10. For practical reasons only three steps are shown. In the first step the optimization algorithm generated an intermediate alternative with low spatial proximity aspiration. In the second step the spatial proximity scores were increased. In the third step both spatial aspiration and objective aspiration levels were raised. Step 1: In the first iteration, land use was restricted to 1/5 of the total area and determined by dividing the total area by the number of land uses. In addition, equal objective weights and low spatial aspiration levels were set. This resulted in an intermediate alternative where land uses were scattered. Step 2: In response, the user selected parcels and restricted these for wet crop algae as allocated by the optimizer (shaded parcels). Also the users raised the spatial aspiration level of nature and wet crops to cluster these land uses. This results in plan 2 with a cluster of wet crops around the restricted parcels and a cluster of nature in the north. This decision led to higher proximity scores and a slight decrease in objective values. Step 3: The users responded by restricting the parcels where farmers invested recently to remain intensive grassland (shaded area). The objective aspiration levels were raised to compensate for the decrease in values due to the second step. In addition, the users decrease the upper limit of nature area. The generated result (plan 3) is an alternative that has both higher objective values as well as higher proximity scores. This example demonstrated how the optimization model can complement the interactive geodesign tool. Stakeholders can either accept, reject or adjust the results of the optimization model and thereby incorporate their preferences and local knowledge.



105

Map Feedback

Plan 1

Start: Five land uses each with the same area. Objectives: Agriculture 0.54, Soil 0.42, Nature 0.28. Average proximity score of 0.49 (SD 0.03).

Plan 2

User input: A number of parcels are restricted to wet crops type algae (shaded area). Selection is based on the optimization result, local preferences and location (e.g. close to water). High spatial aspiration levels for nature and wet crops were added. Result: Plan 2 with equal amount of land uses. Value of the objectives slightly decreased Objectives: Agriculture 0.53, Soil 0.41, Nature 0.26. Average proximity score increased to 0.60 (SD 0.12).

Plan 3

User input: The users restricted parcels in the north to intensive grassland based on local knowledge (recent investments made by the farmer). The objective aspirations were raised. Result: Map 3 shows an increase in both values of the objectives as of the average proximity score. Objectives: Agriculture 0.57, Soil 0.41, Nature 0.27. Average proximity score of 0.69 (SD 0.04).

Fig. 10. Iterative design with the use of optimization in three steps


CHAPTER 6

106

6.7 Conclusions This chapter addressed application of a genetic algorithm that can be used to generate land use plans that maximize both additive and spatial objectives in a vector-based GIS environment. To test the usefulness of the algorithm it was integrated in a geodesign tool and applied to a planning process in a peat meadow area in the Netherlands . The chapter showed how the algorithm can be used to (1) generate a set of alternatives to start a decision process, (2) identify similarities and differences between collaborative planning results and results from optimization, and (3) as an interactive tool using feedback from stakeholders. It proved possible to generate a relevant set of non-dominated solutions to begin the planning process. The results showed that generating a set single objective maximum alternatives combined with compromise alternatives provides the stakeholder with a set of alternatives that covers the range of possible solutions. Previous workshop experiences demonstrated that stakeholders first explore extreme situations to find out the influence of large changes. This exploration stage can be structured and accelerated using the optimization routine. When maximizing objectives, it has to be kept in mind that the spatial and additive objectives compete in achieving the best results for the area. Comparing results from the optimizer with stakeholder results demonstrated that both approaches generated plans with similar values for the objectives but with large differences in the maps that were produced. It is clear that there is no single optimal plan and that more than one solution exists. As the values of the objectives are similar to the stakeholder plans the result from the optimizer could be used as a point of reference for the stakeholders. One of the challenges in using an optimization for interactive design is to be able to explain the model assumptions such that the model results are accepted and found valid by the stakeholders. The model provides the user only with a result instead of the road to it. Integrating the optimizer in a geodesign tool demonstrated how the optimizer can complement stakeholder input if it is used as an interactive geodesign tool. Stakeholders can either accept, reject or adjust the results of the optimization model and thereby incorporate their preferences and local knowledge. The genetic algorithm has short response times which is a major prerequisite for interactive use. To conclude: collaborative planning is based on the assumption that the stakeholders have knowledge that is not, or even cannot be represented in a formal model. The challenge is to combine optimization with stakeholder input in such a way that both approaches complement each other to get the best of both worlds.


Spatial analysis of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change, and adaptation

107

CHAPTER 7 Spatial analysis of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change, and adaptation

Abstract Dutch peatlands have been subsiding due to peat decomposition, shrinkage and compression, since their reclamation in the 11th century. Currently, subsidence amounts to 1–2 cm/year. Water management in these areas is complex and costly, greenhouse gases are being emitted, and surface water quality is relatively poor. Regional and local authorities and landowners responsible for peatland management have recognized these problems. In addition, the Netherlands Royal Meteorological Institute predicts higher temperatures and drier summers, which both are expected to enhance peat decomposition. Stakeholder workshops have been organized in three case study areas in the province of Friesland to exchange knowledge on subsidence and explore future subsidence rates and the effects of land use and management changes on subsidence rates. Subsidence rates were up to 3 cm/year in deeply drained parcels and increased when we included climate change in the modelling exercises. This means that the relatively thin peat layers in this province (ca 1 m) would shrink or even disappear by the end of the century when current practices continue. Adaptation measures were explored, such as extensive dairy farming and the production of new crops in wetter conditions, but little experience has been gained on best practices. The workshops have resulted in useful exchange of ideas on possible measures and their consequences for land use and water management in the three case study areas. The province and the regional water board will use the results to develop land use and water management policies for the next decades. Published as: Brouns K., Eikelboom T., Jansen P.C., Janssen R., Kwakernaak C., Akker J.H.J. van den and Verhoeven J.T. A. (2014). Spatial analysis of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change and adaptation. Environmental Management. 10.1007/s00267-014-0392-x


CHAPTER 7

108

7.1 Introduction Centuries of drainage and peat cutting have resulted in a major loss of peat soils in the Netherlands. Peatland ecosystems once covered a major proportion (40 %) of the Dutch land surface, but the area of peat soils has been reduced to less than 10 % since drainage started in the 11th century (Schothorst, 1977; TNO, 2007). Drainage has enabled agricultural use by optimizing oxygen and nutrient availability and has allowed access of heavy machinery. However, deep drainage to facilitate intensive agriculture currently causes rapid soil subsidence, generally up to 2 cm/year (MNP, 2005; Querner et al., 2012), which has to be followed by regular adjustments of the water level. Such peat soils are also emitting on average 19 tonnes of CO2/ ha/year (van den Akker et al., 2008) and often lead to poor surface water quality (Beek et al., 2007). Subsidence of peat soil is the result of a combination of processes, i.e., shrinkage, compression, and oxidation, which are all caused by lowering of the water table. The shrinkage process is a reduction in volume caused by the withdrawal of water from the upper soil layer. The loss of the buoyant force of water in the upper layers also leads to the compression of deeper layers. The microbial oxidation of soil organic matter under oxic conditions leads to a major peat loss after drainage. In fact, up to 85 % of subsidence can be attributed to oxidation (Schothorst, 1977). It has been predicted that the peat areas will subside between 40 and 60 cm between 1999 and 2050 (Hoogland et al., 2012). Moreover, continuation of the current land use in the Dutch peat meadow areas will lead to the disappearance of most peat within 200 years and all peat within 500 years (Querner et al., 2012; Rienks et al., 2002). The national government, provinces and water boards increasingly realize that a continuation of this management will have problematic side effects such as damage to building foundations, desiccation of nature reserves, emission of greenhouse gases, increasing costs for water management and infrastructural maintenance, deterioration of surface water quality and, finally, loss of the characteristic landscape. In the Dutch peat areas, the regional governments are reaching out to stakeholder groups representing the various interests (farmers, recreation entrepreneurs, nature conservation agencies) to discuss the future of the peat areas and the adaptation measures needed for a sustainable management avoiding very high costs and threat to human settlements from potential flooding at later stages. The problems indicated have been familiar to regional and local stakeholders for decades; however, recent concerns about, and research into, the effects of climate change have suggested that the situation will most probably be aggravated in the near future. Besides a rise in average temperatures of 1–2 °C in the period 1990–2050, two of the climate change scenarios put forward by the Netherlands Royal Meteorological Institute (i.e., the G+ and W+ scenarios) predict drier summers due to a change in atmospheric circulation (Hurk et al,. 2006). The predicted temperature rise, changes in precipitation, and the more frequent occurrence of extreme episodic events will potentially have strong additional effects on organic matter decomposition (Hellmann and Vermaat, 2012; Laiho 2006; Querner et al., 2012; Witte et al., 2009). Experimental studies have overwhelmingly shown that soil organic matter decomposition increases at higher temperatures. A temperature increase of 10 _C usually leads to a tripling of the peat decomposition rates (Berglund et al., 2008; Dorrepaal et al., 2009). Furthermore, it is plausible that dry summers enhance long-term decomposition rates (Fenner and Freeman, 2011). It has been estimated that the combination of temperature rise and lower groundwater levels in summer in peat meadow areas will lead to a 70 % increase in subsidence in the W+ scenario (Querner et al,. 2008). Here, the current management strategy of regularly adapting water levels to the subsiding land surface is being re-evaluated in the context of the development of a policy plan by the province of Friesland for future management of their peat meadow areas. In order to facilitate the decision making and increase



109

the public support for land use and water management changes, spatial visualizations of the effect of climate change on soil surface level were generated and the consequences for agriculture, nature, and environmental targets were analysed and discussed in stakeholder workshops in three case study areas. These meetings involved the use of a spatial model presented on an interactive mapping device (the ‘touch-table’, Fig. 1) which was used as a common interface. This allowed the participants to learn about the mechanisms behind soil subsidence and the consequences. The participants were asked to change land use and water levels to observe the impacts on soil subsidence and to inspect relevant map layers such as expected agricultural yields (Janssen et al., 2013, Janssen et al., 2014). This chapter summarizes the way in which spatial information on peat soil characteristics, subsidence rates, and water resource management was presented to stakeholders. It also summarizes what overall conclusions can be drawn from the peat responses to different climate scenarios and from the effectiveness of various adaptation measures. Moreover, this chapter draws conclusions on future land use options.

Fig. 1. Use of the ‘touch-table’ during the workshop in Hommerts


CHAPTER 7

110

7.2 Materials and Methods

In each study area, a workshop was conducted to provide detailed, spatially explicit knowledge to the local stakeholders. The stakeholders were able to apply measures in the form of land use and water management changes using the ‘touch-table’ tool. A questionnaire was used to identify the opinion and knowledge gains of the stakeholders (Janssen et al., 2014). A total of 19 stakeholders representing different interest groups and authorities (agriculture, recreation, water board, province, municipalities, nature conservation organizations) participated in the workshops.

Fig. 2. Soil surface level in relation to average summer ditch water level (GDL), results of long-term monitoring in Zegveld (van den Akker et al., 2007)

7.2.1 Modeling soil subsidence Soil subsidence maps have been generated by modelling spatially explicit information on soil, land use, and ditch water and groundwater levels. Factors determining the rate of soil subsidence are the presence and thickness of peat, the presence of a clay cover, and the height of the ditch water level relative to the soil surface. The models provided in the Guidelines for Soil Protection were taken as the starting point (Akker et al., 2007). These models are based on long-term data of land subsidence in the western and northern Netherlands (Fig. 2). Furthermore, the soil maps of the province of Friesland are ca. 40–50 years old. For the workshops, we were able to use the still unpublished new soil maps produced by the research institute Alterra, which is linked to the Wageningen University and Research centre.

7.2.2 Groundwater level Detailed information on the relationship between soil subsidence and ditch water level has originated from the experimental farm ‘Zegveld’ in the province of Utrecht (Akker et al., 2007). At the experimental farm, two pairs of blocks were created in 1969 with ditch water levels of 35 and 70 cm below ground surface, respectively. We define this vertical distance between Ground surface and Ditch water Level as GDL. Later on these water levels were adjusted to 20 cm (high GDL) and 55 cm (low GDL), respectively. The monitoring results for the period 1972–2006 (Fig. 2) show that the difference in subsidence between the plots with high and low GDL has been consistent, i.e., 4.4. and 11 mm/year, respectively (Akker et al., 2007). The relationship between subsidence and drainage depth as defined in Zegveld is confirmed by a long-term study of subsiding peat meadows in Friesland. In the period 1920–1960, the average land subsidence was 5 mm/year, while it increased to an average of 12 mm/year between 1960 and 1995 after increasing the GDL (Nieuwenhuis and Schokking, 1997). Another



111

example is a study in which data on surface water levels and land subsidence from the Frisian peat meadow areas were shown to be significantly correlated (Janssen, 1986). In the Netherlands, groundwater levels are generally being monitored twice a month. The three highest (HG3) and the three lowest (LG3) groundwater levels measured in each year are averaged. The mean highest and lowest water table (MHW and MLW, respectively) are then determined as the mean HG3 and LG3 for at least 8 years, respectively. Research has shown that the MLW provides the best correlation with soil subsidence rates (Akker et al,. 2007), because this indicates the thickness of the peat that is being oxygenated in summer (Fenner and Freeman, 2011). In general, the lowest groundwater levels occur in August and September, coinciding with highest soil temperatures, providing optimal conditions for peat oxidation in during these months (Hoving et al., 2004; Wesseling, 1985). Although MLW corresponds best with subsidence rates, we used the average Groud-Ditch Level distance (GDL) in the workshops with local stakeholders. This term is easier to implement and interpret than MLW and still gives reliable estimates mimicking the models based on MLW (van den Akker et al., 2007). Hydrological modelling was done with the SIMGRO model by research institute Alterra; this model quantifies the hydrology on a local scale. In workshops with local stakeholders, however, we used the average Ground-Ditch level distance (GDL). The equations given in Table 1 are based on the long-term monitoring data described above and on expert judgment on the quantitative effects of the major factors driving the subsidence rates, (i.e., presence of a clay cover, MLW, and GDL). For the modelling exercises in this study, the equations expressing the relationship with GDL have been used. Table 1. Equations for land subsidence of grassland parcels (S, mm/year) in relation to the average lowest groundwater level (MLW, cm) and related to the vertical distance between ground level and ditch level (GDL, cm) for peat meadows with or without clay cover

Conditions Relation to Equation Remark Peat groundwater level (ALGL, cm) S(i)=0.2354*ALGL-6.68 Generally applicable in the

Netherlands Peat with clay cover

groundwater level (ALGL, cm) S(i)=0.2354*ALGL-10.47

Peat ground-ditch level (GDL, cm) S(i)=0.538*GDL0.776 Adapted equations used for 'touch table' workshops in Friesland Peat

with clay cover ground-ditch level (GDL, cm) S(i)=-4*10-6*GDL3+12*10-4

*GDL2+439*10-4*GDL)

7.2.3 Clay cover Due to alternating periods of rapid and slower sea level rise during the Holocene, periods of undisturbed peat formation alternated with periods where the peatlands were flooded by nearby rivers or estuaries and covered by a clay layer. Peat meadows with presence of a clay cover have a slower subsidence because oxygen intrusion in the peat soil is limited below the clay layer, and overall less organic material is present to decompose (Fig. 3) (Akker et al., 2007). In the soil subsidence models, a distinction was made between peat and peat with clay cover, where the clay cover thickness is less than 40 cm. Soils with a clay cover thicker than 40 cm are not considered peat soils in these workshops as their subsidence is marginal.


CHAPTER 7

112

Fig. 3. Comparison of subsidence of peat meadows with and without clay cover, adapted from van den Akker et al. (2007)

7.2.4 Land use, peat type and thickness All study areas are characterized by grassland parcels, with occasionally some fields where maize is grown. Based on measurements in Friesland, the subsidence rates of arable land have been estimated to be 1.5 times faster than that of grassland (Janssen, 1986). The peat soils in the Netherlands are composed of various peat types, ranging from fen peat, which was formed in eutrophic conditions, to bog peat, which was formed in oligotrophic conditions. While the fen peat has a slightly faster rate of decomposition, it also has a higher bulk density, so that land subsidence rates of nutrient-rich and nutrient-poor peat are almost equal (Janssen, 1986; Akker et al., 2007). Peat thickness is taken into account in the model predictions of land subsidence. It was assumed that once the MLW has become deeper than the thickness of the peat layer, subsidence would not increase if a further MLW drop occurs.

7.2.5 Climate change scenarios The Royal Netherlands Meteorological Institute has constructed four climate change scenarios for the Netherlands, which have an equal probability of occurrence (Hurk et al., 2006). The G and G+ scenarios predict a moderate temperature change (+1 °C), the W and W+ scenarios predict higher temperature (+2 °C). The ‘+’ indicates a modified atmospheric circulation, resulting in drier summers. As a result of higher temperature and drier summers, the W+ scenario has been modelled to result in an increase up to 70 % for soil subsidence rates in peat meadow areas with peat lacking a clay cover in 2050 (Querner et al., 2012). In addition, substantial amounts of surface water need to be supplied to prevent desiccation of the peat (Querner et al., 2012). Pressures on freshwater resources are, however, increasing and the question is whether sufficient good-quality freshwater will be available in future (MNP, 2005). The study presented here is based on the W+ scenario. For the purpose of our study, we used a 1.5 x faster rate in our model simulations to calculate the subsidence rates at higher temperatures, based on literature of the total effect of higher temperatures (Andriesse 1988; Berglund et al. 2008; Davidson and Janssens 2006; Dorrepaal et al. 2009).



