- Project Overview and integrative approach
- Work Package 1: City Learning Labs
- 1.2.1 Embedded researchers in 3 Tier 1 cities
- 1.2.2 Train the trainer approach
- 1.2.3 Webinars
- 1.2.4 Smartphone app
- 1.2.5 Visualisation of big or complex data
- 1.3.1 Participatory scenario building
- 1.3.2 Participatory video in the M&E process
- 1.3.3 Iterative processes of learning in and between the City Learning Labs
- 1.3.4 Audit of the actor knowledge network
- WP 2: Transdisciplinary research to integrate climate science in decision-making
- Task 2.1: Development of an urban-regional system framework for climate impact analysis
- Task 2.2: Understanding the complex decision making space
- Task 2.3: Analysis and critical assessment of the co-exploration approach
- Task 2.4: Identifying thresholds to uptake of increasingly detailed climate information
- Work package 3: Advances in understanding changes in the regional climate
- Management and coordination
1. Project Overview and integrative approach
What we will address
The project’s overarching aim is to advance scientific knowledge about regional climate responses to anthropogenic forcings, and to enhance the integration of this knowledge into decision making at the co-dependent city-region scale that responsibly contributes to resilient development pathways. This is addressed through an iterative, transdisciplinary co-exploration / co-production approach, worked through a set of deep-focus case studies.
Why this is critical to Africa
Development pathways in Africa are faced by the serious challenge of rapid urban growth that creates a fluid landscape for decision makers. This challenge is fundamentally bound to questions of resilience in the face of intensifying climate stresses on complex multi-stressor environments, affecting resource and infrastructure governance and management. However, the dynamics of current and future climate variability and change on these city-regions are poorly understood, particularly at the regional sub-national scale. Additionally, there is evidence that the available climate data is poorly translated to information and is not well used to inform policy and decision making in cities.
In the face of no clear and binding international treaty to frame a globalized response to climate change, regions are facing a necessary shift from a focus on the science-policy nexus toward critical questions of how climate change impacts decision-making across governance, economics, business, energy, national security planning etc. These decisions carry long-term implications, and are intertwined with complex local, regional and transboundary dependencies.
Our point of departure
We identify three equally critical and co-dependent intellectual challenges:
i. To understand the climate processes driving the African regional climate system’s natural variability and response to global change in the recorded history and climate model simulations;
ii. To distil defensible, scale-relevant climate information, informed by and tailored to urban decision making and risk management within their regional dependencies;
iii. To use co-exploration of climate information with urban partners within the systems-thinking paradigm to integrate climate messages within real-world decisions, and enhance the resilience of development pathways.
To tackle these challenges, the project adopts a unique research focus on the neglected yet vitally important urban scale that is embedded within and intimately coupled to the regional scale. This is arguably one of the most critical societal vulnerabilities given the rapid urbanisation (UN Habitat 2013) that is occurring across the African continent. The impacts associated with an increasing urban footprint are being further accentuated by the changing nature of resource flows and the cross-scale dependencies, which together create unique vulnerabilities and risk. Despite the magnitude of the urban challenge, climate science has yet to adequately respond at the urban scale (and does so only weakly at the regional scale).
Where and why we focus our attentions
This project will focus its primary attention on three Tier 1 cities within their associated regional contexts (Windhoek, Maputo and Lusaka), and to a lesser degree on three Tier 2 cities (Blantyre, Gaborone and Harare) to help explore a range of contexts. These are key cities in the sub-continent, and represent a strong climate gradient from arid to wet sub-tropical, a significant contrast of society and culture, and a range of risk exposures and governance issues with local and regional dependencies. In addition we leverage the added value of research with two self-funded city partners (Cape Town and eThekwini (Durban)) in South Africa. These two cities represent the substantial challenge of how to develop urban centers within developing nation constraints.
Within this context the critical resources for urban centres, and hence vulnerabilities, are water and energy which strongly depend on functional infrastructure.
Informed by the literature and reinforced through our building relationships and initial discussions (during the proposal preparation workshops) with local government officials in our Tier 1, 2 and partner cities, we place our primary attention on water and energy. We recognize food as equally important but also note that this is a complex mix of trade and production that is both local and global, where the local aspects are critically dependent on water and energy. We also acknowledge the key issue of health in cities but note that this vulnerability is critically tied to water supply and quality, and to flooding.
Climate change impacts water resources in cities by stressing the storage, supply, and demand of water, as well as directly affecting livelihoods through extreme events, variability, damage to infrastructure and agriculture. Energy (situated at the heart of enabling potential for development) is characterized by large climate risks to managing capacity for generation, transmission, and demand growth, particularly for city-regions that depend on hydropower generation.
Both water and energy systems are to a large extent spatially clustered within distribution networks, with source, storage and delivery to critical urban centers having exceptionally strong regional coupling. These components represent issues of very high risk through the intersection of existing and future hazards, exposure, and vulnerability (Gemenne et al. 2014), and are compounded by city-scale infrastructure and planning decisions. Effective decision making therefore needs to be able to engage the physical, economic and political drivers of vulnerability.
Framing the research issues and the approaches to implementation
Three fundamental components need to be addressed:
- the climate science (including translation, tailoring, and communication),
- understanding the decision making space, and
- understanding the cross-scale dynamics at play within, between and beyond urban areas. We approach this through hypotheses (Table 1) that are addressed in three research work packages (Figure 1)
Table 1: Framing hypotheses for the Work Packages (WP) and indicative timing of effort.
Note: WP1=Pillar 3; WP2=Pillar 2; WP3=Pillar 1.
1. The climate science
Uptake of climate change information in decision making is most undermined by the inherent spread, and sometimes contradictions, within the spectrum of multiple models, downscaling methods, and observational data sets (as has also been identified as a grand challenge focus by the WCRP). This problem confounds an effective response by society due to a propensity by the climate services community to conflate multiple sources of uncertainty; inherent natural variability of the climate, reducible uncertainty -knowledge, structural error in models, etc. – and manageable uncertainty such as scenarios. This conflation often leads to weakened messages from science being dismissed by decision makers. Failing to engage with this frontier challenge only serves to perpetuate the inadequacies of the status quo. We address these through linked research themes in WP3.