113

Fig. 4. Location of three study areas in Friesland, the Netherlands; 1 Hommerts, 2 Groote Veenpolder, 3 Buitenveld

7.2.6 Modeling the effect of adaptation A new interactive tool, a ‘touch-table’ with implemented GIS applications, was used during the workshops, enabling the visualization of land use, subsidence rates, and groundwater or ditch water levels characteristics in a spatially explicit and readily understandable way for the study area under investigation. The learning-by-doing aspect of this type of tool was found effective in supporting the exchange of information between stakeholders with different backgrounds (Janssen et al., 2014). Soil subsidence rates under the current climate and the W+ scenarios were likewise visualized per parcel. Various adaptation measures, (e.g., higher surface water levels, change of land use, or turning over land into open-water systems) could also be evaluated as they affect GDL and, in turn, affect land subsidence rates which were visually presented. The consequences of climate change and adaptation measures on land subsidence could immediately be calculated and displayed in a spatially evident way by running the models in this interactive tool. Although the model does describe 0 subsidence if the water level is at the soil surface, it does not include the possibility of peat formation for even wetter situations and, hence, excludes rising surface levels. During the workshops, however, we did discuss such options and the risks of adverse effects on water quality after rewetting agriculturally used peat soils (Zak et al., 2008). In addition to the effects of adapted water regimes on subsidence, their effects on agricultural yields were also displayed using the parameters defined in by the Dutch Foundation for Applied Water Research (Stowa, 2005).


CHAPTER 7

114

7.2.7 Study areas in the Province of Friesland In collaboration with the provincial government and the water board (Wetterskip) of Friesland, the Netherlands, three study areas were selected, together covering the range of variation in Friesland in types of peat meadows (peat type, thickness of peat layer, land use, ditch water level management). These study areas were ‘Hommerts’, ‘Groote Veenpolder’, and ‘Buitenveld’, as examples of polders with a peat soil with clay cover, a thick peat layer (>1.50 m) and a thin peat layer (<1 m), respectively (see Fig. 4). Hommerts (1) In the western part of Friesland, there are clay-covered peat areas which in the past have been little cut-over, and thus retain substantial depths of peat in places. The Hommerts polder is an example of such an area covering 2400 ha. A land reallotment scheme in the 1970s restructured the area and resulted in deeper drainage leading to summer groundwater levels deeper than 1 m below soil surface. The whole area consists of ‘clay-covered peat’ soils. According to recent measurements, the thickness of the clay cover ranges between 10 and 40 cm. An east–west orientated sand ridge cuts through the area. The old course of the Drylster Ie, a former bog stream that ran west of the building ribbon and discharged into the former Middelsee, is still recognizable in the landscape. The peat layer beneath the clay cover is continuous and does not contain further clay deposits. The peat thickness is mostly around 40–60 cm, with thicker layers up to 2 m in the northern part of the area (Fig. 5).



115

Fig. 5. Peat thickness for the three study areas Hommerts (1), Groote Veenpolder (2), and Buitenveld (3)

Groote Veenpolder (2) This polder is an example of the area in the southern part of Friesland, where clay cover is mostly absent and large parts of the peatlands have been superficially cut-over for fuel in the mid-19th century. These areas were reclaimed for agriculture at a later stage and turned into peat polders around 1900. Because of the peat extraction, these areas now lie significantly below sea level. The ground level in the Groote Veenpolder is now approximately -2.50 m. Of the 3,450 hectares, over two-thirds is in agricultural use, whereas ca. one-third comprises the nature reserve of the Rottige Meente, which consists of ditches, swamp forest, species-rich grasslands, and associated habitats. The Groote Veenpolder is very diverse in terms of peat soil type: ranging from fen peat formed in eutrophic conditions (including the remains of trees) to bog peat, formed in nutrient-poor conditions, and from very little humified to


CHAPTER 7

116

‘earthified’ peat. The peat thickness of the peat deposit is more than 2 m in the nature reserve, but significantly less than that in the agricultural grasslands (Fig. 5). Buitenveld (3) The Buitenveld (Frisian: Butenfjild), which covers an area of 1,525 ha, is the most northeastern part of the Frisian peat area. It lies between clay and sandy soils in the north and south, respectively. The peat soils have always been relatively thin here, but they have undoubtedly subsided during the course of the past century. The northern part of the area was re-allotted and restructured in the 1950s and is now in agricultural use for dairy production. The southwestern part is a nature reserve and has the highest elevation because surface water levels have been kept higher here. The low groundwater levels due to the drainage of the agricultural part have led to seepage of water from the nature reserve, thus creating desiccation of the reserve as well as inconveniently high groundwater levels in the pastures. The peat layer is thin or even absent, with the thickest layers (up to 1.20 m) in the wetland reserve in the southwest (Fig. 5). In the agricultural pastures, the peat layer is often thinner than 40 cm and is no longer classified as peat soil.

7.3 Results

7.3.1 Land subsidence with different climate scenarios Maps of the three study areas displaying the rate of land subsidence calculated per agricultural parcel are shown in Fig. 6. The parcels are separated by drainage ditches and the colour of the parcels represents current land subsidence rates, whereas the colour of the dots in each parcel represents subsidence rates under climatic conditions of the W+ scenario. It is clear that subsidence rates differ between the three study areas. Rates are relatively low in the Hommerts and Buitenveld areas (1–15 mm/year) compared to the Groote Veenpolder (6–30 mm/year, with many parcels in the more rapid categories). These differences are caused by the different conditions of the peat soils in the three study areas, namely the relatively thick peat layer without clay cover in the Groote Veenpolder, the presence of a clay cover at Hommerts, and the really thin peat layers at Buitenveld (see also Fig. 5). There are also clear differences in subsidence rates within the study areas. These are again associated with differences in thickness of peat and the presence of a clay cover, but are also related to current land use. In the Hommerts area (Fig. 6a), the parcels with the lowest subsidence are those with the thinnest peat layers (i.e., in the southern half of the area). The within-area differences in subsidence rates are the largest in the Groote Veenpolder with rapid subsidence rates in parcels located in the northeast and south of the area (Fig. 6b), where the peat layer is thick and drainage is deep. Parcels in the southwest corner have a lower rate of subsidence because locally high groundwater discharge gives rise to higher groundwater levels. The eastern part of this area is a nature reserve and mostly has a moderate subsidence rate because ditch water levels are deliberately kept higher than in the rest of the polder while soils are not submerged. The Buitenveld (Fig. 6c) shows quite low subsidence rates (lower than 5 mm/year) but this are thin peat layers, hence there is little organic material to decompose. Counter-intuitively, the nature reserve in the southeast, which still has a somewhat higher elevation and has a thicker peat layer, has faster subsidence, up to 15 mm/year. More drastic changes in land surface levels emerge if subsidence is calculated for conditions predicted by the W+ scenario, which implies lower summer groundwater levels and higher mean temperatures (Fig. 6). Subsidence rates show relatively small increases in the Hommerts area, where the clay cover does not change its behavior and in the Buitenveld area, where peat layers are mostly thin. Almost all parcels in the Groote Veenpolder show a distinctly faster subsidence rate in the W+ scenario. In both areas, many parcels reach the highest level of subsidence, i.e., 20–30 mm/year. The average



117

climate-induced increases for the three areas are 0.5 mm/year (Hommerts), 1 mm/ year (Buitenveld), and 2 mm/year (Groote Veenpolder, see Table 2).

Fig. 6. Maps of subsidence rates in current conditions and for the W+ climate change scenario in the study areas Hommerts (a), Groote Veenpolder (b), and Buitenveld (c)

7.3.2 Predicted peat cover in 2050 and 2100 If the high subsidence rates calculated for the W? scenario are extrapolated for the next 35 or 85 years, assuming a gradual increase starting from the current rates over the years, it is clear that the proportion of the areas where peat soil disappeared becomes substantial. In the predictions for 2050, about half of the Buitenveld has lost its peat soil entirely and 20 % will only have a humic mineral soil, with the nature reserve as the sole remaining part with a true peat soil (Fig. 7). In the other two areas, these proportions are distinctly lower and mostly limited to the parcels where the peat soil is thin (see also Fig. 5). If the subsidence rates for the W+ climate are extrapolated to 2100, the areas without peat soils become much larger (Fig. 8). In Buitenveld, more than 80 % of the area would lose its peat soil, with the remnants only in nature reserves. The Groote Veenpolder would lose its peat soils across the major proportion of the dairy meadow areas. Only the nature reserve Rottige Meente in the east and the peat meadow area with high groundwater discharge in the southwest would still have peat soils in 2,100. Hommerts would lose around 70 % of its peat soils, with the greater part of its area even lacking humic remains. Table 2. Mean soil subsidence rates for the study areas for the current situation and under climate change (subsidence rates in 2050)

Mean soil subsidence (mm.yr-1)

Mean soil subsidence under climate scenario W+ (mm.yr-1)

Hommerts 6.3 6.9 Buitenveld 5.8 6.7 Groote Veenpolder 13.6 15.6


CHAPTER 7

118

7.3.3 Effects of adaptation measures to reduce land subsidence The way in which the effectiveness of adaptation measures was investigated during workshops using the ‘touch-table’ is illustrated by the examples generated during the stakeholder workshop on the Groote Veenpolder. Figure 9a shows the current land use on the parcels in this area: the eastern, dark-green coloured area is the nature reserve; all other parcels (light green) are in use as agricultural dairy meadows. The land subsidence data in this map are the same as those in Fig. 6. The stakeholders have raised the ditch water levels in the dairy meadows from 100 cm below surface level toward 40 cm below ground level in Fig. 9b (dark blue area). This measure, which would reduce the grass and maize growth to such an extent that agricultural targets are no longer met, has a strong inhibiting effect on subsidence in all parcels, reducing it from 21 to 9.5 mm/year (Fig. 9b). In some parcels, however, subsidence rates will increase as these sites are currently quite wet, and this measure would, therefore, lower the ditch water levels which lead to increased subsidence rates. Another set of measures was implemented in Fig. 9c: the western section with peat meadows was transferred into ‘extensive’ dairy meadows with a somewhat higher water level in the ditches (70 cm below ground level) and less intensive fertilizer use. The midsection (light blue) was turned into a shallow open-water area as a buffer zone to prevent water loss from the nature reserve in the east. It is clear that the extensive use measure did reduce subsidence rates, but not as strongly as in the situation depicted in Fig. 9b. The open-water measure, however, effectively halted soil subsidence to almost 0. When averaged for these two sections, the subsidence had decreased to 7.5 mm/year, which is similar to the subsidence in the nature reserve in the eastern part of the polder.



119

Fig. 7. Status of peat in 2050 at climate scenario W+ for the three study areas Hommerts (1), Groote Veenpolder (2), and Buitenveld (3)


CHAPTER 7

120

Fig. 8. Status of peat in 2100 at climate scenario W+ for the three study areas Hommerts (1), Groote Veenpolder (2), and Buitenveld (3)



121

Fig. 9. Examples of measures to reduce the rate of subsidence in the Groote Veenpolder; a current land use and water level management (ditch water level 1 m below ground level); b. Higher ditch level (ditch water level 0.4 m below ground level) in the western (blue) part of the polder, and c extensive grassland use (westernmost section) and open water (midsection, blue area) as buffer for the nature area (green). The dots indicate land subsidence rates in each parcel (Colour figure online)

7.4 Discussion

7.4.1 Quality of the spatial information The current rates of soil subsidence in typical peat meadows in Friesland were estimated in this study to be up to 3 cm/year for sites with a thick peat layer without clay cover and a ditch water level 1 m below ground surface. The higher values in the range are for parcels grown with maize. These values have been based on longterm monitoring of land level in peat meadows in this region and on the extensive soil level measurements in the experimental farm in Zegveld. Similar values were measured by analyzing long-term data for peatlands in Norfolk, England (Dawson et al. 2010), extensive drained peatlands south of the Venice lagoon (Camporese et al., 2006; Gambolati et al,. 2005), peat meadow areas in northeast Germany (Egglesman, 1976; Gebhardt et al., 2010), and peatlands in agricultural use in New Zealand (Schipper and McLeod, 2002). A recent evaluation of the actual water levels maintained in the study area, which are adapted every 10–15 years, to follow soil subsidence, in the past 20 years indicates a drop of about 40 cm (J. Schouwenaars, personal communication), which concurs nicely with the data from the soil subsidence monitoring. Therefore, we consider this modelling exercise valuable for the exploration of future predictions of subsidence rates and peat thickness. The data for Zegveld show indications that subsidence is affected by short-term variation in climate: the decline was relatively fast in the dry years 1976 and 1996, whereas in wet years, the surface levels even rose, showing that shrinkage and compression are partially reversible. This is consistent with the observation that subsidence rates have been found to be highest in the period immediately after the reclamation, and to slow down to a constant rate thereafter (Dawson et al., 2010; Gambolati et al., 2005; Schipper and McLeod, 2002). Consolidation and possibly the degradation of easily degradable organic components are responsible for this early response to reclamation or water level drop (Berg and Meentemeyer, 2002; Schothorst, 1977). For the situation in Friesland, the very long history of drainage, as well as the new water management structures implemented during the re-allotment schemes in the 1990s (Schouwenaars, 2002) might have resulted in a slightly more rapid subsidence during the late


CHAPTER 7

122

1990s, later stabilizing to the current rate. Our model did not include short-term changes in subsidence rates associated with weather influences or temporarily high subsidence rates directly after lowering water levels. The spatial information used in the maps indicating current peat soil thickness, land elevation, and land use in the province of Friesland has been derived from recent updates of the information for these characteristics by Alterra research institute and can be considered representative for the situation in 2012. The algorithms used to calculate current soil subsidence in a spatially explicit way have been based on the data for (1) mean lowest ditch level in summer (GDL), (2) thickness of the peat layer, (3) presence or absence of a clay cover, and (4) agricultural land use. Furthermore, Alterra has modelled lower water levels in summer, and the effects of higher temperatures on peat decomposition were incorporated. These data were presented for each parcel, thereby covering part of the spatial variability. This level of spatial differentiation was of fundamental importance in the stakeholder workshops, because measures can often be taken at the scale of individual parcels, and land ownership boundaries generally coincide with boundaries between parcels.

7.4.2 Characteristic aspects of soil subsidence in Friesland and climate change effects Compared to the peat meadows in the western part of the Netherlands, the Frisian peat meadow areas are characterized by larger parcels and deeper drainage. While in the western Netherlands peat meadow use often has combined targets for agricultural production and biodiversity enhancement, agricultural production is the main or even the only pursue in peat meadows in Friesland. Land use planning policies in the province of Friesland have sought for spatial separation of agricultural use and nature management, so that large sections are being managed for dairy production, while other areas have become nature reserves with targets for the European Natura-2000 framework and tailored land and water management. The deeper drainage has led to a 150 % faster rate of subsidence and the much smaller populations of meadow birds than in the peat meadows in the provinces of Noord- and Zuid-Holland. The acceleration of soil subsidence due to climate change will clearly lead to distinctly lower soil levels and loss of the entire remaining peat layer in parts of the study areas during the course of this century. Areas with thin peat layers, such as most of Buitenveld and some parts of the centre of the Groote Veenpolder, will only have mineral or humic soils by 2050. Subsidence will continue in other areas and by 2100 large proportions of the Groote Veenpolder and Hommerts will also have lost their peat layers. The central part of the Groote Veenpolder will be deprived of peat by the end of the century. The nature reserve Rottige Meente will still have peat, and differences in elevation between the nature reserve and the neighbouring agricultural area will increase. The local variations in peat thickness will, in this way, finally result in differences in land elevation, when parts of a polder will stop subsiding while the remainder continues to sink. This will create problems for infrastructure (roads, water control structures) and buildings and will certainly also lead to desiccation of nature reserves, which will have a much slower rate of soil subsidence and will lose water toward the lower sinking agricultural areas surrounding them. Issues related to the greenhouse gas balance of peat meadow areas were also discussed, but not in a spatially explicit, quantitative sense. Although there are reasonably reliable estimates of the CO2 emissions associated with subsidence, much less is known on the emissions of methane and nitrous oxide. For interpreting environmental effects of subsidence, we assumed a CO2 emission of 22,600 kg CO2 ha/year to be associated with a subsidence of 10 mm/year (Akker et al., 2008). Although an elaborate study of GHG emissions in the Dutch peat meadow areas has resulted in well-based insights on the GHG balance of a number of polders with different management in the western peat district (Schrier-Uijl et al., 2010), these results cannot be transferred easily to the proposed changes in land use in the polders in Friesland that we studied.