Figure 1: Representation of the project structure: Work packages and tasks
2. The policy and decision making space
Decision makers are not effectively using available climate information. This research posits that this apparent failure is not simply a failure of information, or communication, or even of individual decision makers, but rests in the meeting of these factors in a particular institutional and knowledge context (e.g. Daron et al. 2014a). There is a need to understand the decision making space in order to facilitate the effective uptake of climate information for decision making. Or in the words of Berkhout (2012, p.92): “Many factors play a role in shaping the decisions and actions of organizations, of which perceptions about climate change will be but one. An analysis of organizational adaptation therefore needs to start with the complex reality of organizations themselves, rather than starting with the climate signal and then seeking to trace its presumed influence on organizational behavior. The analysis needs to be done inside-out, rather than outside-in.” This was demonstrated through a real world infrastructure decision in Cape Town (Daron 2014b).
Critical analysis of the decision making space is conducted in WP2 and explored in depth with city case studies (WP1) to develop novel and effective approaches to enhancing the uptake of climate information by decision makers.
3. Cross-scale dynamics
Scales of municipal decision making are often poorly aligned to ecological spatial, temporal or functional scales (Cumming et al. 2006). This research therefore seeks to understand the cross-scale dynamics of energy, water, with the intention of developing strategies to overcome governance-based scale mismatches. The temporal scale mismatch is often acute with short-term decisions (5 – 10 years) often in conflict with longer term goals (10-40 years). Spatially, decision making at the municipal scale is informed by national policies and frameworks, shaped by the influence of international development and donor agencies (amongst others). However, the municipal government’s sphere of influence is generally bound by specific mandates within the bounds of physical boundaries, which are met through formal and informal governance arrangements across multiple scales.
Water and energy systems, driven by the regional climate, transcend political and administrative boundaries, and operate on time frames beyond the usual consideration of decision makers (Folke et al. 1997). This problem is addressed through systems and institutional analysis in WP2, drawing on understanding of the regional climate process dynamics from WP3, and tested in the case studies of WP1. In turn this is used to inform the scalar assumptions in WP2 and 3, and takes a transdisciplinary, co-exploration approach that mixes qualitative and quantitative approaches.
Integration within and beyond the consortium
Addressing key question of climate (WP3) and systems (WP2) in an integrated fashion requires research that is highly participatory and that challenges the prevalent persistence of disciplinary barriers. Such efforts will be fundamental to achieving political and technical buy-in to produce tangible outcomes.
We explicitly construct our activities to engage the cross-institutional collaboration through framing hypotheses that cannot be adequately addressed by one institution alone. The consortium builds its research tasks with explicit collaboration with academics from partner universities in Tier 1 and 2 cities and with government officials. Tier 1 and Tier 2 participants are also twinned to develop opportunities for peer-to-peer learning. These individuals have already played a role in scoping and preparing the research and practice agenda during the project proposal preparation and remain engaged. The academic partners and their city colleagues are pivotal in effecting the relationship with the core research, and especially in designing the evolving research foci. Likewise the two self-funded city partners have a long relationship with the consortium members, and bring valuable lessons learned, with a commitment to substantially collaborate and invest resources in achieving the real world impacts.
The three Tier 2 cities (Blantyre, Gaborone and Harare) and the valuable self-funded partner cities in South Africa will allow for cross-learning with Tier 1. To avoid the assumption that approaches that work in one context can be simply “scaled up/across” to another, we target transferability of the knowledge production from Tier 1 cities, and of the city-region system understanding (see WP2).
A project researcher is embedded in local government of each Tier 1 city and each self-funded city (see Section 1.2.1) – an approach that has already been tested in a UCT/City of Cape Town research project. These individuals will be a link to sustaining the relationship with the consortium, and will act as permanent connection points between the various work package researchers and city decision makers. These relationships lay the foundation for long term sustainable partnerships within the region.
Small opportunity grants (60,000 GBP budgeted) are a key strategic enabler of research activities that can address emerging questions throughout the project. The project strongly emphasizes the value of emerging research questions that are not anticipated at the outset. These grants will be allocated to researchers within Tier 1 or Tier 2 cities. Research activities under these grants will be driven by the in country researchers but significant cross consortium and cross work package engagement will be an explicit requirement to avoid isolated and unconnected research.
A critical mode of co-exploration will be the establishment of City Learning Labs (WP1), which respond to Pillar 3 in the project call. This is the application of the knowledge intersection between WP2 and WP3 (See Figure 1). The activities within these Labs serve both to test the findings through case studies, and from this to feed back to the core research to refine the work of WP2 and 3. Although co-exploration activities occur at specific moments in the project, co-exploration is also a continuous mode within the Tier 1 cities through the embedded researchers.
We also intend developing and applying a range of experimental and innovative methods and tools to facilitate capacity building, doing so in partnership with the CCKE, and include monitoring the changes in the uptake of climate knowledge over time. An experimental online knowledge sharing and researcher-networking environment that builds on the existing platforms will mitigate the isolating effects of geographical distances between in-country research partners, and facilitate real time engagement. These efforts will develop a community of practice.
When investigating decision making in complex systems, an overemphasis on causality and linearity can overlook the more tacit decision making drivers, as they are not easily articulated and thus difficult to disentangle. This can assume human action is more rational and planned than it actually is (Mowles 2009). To elicit this level of understanding, FRACTAL will ensure regular and iterative reflection on the processes and interactions between the Tier 1 and Tier 2 cities and across and between all 3 work packages, by adopting an overarching and integrated approach to both internal and external learning. On an external level, this will contribute to the ‘indicators of progress’, which will be identified in collaboration with local partners in the monitoring and evaluation process and to help evaluate the ‘most significant changes’ occurring throughout the project. On an internal level, iterative learning will facilitate the sharing of failures and lessons learnt as well as successes to continually feed into the project cycle.