123

7.4.3 Adaptation measures Designing a climate-proof and economically feasible plan for the Frisian peat areas was outside the scope of these workshops. The main objective was to provide knowledge on consequences of climate change and the effectiveness of adaptive water level and land use measures. Decision-making on regional policies in the Netherlands is usually done in a consensus-oriented way, which requires that local stakeholders have sufficient knowledge on the topics being discussed. These workshops and the ‘touch-table’ did prove effective in exchanging, validating, and correcting information among local and regional stakeholders (Janssen et al., 2014). During the workshops, the effect of adaptation measures on soil subsidence rates were explored using the ‘touch-table’ and discussed among the stakeholders. The ‘touch-table’ simulations during workshops showed to what degree subsidence rates can be slowed down by specific measures, such as raising ditch water levels is a spatially explicit way. In the examples of measures explored by stakeholders, a complete halt to subsidence only occurred after transforming a drained peat meadow area into an open-water lake. Actually, peat formation could take place again with such high-water levels, although the history of drainage possibly has a long-term stimulating effect on decomposition (Best and Jacobs, 1997; Brouns et al., 2014; Fenner and Freeman, 2011). Furthermore, peat soils with a history of fertilization can release vast amounts of phosphate and ammonium (Olde-Venterink et al., 2002; Riet et al., 2013), which poses a risk of eutrophication of the regional surface water. In the current situation, the micro-economic gains at the level of individual farms are in contrast with macro-economic (societal) costs of upholding current forms of land use. Rewetting to create optimal conditions for peat formation would be a costly measure initially, because all farmers have to be bought out. However, calculations of long-term costs and benefits of rewetting peat meadows and buying out the farmers in the central part of the province of Noord-Holland have shown that the economic balance could be positive over a period of 50 years because of avoided costs, if the costs of mitigation of CO2 emissions were to be included (Provincie Noord-Holland, 2012). In Hommerts and Groote Veenpolder, the areas with thickest peat layers, the participants concluded that only drastic increases in GDL in combination with alternative crops such as reed, hemp, and duckweed would significantly reduce subsidence rates. However, a possible change Environmental Management in income is considered worrisome. The conclusion of the workshop in Buitenveld was that raising water levels to reduce the degradation of the already thin peat layers would be troublesome for infrastructure, especially houses located at the banks of watercourses. Participants suggest to accept subsidence, but at the same time, stop adapting water levels to the subsided soil surface levels. Doing so, subsidence is gradually reduced and conditions gradually become wetter. In the meantime, farms can change their main source of income. Solutions where ditch water levels would be raised would substantially reduce subsidence rates. However, it is uncertain whether formerly drained peat soils would recover into peat-forming systems in a short time-span. An example of rewetting is found in De Veenkampen, Gelderland, and the Netherlands, where an intensively used dairy meadow was rewetted from water levels >40 cm below soil surface to an average water level within 20 cm below soil surface. 20 years after rewetting, combined CO2 and CH4 emission rates were 2,456 kg C/ha/year in the rewetted field and 2,830 kg C/ha/year in the control field and (Best and Jacobs, 1997). This indicates that substantially reducing drainage depth does not lead to equivalent reductions in carbon loss, and hence decomposition and subsidence. This might be caused by the long-term effect of oxygenation (Brouns et al., 2014; Fenner and Freeman, 2011). Raising ditch water levels would necessitate a restructuring of the agricultural land use. Farmers would need to find alternative practices to compensate for the lower productivity of their grasslands. Such innovations could include (1) the use of new crops adapted to the wetter conditions (e.g., those that can be used as a resource in bioplastics), (2) the combination with non-agricultural activities such as


CHAPTER 7

124

facilitating recreation or medical care. The attempt to test a water level management still enabling agricultural practice (average summer ditch water level 70 cm below ground surface) resulted in only a 20 % reduction in subsidence rates in the Groote Veenpolder. Agricultural targets are at risk of not being met for most parcels in this exercise. It can be concluded that raising ditch water levels to strongly reduce subsidence will inevitably result in quite drastic changes in the forms of economic land use to which these areas may be put in the future. The participating stakeholders were enthusiastic about the workshops; they appreciated the format of information exchange and the ability to explore the future of the peat meadow areas. The results of the questionnaire showed that the level of knowledge about soil subsidence was higher after the workshops. The stakeholders also stated that they were able to provide their own knowledge and ideas. The combination of the knowledge from researchers, stakeholders, and from the maps was considered useful. Several participants mentioned that this approach to the problem of climate change and soil subsidence was instructive and that especially experimenting with different solutions provided new insights. More information on the exchange of thoughts and opinions among stakeholders regarding the various simulations can be found in other publications (Janssen et al., 2013; Janssen et al., 2014).

7.5 Conclusion In this study, spatially explicit information on the effect of peatland management and climate change on subsidence rates were provided and validated in stakeholder workshops using an interactive mapping device. Subsidence rates were up to 3 cm/year in deeply drained parcels and increased when we included climate change in the modelling exercises. Because peat layers in Friesland are generally relatively thin (less than 1.5 m), most peat will have disappeared from the province by the end of the century when current practices continue. This would lead to the loss of characteristic landscape features with a long cultural history. The national government, provinces and water boards increasingly realize that a continuation of this management will have problematic side effects such as damage to building foundations, desiccation of nature reserves, emission of greenhouse gases, increasing costs for water management and infrastructural maintenance, and deterioration of surface water quality. In peat polders with thin peat layers (several decimeters) such as Buitenveld, moderate changes in drainage depth do not reduce subsidence rates substantially. In peat polders with thicker peat layers (over 1 m in this case) such as Hommerts or Groote Veenpolder, the degradation rate of the organic soil could be reduced. There, the use of new crops and exploring other sources of income in peat polders with thicker peat layers (over 1 m in this case) needs deliberation.


Using geodesign to develop a spatial adaptation strategy for Friesland

125

CHAPTER 8 Using geodesign to develop a spatial adaption strategy for Friesland.

Abstract The Province and Waterboard of Friesland have decided to develop a long-term adaptation strategy for the Frisian peat meadow area. A planning process with all stakeholders has been started to develop this strategy. In a workshop setting, the participants were asked to design spatial plans for the region. A large amount of spatial information was available to support the participants. A geodesign tool was developed to support the stakeholder workshops. This tool allowed the participants to change land use and water management while providing immediate feedback on policy objectives. The application proved effective in exchanging, validating and correcting information. The application was also effective in supporting the participants to jointly design the spatial plans.

Published as: Janssen R., Eikelboom T., Verhoeven J.T.A. and Brouns K. (2014) Using geodesign to develop a spatial adaption strategy for Friesland. In D. Lee, E. Dias and H. Scholten (Eds.), Geodesign by integrating design and geospatial sciences (GeoJournal Library), New York: Springer, 103-116.


CHAPTER 8

126

8.1 Introduction Collaborative workshops are common in land use planning. Stakeholders using maps to design plans is not something that started in recent years. Initially, workshops were supported using large hard copies of maps in combination with sheets of tracing paper maps representing attributes of the proposed plan or plan area (Burrough and McDonnell, 1998). In following years, with the arrival of Geographical Information Systems (GIS), the transparent tracing map sheets were replaced by map layers presented within a GIS on a computer screen (Longley et al., 2005). This has now proceeded further towards interactive map interfaces with direct interaction between participants and information. Along with these technical developments, the involvement of stakeholders in spatial planning has changed over the years. In the early years, the emphasis was on communication; in later years this shifted to participation where active involvement of stakeholders was required (Sieber, 2006). At present, the focus is on collaboration, with stakeholders actively working together to reach the best plan. Geodesign tools can be used to support collaborative processes. Typical tools combine different methods, such as simulation models, spatial multi-criteria analysis, visualization, and optimization. User-friendly interfaces allow multiple users to provide input and generate real-time output to support negotiated spatial decisions (Geertman and Stillwell, 2009; Pelzer et al., 2013; Petit, 2011). The Province and Water board of Friesland have decided to develop a long term adaptation strategy for the peat meadow area of the province. Primary activities in this region are highly productive dairy farming, nature conservation, recreation and housing. The region is currently mainly used for commercial dairy farming but is also important for its high natural, cultural and historical values. Important problems in the region are soil subsidence causing damage to buildings and infrastructure, deterioration of landscape values, inefficient water management, poor water quality, and the changing perspectives for dairy farming (Janssen et al., 2013). A planning process with all stakeholders has been started to develop an adaptation strategy. The Province and Water board have described three scenarios in general terms: 1. Business as usual; 2. Parallel tracks and 3. New horizons. The Business as usual strategy included technical measures to reduce soil decline. These technical measures could only be applied if they did not create limitations for agriculture. Parallel tracks was based on zoning and separation of different types of land use. Buffer zones were created to separate the different types of land use. New horizons involved a major transformation of agriculture. New types of crops were introduced that could adapt to the wet conditions needed to prevent soil decline. None of these strategies included a spatial allocation of the proposed measures. As part of this policy process stakeholder workshops were conducted that had the following objectives:

Exchange of information Validate information Design three spatial adaptation scenarios.

The workshops are exploratory and do not involve a choice for one of the scenarios. The workshops were held in three regions, Hommerts, Groote Veenpolder and Buitenveld. These regions were assumed to be representative for the peat meadow area in the province.



127

Fig. 1. workshop participants around the Touch Table

An interactive mapping device (the Touch Table) was used as a common interface (Fig. 1). The use of the ‘Touch Table’ made it possible for participants to have direct access to tools and information and provided a common platform for discussion. The geodesign application developed for these workshops included an evaluation tool and a design tool. Both tools were dynamic and provided immediate feedback to any change made by the participants. These tools were used in the workshops to design spatial strategies by allocating measures and land use types specific to each scenario. This chapter describes the use of geodesign tools to support the development of a spatial adaption strategy for peat meadow areas in Friesland. The evaluation and design tools are described in the next two sections. The third section describes the use of these tools in the workshops. The chapter concludes with a discussion of the effectiveness of the tools. A full report of the workshops is available in Janssen et al. (2013).

8.2 The approach The objective of the workshops was for the stakeholders to develop three spatial strategies at the local scale based on an implementation of the three general strategies for the whole region (Business as usual, Parallel tracks, New horizons). The workshops were exploratory in nature and no strategy was selected as the preferred choice. The workshops followed the steps of the geodesign framework as defined by Steinitz (2012). Steinitz distinguishes three iterations through the framework phrasing the questions as ‘Why?’ question in the first iteration, as ‘How?’ questions in the second iteration and as ‘What?, where? and when?’ questions in the third and final iterations. The workshops go through these iterations in one afternoon (Fig. 2). To be able to do this so quickly, the workshops relied heavily on stakeholder input. No extensive field work


CHAPTER 8

128

was done prior to the workshops. It was accepted that the information about the region was not complete and maybe even incorrect but that stakeholders will correct this information. Participants of the workshops were invited by the Province and Water board of Friesland and were based on prior involvement in the planning process. Some participants had a direct stake in the planning process as they owned a farm or house in the region, other represented interest groups such as farmers or nature conservationists.

Fig. 2. The stakeholders, the geodesign team, and the framework for geodesign (Steinitz 2012)

The workshop started with an introduction to the region and communication of information needed for the assignments (representation). The complicated relations between land use, ground water level and soil subsidence were communicated by an exercise where participants experimented by changing land use and water levels and observe the result on soil decline. Only very simple models were used to model the relations between water management, soil decline, nature and agriculture. This made it possible to explore and even adapt all model relations during the workshop. About half the available time was used to communicate all information. The second half was used for three assignments where participants were asked to design three scenarios for the region. The first assignment around the Touch Table was an introductory assignment to get a feel for the Touch Table and to learn about the mechanics behind soil subsidence. The participants were asked to change land use and water levels, to observe the impacts on soil subsidence and to inspect relevant map layers. This was followed by three assignments linked to the three strategies. The participants were asked to design a land use and water management plan for the following three policy strategies:

Business as usual: low impact technical measures only, no changes in land use; Parallel tracks: create buffers to separate conflicting functions such as agriculture, housing and

nature;



129

New Horizons: introduction of new crops, large changes in land use and water management. The evaluation, change, impact and decision model were integrated in the design tool. The design tool allows to make changes and to get immediate feedback on the resulting changes on policy objectives such as agricultural production, prevention of soil decline and quality of nature. The physical impacts were calculated but not presented to the participants as they would not be immediately understandable. To provide more easy to use feedback to stakeholders, impact and evaluation were combined. The underlying model for evaluation was based on multi-criteria analysis (Arciniegas et al. 2011). This model can be used on various spatial scales and can be adapted to specific decision conditions (Eikelboom and Janssen, 2015). The value of the objectives had a linear relation with the water level in the ditches. This relation was different for each type of land use. Each objective value responded differently to increasing water levels. A high water level results in high objective values for soil subsidence and nature, but in low values for agriculture. Similarly, extensive grasslands have a higher value for nature due to land management compared to intensive grassland and with similar water level conditions nature areas have the highest values for nature. A high value for soil subsidence means low subsidence rates, high values for nature means high quality and low values for agriculture means low productivity. During the process the stakeholders zoomed in to different scales and locations within the area. They kept iterating between evaluation, change, impact and decision until there was consensus about the plan. In doing this they have answered the why, how and what/where questions. The when question was not represented on the map but appeared in the notes of the meeting. No field research was conducted in preparation of the workshops as the approach is based on input from the stakeholders. This made it possible to move quickly but made the approach very dependent on the available knowledge of the stakeholders present during the workshop.

8.3 Geodesign tools The workshops were supported with an evaluation and a design tool. The tools were developed using Community Viz 4.3 which is an extension to ArcGIS 10. Community Viz is a software package that allows for spatial programming of dynamic attributes and indicators (http://placeways.com/communityviz/ last accessed on October 10, 2013). The Samsung SUR40 with Microsoft Surface 2.0. provided the interface between the participants and the tools. A large amount of spatial information is available for the regions. Current and historical land use, and land ownership were obtained from the Provincial authorities. Water levels and an elevation map were obtained from the Water board. Spatial information about the soil was supplied by Alterra University and Research Centre, Wageningen. Most of the information is collected in the years 2012 and 2013. The information is first translated to value maps based on expert judgement.


CHAPTER 8

130

Fig. 3. Performance values for objective agriculture (a) and for objectives agriculture, soil and nature and land use (b) in the ‘Groote Veenpolder’ (current situation)

Figure 3a shows a value map for agriculture (see also Arciniegas et al., 2011). The value in Figure 3a represents the production conditions for agriculture. Values above 0.80 are considered acceptable and therefore green. Values between 0.70 and 0.80 are considered problematic (orange) and values below 0.70 unacceptable (red). In Figure 3b the value maps for agriculture, soil and nature are combined. The value of the objectives is presented as a traffic light where red is low, yellow is average, and green represents a high value. The traffic light shows for each parcel the value for agriculture (left) soil (middle) and nature (right). For example a traffic light with three red boxes means a low value for all three objectives. The traffic lights makes it possible to project the main indicators on top of other maps. During the workshops, the objective values were mainly shown in combination with land use and water level, because the objective values change when water level or land use is changed. This tool enabled participants to monitor the performance for each objective under different circumstances. The evaluation tool is used in combination with the design tool. The design tool has a list of potential measures and types of land use next to the map. Figure 4 shows the design tool for one of regions. The map on the left shows land use and the map on the right water levels on the right. The tool provides a list of measures that affect water levels and land use types. Participants can apply these measures to one or more parcels to improve one or more objectives. Changes in values are shown immediately as changes in the colours of the traffic lights in each parcel. Participants can also change the land use of each parcel. Changes in land use and water management are shown on the map. Any map participants consider relevant can be used as a background for the traffic lights. A division is made between intensive and extensive grassland. Intensive grasslands are characterized by a high density of cattle, use of heavy machinery and use of pesticides and fertilizers. Extensive grassland have a lower production, do not



131

need heavy machinery and no or little fertilizer and pesticides. As a result extensive grassland are more suitable for higher water levels.

Fig. 4. The use of the design tool for the region ‘Hommerts’: combined with land use (a) and water levels (b)


CHAPTER 8

132

8.4 Workshop results Workshops were organized in three regions, ‘Hommerts’, ‘Groote Veenpolder’ and ‘Buitenveld’. These regions were assumed to be representative for southeast Friesland. The workshops started with an introduction to the region and communication of information needed for the assignments. An effective way to introduce the region is a comparison of the current and historic topographical maps (Fig. 5). Swiping the two maps showed how the current map has evolved from the past. Especially in peat meadow areas many of the current issues can be traced back to the past.

Fig. 5. ’De Groote Veenpolder’: in 1860 (a) and 2012 (b)

The introduction was also used to validate the information presented. As the preparation did not involve extensive field visits not all information was up to date. Figure 6 (left) shows the information as available from the topographical map and aerial pictures. Figure 6 (right) shows the same map after corrections from the participants. It appeared that, recently, open water had been added, more nature had been created and some parcels had been changed to extensive agriculture. Also the former waste dump was not included in the original map.



133

Fig. 6. Land use map ‘Buitenveld’ before (a) and after the workshop (b)


CHAPTER 8

134

Fig. 7. Creating a buffer in scenario ‘Parallel tracks’: a) Change in land use, b) Change in water level and performance

The next map shows the result for scenario ‘Business as usual’ for the ‘Groote Veenpolder’. Before applying any measures, as a first step, participants zoomed in to a specific area. To do this they used the map library to identify parts of the region with specific characteristics. As one of the available measures could only be applied in areas without upward seepage participants used the hydrology map to identify the relevant sub region (Fig. 8). The map showed that for this part of the region the measure applied was beneficial both for agriculture (left box) and prevention of soil subsidence (middle box).



135

Fig. 8. The effect of underwater drains for the south east corner of the’ Groote Veenpolder’

The design tool allows for detailed, small scale, changes but can also support a total redesign of the region. This is what is called for by the ‘New horizons’ scenario. This scenario assumes substantial changes from the current situation. A good example was the ‘New Horizon’ plan designed for the ‘Groote Veenpolder’ (Fig. 9).


CHAPTER 8

136

Fig. 9. Scenario ‘New Horizons’ for the ‘Groote Veenpolder’: a) measures, b) land use change

A major problem in this polder is the net loss of water from the nature reserve located east of the lower situated agricultural area in the west. Participants decided that only a radical measure could solve this problem. As can be seen in the maps in this scenario (Fig. 9), a large area to the west of the nature reserve will be flooded to be used for water sports or aquaculture. In the agricultural area to the west of this buffer, water levels were increased and the grassland use was changed from intensive to extensive. Wet crops and common reed were introduced along the border between water and agriculture. This radical change leads to a total stop of soil subsidence and creates high values for nature, while intensive agriculture is no longer possible. Surveys were conducted before and after the workshops. Results from these surveys suggested that participants find the traffic light presentation easy to understand and are able to use feedback provided by the tool to perform tasks such as changing land use or changing water management. Results also indicated that no more than three indicators should be presented and that the calculation of the scores should be simple (Eikelboom and Janssen, 2015). The surveys also showed that eight persons considered information from the Touch Table most important, while also eight persons valued information from participants and seven from the experts present during the workshop. This shows that workshop did well in facilitating exchange of information (Fig. 10). An important objective of the workshops was to bring the strategies to life. It is remarkable that only a small proportion of the participants think that Business as usual is the most realistic. This proportion is even smaller after the workshop (Fig. 11 right). Feedback from the participants indicated that they found the exchange of information very important. The opportunities to switch between maps to allowed learning by doing were considered useful. Participants reported an increased understanding of these relations and a raised awareness of the perspectives of other participants. These results are consistent with studies that stressed that



137

presenting spatial information through interactive map interfaces during planning workshops facilitated exchange of views on spatial decision problems (e.g. Andrienko et al. 2007; Bacic et al. 2006).