What FRACTAL will deliver, and the fit to call
FRACTAL’s aim (outlined above) is strongly African designed and African led (by regional researchers, who are centrally engaged in the research of all work packages – see section on management). This builds on strong established partnerships of individuals and institutions of international excellence in research, and will advance the within-continent capacity to maximize value from science, minimize climate change consequence for society, and optimize the near term trajectory of regional development pathways for long term resilience of the majority population.
The research activities on natural and social system are founded on an innovative co-exploration framework and in depth case studies (implemented through City Learning Labs – WP1) that is centered on and informed by a sustained dialogue between decision makers, and leverages embedded researchers within governance institutions and academic research partnerships (Task 1.1). Through this we create a legacy of increased capacity among the regional academic and decision making communities, supported by new relational collaboration within and between disciplines and stakeholders (Task 1.2). The integration with urban social systems research and the physical climate research serves to inform, optimize, and steer the decision making (Task 1.3).
The project delivers research excellence through innovation in each WP. WP3 addressing the requisite scientific knowledge frontiers about the African regional climate system dynamics (WP3, Task 3.1 – 3.5), while the urban-regional social systems and attendant infrastructural issues are addressed in WP2. WP2 and WP3 have points of joint activities (within Tasks 2.3, 2.4, 3.3, 3.4, 3.5), and WP1 is the primary means for integration that informs and impacts the decisions that steer the development pathways toward resilience (WP1, and drawing on WP2 Task 2.1 – 2.4 and WP3 Tasks 3.3-3.5). Fundamental research priorities are continually informed by the dialogue within the case studies (WP1), and from the scientific literature and the internationally framed grand challenge research (e.g. from the WCRP and Future Earth). By leveraging the range of activities by the consortium partners we add significant value to the inter-disciplinary partnership of regional and international scientists which will impact development pathways.
FRACTAL targets a step advance in understanding the dynamics of the regional climate system processes (WP3) through exploring the co-behavior of the driving system processes in historical and model data. This is integrated with the understanding of climate sensitive social system dynamics, related thresholds and dependencies of the coupled urban-region system linkages (WP2).
The work provides innovation in methods, tools and communication of information products (Tasks 3.2-3.4) in relation to vulnerabilities, thresholds, and climate impacts (Tasks 2.4, 3.3 – 3.5). These are tailored to inform the decision-making community (Task 2.2) with defensible and actionable climate information (Task 3.3), with an understanding of the attributes of uncertainty and probability in order to best manage risk. We leverage the voluminous data sets already available and augment these with new data tailored to critical questions of regional vulnerabilities (Task 3.1).
We bring to bear our expertise in all relevant facets of physical climate science, infrastructural engineering, environmental economics, decision making, and the social dimensions of institutional governance and urban geography. In each area we have world class scientists with long track records and excellent stakeholder engagement experience in Africa, with strong capacity in climate services (CSAG is arguably Africa’s leader in this).
Capacity building is integral to the consortium’s work and will occur through workshop-based learning, inter-institutional exchange of researchers for targeted skill building, interdisciplinary research collaborations, the embedding of researchers, and strengthening of science communication skills that inform and animate the decision making space.
2. Work Package 1: City Learning Labs
This work package largely maps to Pillar 3 of the call and is guided by hypotheses 1, 2 and 3 of WP2 and will facilitate transformative learning and engagement within Tier 1 cities regarding the use of climate information for the support of decision making. This will happen through implementing a “City Learning Labs” approach which involves; a) the transdisciplinary co-exploration of specific deep case studies in each Tier 1 city b) building capacity and communication channels for better informed decision making, and c) monitoring the impact and change in policy and decision making.
The following cities have been identified as Tier 1 cities and will host a “City Learning Lab”: Maputo, Lusaka, Windhoek. While the main focus of the project will be on these 3 cities, Tier 2 cities (Blantyre, Gaborone and Harare) will apply some of the lessons learnt adapted to their context. The cities have expressed a commitment to the process. While the focus in on cities, the learning and dialogue process will explicitly integrate regional connectivity and dependencies within the systems-thinking framework.
The integrated City Learning Labs are fundamental to the project through: i) linking research from WP 2 & 3 to a real world iterative dialogue and case study, and ii) informing, directing and narrowing the research questions on which WP 2 & 3 will focus. In particular, urban system thresholds (e.g. infrastructure design specifications, rainfall-sea level-flood tolerances, heat stress etc.) and associated climate sensitivities will inform the research foci of WP3. Additionally WP1 activities and outputs will steer the research foci in WP2, as well as respond to and apply the knowledge from WP2 in real case study decisions. The two self-funded cities in South Africa will engage with the project team on a level similar to the Tier 1 cities through an embedded researcher.
Based on interacting and twinned cities (Tier 1 & 2) this WP will also extend knowledge gained through Tier 1 cities into Tier 2 cities and test the limits and constraints of applying this knowledge more widely. Tier 2 activities will be primarily through desktop studies and expert interviews, though there are opportunities for targeted research through small opportunity grants.
Tier 1—Tier 2 cross learning will be accomplished through bilateral exchanges. These allow exchanges of researchers between institutions for up to one year to gain specific analytical skills and/or to observe transdisciplinary engagement through WP1 activities. In addition to direct capacity building gained through cross-city engagement, ancillary benefits created through the opportunity grants will include participation of Tier 2 cities.
WP1 will focus on three tasks that will be integrated within the research undertaken to address the three hypotheses under WP2. These are listed below with (non-exhaustive) examples of methods that can be used for each:
Task 1.1: Development of the case studies
WP1 is primarily implemented through real world decision case studies within each Tier 1 city. Through the initial proposal workshop including city officials we have identified water and energy as key issues and a number of emerging case studies have been identified (eg. ground water in Windhoek). Initially the project will, through further engagement with the cities, identify suitable focused decision case studies that will allow a deep exploration of the overarching hypotheses.