Fig. 10. Participants evaluation of the usefulness of different types of information (n=25)

Fig. 11. Most realistic scenario according to the participants before (left) and after (right) the workshop (n=25)


CHAPTER 8

138

8.5 Conclusions The decision support tools provided in the design workshops combined maps, a simple simulation model, an evaluation tool, a design tool, basic GIS functions and a shared interactive map interface. Maps were used to obtain feedback from participants. The interactive map interface made it possible to learn about the complex relations in the model by trial and error. Organizing policy workshops is a learning process that requires substantial effort in terms of preparation, logistics and technical challenges. Preparation is especially crucial as there should be no technical problems and all relevant information must be readily available. There is little tolerance from the participants for technical or methodological errors. A first lesson is that in the Netherlands it is not so easy to convince policy-makers to include workshops in their planning process. There is fear for the unknown and fear that bringing the problem outside the usual setting would change the level of control on the process. In this process the workshops were actively supported by the Province and Water board. Results of this phase are currently being used as the starting point of the next phase of the process. This phase will involve selection of the most appropriate strategy for each region. A second lesson is that selection of the participants is essential. Participants of the workshops were invited by the Province and Water board of Friesland. Although this increased commitment of the participants it was not always easy to attract participants to the workshops. It proved important to ensure that there was some benefit for all the participants invited and that the workshop was a positive experience. The main advantage of the presented approach was to get different types of people in the same conversation to optimize opportunities for knowledge exchange between different expertise. The research results suggest that the approach depends highly on a cooperative attitude by participants. This worked well in a Dutch context as it suits the Dutch consensus-oriented way of decision-making. Other examples of this approach can be found in Arciniegas and Janssen (2012) and Alexander et al. (2012). However, it is uncertain if the same approach would also work in contexts of sharp conflict or with a more power-based style of decision making. For this study the use of geodesign as part of interactive workshops proved to be a useful instrument for facilitating group work around spatial information. Using interactive maps as tools for communication and interaction proved to be an effective method to support planning meetings with stakeholders with different backgrounds.


Synthesis and conclusions

139

CHAPTER 9 Synthesis and conclusions Adaptation planning consists of multiple stages with complex tasks, and involves multiple sectors and stakeholders. This thesis contributes to improving our understanding of the use of geodesign tools for collaborative planning at the regional scale, and provides tools to support decision making with multiple stakeholders and objectives. The main objective of this thesis was to develop and evaluate geodesign tools that support the development of regional adaptation strategies. In order to achieve this goal the following research questions were defined in Chapter 1, and addressed in Chapters 2 to 8:

1. How is spatial information used in regional adaptation strategies? 2. How can adaptation planning tasks be linked to spatial decision support tools? 3. What is the performance of a new set of geodesign tools? 4. How are these geodesign tools used in collaborative planning? 5. What is the potential for integrating an optimization algorithm in a geodesign tool? 6. How can expert knowledge be integrated in a geodesign tool? 7. What can be learned from the application of the tools in planning workshops?

9.1 Geo-information in regional adaptation strategies In order to develop tools that are perceived to be useful in practice, the needs for geodesign tools have to be identified. First, it was considered relevant to assess how spatial information has recently been used in regional adaptation strategies. The adaptation framework of Willows and Connell (2003) and the evaluation matrix of Preston et al. (2011) were used to evaluate the quality and the use of spatial information in regional adaptation strategies. Although a regional adaptations strategy is defined as encompassing a range of adaptation options, reviewing 25 strategies revealed that the quality of the strategies differs greatly and can be explained by multiple factors. For many strategies the inclusion of maps increased the quality, but high quality strategies also exist with only a few maps. Some of these strategies included detailed lists of adaptation options, or were developed with high stakeholder involvement. Currently, the spatial information and the adaptation options are only listed in the strategies, and none of the strategies report on options appraisal. It is expected that the need to make trade-offs on where to apply which adaptation options could benefit from the use of spatial information and detailed descriptions of options. To include the appraisal of options in the strategies, spatial decision support tools can have a high potential as the spatial dimension of climate change can serve as an integrating factor for different sectors and multiple stakeholders in adaptation planning.

9.2 Linking adaptation planning tasks with spatial decision support tools The identification of tasks that need support is necessary In order to develop tools that suit adaptation planning. The adaptation framework of Willows and Connell (2003) was selected to identify these tasks from three looped development stages: assess risks; identify options; and appraise options. Each of the three stages was assumed to contain two task categories: risk assessment includes analysis and validation; identification of options consists of exploration and design; appraise options consists of evaluation and negotiation as tasks needed to make improvements for the next round. These tasks were used to link the different types of tools. Three types of tools were tested in three separate workshops: 1) drawing tools; 2) simulation tools; and 3) evaluation tools. The participants had no difficulty in identifying their own parcels based on the aerial photograph, and they drew the outlines of their property with high accuracy. Next, the validation task reveals that participants’ expertise is very valuable to evaluate model results. However, the participants


CHAPTER 9

140

were only prepared to comment on the model results for their own parcels, and, as a result, they lacked the confidence to extrapolate their expertise to similar adjacent parcels. As a first step, the participants used the tool to test whether the model produced plausible results. After passing this test, the tool was used to interactively design a plan. The possibility to interactively test the model combined with short response times is essential for the successful use of the tool. Value maps where the main component for the evaluation tool. These maps proved useful in communicating the relative qualities of the plans, and also triggered specific questions related to underlying information. Furthermore, individual value maps were considered more useful than the aggregated value scores for the whole area. In all three case studies, it proved necessary to tailor the spatial support tools to the regional context. The time and effort this takes may be a limitation to the use of these tools in practice. All three workshops showed that interactive participation promotes stakeholder involvement and encouraged knowledge exchange and the acceptance of workshop products. However, it is important to allow the participants to play around with and test the tools before the actual planning session starts.

9.3 Evaluation of geodesign tools Four types of geodesign tools were developed that varied both in the inclusion of a ranking and aggregation step and in their underlying rationality, varying from individual and analytical to collective and political. The performance of the tools was evaluated in two ways: in an online survey and in experimental workshops in order to evaluate both the individual use and the collective use of the tools with multiple stakeholders. On the one hand, adding ranking and aggregation steps made the tools more difficult to understand. On the other hand, tools should also limit the amount of information to be processed by the user of the tool, which may well call for including ranking and aggregation steps. All this stresses the importance of tailoring specific methods to specific tasks. Furthermore, the ability to allow trial and error use within the tools was also shown to be of importance as well.

9.4 Collaborative use of geodesign tools The collaborative use of the tools in experimental workshops resulted in different spatial designs. Both planners and researchers considered the tools developed in this study useful at the scoping stage of an adaptation planning process. The learning-by-doing aspect of the tools was reflected upon as very effective. It was concluded that the tools were easy to use, and their application positively contributed to extending the knowledge of the participants. Furthermore, results indicated that the choice for a tool influences the decision making process, as each tool yielded different designs of adaptation measures. During the sessions, discussions revealed that for each tool similar logic was used by each group to decide on the appropriate measures, and that the participants tended to cluster the same changes. Only a little time was spent on tool interpretation for tools tailored to individual analytical and collective political rationalities, whereas a great deal of time was spent on discussing what could be interpreted from the individual political and collective analytical tools. The application of the individual analytical tool induced many changes, though not much time was spent on negotiating the necessary measures. Less changes were made using the remaining tools, but here more time was spent on negotiating the allocation of measures. Moreover, the individual analytical tool was preferred by both researchers and planners, and was also found to induce many measures in line with tool information. Therefore, it is argued to carefully select methods and tools to support the development of adaptation plans, and rationality can be used to choose between different geodesign tools. If the rationality behind the decision process is unclear, then the analytical and individual approach would be best, as the interpretation and use of this tool was found to be quick and easy.



141

9.5 Optimization and geodesign The design of different strategies depends on multiple objectives and constraints. An optimization model can support the fast generation of alternatives using these constraints. The integration of an optimization model in a geodesign tool showed that a genetic algorithm can be used to generate land-use plans that maximize both additive and spatial objectives in a vector-based GIS environment. The optimization geodesign tool was used to: (1) generate a set of alternatives to start a decision process; (2) identify similarities and differences between collaborative planning results and results from optimization; and (3) serve as an interactive tool using feedback from stakeholders. Also, it proved possible to generate a relevant set of non-dominated solutions to begin the planning process. The results showed that generating a set of single objective maximum alternatives, combined with compromise alternatives, provides the stakeholder with a set of alternatives that covers the range of possible solutions. Previous workshop experience demonstrated that stakeholders first explore extreme situations to find out the influence of large changes. This exploration stage can be structured and accelerated using the optimization routine. When maximizing objectives, it has to be kept in mind that the spatial and additive objectives compete in achieving the best results for the area. Comparing results from the optimizer with the stakeholder results demonstrated that both approaches generated plans with similar values for the objectives, but with large differences in the changes made to the maps that were produced. It is clear that there is no single optimal plan, and that more than one solution can exist. As the values of the objectives are similar to those of the stakeholder plans, the result from the optimizer could be used as a point of reference for the stakeholders. One of the challenges in using an optimization model for interactive design is to be able to explain the model assumptions, such that the model results are accepted and found valid by the stakeholders. The model provides the user only with a result, but not the road leading to it. Integrating the optimizer in a geodesign tool demonstrated how the optimizer can complement stakeholder input if it is used as an interactive geodesign tool. Stakeholders can either accept, reject, or adjust the results of the optimization model, and thereby incorporate their preferences and local knowledge. The genetic algorithm has short response times, which is a major prerequisite for interactive use. In conclusion, collaborative planning is based on the assumption that the stakeholders have knowledge that is not, or even cannot be represented in a formal model. The challenge is to combine optimization with stakeholder input in such a way that both approaches complement each other to get the best of both worlds.

9.6 Tool application in planning practice In order to draw conclusions on a new set of tools, the application of the tools in practice is essential. Applying the tools to specific case studies requires accurate data that is tailored to the regions’ characteristics. In Chapter 7, spatially explicit information on the effect of peatland management and climate change on subsidence rates was provided and validated in stakeholder workshops using an interactive mapping device. Subsidence rates were up to 3 cm/year in deeply-drained parcels, and increased when we included climate change in the modelling exercises. Because peat layers in Friesland are generally relatively thin (less than 1.5 m), most peat will have disappeared from the province by the end of the century if current practices continue. This would lead to the loss of characteristic landscape features with a long cultural history. The national government, provinces, and Water Boards are increasingly realizing that a continuation of this management will have problematic side effects such as damage to building foundations, desiccation of nature reserves, emission of greenhouse gases, increasing costs for water management and infrastructural maintenance, and deterioration of surface water quality. In peat polders with thin peat layers (several decimeters),


CHAPTER 9

142

such as Buitenveld, moderate changes in drainage depth do not reduce subsidence rates substantially. In peat polders with thicker peat layers (over 1 m in this case), such as Hommerts or Groote Veenpolder, the degradation rate of the organic soil could be reduced. There, the use of new crops and exploring other sources of income in peat polders with thicker peat layers (over 1 m in this case) needs further deliberation. Organizing policy workshops is a learning process that requires substantial effort in terms of preparation, logistics, and technical challenges. Preparation is especially crucial: everything has to work perfectly, and all the relevant information must be readily available. There is little tolerance from the participants of technical or methodological errors. A first lesson is that it is not so easy to convince policy makers to include workshops in their planning process. There is fear of the unknown, and fear that bringing the problem outside its usual setting would change the level of control in the process. A second lesson is that the selection of the participants is essential, yet it is not always easy to attract participants to the workshops. It is therefore important to ensure that there is something in it for all the participants invited, and that the workshop is a pleasant experience. Every workshop should, therefore, start in a light way to get the participants going. The main plus point of the present approach is to get different types of people talking, so that opportunities for exchange between difference sources of expertise can be maximized.

9.7 Conclusions The results of the surveys as well as the results from the workshops confirmed the need to carefully match specific planning tasks and tool characteristics. A tool can easily become too complex for the task at hand. The evaluations confirmed that the more simple tools were perceived to be those which were the most easy to use, and these were found especially valuable for the scoping stage of planning. For climate change adaptation, the spatial identification and appraisal of options were tasks that were found to benefit from the use of interactive geodesign tools. Geo-information proved a valuable common language to support communication and interaction between stakeholders from different backgrounds. Furthermore, geo-information can be used to convert complex scientific knowledge to information that is easy to interpret and valuable for local stakeholders. Incorporating this information in geodesign tools not only makes it accessible for users, but means that users can also interact with the data. The visualization of three objectives in traffic light colours was found an effective method to combine multiple stakeholders in a single map, as they provided immediate feedback on the effects of land use changes and adaptation measures, such as water management changes. Interaction based on spatial information and between stakeholders from a range of backgrounds encourages knowledge exchange and acceptance of workshop products. This is enhanced by the fact that stakeholders can add to the system their local and specific knowledge that was not in the models beforehand. The combination of an optimization algorithm with the ability to interactively apply changes to the map indicates the potential of interactive geodesign tools to support decision making.

9.8 Lessons learned As part of this thesis the geodesign tools were applied and evaluated in 13 stakeholder workshops. From this experience, lessons for further tool development, evaluation and application were learned. The use of symbols to visualize multiple objectives simultaneously was well received. The advantage of symbols is the ability to place them on top of different map layers depending on the subject of interest. However, the number of symbols should be kept limited, depending on the scale of the regions. In the case study areas of this thesis, the parcel level proved suitable for a maximum of three objectives.



143

Secondly, the strength of the interactive geodesign approach was found to be the ‘learning-by-doing’ element. This allowed participants to experiment with the model in order to validate the results themselves. No detailed explanations were needed prior to the workshops, but time to experiment with the tools before the planning exercise should be allowed. Furthermore, workshop success did not only depend on having a leading supervisor who keeps to the schedule and provides technical support to solve small technical problems or to make small adjustments at the location, but also on inviting the relevant stakeholders. For some stakeholder groups it is difficult to find time for an afternoon of interactive planning. This means one should make a workshop session an attractive outing by elaborating on the innovativeness of the approach and the relevance of the workshop for each stakeholder.

9.9 Implications for future planning Although the tools were tested to support the design of adaptation plans in a Dutch setting, the tools could also be used for regional adaptation planning in other countries, such as the development of Regional Adaptation Strategies (RAS) as required by the European Union, or on a national scale to support the development of National Adaptation Plans of Action (NAPAs), as initiated by the United Nations Framework Convention on Climate Change (UNFCCC) for Least Developed Countries. Nonetheless, the collaborative approach depends highly on a cooperative attitude by participants. This worked well in a Dutch context, as it suits the Dutch consensus-oriented way of decision making. However, it is uncertain whether the same approach can be readily applied in situations of sharp conflict, or in contexts with a more power-based style of decision making. In recent times, communication through social media has increased greatly, partly as a result of the development of devices such as smartphones and tablets, the development of apps, and because of the availability of information in the cloud. GIS systems have evolved from a ‘close’ expert-oriented to an ‘open’ user-oriented technology (Malczewski, 2006). The scientific community is also discovering the value of these technological developments for research purposes. On the other hand, this growing convergence of GIS and social media poses new challenges for GIScience (Sui and Goodchild, 2011). One of these challenges for the future is how to elicit, represent, and handle user-defined fuzzy information which is in people’s minds, but is difficult to represent on a map. In addition, such data can contain sensitive or incorrect information. Regional adaptation plans can serve as ingredients for national, European, and global adaptation. Improvement of geodesign tools, as well as the users’ learning process must be seen as an interactive and iterative process. The communication between science and policy can benefit from further tool development by improving user-friendliness, for example the integration of tools such as Urban Strategy and Phoenix (Dias et al., 2013); including downscaled climate scenarios; adding information on costs; providing filters to reduce the amount of information on the maps; creating scenarios of measures; and improving modelling and visualization techniques to further tailor tools to specific planning tasks. More improvements are needed in terms of the availability of the tools for a wider audience (e.g. the web tool), and making the tool flexible for different scales and users. Depending on the tool, it is also possible to apply the underlying methodology of the tools in a setting without a Touch Table. However, without it the perception and performance of the tools could differ, as the Touch Table was positively received as a communication platform by all participants. Further research is recommended to experiment with the tools at different scales and to test the tools in different contexts (see, for example, Alexander et al., 2012). Furthermore, the tools need constant updates on the latest findings and best available data. To conclude, the development and application of this type of tools is an ongoing process rather than a finished product (Wenkel et al., 2013).


CHAPTER 9

144


Summary

145

Summary

Climate change and sustainable development are presenting decision makers with unprecedented challenges, such as interactions at spatial scales, uncertainty, dynamic changes, and contrasting views. These challenges make climate adaptation a complex task from both an information processing and a process point of view. Adapting to climate change involves multiple stakeholders, ranging from individuals, firms and civil society to public bodies and governments at local, regional and national scales, and international agencies. Climate change is only one of multiple factors considered in current decision making. In deciding how to incorporate climate adaptation in spatial planning, access to relevant spatial information is needed. Geo-information-based tools can make spatial information available for stakeholders. The advent of methods to host and manage large geo-spatial databases, improvements in computer processing performances and the development of intuitive natural user-interfaces now offer new opportunities for geo-analytical tools to support the design professionals in real time (Dias et al., 2013). Integrating natural user-interfaces with numerical analysis has resulted in a new approach, called ‘Geodesign’. Using a combination of previous definitions, Lee et al. (2014) defined geodesign as an iterative design and planning method whereby an emerging solution is influenced by (scientific) geospatial knowledge derived from geospatial technologies. In contrast to traditional planning processes, where analysis, design and evaluation are executed in separate steps, geodesign integrates the exploration of ideas with direct evaluation at the same moment (Lee et al., 2014). This thesis contributes to improving our understanding of the use of geodesign tools for collaborative planning at a regional scale, and provides tools to support decision making with multiple stakeholders and objectives. The main aim of this research is to develop and evaluate geodesign tools that support the development of regional adaptation strategies (RAS). Regional adaptation strategies are plans that consist of feasible measures to shift a region towards a system that is flexible and robust with respect to future climate changes. Integrating geodesign in an interactive way for use in the development of regional adaptation strategies introduces a new methodology in participatory decision making which supports the exchange of information between stakeholders. The developed geodesign tools were made directly available for practice, in order to evaluate the tools and learn from stakeholders. In total, 13 workshops (from 2010-2014) were organized that gathered feedback from scientists, regional planners, government officials and local farmers. This extensive feedback was used for the further development of the tools and to draw conclusions on their usability. An interactive mapping device, called the ‘Touch Table’, was used to support collaborative planning workshops. The ‘Touch Table’ was used in a series of workshops with various stakeholders to generate, assess, and discuss adaptation strategies for peat meadow areas in the Netherlands.