Task 1.2: Participatory co-exploration of climate related challenges:
1.2.1 Participatory action research
A facilitated participatory action research process informed by the work of Kemmis & McTaggart (2000) and Senge & Scharmer (2001) will allow a range of stakeholders to engage in exploring the complexity which will emerge from co-exploration processes undertake in WP 2. While the action research process allows the “co-researchers” in the process to use their expertise and share the results with the group, they are engaging in an iterative learning process, actively exploring possible solutions in this joint process. The facilitation of ‘competency groups’ is a further method that will be used at the interface between decision making and environmental problems. This method involves engaging a range of ‘knowledge-holders’ to work together to co-produce new knowledge that ‘redistributes expertise’ (e.g. Lane et al. 2011).
1.1.3 Formal methods for decision-making
Using the joint co-exploration approach, multiple formal and semi-formal (combined with participatory approaches) methods for decision making will be explored, based on the assumption that there is no ‘one size fits all’ approach to understanding and supporting decision making in this diversity of decision contexts. Methods that focus on medium-term decision making and can utilize economics and systems framework (WP2) and model based inputs (WP3) such as Real Options Analysis, Robust Decision Making or Portfolio Analysis. In addition, other methods which can incorporate more qualitative information (e.g. WP2), such as the Analytical Hierarchy Process (AHP) and multi-criteria analysis will also be tested within the case studies through the City Learning Labs based on their conditions of applicability.
Task 1.2: Building capacity and communication channels for transdisciplinary engagement
1.2.1 Embedded researchers in 3 Tier 1 cities
While social learning is an important goal, it does not necessarily lead to social transformation (Shaw & Kristjanson 2013). It is necessary to engage with structures and power that underpin and embed social learning. By paying attention to the political economy of knowledge production and use, knowledge is made transformative and co-learning and co-production become possible. The project proposes to embed researchers within local government in order to better understand the un-codified everyday practices of governance that influence the ability to engage with climate information and the perceived limitations to existing climate information, as guided by the WP2 hypotheses. The embedded researcher and the wider research team will then work with municipal officials to develop realistic strategies, focused on the case study,to circumvent the identified barriers and to use the points of leverage to improve the uptake of climate information, where appropriate and desirable. The identification of un-codified knowledge and bridging and highly connected actors in the analysed networks (Task 1.1.2) as well as the embedding of researchers within local government will help strengthen existing urban multi-stakeholder platforms. This will bring together representatives of local governments and civil society that could play a key role in further engaging with climate information in decision making. These platforms could be used as a further basis for the co-exploration approach (WP 2, Hypothesis 1).
1.2.2 Train the trainer approach
Train the trainer approaches will be used to scale up the capacity building activities of the project. By providing training to a core set of embedded practitioners and co-developing a wider transference of knowledge plan, there will be a facilitated transfer of knowledge and sense of ownership developed in-country that has the potential to enhance the impact of the project beyond the limitations of the project bounds.
When the context is appropriate, webinars will be used to facilitate interaction between the cities on learning and challenges resulting from the project. Webinars will be used as one technique for developing a supportive community of practice amongst the Tier 1 and 2 cities, and a means for informing the Work Package 2 research.
1.2.4 Smartphone app
The motivation for development of a smartphone app is supported by the wide use of mobile technology in Africa. Currently SMS services are used as an effective mechanism for dissemination of forecast information due to internet bandwidth constraints. Through this experimental task, an app will be developed (in collaboration with the CCKE unit) that will incorporate the concept of virtual gaming to build capacity and support and simplify the use of climate information within an urban decision making context on the 5-40 year time scale.
1.2.5 Visualisation of ‘big’ or complex data
Using outputs from various tasks in WP 2 and 3, FRACTAL will apply innovative visualisation methods, such as controversy mapping, to explore the complex qualitative and quantitative data produced (Venturini, 2012), revealing relationships about the barriers and opportunities for the uptake of climate information. For example, analysis of these visualisations will provide entry points which can be leveraged to improve the robustness of climate messages contributing to Task 3.4. Output from WP3 will iteratively provide new data for these visualizations and provide further learning opportunities. And, translating this information to simple, communicable infographics will contribute to the creation of new channels for transdisciplinary engagement (Task 3.5b).
Task 1.3: Monitoring the impact and change in policy and decision making
1.3.1 Participatory scenario building
This tool allows stakeholder to conceptualise adaptation plans based on their current capacities and assets, the climate disturbances they are facing, their current vulnerability and perceptions of climate change. Participants are led through a brainstorming exercise to develop road maps/ adaptation alternatives towards the desired state of the community in the future, exploring opportunities for decision-making processes in innovative ways. Both real and hypothetical climate change information will also be introduced to explore the contribution of climate information to the process. Relates to Task 2.3.2 in WP2.
1.3.2 Participatory video in the M&E process
An active learning process will include regular moments of reflection on city level throughout the process. In order to explore how process of dialogue is impacting on decision-making processes on city level, we suggest to conduct a participatory video baseline process at the inception phase. This process would include documenting the actual situation within the cities of integrating climate information in medium term decision making while defining indicators for assessing potential change in decision making. This baseline can be used to monitor project impact and will feed into the project monitoring and evaluation process. While it is always difficult to attribute change to one particular project or intervention, several monitoring, evaluation and learning processes will be implemented (see below), to capture as much of the drivers of change throughout the project as possible.
1.3.3 Iterative processes of learning in and between the City Learning Labs
The project will host bi-annual learning events between the City Learning Labs to deepen the learning on city level and to contribute towards a process exploring different methods for transdisciplinary co-exploration on the meta level. As part of the learning processes the City Learning Labs will develop a simulation game specific to their identified medium term challenges (see for example Red Cross Climate Centre, START and UNISDRA-Africa 2013). These games will be played in a tournament with all City Learning Labs – opening and deepening the discussions of transdisciplinary co-exploration and a way of creatively sharing research results.