Geo-information in regional adaptation strategies Different levels of government initiate the development of strategies to mitigate and adapt to climate change. Climate change is affecting decision making and shaping its priorities as an additional factor in the often already complex domain of environmental governance. Spatial planning has a key role in addressing the causes and impacts of climate change. Chapter 2 reports the result from a review of 25 European regional adaptation strategies in order to explore relations between their quality and the use geo-information. The quality of the studies was assessed with use of the evaluation matrix from Preston et al. (2011). The results show a wide variety in the quality of the strategies, and in the type of geo-information included in these strategies. Most of the included strategies turned out to be climate


Summary

146

change effect and climate change impact studies, with little information on adaptation options. None of the studied strategies included the appraisal of adaptation options. For many strategies the inclusion of maps increased the quality, however high quality strategies also exist with only a few maps. The engagement of stakeholders and inclusion of decision support tools were found to increase the overall quality of strategies. The expectation is that the need to make trade-offs on where to apply which adaptation option can benefit from the use of spatial information and detailed descriptions of options. Spatial decision support tools have much potential to support the appraisal of options, as the spatial dimension of climate change can serve as an integrating factor for different sectors and multiple stakeholders in adaptation planning. The findings intend to contribute to the ongoing discussion on good practices for adaptation planning, exchange of knowledge and experiences, and results of adaptation practices between countries.

Interactive spatial tools for the design of regional adaptation strategies Multiple adaptation frameworks and guidelines exist that describe the development stages of regional adaptation strategies. Spatial information plays a key role in the design of adaptation measures as both the effects of climate change as well as many adaptation measures have spatial impacts. Interactive spatial support tools such as drawing, simulation and evaluation tools can assist the development process. Chapter 3 presents how to link tasks derived from the actual development stages to spatial support tools in an interactive multi-stakeholder context. This link helps to decide what spatial tools are suited to support which stages in the development process of regional adaptation strategies. The practical implication of the link is illustrated for three case study workshops in the Netherlands. The regional planning workshops combine expertise from both scientists and stakeholders with an interactive mapping device. This approach triggered participants to share their expertise and stimulated integration of knowledge.

Comparison of geodesign tools to communicate stakeholder values Geodesign tools are increasingly used in collaborative planning. An important element in these tools is the communication of stakeholder values. As there are many ways to present these values it is important to know how these tools should be designed to communicate these values effectively. The objective of this study is to analyse how the design of the tool influences its effectiveness. To do this stakeholder values were included in four different geodesign tools, using different ways of ranking and aggregation. The communication performances of these tools were evaluated in an online survey to assess their ability to communicate information effectively. The survey assessed how complexity influence user performance. Performance was considered high if a user is able to complete an assignment correctly using the information presented. Knowledge on tool performance is important for selecting the right tool use and for tool design. The survey showed that tools should be as simple as possible. Adding ranking and aggregation steps makes the tools more difficult to understand and reduces performance. However, an increase in the amount of information to be processed by the user also has a negative effect on performance. Ranking and aggregation steps may be needed to limit this amount. This calls for careful tailoring of the tool to the task to be performed. For all tools it was found maybe the most important characteristic of the tools is that they allow for trial and error as this increases the opportunity for experimentation and learning by doing.


Summary

147

Collaborative use of geodesign tools to support decision making on adaptation to climate change Spatial planners around the world need to make adaptation plans. Climate adaptation planning requires combining spatial information with stakeholder values. This study demonstrates the potential of geodesign tools as a mean to integrate spatial analysis with stakeholder participation in adaptation planning. The tools are interactive and provide dynamic feedback on stakeholder objectives in response to the application of spatial measures. Different rationalities formed by underlying internalized values influence the reasoning of decision making. Four tools were developed, each tailored to different rationalities varying between a collective or individual viewpoint and analytical or political arguments. The tools were evaluated in an experiment with four groups of participants that were set around an interactive mapping device: the Touch Table. To study how local decision making on adaptation can be supported, this study focuses on a specific case study in the Netherlands. In this case study, multiple different stakeholders need to make spatial decisions on land use and water management planning in response to climate change. The collaborative use of four geodesign tools was evaluated in an interactive experiment. The results show that the geodesign tools were able to integrate the engagement of stakeholders and assessment of measures. The experiment showed that decision-making on adaptation to climate change can benefit from the use of geodesign tools as long as the tool is carefully matched to the rationality that applies to the adaptation issue. Although the tools were tested to support the design of adaptation plans in a Dutch setting, the tools could be used for regional adaptation planning in other countries such as the development of RAS as required by the European Union or on a national scale to support developing NAPAs (National adaptation plans of action) as initiated by the United Nations Framework Convention on Climate Change (UNFCCC) for Least Developed Countries.

A spatial optimization algorithm for geodesign Chapter 6 describes a genetic algorithm that can be used to generate land use plans that maximize both additive and spatial objectives in a vector-based GIS environment. To test the usefulness of the algorithm it was integrated in a geodesign tool and applied to a planning process in a peat meadow area in the Netherlands. The objective of this chapter is to demonstrate the potential and limitations of a genetic optimization algorithm to support collaborative land use planning workshops. The chapter shows how the algorithm can be used to 1. generate a set of alternatives to start a decision process, 2. To identify similarities and differences between collaborative planning results and results from optimization and 3. As an interactive tool using feedback from stakeholders. It proved possible to generate a relevant set of non-dominated solutions to begin the planning process. Comparing results from the optimizer with stakeholder results demonstrated that both approaches generated plans with similar values for the objectives but with large differences in the maps that were produced. Integrating the optimizer in a geodesign tool demonstrated how the optimizer can complement stakeholder input if it is used as an interactive geodesign tool. Collaborative planning is based on the assumption that the stakeholders have knowledge that is not, or even cannot be represented in a formal model. The challenge is to combine optimization with stakeholder input in such a way that both approaches complement each other to get the best of both worlds.

Spatial analysis of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change, and adaptation Dutch peatlands have been subsiding due to peat decomposition, shrinkage and compression, since their reclamation in the 11th century. Currently, subsidence amounts to 1–2 cm/year. Water management in these areas is complex and costly, greenhouse gases are being emitted, and surface


Summary

148

water quality is relatively poor. Regional and local authorities and landowners responsible for peatland management have recognized these problems. In addition, the Netherlands Royal Meteorological Institute predicts higher temperatures and drier summers, which both are expected to enhance peat decomposition. Stakeholder workshops have been organized in three case study areas in the province of Friesland to exchange knowledge on subsidence and explore future subsidence rates and the effects of land use and management changes on subsidence rates. Subsidence rates were up to 3 cm/year in deeply drained parcels and increased when we included climate change in the modelling exercises. This means that the relatively thin peat layers in this province (ca 1 m) would shrink or even disappear by the end of the century when current practices continue. Adaptation measures were explored, such as extensive dairy farming and the production of new crops in wetter conditions, but little experience has been gained on best practices. The workshops have resulted in useful exchange of ideas on possible measures and their consequences for land use and water management in the three case study areas. The province and the regional water board will use the results to develop land use and water management policies for the next decades.

Using geodesign to develop a spatial adaption strategy for Friesland The Province and Water board of Friesland have decided to develop a long-term adaptation strategy for the Frisian peat meadow area. A planning process with all stakeholders has been started to develop this strategy. In a workshop setting, the participants were asked to design spatial plans for the region. A large amount of spatial information was available to support the participants. A geodesign tool was developed to support the stakeholder workshops. This tool allowed the participants to change land use and water management while providing immediate feedback on policy objectives. The application proved effective in exchanging, validating and correcting information. The application was also effective in supporting the participants to jointly design the spatial plans.

Conclusions In order to draw conclusions on a new set of tools, the application of the tools in practice is essential. Applying the tools to specific case studies requires accurate data that is tailored to the regions’ characteristics. The time and effort this takes may be a limitation to the use of these tools in practice. All workshops showed that interactive participation promotes stakeholder involvement and encouraged knowledge exchange and the acceptance of workshop products. However, it is important to allow the participants to play around with and test the tools before the actual planning session starts. The learning-by-doing aspect of the tools was reflected upon as very effective. It was concluded that the tools were easy to use, and their application positively contributed to extending the knowledge of the participants. Organizing policy workshops is a learning process that requires substantial effort in terms of preparation, logistics, and technical challenges. Preparation is especially crucial: everything has to work perfectly, and all the relevant information must be readily available. There is little tolerance from the participants of technical or methodological errors. The results of the surveys as well as the results from the workshops confirmed the need to carefully match specific planning tasks and tool characteristics. A tool can easily become too complex for the task at hand. The evaluations confirmed that the more simple tools were perceived to be those which were the most easy to use, and these were found especially valuable for the scoping stage of planning. For climate change adaptation, the spatial identification and appraisal of options were tasks that were found to benefit from the use of interactive geodesign tools. Spatial information proved a valuable common language to support communication and interaction between stakeholders from different backgrounds. Spatial information can be used to convert complex scientific knowledge to information that is easy to interpret and valuable for local stakeholders.


Summary

149

Incorporating this information in geodesign tools not only makes it accessible for users, but means that users can also interact with the data. The visualization of three objectives in traffic light colours was found an effective method to combine multiple stakeholders in a single map. The traffic lights provided immediate feedback on the effects of land use changes and adaptation measures, such as water management changes. Interaction based on spatial information and between stakeholders from a range of backgrounds encourages knowledge exchange and acceptance of workshop products. This is enhanced by the fact that stakeholders can add to the system their local and specific knowledge that was not in the models beforehand. The combination of an optimization algorithm with the ability to interactively apply changes to the map indicates the potential of interactive geodesign tools to support decision making.


Samenvatting

150


Samenvatting

151

Samenvatting Klimaatverandering en duurzaamheid zorgen voor nieuwe uitdagingen in besluitvorming, zoals schaalinteractie, onzekerheden, dynamische processen en tegengestelde belangen. Klimaatverandering is slechts een van de vele factoren die van belang zijn in de huidige besluitvorming. Adaptatie is een manier om te reageren op de te verwachten effecten van klimaatverandering en kan uitgevoerd worden door verschillende partijen, variërend van individuele burgers of bedrijven tot overheden. Daarnaast kan het plaatsvinden op lokaal, regionaal, nationaal of internationale schaal. Het integreren van klimaatadaptie in ruimtelijke planning vereist de beschikbaarheid van ruimtelijke informatie. Geo-informatie instrumenten kunnen ruimtelijke informatie beschikbaar maken voor belanghebbenden. Recente ontwikkelingen maken het mogelijk om grote hoeveelheden aan data real-time beschikbaar te maken via gebruiksvriendelijke instrumenten. De combinatie van gebruiksvriendelijke interfaces met dynamische rekenmodellen met behulp van geo-informatie wordt ook wel ‘geodesign’ genoemd. In tegenstelling tot traditionele planningsprocessen waar analyse, ontwerp en evaluatie apart plaatsvinden, wordt dit in deze methode geïntegreerd, waardoor er direct terugkoppeling plaatsvindt bij het toepassen van veranderingen. Dit proefschrift draagt bij aan het verbreden van de kennis over het gebruik van geodesign instrumenten voor regionale planning met meerdere belangen en doelen. Het doel van het onderzoek is het ontwikkelen en evalueren van instrumenten om de ontwikkeling van regionale adaptatie strategieën te ondersteunen. Het ontwikkelen van een adaptatiestrategie op regionale schaal bestaat uit het ontwerpen van adaptatie opties die bijdragen aan het klimaatbestendig maken van de regio. Deze opties moeten aan verschillende eisen voldoen. Zo moeten ze voldoende draagkracht hebben bij de lokaal betrokken partijen, flexibel zijn voor veranderende scenario’s (zowel klimaat als sociaal-economisch) en technisch haalbaar zijn. Het gebruik van geodesign instrumenten voor klimaatadaptatie levert een nieuwe benadering in participatieve besluitvormingsprocessen dat gebruik maakt van het uitwisselen van kennis van verschillende betrokkenen. De ontwikkelde geodesign instrumenten zijn toegepast in de praktijk zodat deze geëvalueerd konden worden en geleerd kon worden van de ervaringen van planners en besluitvormers. In totaal zijn 13 workshops georganiseerd voor wetenschappers, regionale planners, overheden en agrariërs. De terugkoppeling hiervan is gebruikt om de methodes te verbeteren en om conclusies te trekken ten aanzien van de bruikbaarheid. Een zogenaamde ‘Touch Table’, is gebruikt als communicatie platform in workshops die onderdeel waren van een planningsproces voor het veenweidegebied in de provincie Friesland, Nederland.

Geo-informatie in regionale adaptatie strategieën In hoofdstuk 2 is een review gedaan naar 25 regionale adaptatie strategieën in Europa om te onderzoeken wat de rol van ruimtelijke informatie en de kwaliteit van de strategieën is op dit moment. De kwaliteit van de studies is gemeten aan de hand van de evaluatie matrix van Preston et al. (2011). Het resultaat laat een wijde variatie in kwaliteit en gebruik van ruimtelijke informatie zien. De meeste strategieën zijn voornamelijk effect en impact studies met weinig informatie over adaptatie opties. Geen enkele strategie beschrijft het afwegen van opties. Het gebruik van kaarten verhoogd in veel gevallen de kwaliteit. Daarentegen zijn er ook kwalitatief hoog scorende strategieën met daarin weinig kaarten. Het samenwerken met betrokkenen en het gebruik van ondersteunende instrumenten bevorderen beide de kwaliteit. De verwachting is verder dat de strategieën in een volgende stap, waarin afwegingen tussen opties gemaakt moeten worden, verder kunnen profiteren van het gebruik van ruimtelijke informatie en


Samenvatting

152

instrument om dit proces te ondersteunen. Het ruimtelijke aspect van klimaatverandering kan op deze manier dienen als integrerende factor om verschillende belangen en doelen bij elkaar te brengen.

Interactieve ruimtelijke instrumenten voor het ontwerpen van regionale adaptatie strategieën. Op basis van het adaptatie kader van Willows and Connels (2003) is een indeling gemaakt in fasen voor het ontwerpen van een gebied specifieke adaptatie strategie. Bij deze fasen zijn verschillende taken te bedenken waarbij ondersteuning mogelijk is. Deze fasen zijn gelinkt aan verschillenden type ruimtelijke ondersteunende methodes en aan de hand van drie voorbeeldgebieden zijn een aantal passende instrumenten gedemonstreerd in regionale planningsworkshops.

Vergelijken van geodesign instrumenten om verschillende belangen te communiceren In een studie gebaseerd op online vragenlijsten zijn verschillende typen visualisaties vergeleken om meerdere belangen in een enkele kaart zichtbaar te maken. Door gebruik te maken van symbolen is het mogelijk om de achtergrond kaart te wisselen. Daarnaast blijkt het gebruik van intuïtieve kleuren belangrijk voor een snelle en juiste interpretatie. Er zijn vier verschillende methoden om ruimtelijke gerelateerde doelen te presenteren ontwikkeld en geëvalueerd. Uitgangspunt voor deze instrumenten is het gebruik van stoplichten om meerdere doelen of belangen te visualiseren. De instrumenten verschilden in het gebruik van ranking en aggregatie. Het functioneren van de instrumenten met verschillend niveau aan complexiteit was getest in een online vragenlijst waarbij beoordeeld werd of een gebruiker de beschikbare informatie in de kaart gebruikte om een opdracht uit te voeren. De resultaten lieten zien dat instrumenten met minder complexiteit beter begrepen worden en daarnaast ook de voorkeur hebben. Daarentegen bleken de complexere instrumenten wel geschikt voor situaties met een specifieke vraagstelling waarop de visualisatie aangepast was. Tezamen vraagt dit om het aanpassen van een instrument op de taak van de gebruiker. Voor alle methoden bleek het element leren door doen en de mogelijkheid om veranderingen uit te kunnen proberen het sterkst mee te wegen bij de positieve beoordeling.

Gezamenlijk gebruik van geodesign In een volgende studie zijn de vier methodes gebruikt in experimentele workshops en lag de focus op de onderliggende rationaliteit achter het nemen van beslissingen. De rationaliteit achter de methodes varieert tussen collectieve en individuele belangen en tussen analytische en politieke argumenten. Tijdens de workshops werden belanghebbenden gevraagd om landgebruik- en watermanagement veranderingen toe te passen met behulp van de verschillende instrumenten. In het bredere perspectief van klimaatadaptatie toonden de resultaten van de workshops aan op welke manier deze instrumenten ingezet kunnen worden voor adaptatie in andere regio’s en landen.

Een ruimtelijke optimalisatiemodel voor geodesign Een optimalisatiemodel is ontwikkeld dat gebruikt kan worden als integraal onderdeel van een interactief ruimtelijk instrument voor besluitvorming. De studie laat zien hoe het model gebruikt kan worden om (1) een set alternatieven te genereren voor het starten van een planningsproces, (2) om gelijkenissen en verschillen te identificeren tussen resultaten van een participatief planningsproces en het optimalisatiemodel, en (3) als een interactief instrument met directe terugkoppeling van belanghebbenden op de modeluitkomsten. Participatief plannen is gebaseerd op de aanname dat belanghebbenden over kennis beschikken die niet in een model beschikbaar is. Uitdaging in het combineren van lokale kennis met een optimalisatiemodel op een manier waarbij beide methodes elkaar aanvullen om zo het beste uit twee werelden te verkrijgen.


Samenvatting

153

Ruimtelijke analyse van bodemdaling in veenweidegebieden in Friesland Nederlandse veengebieden dalen sinds de verkaveling in de 11e eeuw. Bodemdaling vind plaats door afbraak van veen, krimp en compressie. Op dit moment daalt de bodem met 1-2 cm/jaar. Watermanagement in deze gebieden is complex en kostbaar. Er komen broeikasgassen vrij en de oppervlaktewaterkwaliteit is relatief slecht. De regionale en lokale overheden en landeigenaren die verantwoordelijk zijn voor het gebruik hebben deze problemen erkend. Daarnaast voorspelt het KNMI hogere temperaturen en drogere zomers waardoor bodemdaling versterkt wordt. Voor drie deelgebieden zijn workshops georganiseerd om kennis over bodemdaling uit te wisselen en toekomstig te verwachte bodemdaling te onderzoeken ten aanzien van veranderingen in landgebruik en watermanagement. De bodemdaling bleek toe te nemen tot 3 cm/jaar en nam verder toe na het meenemen van de effecten van klimaatverandering. Dit betekend dat de dunne veenlaag (ca 1m) zou verdwijnen voor het eind van deze eeuw met het huidige beleid. Adaptatiemaatregelen zijn verkend zoals extensieve landbouw en de productie van natte gewassen.