1.3.4 Audit of the actor knowledge network
A socio-institutional actor network analysis (WP2) will be carried out at the start of the project to capture a baseline regarding the decision-making landscape and then repeated near the close of the project as a way of assessing impact of the learning exchanges between research and government partners in Tier 1 and Tier 2 cities to identify changes in the network structure and/or the strength of the links after the implementation of the FCFA programme.
3. WP 2: Transdisciplinary research to integrate climate science in decision-making
This work package delivers the key research on the use of climate information in decision making. The research activities of WP2 are guided by the posed hypotheses and informed by the learning and knowledge generated in WP1. Likewise, WP2 stands as a critical analytical bridge to WP3 by exploring the integration of climate science and impacts modeling into decision-making processes within the systems-thinking paradigm. WP2 also explicitly undertakes the role of critiquing of and reflecting on the iterative, transdisciplinary co-exploration / co-production approach adopted in the project.
Contemporary climate impacts work in Africa has mostly not probed these complex place-based linkages but has largely adopted a sectoral approach. An impacts driven approach explicitly places sector-based climate impacts at the top of the agenda. This approach aligns well with economic planning and policy, which informs significant investment decisions across key urban sectors. However, the sectoral approach does not always work, and there is an emerging realization that risk and impacts manifest within a system of interacting networks; social, bio-geophysical, political and economic, and thus systems thinking is crucial to addressing vulnerability/resilience questions (Berkes & Ross 2012).
The research foci of this work package involve analysing the extent to which participatory co-exploration processes and an understanding of the ‘complex decision space’ are necessary precursors to maximising the uptake of climate information; and to determine the threshold at which further additions of climate information no longer enhance decision making. A central tenet of this WP is that climate information is contextual and that it is critical to recognise that context within the wider decision-making space. The first two tasks therefore approach the description of the context via two different avenues, a systems approach, and an analysis of the urban configurations in each city. The second two tasks analyse the co-exploration approach and the value of increasingly detailed climate information within the process respectively. This analytical research focus is iteratively interdependent with WP1 and WP3 (See Figure 1).
The tasks within this work package will be led by UCT, Tier 1 city-based researchers, START, SEI, Red Cross and ICLEI, with participation by Aurecon and NASA on the infrastructure components. In line with the project’s co-exploration approach, the regional partners will co-develop the research tools and be involved in all phases of the research. Acknowledging the key role that engineering consultancies play in both impacts modeling, setting of design standards, and design and implementation of infrastructure, the consortium has explicitly included a partnership with a major large infrastructure engineering consulting company with a presence and prior experience in a number of large African cities. Likewise the strategic inclusion of the discipline of economics is crucial for engaging with dominant economics principles such as decision discounting. This work package is also strongly informed by urban geographers with expertise in understanding institutional and governance aspects of decision making. Equally critical, through WP1, this work package will engage with multiple disciplinary experts within both the Tier 1 and Tier 2 cities themselves and draw on the expertise, experience, and competency within these cities.
Task 2.1: Development of an urban-regional system framework for climate impact analysis
While city level decision-making is mostly concerned with local-scale processes and dynamics, the regional multi-sectoral dependencies should be taken into account, particularly at mid to longer term decision consequence horizon, and if they display climate sensitivities. To facilitate this process, the urban-regional (perhaps even global) co-dependencies will be formalized in a systems framework. Building a comprehensive quantitative model of complex multi-scale, multi-sectoral urban-regional system is a task exceeding the scope of this project. We will, however, create a variable-complexity model, focusing on sub-system dependencies which are explicitly articulated, and where climate sensitivities and responses to change are described initially qualitatively, and if identified as key constraints, formalized in a quantitative sub-model of complexity warranted by perceived significance and available information. Considering the focus of the project, water resource and agricultural sub-systems will be modeled explicitly within this framework, using JULES land surface model. Other aspects, such as regional energy constraints, economic growth, technological progress, migration will be considered in a semi-quantitative way, through a number of scenarios, drawn from existing body of literature, experts opinion or modeled using time and resource expedient approaches. Analyses and simulations within this framework will be driven by climate data generated in WP3 and will interactively guide development of climate messages of value within that WP. The development of the framework will be an integral part of (i.e. it will be created during, and informed by) co-exploration process carried within WP1, and will be used to frame remaining activities in WP2, speaking directly to hypothesis WP2a (Table 1).
Task 2.2: Understanding the ‘complex decision making space’
While information pathways are important in shaping municipal responses to the projected impacts of climate change, this project hypothesizes that in order to understand the response of decision makers to climate information it is essential to develop an understanding of the broader institutional context and political economy in which decisions are taken. This includes understanding what legislation, fiscal regulation and sets of norms and standards frame the potential decision space, what technical lock-in limits innovation, and what are the embedded power relations and institutional culture that shape individual decision-maker and departmental responses to external information and flow of information across institutions and scales of governance. This component of the project therefore seeks to provide an understanding of the factors that shape the decision-making space within particular sectors of local government through the following tasks:
2.2.1 Document analysis
Collect and analyse regulatory and legislative documentation related to the critical municipal mandates, such as spatial and strategic planning, relevant legislation and regulatory frameworks, engineering design standards, municipal budgeting processes, and related sectoral programmes and projects.
2.2.2 Socio-institutional network analysis
Building upon and triangulating with Task 1.1.1, a socio-institutional actor network mapping approach will incorporate different sectors of local government, working at different scales, i.e. local (community, neighbourhood, ward or district), city, city-region, provincial, national and international). The analysis will focus on the contemporary context, however this will inform the limits and opportunities that climate projections have on effectively informing medium term cross-sectoral decisions. This will help visualise formal and informal power dynamics, information exchange, and barriers to decision-making such as weak collaborations or institutional gaps, as well as potential problems that may emerge if information flows are disrupted between critical nodes.