Gebruik van geodesign voor de ontwikkeling van een adaptatie strategie voor Friesland De provincie en het waterschap hebben besloten een lange termijn visie te ontwikkelen voor de veenweidegebieden. Tijdens de workshops in drie deelgebieden zijn steeds drie scenario’s uitgewerkt (huidige beleid, parallelle sporen, en nieuwe wegen) om de consequenties van verschillende vormen van landgebruik en water management door te rekenen voor de doelen landbouw, bodemdaling en natuur. De methode bleek effectief in het uitwisselen, valideren en corrigeren van informatie. Ook maakte de methode het mogelijk om gezamenlijk tot een ruimtelijk plan te komen.

Conclusies Het praktisch toepassen van instrumenten is essentieel voor het evalueren van nieuwe methoden. Dit vereist de beschikbaarheid van accurate en gebied specifieke informatie. De tijd en moeite die het kost om instrumenten geschikt te maken voor een gebied kunnen een belemmering vorming. De interactieve benadering vergroot het vertrouwen in de ruimtelijke informatie en de betrokkenheid van beleidsmakers. Daarnaast stimuleert het de uitwisseling van kennis en vergroot het de kans op acceptatie van het plan. Het heeft daardoor een positieve invloed op het planningsproces. Dit komt o.a. doordat betrokkenen hun eigen kennis konden toevoegen aan het systeem. Het is wel van belang dat deelnemers eerst het instrument kunnen uittesten in een voorronde alvorens het planningsproces gestart wordt. Het organiseren van interactieve ontwerpsessies is een leerproces dat vraagt om een gedegen voorbereiding, aangezien de praktijk over weinig tolerantie ruimte beschikt voor methodische of technische problemen. Zowel de workshops als vragenlijsten toonden aan dat de eenvoudige instrumenten gemakkelijker in gebruik zijn en vooral geschikt bevonden zijn voor de oriënterende fase. De afgenomen vragenlijsten en workshops onderstrepen het belang van eenvoudige instrumenten die meerdere belangen inzichtelijk maken. De resultaten benadrukken de positieve bijdrage van ‘leren door doen’ op acceptatie en draagvlak en het nut van aanpassen van instrumenten aan specifieke taken. Voor het plannen van klimaatadaptatie kunnen ook het ruimtelijke identificeren en afwegen van maatregelen goed ondersteund worden. Ruimtelijke informatie kan dienen als waardevol communicatie middel waar betrokkenen van verschillende disciplines elkaar kunnen vinden. Ruimtelijke informatie zichtbaar maken in kaarten kan helpen om complexe wetenschappelijke informatie te vertalen naar informatie die gemakkelijker te interpreteren is en van waarde is voor lokale partijen. Het gebruik van deze kaarten in geodesign maakt de informatie niet alleen toegankelijk, maar gebruikers kunnen er ook interactief gebruik van maken. Er is gekozen voor een visualisatie van drie doelen in stoplichtkleuren. Deze manier om drie belangen tegelijk inzichtelijk te maken op een kaart is effectief


Samenvatting

154

gebleken. De stoplichten tonen direct de effecten van veranderingen in landgebruik en adaptatiemaatregelen zoals wijzigingen in het watersysteem. Ter aanvulling is een optimalisatie model ontwikkeld en geïntegreerd om het brede potentieel van geodesign te tonen. De resultaten van dit proefschrift versterken de kennis van interactieve instrumenten om ruimtelijke planning te ondersteunen. Ontwikkelingen in het visualiseren van meerdere belangen en het gebruik van optimalisatie modellen kunnen waardevol zijn voor het verbeteren van plannings- en besluitvormingsprocessen.


Bibliography

155

Bibliography Adger, W. N., Agrawala, S., Mirza, M. M. Q., Conde, C., O'Brien, K., Pulhin, J., Pulwarty, R., Smit, B., Takahashi, K. (2007)

Assessment of adaptation practices, options, constraints and capacity. Climate change 2007: Impacts, Adaptation and Vulnerability.

Adger, W. N., Arnell, N. W., Tompkins, E. L. , (2005) Successful adaptation to climate change across scales. Global Environmental Change, 15:77-86.

Aerts, J. C. J. H., Botzen, W. J. W., Emanuel, K., Lin, N., de Moel, H., and Michel-Kerjan, E. O. (2014) Evaluating Flood Resilience Strategies for Coastal Megacities. Science, 344:473-475.

Aerts, J. C. J. H., Eisinger, E., Heuvelink, G. B. M., and Stewart, T. J. (2003) Using Linear Integer Programming for Multi-Site Land-Use Allocation. Geographical Analysis, 35:148-169.

Akker, van den, J.J.H., Beuving, J., Hendriks, R.F.A., Wolleswinkel R.J. (2007) Maaivelddaling, afbraak en CO2 emissie van Nederlandse veenweidegebieden. Leidraad Bodembescherming, afl. 83, Sdu, Den Haag, 32 (in Dutch).

Akker, van den, J.J.H., Kuikman, P.J., De Vries, F., Hoving, I., Pleijter, M., Hendriks, R.F.A., Wolleswinkel, R.J., Simões, R.T.L., Kwakernaak, C. (2008) Emission of CO2 from agricultural peat soils in the Netherlands and ways to limit this emission. In: Farrel C, Feehan J (eds) Proceedings of the 13th International Peat Congress. International Peat Society, Jyva¨skyla¨, 645–648.

Alexander, K.A., Janssen, R., Arciniegas, G.A., O'Higgens, T.G., Eikelboom, T., and Wilding, T.A. (2012) Interactive Marine Spatial Planning: Siting tidal energy arrays around the Mull of Kintyre . PLoS ONE, 7(1):1-8.

Al-Kodmany, K. (2002) Visualization Tools and Methods in Community Planning: From Freehand Sketches to Virtual Reality. Journal of Planning Literature, 17:189-211.

Andrienko, G., Andrienko, N., and Jankowski, P. (2002) Building Spatial Decision Support Tools for Individuals and Groups, Journal of Decision Systems, 2003, 12(2):193-208.

Andrienko, G., Andrienko, N., Jankowski, P., Keim, D., Kraak, M. J., Maceachren, A., and Wrobel, S. (2007) Geovisual analytics for spatial decision support: Setting the research agenda. International Journal of Geographical Information Science, 21: 839-857.

Andrienko, N., Andrienko, G., Voss, H., Bernardo, F., Hipolito, J., Kretchmer, U. (2002) Testing the usability of interactive maps in CommonGIS. Cartography and Geographic Information Science, 29:325-342.

Andriesse, J.P. (1988) Nature and management of tropical peat soils. FAO. Antunes, P., Zurita, G., Baloian, N., and Sapateiro, C. (2013) Integrating Decision-Making Support in Geocollaboration Tools.

Group Decision and Negotiation, 1-23. Appleton, K., Lovett, A. (2005) GIS-based visualization of development proposals: reactions from planning and related

professionals. Computers, Environment and Urban Systems, 29:321-339. Arciniegas, G. A. and Janssen, R. (2009) Using a touch table to support participatory land use planning. 18th IMACS World

Congress - MODSIM09 International congress on Modeling and Simulation 13-17 July 2009, Cairns, Australia, 2206-2212.

Arciniegas, G.A., Janssen, R., (2012) Spatial decision support for collaborative land use planning workshops, Landscape and Urban Planning, 107(3):332-342.

Arciniegas, G.A., Janssen, R., and Omtzigt, N. (2011) Map-based multicriteria analysis to support interactive land use allocation. International Journal of Geographical Information Science 25(12):1931-1947.

Arciniegas, G.A., Janssen, R., and Rietveld, P. (2012) Effectiveness of collaborative map-based decision support tools: Results of an experiment. Environmental Modelling and Software, 1-17.

Bacic, I. L. Z., Rossiter, D. G., and Bregt, A. K. (2006) Using spatial information to improve collective understanding of shared environmental problems at watershed level. Landscape and Urban Planning, 77:54-66.

Baker, I., Peterson, A., Brown, G., and McAlpine, C. (2012) Local government response to the impacts of climate change: an evaluation of local climate adaptation plans. Landscape and Urban Planning, 107:127-136.

Batty, M. (2008) Progress, Predictions, and Speculations on the Shape of Things to Come. In R. K. Brail (Ed.), Planning Support Systems for Cities and Regions (pp. 3-30). Cambridge: Lincoln Institute of Land Policy.

Beek, van, C.L., Droogers, P., van Hardeveld, H.A., van Den Eertwegh, G.A.P.H., Velthof, G.L., Oenema, O. (2007) Leaching of solutes from an intensively managed peat soil to surface water. Water Air Soil Pollution, 182:291–301.

Belton, V. and Stewart, T. J. (2002) Multiple criteria decision analysis: an integrated approach. Springer. Berglund, Ö , Berglund, K., Klemedtsson, L. (2008) A lysimeter study on the effect of temperature on CO2 emission from

cultivated peat soils. Geoderma, 154:211–218. Bertin, J. (1983) Semiology of Graphics: Diagrams, Networks, Maps. University of Wisconsin Press.


Bibliography

156

Best, E.P.H., Jacobs, F.H.H. (1997) The influence of raised water table levels on carbon dioxide and methane production in ditch dissected peat grasslands in the Netherlands. Ecological Engineering, 8:129–144.

Beunen, R. and Opdam, P. (2011) When landscape planning becomes landscape governance, what happens to the science? Landscape and Urban Planning, 100:324-326.

Biesbroek, G. R., Swart, R. J., and van der Knaap, W. G. (2009) The mitigation adaptation dichotomy and the role of spatial planning. Habitat International, 33:230-237.

Biesbroek, G. R., Swart, R. J., Carter, T. R., Cowan, C., Henrichs, T., Mela, H., Morecroft, M. D., and Rey, D. (2010) Europe adapts to climate change: Comparing National Adaptation Strategies. Global Environmental Change, 20:440-450.

Bishop, I. (2013) Optimization in Geodesign. Landscape Architecture Frontiers, 1:64-75. Boroushaki, S., Malczewski, J. (2010) Measuring consensus for collaborative decision-making: A GIS-based approach.

Computers, Environment and Urban Systems, 34:322-332. Borst, J. (2010) Urban Strategy: Interactive spatial planning for sustainable cities. Infrastructure Systems and Services: Next

Generation Infrastructure Systems for Eco-Cities (INFRA) Third International Conference on , 1-5. IEEE. Brouns, K., Eikelboom T., Jansen P.C., Janssen R., Kwakernaak C., van den Akker J.J.H., Verhoeven J.T.A. (2014). Spatial analysis

of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change and adaptation. Journal for environmental management and planning, 1-13.

Brouns, K., Verhoeven, J.T.A., Hefting, M.M. (2014) Short period of oxygenation releases latch on peat decomposition. Science of the Total Environment, 481:61–68.

Brown, A., Gawith, M., Lonsdale, K., Pringle, P. (2011) Managing adaptation: linking theory and practice, UKCIP. Brundtland, G. H. and World Commission on Environment and Development (1987) Our common future. Oxford University

Press Oxford. Bruin, K., Dellink, R., Ruijs, A., Bolwidt, L., van Buuren, A., Graveland, J., de Groot, R., Kuikman, P., Reinhard, S., Roetter, R.,

Tassone, V., Verhagen, A., van Ierland, E. (2009) Adapting to climate change in The Netherlands: an inventory of climate adaptation options and ranking of alternatives. Climatic Change, 95:23-45.

Budic, I.Z. (1994) Effectiveness of geographic information systems in local planning. Journal of the American Planning Association, 60(2):244-263.

Burrough, P. A., McDonnell, R., Burrough, P. A., and McDonnell, R. (1998) Principles of geographical information systems. (333 Edition). Oxford university press Oxford.

Burton, I., Huq, S., Lim, B., Pilifosova, O., Schipper, E. L., (2002) From impacts assessment to adaptation priorities: the shaping of adaptation policy. Climate Policy, 2:145-159.

Buuren, van A., Driessen P.P., van Rijswick, M., Rietveld, P., Salet, W., Spit, T., and Teisman, G. (2013) Towards adaptive spatial planning for climate change: balancing between robustness and flexibility. Journal for European Environmental and Planning Law 10(1):29-53.

Campagna, M. and Matta, A. (2014) Geoinformation technologies in sustainable spatial planning: a Geodesign approach to local land-use planning. International Society for Optics and Photonics.

Camporese, M., Gambolati, G., Putti, M., Teatini, P. (2006) Peatland subsidence in the Venice watershed. In: Martini IP, Mart nez Cortizas A, Chesworth W. Peatlands evolution and records of environmental and climate changes. Elsevier, Amsterdam, 529–550.

Cao, K., Batty, M., Huang, B., Liu, Y., Yu, L., and Chen, J. (2011) Spatial multi-objective land use optimization: extensions to the non-dominated sorting genetic algorithm-II. International Journal of Geographical Information Science, 25:1949-1969.

Carsjens, G. J., Ligtenberg, A., (2007) A GIS-based support tool for sustainable spatial planning in metropolitan areas, Landscape and Urban Planning, 80(1):72–83.

Carton, L.J. (2007) Map making and map use in a multi-actor context. Spatial visualizations and frame conflicts in regional policymaking in the Netherlands. PhD thesis, TU Delft, Delft.

Carton, L.J., Thissen, W.A.H. (2009) Emerging conflict in collaborative mapping: Towards a deeper understanding? Journal of Environmental Management 90:1991-2001.

Cash, D. W., Clark, W. C., Alcock, F., Dickson, N. M., Eckley, N., Guston, D. H., Jager, J., and Mitchell, R. B. (2003) Knowledge systems for sustainable development. Proceedings of the National Academy of Sciences of the United States of America, 100:8086-8091.

CEC. White paper - Adapting to climate change: Towards a European framework for action. COM, (2009) 147, Brussels, Commission of the European Communities.

Ceccato, L., Giannini, V., and Giupponi, C. (2011) Participatory assessment of adaptation strategies to flood risk in the Upper Brahmaputra and Danube river basins. Environmental Science and Policy, 14(8):1163-1174.

Celino, A. (2011) Explorative Nature of Negotiation in Participatory Decision Making for Sustainability. Group Decision and Negotiation, 20(2):255-270.

Climate-adapt (2014) http://climate-adapt.eea.europa.eu, last accessed February 2015. COM 216 (2013) An EU Strategy on adaptation to climate change. Brussels, European Commission. Community Viz software version 4.3 (http://placeways.com/communityviz, last accessed December 2014).


Bibliography

157

Crossland M.D., Wynne B.E., Perkins W.C. (1995) Spatial decision support systems: An overview of technology and a test of efficacy. Decision Support Systems, 14:219-235.

Davidson, E.A., Janssens, I.A. (2006) Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature, 440:165–173.

Davis, F.D. (1985) A technology acceptance model for empirically testing new end-user information systems: Theory and results, Dissertation, Massachusetts Institute of Technology.

Davoudi, S. (2013) Climate Change and the Role of Spatial Planning in England. Climate Change Governance, Springer, 153-169. Dawson, Q., Kechavarzi, C., Leeds-Harrison, P.B., Burton, R.G.O. (2010) Subsidence and degradation of agricultural peatlands in

the Fenlands of Norfolk, UK. Geoderma, 154:181–187. Densham, P.J. (1991) Spatial decision support systems. Geographical information systems, 1:403-412. Dessai, S., Lu, X., Risbey, J. S. (2005) On the role of climate scenarios for adaptation planning. Global Environmental Change Part

A, 15:87-97. Dias E. and Beinat E. (2009) Measuring the Impact of Location-Awareness in the Acceptance of Mobile Systems. Location Based

Services and TeleCartography II: 137-154. Dias E., Linde M., Rafiee A., Koomen E., and Scholten H. (2013) Beauty and Brains: Integrating Easy Spatial Design and Advanced

Urban Sustainability Models. In: Planning Support Systems for Sustainable Urban Development. Springer Berlin Heidelberg, 469-484.

Dong W., Ran J., Wang J. (2012) Effectiveness and Efficiency of Map Symbols for Dynamic Geographic Information Visualization. Cartography and Geographic Information Science, 39:98-106.

Dorrepaal, E., Toet, S., van Logtestijn, R.S.P., Swart, E., van de Weg, M.J., Callaghan, T.V., Aerts, R. (2009) Carbon respiration from subsurface peat accelerated by climate warming in the subarctic. Nature, 460:616–619.

Dragicevic, S. (2004) The potential of Web-based GIS. Journal of Geographical systems, 6:79-81. Dühr, S. (2004) The form, style, and use of cartographic visualizations in European spatial planning: examples from England and

Germany. Environment and Planning A, 36:1961-1989. Dühr, S. (2007) The visual language of spatial planning. London: Routledge. Egglesman, R. (1976) Peat consumption under the influence of climate, soil condition, and utilisation. In Proceedings of the 5th

International Peat Congress, Poznan, 233–247. Egmond van, K. (2010) Een vorm van beschaving, Christofoor. Eikelboom, T. and Janssen, R. (2013) Interactive spatial tools for the design of regional adaptation strategies. Journal of

Environmental Management, 127:S6-S14. Eikelboom, T. and Janssen, R. (2015) Collaborative use of geodesign tools to support adaptation planning, Journal of Mitigation

and Adaptation Strategies for Global Change. Eikelboom, T. and Janssen, R. (2015) Comparison of geodesign tools to communicate stakeholder values. Group Decision and

Negotiation. Feick, R. D., Hall, G. B. (2002) Balancing consensus and conflict with a GIS-based multi-participant, multi-criteria decision

support tool. GeoJournal, 53:391-406. Fenner, N., Freeman, C. (2011) Drought-induced carbon loss in peatlands. Nature Geoscience, 4:895–900. Fisher-Gewirtzman, D. (2012) 3D models as a platform for urban analysis and studies on human perception of space. Usage,

usability, and utility of 3D city models Nantes. Flaxman, M. (2010) Fundamentals of Geodesign. In: Buhmann E, Pietsch M, Kretzler E (eds) Proceedings of digital landscape

architecture conference 2010. Wichmann Verlag BerlinWichmann, 28-41. Ford, J. D., Berrang-Ford, L., Paterson, J., (2011) A systematic review of observed climate change adaptation in developed

nations A letter. Climatic Change, 106:327-336. Fuchs, S., Spachinger, K., Dorner, W., Rochman, J., and Serrhini, K. (2009) Evaluating cartographic design in flood risk mapping.