2.2.3 Interview and focus groups
Conduct focus groups and individual interviews with municipal officials, politicians and other relevant stakeholders in order to understand the municipalities’ perceptions of climate risks; how climate risks and adaptation decisions are managed in the institutional framework and decision making processes. Added to this will be the use of economic experiments (games) which facilitate testing how perceptions of risk interact with information provision and the role of social norms on uptake of information.
2.2.4 Urban climate adaptation configuration analysis
Drawing on the outputs of WP1 (particularly the embedded researchers) and Tasks 2.1 and 4.2, undertake a qualitative analysis of the specific climate adaptation configuration/urban governance arrangements and outcomes in each Tier 1 city.
2.2.5 Local community engagement
Conduct focus group discussions with relevant local actors to develop an understanding of local community coping and adaptation responses, particularly the use of indigenous knowledge
Task 2.3: Analysis and critical assessment of the co-exploration approach
The project proposes that a transdisciplinary co-exploration approach maximises the integration of climate information into decision making and thereby delivers measurable positive impacts on decision making. The tasks below are designed to support and critique this hypothesis by drawing strongly on the parallel case study learning of WP1 as well as undertaking further research and analysis of the co-exploration approach guided by questions emerging from WP1. It is also important to note that the co-exploration, or similar transdisciplinary approaches form significant components of other research that consortium members have been and are currently engaged in. This provides a significant point of leverage for broader learning. There is also a an existing pool of literature addressing various aspects of ’experimental governance’ which will contribute value to the research (e.g. Bulkeley & Castan Broto 2012).
2.3.1 Documentation, analysis, and generalisation of co-exploration approach
Drawing on the outputs of WP1, we will analyse the extent to which participatory co-exploration processes increases the potential for climate information uptake in decision making in each city throughout the timeline of the project. This task is firstly concerned with documenting and analysing the co-exploration process with a focus on the role of key aspects of the process such as varying discourses (climate change, development, resilience), their visualisation and translation, climate literacy, decision science literacy, trust and authority, power hierarchies and the readiness to engage in critical dialogue, and face to face versus other modes of engagement. Secondly, the task will describe how the learning around these aspects of the process relate to the larger picture of urban decision making within each Tier 1 city, the Tier 2 cities, and where possible, African cities in general. This will draw strongly on the tasks related to the second hypothesis below as well leverage parallel research projects and activities such as the recently funded ESRC/DFID Cities at Risk.
2.3.2 Critique, gap analysis and refining
This task captures the critical analysis of the co-exploration approach by unpacking the learning of Task 2.1 and identifying key failures or challenges. It is expected that the early stages of the project and the co-exploration activities will highlight lack of relevant or appropriate climate information as a critical gap. Further analysis of this gap will be required to determine if the knowledge gap is one of translation, communication, visualisation, or a fundamental gap in the climate science. This will strongly inform the evolving research agenda of WP3, specifically Tasks 3.3, 3.4 and 3.5.
It is also anticipated, given prior experience of the co-exploration approach, including during the FCFA pilot phase, that certain challenges will emerge such as: the significant time required to undertake new climate analysis in response to evolving decision demands for information; the strong tendency to defer to the perceived authority and importance of science; and the challenges of extrapolating co-exploration beyond a workshop context and into real world realities. The critique process will be ongoing and will provide an important input into the evolving design of the co-exploration process and associated activities in WP1.
Task 2.4: Identifying thresholds to uptake of increasingly detailed climate information
There is a seemingly ever-growing demand on climate scientists to produce more detailed climate information and projections. This can mean higher resolution but can also refer to the need for greater detail around climate characteristics such as seasonal onsets and extremes. In reality, often very rudimentary types of climate information are used (annual averages, regional averages etc.) when more detailed information could provide better guidance to decision makers (this is the object of the prior tasks of this work package). However, many studies including the CSAG-led FCFA pilot study for Accra and Maputo and a number of previous consortium activities, suggest that there is a limit to the amount of detail that decision processes are able to use due to the inherent nature or limitations of the decision space. This threshold or point of diminishing returns is going to be context dependent and qualitative but identifying this threshold is a valuable activity as it could avoid wasting considerable effort producing new and more detailed climate information that cannot actually be used. By tracking the use and impact of advanced climate science (developed in WP3) in decision-making across multiple decisions at the city scale, a threshold at which further information no longer changes the outcome of any adaptation decision-making process can be determined.
2.4.1 Tracking the evolving real world decision space
Through iterative reflection via WP1 activities (participatory research, network analysis and decision support) (referring to the 3 tasks under Task 1 in WP2), and drawing on a pathways approach (Wise et al. 2014), the decision making process will be tracked over the duration of the project to map the evolution of decision making given the availability of more detailed or tailored climate information and in relation to the following: evolving climate literacy, shifting information needs, increased uptake of climate information, climate information deficits.
2.4.2 Participatory scenario developments and simulated decision making
To supplement the previous task, which may offer some analysis challenges, the Participatory Scenario Building activities of Task 1.3.1 in WP1 will be used to further explore the thresholds to information uptake with the possibility of introducing hypothetical climate information if sufficiently detailed climate information is not yet available or if the information demand threshold in fact exceeds the available climate information.
4. Work package 3: Advances in understanding changes in the regional climate
This work package leads from the question “What gaps in our scientific understanding, if addressed, would maximise the value content of regional climate information for risk management and decision makers?” To address this we engage on two fronts in an iterative cycle of evolving research. First, through co-design with WP2 and 3, we identify the critical knowledge gaps in the climate science that constrain the effective value of the information, and from this develop approaches to maximise the skill of defensible information that is tailored to the decision needs. Second, this engagement is used to structure the core physical climate research questions to advance our understanding of the regional climate system, and so provide the necessary knowledge foundations for the development of robust and scale relevant climate information.
The tasks within this work package will be led by UCT, UKMO, and SMHI. Regional partner contributions involve the CSIR, and academic/NMS from the nations of the Tier 1 cities. Additional self-funded contributions include the Lawrence Berkeley National Laboratory (LBNL) and the EC’s Joint Research Center. In line with the project’s co-exploration approach, WP2 and the regional partners will co-evolve the research priorities over the projects lifetime. LBNL also contribute extensive computer time for the regional partners to undertake targeted model-based experiments.
Foci of priority research: This work package situates the research within four hypotheses (see Table 1). These consider what knowledge areas will best advance the value for impact on development, and put forward the position that the value will be maximised through (i) evolving existing metrics and data to best assess the driving multi-scale processes, (ii) understanding the co-behaviour of processes in the real world and in models, (iii) finding new approaches to identifying the real skill in climate projections, and (iv) gaining insight into the sources and scale dependencies of uncertainty. To tackle these we formulate a set of co-dependent research tasks, each with specific research questions informed through co-design with WP2. These require strong cross-institutional and cross-disciplinary collaboration, and are intended to synergistically evolve over the project duration as understanding is advanced. The tasks are:
Task 3.1: Research foundations:
To support the tasks described further below, here we target three enabling activities;
i) Data. In conjunction with Tasks 3.3 and 3.4 below we will develop new data products tailored for relevance to users. These will include new quality controlled station data, a gridded best-estimate historical climate data set with associated uncertainty measures obtained through integration of station and gridded products, and model/downscaling integrated products suitable for application in impacts modeling and which span the representative future climate space. Data sources will draw on all available observation, reanalysis, and satellite data, including the extensive archives held by CSAG and the UKMO. We will also access data that are not already in the public domain through direct engagement with the region’s national meteorological services. For model data we will use CMIP3/CMIP5, as well as the WCRP’s CORDEX-Africa downscaling data (Regional Climate Models (RCMs), empirical-statistical downscaling (ESD)).
ii) Metrics. We recognize that the conventional metrics of models and their simulation of multi-scale processes, as well as associated measures of uncertainty, are inadequate to properly assess the dynamics of the regional climate system response to forcing (Klocke 2011). This task thus focuses on definitions and metrics of added value and skill for use in understanding the regional process dynamics, for process-based model evaluation, and for identifying and assessing sources of uncertainty. The development of these metrics is designed for, and will co-evolve with, the ongoing research.
iii) Methods. We note how sophistication of the climate community’s analyses seriously lag the voluminous production of data which has led to a “Big Data” conundrum. As noted by Faghmous and Kumar (2014, p.161), “What is needed is the development of data-driven methodologies that are guided by theory to constrain search, discover more meaningful patterns, and produce more accurate models”. Consequently, framed by the scale-relevant and decision-centric information needs, this task will evolve new methodologies to help elucidate process behaviour and relevant defensible messages on change from the multiplicity of data. For this we draw strongly on the perspective of user-relevance, and leverage the consortium member’s participation in complementary projects in this area of research. Development of new analytical methods will begin with exploratory ideas from consortium member’s existing projects that include the use of high-dimensional phase space analysis, ideas on signal analysis and multivariate signal identification techniques, and a suite of traditional analytical methods applied in a new context of mapping uncertainty and skill as a function of scale in time and space, and of driving processes. We expect to leverage the outputs from the forthcoming WCRP Expert Meeting on Climate Information “Distillation” which explicitly sets out to explore research ideas to address these specific topics (Consortium members are the lead for this meeting).
Task 3.2: Process dynamics
This task explores the dynamics, responses, natural variability, and especially co-behavior of key components of the climate system. We do this recognizing the range of relevant scales that spanning local processes to global modes of variability, including teleconnections and feedbacks. The objective is to understand the roles of the relevant local and regional controls in the context of the conditioning global modes and teleconnections as together these deliver the local scale expression of the climate system. These activities will identify key processes that modulate or translate global scale forcing to the scales relevant for urban decision making. In doing so this will also identify the key processes of concern in the model simulation, and to what extent these are skillfully simulated, or even absent. In turn this will help inform the evolving design of user-relevant and process-based metrics in Task 1 to evaluate model performance.
In addition to using observational and model data, in this task we will also undertake custom model simulations designed to test our interpretations and explore the simulation of processes critical to conditioning or driving the local climate. Examples of important local processes include antecedent soil moisture, aerosols (especially as pertains to fire), regional processes such as organized convection, tropical-temperate troughs, cut-off lows, and land cover change and degradation on the longer time scales. The central questions are the drivers of trends and variability, the co-behavior of local processes alongside other large-scale processes, surface responses to the combination of process, simulation skill and error in GCMs and RCMs of the processes critical to local climate change, and the impact of unresolved processes in models.
A initial set of modeling experiments will include a suite of very high resolution (~4km) multi-model (RCM and stretched grid GCM) simulations with the objective of determining the value of (a) including key processes such as aerosols and land surface dynamics and (b) very high resolution modeling, in capturing key regional processes.
Task 3.3: Defensible regional and local scale climate messages
This task focuses on the spatial, temporal and co-variate attributes of the climate system that impact the threshold vulnerabilities of the coupled urban-regional socio-economic system. Central to this task is the distillation dilemma – how to understand and analyze the multi-model multi-method and multi-scale (M5S) spread of data that contains inherently contradictory messages (this also directly contributes to WCRP’s Grand Challenges). This will start from three concepts; integration across model outputs through process-related weighting approaches, selection and exclusion of data sets based on processed-based metrics of skill, and signal processing techniques to identify and isolate signals of value. We target the understanding of the spatial-temporal co-behavior of relevant variables and related extremes as a function of process dynamics, and characterize the added value of different analytical and downscaling methods. Core questions include how the local and regional response relates to the mix of multi-scale processes, and building on this in what way downscaling and integration of multiple variables within impact models can compensate for GCM deficiencies (including consideration of bias correction issues). This work will directly engage with the co-exploration and systems-thinking framework through engagement in WP 2 and 3.