Environmental Hazards, 8:52-70. Füssel, H. M. (2007) Adaptation planning for climate change: concepts, assessment approaches, and key lessons. Sustainability

Science, 2:265-275. Gambolati, G., Putti, M., Teatini, P., Camporese, M., Ferraris, S., Gasparetto Stori, G., Nicoletti, V., Silvestri, S., Rizzetto, F., Tosi,

L. (2005) Peat land oxidation enhances subsidence in the Venice watershed. Eos, 86:217–224. Garlandini, S. and Fabrikant, S. (2009) Evaluating the Effectiveness and Efficiency of Visual Variables for Geographic Information

Visualization. In K. Hornsby, C. Claramunt, M. Denis, and G. Ligozat (Eds.), Spatial Information Theory, Springer Berlin Heidelberg, 195-211.

Gebhardt, S., Fleige, H., Horn, R. (2010) Shrinkage processes of a drained riparian peatland with subsidence morphology. J Soils Sediments 10:484–493.

Geertman, S. (2006) Potentials for planning support: a planning-conceptual approach. Environment and Planning B-Planning and Design, 33:863-880.

Geertman, S. and Stillwell, J. (2009) Planning Support Systems Best Practice and New Methods (Advances in Spatial Science). New York: Springer Publishers.


Bibliography

158

Geertman, S., and Stillwell, J. (2004) Planning support systems: an inventory of current practice. Computers, Environment and Urban Systems 28:291-310.

Geertman, S., Toppen F., and Stillwell J. (2013) Planning Support Systems for Sustainable Urban Development. Springer, London Gidhagen, L., Denzer, R., Schlobinski, S., Michel, F., Kutschera, P., and Havlik, D. (2013) Sustainable Urban Development Planner

for Climate Change Adaptation (SUDPLAN), 6-8, Bonn. Giupponi, C., Giove, S., and Giannini, V. (2013) A dynamic assessment tool for exploring and communicating vulnerability to

floods and climate change. Environmental Modelling and Software, 44:136-147. Goodhue, D.L., Thompson, R.L. (1995) Task-technology fit and individual performance. MIS quarterly, 213-236. Goosen, H., Groot-Reichwein, M. A. M., Masselink, L., Koekoek, A., Swart, R., Bessembinder, J., Witte, J. M. P., Stuyt, L., Blom-

Zandstra, G., and Immerzeel, W. (2013) Climate Adaptation Services for the Netherlands: an operational approach to support spatial adaptation planning. Reg Environ Change, 1-14.

Goosen, H., Janssen, R., Vermaat, J. E., (2007) Decision support for participatory wetland decision-making. Ecological Engineering, 30:187-199.

Grêt-Regamey, A., Celio, E., Klein, T. M., and Wissen Hayek, U. (2013) Understanding ecosystem services trade-offs with interactive procedural modeling for sustainable urban planning. Landscape and Urban Planning, 109:107-116.

Gudmundsson, H. (2011) Analysing models as a knowledge technology in transport planning. Transport Reviews, 31(2):145-159.

Hage, M., Leroy, P., and Petersen, A.C. (2010) Stakeholder participation in environmental knowledge production. Futures, 42(3):254-264.

Hellmann, F., Vermaat, J.E. (2012) Impact of climate change on water management in Dutch peat polders. Ecological Modelling, 240:74–83.

Hemmati, M. and Enayati, J. (2002) Multi Stakeholder Processes for Governance and Sustainability: Beyond Deadlock and Conflict, Routledge.

Herwijnen, M. van (1999) Spatial decision support for environmental management. PhD thesis VU University, Amsterdam. Hoogland, T., van den Akker, J.J.H., Brus, D.J. (2012) Modelling the subsidence of peat soils in the Dutch coastal area. Geoderma

171:92–97. Hoving, I., van den Akker, J.J.H., Hendriks, R.F.A. (2004) Zegveld gaat zakken veengrond te lijf. PraktijkKompas Rundvee (in

Dutch). Hunt, A. and Watkiss, P. (2011) Climate change impacts and adaptation in cities: a review of the literature. Climatic Change,

104:13-49. Hurk, van den, B., Tank, A.K., Lenderink, G., van Oldenborgh, G.J., Katsman, C., van den Brink, K.F., Bessembinder, J., Burgers, G.,

Komen, G., Hazeleger, W., and Drijfhout, S. (2006) KNMI Climate Change Scenarios 2006 for the Netherlands, De Bilt KNMI.

Inman, D., Blind, M., Ribarova, I., Krause, A., Roosenschoon, O., Kassahun, A., Scholten, H., Arampatzis, G., Abrami, G., McIntosh, B., Jeffrey, P. (2011) Perceived effectiveness of environmental decision support systems in participatory planning: Evidence from small groups of end-users. Environmental Modelling and Software, 26:302-309.

IPCC (2001) McCarthy, J. J., Canziani, O. F., Leary , N. A., Dokken, D. J., White, K. S. (eds), Climate Change 2001: Impacts, Adaptation and Vulnerability, Contribution of theWorking Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge.

IPCC (2007) Climate change 2007: impact, adaptation and vulnerability. Summary for policymakers, working group II contribution to the intergovernmental panel on climate change, fourth assessment report University Press, Cambridge, U.K., 1032 pp.IPCC, 2007. Climate change: impact, adaptation and vulnerability. Working group II contribution to the intergovernmental panel on climate change, fourth assessment report.

IPCC (2014) Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Chan. 1-32. Cambridge, United Kingdom and New York, NY, USA, Cambridge University Press.

Jankowski, P. (2009) Towards participatory geographic information systems for community-based environmental decision making. Journal of Environmental Management, 90:1966-1971.

Jankowski, P., Andrienko, N., Andrienko, G. (2001) Map-centred exploratory approach to multiple criteria spatial decision making. International Journal of Geographical Information Science, 15:101-127.

Janssen, F.B. (1986) Maaivelddalingen in het Friese veenweidegebied. Cultuurtechnisch Tijdschrift, 26:245–252 (in Dutch). Janssen, R. (1992) Multiobjective decision support for environmental management, Dissertation, VU Amsterdam. Janssen, R., Eikelboom, T., Brouns, K., Jansen, P., Kwakernaak, C., and Verhoeven, J. T. A. (2013) Verslag workshops Friese

Veenweidevisie, Kennis voor Klimaat, Utrecht (in Dutch). Janssen, R., Eikelboom, T., Brouns, K., Verhoeven, J.T.A. (2014) Using geodesign to develop a spatial adaption strategy for south

east Friesland, In Geodesign for land use planning, E. Dias, D. Lee, and H. J. Scholten, eds., Springer, New York, 103-116.


Bibliography

159

Janssen, R., Uran, O. (2003) Presentation of information for spatial decision support A survey on the use of maps by participants in quantitative water management in the IJsselmeer region, The Netherlands. Physics and Chemistry of the Earth, Parts A/B/C 28:611-620.

Janssen, R., van Herwijnen, M., Stewart, T. J., and Aerts, J. C. J. H. (2008). Multiobjective decision support for land-use planning. Environment and Planning B-Planning and Design, 35:740-756.

Jonsson D, Berglund S, Almström P, Algers S (2011) The usefulness of transport models in Swedish planning practice. Transport Reviews, 31(2):251-265.

Karakostas, S. and Economou, D. (2013) Enhanced multi-objective optimization algorithm for renewable energy sources: optimal spatial development of wind farms. International Journal of Geographical Information Science, 28, 83-103.

Kingston, R. (2000) Web-based public participation geographical information systems: an aid to local environmental decision-making. Computers, environment and urban systems, 24(2):109-125.

Kirchhoff, C.J., Lemos, M.C., and Dessai, S. (2013) Actionable Knowledge for Environmental Decision Making: Broadening the Usability of Climate Science. Annual Review of Environment and Resources, 38(1):393-414.

Klein, T., Holzkamper, A., Calanca, P., Seppelt, R., and Fuhrer, J. (2013) Adapting agricultural land management to climate change: a regional multi-objective optimization approach. Landscape Ecology, 28:2029-2047.

Ko, T. T. and Chang, Y. C. (2012) An integrated spatial planning model for climate change adaptation in coastal zones. Ocean and Coastal Management, 66:36-45.

Kraak, M. (2004) The role of the map in a Web-GIS environment. Journal of Geographical systems, 6:83-93. Krysanova, V., Dickens, C., Timmerman, J., Varela-Ortega, C., Schlüter, M., Roest, K., Huntjens, P., Jaspers, F., Buiteveld, H., and

Moreno, E. (2010) Cross-comparison of climate change adaptation strategies across large river basins in Europe, Africa and Asia. Water Resources Management, 24:4121-4160.

Kwartler, M. and Bernard, R. N. (2001) CommunityViz: an integrated planning support system. Planning Support Systems: integrating geographic information systems, models, and visualization tools, 285-308.

La Rosa, D. (2014) Geodesign for Urban Ecosystem Services. Tema Journal of Land Use, Mobility and Environment. Laiho, R. (2006) Decomposition in peatlands: reconciling seemingly contrasting results on the impacts of lowered water levels.

Soil Biol Biochem 38:2011–2024 MNP (2005) Effecten van klimaatverandering in Nederland. MNP, Bilthoven, 112. Lasage, R., Aerts, J., Verburg, P., and Sileshi, A. (2013) The role of small scale sand dams in securing water supply under climate

change in Ethiopia. Mitigation and Adaption Strategies for Global Change, 1-23. Lasage, R., Veldkamp, T. I. E., de Moel, H., Van, T. C., Phi, H. L., Vellinga, P., and Aerts, J. C. J. H. (2014) Assessment of the

effectiveness of participatory developed adaptation strategies for HCMC. Natural Hazards and Earth System Sciences Discussions, 2:385-425.

Lee, D. J., Dias, E., and Scholten, H. J. (2014) Introduction to Geodesign Developments in Europe. In Geodesign by Integrating Design and Geospatial Sciences, Springer International Publishing, 3-9.

Ligmann-Zielinska, A., Church, R. L., and Jankowski, P. (2008) Spatial optimization as a generative technique for sustainable multiobjective landuse allocation. International Journal of Geographical Information Science, 22:601-622.

Longley, P. A., Goodchild, M. F., Maguire, D. J., and Rhind, D. W. (2005) Geographic Information Systems and Sciense. John Wiley and Sons.

Malczewski, J. (2006) GIS-based multicriteria decision analysis: a survey of the literature. International Journal of Geographical Information Science, 20:703-726.

Margerum, R.D. (2002) Collaborative Planning: Building Consensus and Building a Distinct Model for Practice. Journal of Planning Education and Research 21(3):237-253.

Massey, E. and Huitema, D. (2013) The emergence of climate change adaptation as a policy field: the case of England. Regional Environmental Change, 13:341-352.

McCall, M.K., Dunn, C.E. (2012) Geo-information tools for participatory spatial planning: Fulfilling the criteria for 'good'governance? Geoforum, 43:81-94.

McClintock, W. (2013) Geodesign: Optimizing stakeholder-driven marine spatial planning. Coast Guard Journal of Safety and Security at Sea, Proceedings of the Marine Safety and Security Council, 70.

McIntosh, B.S., Giupponi, C., Voinov, A.A., Smith, C., Matthews, K.B., Monticino, M., Kolkman, M.J., Crossman, N., van Ittersum, M., Haase, D., Haase, A., Mysiak, J., Groot, J.C.J., Sieber, S., Verweij, P., Quinn, N., Waeger, P., Gaber, N., Hepting, D., Scholten, H., Sulis, A., van Delden, H., Gaddis, E,. and Assaf, H. (2008) Chapter Three Bridging the Gaps Between Design and Use: Developing Tools to Support Environmental Management and Policy. In: Developments in Integrated Environmental Assessment, Environmental Modelling, Software and Decision Support. Elsevier, 33-48.

Meng, Y., Malczewski, J. (2009) Usability evaluation for a web-based public participatory GIS: A case study in in Canmore, Alberta, Cybergeo: European Journal of Geography.

Miettinen, K. (2008) Introduction to Multiobjective Optimization: Noninteractive Approaches. In J. Branke, D. Kalyanmoy, K. Miettinen, and R. Slowinski (Eds.), Multiobjective Optimization, Springer Berlin Heidelberg, 1-26.

Moser, S.C. and Ekstrom, J.A. (2010) A framework to diagnose barriers to climate change adaptation. Proceedings of the National Academy of Sciences 107(51):22026-22031.


Bibliography

160

Mukheibir, P. and Ziervogel, G. (2007) Developing a Municipal Adaptation Plan (MAP) for climate change: the city of Cape Town. Environment and Urbanization, 19:143-158.

Niang-Diop, I. Bosch, H. (2004) Formulating an adaptation strategy. Adaptation Policy Frameworks for Climate Change: Developing Strategies, Policies and Measures. Cambridge University Press.

Nieuwenhuis, H.S., Schokking, F. (1997) Land subsidence in drained peat areas of the Province of Friesland, the Netherlands. Quarterly Journal of Engineering Geology and Hydrogeology, 30:37–48.

Nyerges, T., Jankowski, P., Tuthill, D., Ramsey, K. (2006) Collaborative Water Resource Decision Support: Results of a Field Experiment. Annals of the Association of American Geographers, 96(4):699-725.

Olde-Venterink, H., Davidsson, T.E., Kiehl, K., Leonardson, L. (2002) Impact of drying and re-wetting on N, P and K dynamics in a wetland soil. Plant and Soil, 243:119–130.

Opdam, P. (2010) Learning science from practice. Landscape Ecology, 25:821-823. Opdam, P., Nassauer J.I., Wang, Z., Albert, C., Bentrup, G., Castella, J.C., McAlpine, C., Lui, J., Sheppard, S., and Swaffield, S.

(2013) Science for action at the local landscape scale. Landscape ecology, 28(8):1439-1445. Ozimec, A.M., Natter, M., Reutterer, T. (2010) Geographical Information Systems-Based Marketing Decisions: Effects of

Alternative Visualizations on Decision Quality. Journal of Marketing, 74:94-110. Pelzer, P., Arciniegas, G., Geertman, S., and Kroes, J. (2013) Using MapTable to Learn About Sustainable Urban Development. In:

Planning Support Systems for Sustainable Urban Development. Springer Berlin Heidelberg, 167-186. Pettit, C., Klosterman, R., Nino-Ruiz, M., Widjaja, I., Russo, P., Tomko, M., Sinnott, R., and Stimson, R. (2013) The Online What

if? Planning Support System. In: Planning Support Systems for Sustainable Urban Development. Springer Berlin Heidelberg, 349-362.

Pettit, C.J., Raymond, C.M., Bryan, B.A., Lewis, H. (2011) Identifying strengths and weaknesses of landscape visualization for effective communication of future alternatives. Landscape and Urban Planning, 100(3), 231-241.

Porta, J., Parapar, J., Doallo, R., Rivera, F. F., Sante, I., and Crecente, R. (2013) High performance genetic algorithm for land use planning. Computers, Environment and Urban Systems, 37:45-58.

Pouwels R, Opdam P, and Jochem R (2011) Reconsidering the Effectiveness of Scientific Tools for Negotiating Local Solutions to Conflicts between Recreation and Conservation with Stakeholders. Ecology and Society, 16(4).

Preston, B.L., Westaway, R.M., and Yuen, E.J. (2011) Climate adaptation planning in practice: an evaluation of adaptation plans from three developed nations. Mitigation and Adaptation Strategies for Global Change, 16(4):407-438.

Prutsch, A., Grothmann, T., Schauser, I., Otto, S., McCallum, S. (2010) Guiding principles for adaptation to climate change in Europe. Technical Paper 2010/6. The European Topic Centre on Air and Climate Change (ETC/ACC).

Querner, E.P., Jansen, P.C., Kwakernaak, C. (2008) Effects of water level strategies in dutch peatlands: a scenario study for the polder Zegveld. In: Proceedings of the 13th International Peat Congress: after wise use—the future of Peatlands, Tullamore, Ireland, 2008:620–623.

Querner, E.P., Jansen, P.C., van den Akker, J.J.H., Kwakernaak, C. (2012) Analysing water level strategies to reduce soil subsidence in Dutch peat meadows. Journal of hydrology, 446:59–69.

Reckien, D., Flacke, J., Dawson, R. J., Heidrich, O., Olazabal, M., Foley, A., J. J.-P. Hamann, H. Orru, M. Salvia, S. De Gregorio Hurtado, Geneletti, D., Pietrapertosa, F. (2014) Climate change response in Europe: what’s the reality? Analysis of adaptation and mitigation plans from 200 urban areas in 11 countries. Climatic Change, 122(1-2):331-340.

Reed, M. S. (2008) Stakeholder participation for environmental management: A literature review. Biological Conservation, 141: 2417-2431.

Ribeiro, M., Losenno, C., Dworak, T., Massey, E., Swart, R., Benzie, M., Laaser, C. (2009) Design of guidelines for the elaboration of Regional Climate Change Adaptations Strategies. Study for European Commission - DG Environment - Tender DG ENV. G.1/ETU/2008/0093r. Vienna, Ecologic Institute.

Rienks, W.A., Gerritsen, A.L., Meulenkamp, W.J.H. (2002) Behoud veenweidegebied: een ruimtelijke verkenning. Alterra-rapport 563, Alterra, Research Institute voor de Groene Ruimte, Wageningen.

Riet, van de, B.P., Hefting, M.M., Verhoeven, J.T.A. (2013) Rewetting drained peat meadows: risks and benefits in terms of nutrient release and greenhouse gas exchange. Water Air and Soil Pollution, 224:1–12.

Rinner, C., Kessler, C., Andrulis, S. (2008) The use of Web 2.0 concepts to support deliberation in spatial decision-making. Computers Environment and Urban Systems, 32:386-395.

Roggema, R. (2009) Adaptation to climate change: A spatial challenge. Springer, 59-111. Salter, J. D., Campbell, C., Journeay, M., and Sheppard, S. R. J. (2009) The digital workshop: Exploring the use of interactive and

immersive visualization tools in participatory planning. Journal of Environmental Management, 90:2090-2101. Santé-Riveira, I., Boullón-Magán, M., Crecente-Maseda, R., and Miranda-Barrós, D. (2008) Algorithm based on simulated

annealing for land-use allocation. Computers and Geosciences, 34:259-268. Schatz R, Walker D, Kroes J, Arciniegas G, Getz P, and Khare A (2013) Geographic Information Systems (GIS) as a Tool in

Reducing a Community's Ecological Footprint. In: Mitigating Climate Change. Springer Berlin Heidelberg, 25-43. Schipper, L.A., McLeod, M. (2002) Subsidence rates and carbon loss in peat soils following conversion to pasture in the Waikato

Region, New Zealand. Soil Use and Management, 18:91–93.


Bibliography

161

Scholten, H. J., van de Velde, R. J., van Manen, N. (2009) The role of geo-ICT and spatial approaches in Science. Springer. Schothorst, C.J. (1977) Subsidence of low moor peat soils in the western Netherlands. Geoderma, 17:265–291 Schouwenaars, J.M. (2002) Water levels in the Echten polder: improving agriculture and reducing land subsidence 273.

International Association of Hydrologcial Sciences, Wageningen. Schrier-Uijl, A.P., Kroon, P.S., Leffelaar, P.A., van Huissteden, J.C., Berendse, F., Veenendaal, E.M. (2010) Methane emissions in

two drained peat agro-ecosystems with high and low agricultural intensity. Plant and Soil, 329:509–520. Schröter, D., Polsky, C., and Patt, A. G. (2005) Assessing vulnerabilities to the effects of global change: an eight step approach.

Mitigation and Adaptation Strategies for Global Change, 10:573-595. Seppelt, R., Lautenbach, S., and Volk, M. (2013) Identifying trade-offs between ecosystem services, land use, and biodiversity: a

plea for combining scenario analysis and optimization on different spatial scales. Current Opinion in Environmental Sustainability, 5:458-463.

Shaw, R., Colley, M., and Connel, R. (2007) Climate change adaptation by design: a guide for sustainable communities, London, TCPA.

Sidlar C.L, Rinner, C. (2009) Utility assessment of a map-based online geo-collaboration tool. Journal of Environmental Management, 90(6):2020-2026.

Sieber, R. (2006) Public Participation Geographic Information Systems: A Literature Review and Framework. Annals of the Association of American Geographers, 96(3):491-507.

Slocum, T.A., McMaster, R.B., Kessler, F.C., Howard, H.H. (2009) Thematic cartography and geovisualization. Pearson Prentice Hall Upper Saddle River, NJ.

Smit, B. and Wandel, J. (2006) Adaptation, adaptive capacity and vulnerability. Global Environmental Change-Human and Policy Dimensions, 16:282-292.

Smit, B., Burton, I., Klein, R. J. T., Street, R. (1999) The science of adaptation: A framework for assessment. Mitigation and Adaptation Strategies for Global Change, 4:199-213.

Smit, B., Burton, I., Klein, R. J. T., Wandel, J. (2000) An anatomy of adaptation to climate change and variability. Climatic Change, 45:223-251.

Smit, B., Wandel, J. (2006) Adaptation, adaptive capacity and vulnerability. Global environmental change 16(3):282-292. Steinitz, C. (2012) A framework for Geodesign: changing geography by design. ESRI Press Redmond. Stewart, T. J. (2009) A critical survey on the status of multiple criteria decision making theory and practice. Omega, 20:569-586. Stewart, T. J. and Janssen, R. (2014) A multiobjective GIS-based land use planning algorithm. Computers, Environment and

Urban Systems, 46:25-34. Stewart, T. J., Janssen, R., van Herwijnen, M. (2004) A genetic algorithm approach to multiobjective land use planning.

Computers and Operations Research, 31:2293-2313. Stowa (2005) Help-2005, Uitbreiding en actualisering van de helptabellen ten behoeve van het waternood-instrumentarium.

Rapport 2005–16 STOWA (in Dutch). Sui, D. and Goodchild, M. (2011) The convergence of GIS and social media: challenges for GIScience. International Journal of

Geographical Information Science, 25:1737-1748. SWD 134 (2013) Guidelines on developing adaptation strategies. Brussels, European Commission. Te Brömmelstroet, M. (2012) Performance of Planning Support Systems: What is it, and how do we report on it? Computers,

Environment and Urban Systems. Turner, B. L., Kasperson, R. E., Matson, P. A., McCarthy, J. J., Corell, R. W., Christensen, L., Eckley, N., Kasperson, J. X., Luers, A.,

and Martello, M. L. (2003) A framework for vulnerability analysis in sustainability science. Proceedings of the National Academy of Sciences, 100:8074-8079.

UKCIP (2005) UKCIP Tools: Adaptation Wizard, http://www.ukcip.org.uk/wizard/ last accessed February 2015. UNFCCC (2015) United Nations Framework Convention on Climate Change, http://unfccc.int/adaptation/items/7006.php last

accessed on January 10, 2015. Uran, O. and Janssen, R. (2003) Why are spatial decision support systems not used? Some experiences from the Netherlands.

Computers, Environment and Urban Systems, 27(5):511-526. Vanegas, P., Cattrysse, D., and Van Orshoven, J. (2011) A Multiple Criteria Heuristic Solution Method for Locating Near to

Optimal Contiguous and Compact Sites in Raster Maps. In B. Murgante, G. Borruso, and A. Lapucci (Eds.), Geocomputation, Sustainability and Environmental Planning. Springer Berlin Heidelberg, 35-56.

Vervoort, J. M., Kok, K., van Lammeren, R., and Veldkamp, T. (2010) Stepping into futures: Exploring the potential of interactive media for participatory scenarios on social-ecological systems. Futures, 42:604-616.

Vonk, G., Geertman S., Schot, P. (2006) Usage of planning support systems. Innovations in Design and Decision Support Systems in Architecture and Urban Planning, 263-274.

Vonk, G., Geertman, S., Schot, P. (2005) Bottlenecks blocking widespread usage of planning support systems. Environment and Planning A, 37:909-924.

Vonk, G., Geertman, S., Schot, P. (2007) A SWOT analysis of planning support systems. Environment and Planning A, 39:1699-1714.


Bibliography

162

Vonk, G., Ligtenberg, A. (2010) Socio-technical PSS development to improve functionality and usability-Sketch planning using a Maptable. Landscape and Urban Planning, 94:166-174.

Waddell, P. (2002) UrbanSim: Modeling urban development for land use, transportation, and environmental planning. Journal of the American Planning Association, 68:297-314.

Weaver, C.P., Lempert, R.J., Brown, C., Hall, J.A., Revell, D., and Sarewitz, D. (2013) Improving the contribution of climate model information to decision making: the value and demands of robust decision frameworks. Wiley Interdisciplinary Reviews: Climate Change, 4(1):39-60.

Wenkel, K.O., Berg, M., Mirschel, W., Wieland R., Nendel, C., and Kostner, B. (2013) LandCaRe DSS: An interactive decision support system for climate change impact assessment and the analysis of potential agricultural land use adaptation strategies. Journal of Environmental Management, 127:S168-S183.

Wesseling, J.G. (1985) De invloed van bodemsoort en vochtgehalte op de bodemtemperatuur nota 1645:27. Instituut voor Cultuurtechniek en Waterhuishouding, Wageningen (in Dutch).

Willows, R. I and Connell, R. K. (2003) Climate adaptation, riks, uncertainty and decision-making, UK-CIP technical report. Wilson, C. and McDaniels, T. (2007) Structured decision-making to link climate change and sustainable development. Climate

Policy, 7:353-370. Wilson, E. (2006) Adapting to Climate Change at the Local Level: The Spatial Planning Response. Local Environment, 11(6):609-

625. Wilson, E. and Piper, J. (2010) Spatial planning and climate change. Routledge London. Witjes, Th. G. J. (2011) Climate research Texel, Hoogheemraadschap Hollands Noorderkwartier, Witteveen Bos (in Dutch). Witte, J., Runhaar, H., van Ek, R., van der Hoek, D. (2009) Eerste landelijke schets van de ecohydrologische effecten van een

warmer en grilliger klimaat. H2O, 16–17:37–40 (in Dutch). Wyss, R., Luthe, T., and Abegg, B. (2014) Building resilience to climate change - the role of cooperation in alpine tourism

networks. Local Environment, 1-15. Xiao, N., Armstrong, M.P. (2012) Towards a Multiobjective View of Cartographic Design. Cartography and Geographic

Information Science, 39:76-87. Zak, D., Gelbrecht, J., Wagner, C., Steinberg, C.E.W. (2008) Evaluation of phosphorus mobilization potential in rewetted fens by

an improved sequential chemical extraction procedure. European Journal of Soil Science, 59:1191–1201.


Dankwoord

163

Dankwoord Aan mijn PhD ben ik net zoals mijn master bescheiden en terughoudend begonnen. Uitgangspunt zoals van huis uit meegekregen, was: ‘Meer dan je best doen kun je niet doen’. Erg trots ben ik nu ook dat het gelukt is. Ik weet van mezelf dat ik houd van afwisseling, een 4-jarig project ligt dan niet direct voor de hand. Gelukkig was dit project ook erg praktisch en is er in de afgelopen jaren zowel in het werk als thuis genoeg afwisseling geweest. De komst van onze lieve zoon Tobian halverwege het traject heeft er zeker voor gezorgd dat ik weer verder kon met de tweede helft. Het afronden van mijn proefschrift was extra leuk wetende dat er een zusje op komst is. De afwisseling tussen wetenschap en thuis zorgt voor mij voor een fijne harmonie tussen denken en doen. Hiervoor wil ik graag Eef bedanken, die thuis altijd voor me klaar staat en regelmatig de moeite nam om buiten zijn eigen vakgebied te gaan. Mijn zus wil ik bedanken voor haar vrolijke en relativerende noot die me liet onthouden dat er meer is dan wetenschap. Maar ook mijn ouders wil ik heel erg bedanken. Zij gaven mij mee dat wat een ander kon ik ook kon en maakte een doorzetter van mij. Daarnaast zorgden ze ook voor een fijne uitvalsbasis op de camping waar wandelen, fietsen en gezellig samen zijn voor mij de perfecte combinatie vormen om bij te komen van of op te laden voor een werkweek. Ook wil ik verdere familie, vrienden en buren bedanken voor hun interesse en steun. Het IVM als instituut en als plek was voor mij een thuis als het gaat om het bij elkaar komen van verschillende interesse gebieden. Met plezier heb ik tijdens de vele koffie en lunchpauzes geluisterd naar namen van chemische stoffen en apparaten om allerlei analyses mee te doen. Het deed me terug denken aan mijn bachelor milieuwetenschappen, een Summerschool toxicologie, stages en werk bij TNO en RIVM, toxicologie in Innsbruck en de studieverhalen van Eef. Maar vooral wil ik mijn collega’s bij C&B bedanken voor de vele gezellige sociale uurtjes. In het bijzonder, Petra, bedankt, je was een erg fijne kamergenoot. Ook bij SPACE kon ik mijn plek prima vinden en ik wil ook al mijn SPACE collega’s bedanken voor hun gezelligheid. Peter Verburg, Philip Ward en natuurlijk Gustavo Arciniegas bedankt voor jullie input tijdens het opstarten van mijn PhD en Jasper van Vliet, Ralph Lasage en Eric Massey bedankt voor de input bij het schrijven later. Next, thank you Paul, David and Belinda for correcting my English. In het bijzonder wil ik Nancy Omtzigt en Alfred Wagtendonk bedanken voor hun geduldige uitleg om mijn GIS vaardigheden bij te schaven en om de vele workshops voor te bereiden en te ondersteunen. Denkend aan de workshops wil ik ook graag Fritz Hellmann bedanken en natuurlijk Karlijn Brouns. Samen maakten we wat leuks van de voorbereidingen en de workshops. Daarnaast wil ik alle collega’s en beleidsmakers bedanken die meegedaan hebben aan de workshops en (online) vragenlijsten. Het was daarnaast ook erg fijn om bij het onderzoeksprogramma kennis voor klimaat en de onderzoeksschool sense te horen. Het was erg leuk om via deze twee netwerken regelmatig bekenden en gelijkgestemden tegen te komen op cursussen, conferenties en symposia. Eén daarvan is Marjolein Mens en haar wil ik graag bedanken voor onze eensgezindheid die leidde tot de succesvolle Sense schrijfweek en een H2O artikel. De laatste maanden heb ik gewerkt bij FEWEB en ook alle collega’s daar wil ik bedanken voor hun hartelijke ontvangst zo op het einde van mijn PhD. Maurice, het was leuk om elkaar daar weer tegen te komen. Henk, bedankt voor het aanvaarden van de promotorrol en de hulp bij het afronden.


Dankwoord

164

Uiteindelijk staat of valt een PhD project bij het hebben van een goede begeleider. Hoewel het even wennen was aan de Amsterdamse directheid heb ik deze al snel kunnen waarderen en samen waren we een leuk duo. Ook de Friese stakeholders konden de combinatie van het Brabantse meisje en haar Noord-Hollandse baas wel waarderen. Samen sleepten we een heel circus mee voor onze voorstelling met de Touch Table en sloten we deze regelmatig af met een uitstekend hapje eten. Ron, bedankt voor je vertrouwen en kritische houding, het heeft me veel geleerd.


About the author

165

About the author Tessa Eikelboom was born on April 15th, 1987 in Waalwijk, the Netherlands. She studied at Utrecht University, receiving a BSc in Environmental Sciences and a MSc in Physical Geography. During her BSc she studied at Innsbruck University as part of the Erasmus exchange program. She did an internship on risk assessment of chemicals at TNO Quality of life, the Netherlands. Next, she worked on a project on electromagnetic fields at National institute for public health and the environment (RIVM) de Bilt, the Netherlands. After her study, Tessa Eikelboom joined the Institute for Environmental Sciences (IVM) at VU University as a PhD researcher. Tessa’s research interests are in the field of environmental risk, sustainability, climate change, earth sciences and knowledge exchange between science and practice.

Publications Alexander, K.A., Janssen, R., Arciniegas Lopez, G.A., O'Higgens, T.G., Eikelboom, T. and Wilding, T.A. (2012) Interactive Marine

Spatial Planning: Siting tidal energy arrays around the Mull of Kintyre. PLoS One, 7(1):1-9. doi:10.1371/journal.pone.0030031.

Beek, L.P.H. van, Eikelboom, T., Vliet, M.T.H. van and Bierkens, M.F.P. (2012) A physically-based model of global freshwater surface temperature. Water Resources Research, 48(9), W09530. doi:10.1029/2012WR011819.

Bolte, J.F.B. and Eikelboom, T. (2012) Personal radiofrequency electromagnetic field measurements in the Netherlands: exposure level and variability for everyday activities, times of day and types of area. Environment International, 48: 133-142. doi:10.1016/j.envint.2012.07.006.

Bolte, J.F.B., Baliatsas, C., Eikelboom, T., van Kamp, I. (2015) Everyday exposure to power frequency magnetic fields and associations with non-specific physical symptoms. Environmental Pollution, 196:224-229.

Brouns, K., Eikelboom, T., Jansen, P.C., Janssen, R., Kwakernaak, C., Akker, J.H.J. van den and Verhoeven, J.T.A. (2014) Spatial analysis of soil subsidence in peat meadow areas in Friesland in relation to land and water management, climate change and adaptation. Environmental Management. doi:10.1007/s00267-014-0392-x.

Eikelboom, T. and Janssen, R. (2013) Interactive spatial tools for the design of regional adaptation strategies. Journal of Environmental Management, 127: S6-S14. doi:10.1016/j.jenvman.2012.09.019.

Eikelboom, T. and Janssen, R. (2014) Evaluation of geodesign maps for spatial planning. Peer-reviewed Proceedings, DLA Conference 2014, Zurich, Switzerland, http://dla2014.ethz.ch/talk_pdfs/DLA_2014_8_Eickelboom.pdf.

Eikelboom, T. and Janssen, R. (2015) Comparison of geodesign tools to communicate stakeholder values, Group Decision and Negotiation, doi:10.1007/s10726-015-9429-7.

Eikelboom, T. and Janssen, R. (2015) Collaborative use of geodesign tool to support decision making on climate adaptation, Mitigation and Adaptation strategies for global change, doi:10.1007/s11027-015-9633-4.

Eikelboom, T., Janssen, R., Stewart, T.J. (2015) A spatial optimization algorithm for geodesign, Landscape and Urban planning. Haas, E.M. de, Eikelboom, T. and Bouwman, T. (2011) Internal and external validation of the long-term QSARs for neutral

organics to fish from ECOSARTM. SAR and QSAR in Environmental Research, 22(5-6), 545-559. doi:10.1080/1062936X.2011.569949.

Janssen, R. and Eikelboom, T. (2014) Using geodesign to support collaborative planning workshops. Peer-reviewed Proceedings, DLA Conference 2014, Zurich, Switzerland, http://dla2014.ethz.ch/talk_pdfs/DLA_2014_1_Janssen.pdf.

Janssen, R., Eikelboom, T., Verhoeven, J.T.A. and Brouns, K. (2014) Using geodesign to develop a spatial adaption strategy for Friesland. In D. Lee, E. Dias and H. Scholten (Eds.), Geodesign by integrating design and geospatial sciences (GeoJournal Library), New York: Springer, 103-116. doi:10.1007/978-3-319-08299-8_7.

Download - Interactive geodesign tools to support regional adaptation planning · Fig. 1. Board game Settlers of Catan (©Catan GmbH, 2015) Similar to the tasks in the game, the challenges for

Top Related