Task 3.4: Uncertainty and probability of scale-relevant messages
Uncertainty, as is commonly used in climate services, conflates natural variability with reducible uncertainty (knowledge, structural error, etc) and manageable uncertainty (scenarios, pathways etc.). This contributes to weak messages of probable change and can be used by decision makers to avoid taking climate-informed decisions. For example, the growing evidence of contradictions between models and observations of process response to forcing (e.g. Shin & Sardeshmukh, 2011) can seriously undermine confidence in model output for local scales. In this task we addresses this substantial problem through quantifying and understanding the relative sources of disparity between models, methods, and observations in the context of natural variability, recognizing the scale dependencies in space and time.
By using the breadth of observed and model / downscaling outputs we first explore the reasons for the contradictions and spread of data, and at which spatial and temporal scales of processes the evident disparity between models and observations is most critical (drawing on cross-evaluation of simulations at different scales). Second is the question of at what time and space scales are skillful decision-relevant signals emergent from natural variability. Third is to understand to what extent processes that are key to determining the local/regional response contribute to the uncertainty of the derived messages. From these beginnings we explore the challenging question of the relationship between simulation skill of historical process behavior and the forced climate change response for future projections. Within these we include foci on how does the mixed coupling of different downscaling-GCM pairing influence the skill of user-relevant signals, and how impacts modeling integrate uncertainty in the climate signal to provide more robust message to user?
Task 3.5: Cross-cutting research
Grounded in the above tasks are three cross-cutting actions that are closely integrated with the WP1&2 activities.
a) Using impact models coupled to the climate information. This has two main foci, to explore how impact models can integrate or exacerbate the uncertainty in the climate change messages, and to explore the coupling (Koster et al. 2004) between the critical surface ecosystem responses and the driving atmospheric dynamics. This modeling will be embedded in and will contribute to the analytical systems framework developed in WP2.
b) Research on the communication and visualization of the climate information that leverages Tasks 3 and 4 above, and works in close collaboration with related work in WP 1 & 2. We will build on existing momentum by CSAG where we note that “Choices about the visualisation style and information content form part of the interpretation process, and climate data providers need to recognise the possible consequences of such choices in providing climate data and communicating climate change messages to a diverse user community” (Daron et. al, in review).
c) Third will be the development of online tools to facilitate the reach of climate data and information to as broad an audience as possible. This leverages substantial experience with the CIP and weADAPT websites, and uses these to incorporate the learning from the communication and visualization research in conjunction with the insights from WP1&2.
h) Modeling resources: For the targeted simulation experiments we draw will use appropriate selections of the available climate models: SMHI RCA4 and HARMONIE regional climate models (on SMHI computers), the Hadley HadGEM3-RA model (on UKMOP computers, and including the 4km resolution African simulations planned as part of the proposed FCFA IMPALA project – if that is awarded), the NCAR/NCEP WRF model (on CSAG computers as well as extensive free computer time through LBNL), and the NCAR CAM-EULAG and CSIRO CCAM stretched grid global models (on the South African national HPC facility). The tailored model experiments will be conducted with the additions of on-line and off-line land surface, hydrological and hydro-ecological, and agricultural models.
5. Management and coordination
The work packages will be strategically managed overall by the Lead PI with support from the Project Research Coordinator based in CSAG at the University of Cape Town. Each work package will be co-chaired by a two Co-PIs based in different institutions. These individuals will form the basis of the Project Management Committee. This committee will have oversight of project sequencing and progress, and have responsibility for reporting. The Lead PI is responsible for the overall coherence of the research and for the final deliverables. Together with the Project Research Coordinator he will oversee the activities within the work packages and act as the liaison point with NERC/DFID.
One Co-I will be paired with each embedded researchers in the cities and act as the point of contact, sounding board and advisor. The embedded researchers will meet once a year to share experiences and lessons learned. Additionally, one Co-I will act as the liaison point for each Tier 1 partner, and one to be liaison point for all Tier 2 partners. One Co-PI to oversee Tier 1 and 2 engagements. These Co-Is together with one representative from each Tier 1 and 2 city, and a member of the Project Management team will constitute the Partner Coordination Committee. This committee will coordinate activities across organizations, develop annual work plans and timelines, contribute to annual report, review progress to date and monitor project implementation.
The project will employ an Academic Coordinator to handle project logistics, lead the communication with CCKE, and work with CCKE to develop appropriate cross-consortium and joint capacity development activities. The Academic Coordinator will engage on a day-to-day basis with FCFA. The project’s Finance Officer will manage details of financial reporting. Each Tier 1 and Tier 2 city will have one dedicated academic team leader, who is based in the Tier city and will coordinate a local research team. Tier 1 cities will also have municipal officials (or a national government official in the case of Maputo) as key parts of the project team. All these individuals have been identified and have committed to the project.
Participatory Advisory Team
The entire project will be overseen by the Participatory Advisory Team, and candidates for this team have been identified. This group, comprising of leading experts in climate science and urban governance, will aim to meet bi-annually to review progress, are expected to engage in some of the in-city activities, and can be called on at any time by the PI to deal with project developments and difficulties.
Capacity development will occur through a series of integrative activities associated with the three work packages. The partners’ approach to capacity development posits that sustained capacity mobilization best occurs through iterative learning embedded in the context of local needs and priorities, rather than being circumscribed by externally driven and designed workshops, which too often serve as a proxy for capacity building without serious consideration of impact.
We aim to achieve impact in capacity development through i) peer-to-peer learning via the sustained twinning of researchers with city government partners at their institutions; ii) cross-disciplinary engagement embedded within a transdisciplinary framework that allows for learning through knowledge sharing across the research community as well as at key research-policy-practice interfaces; iii) animation of learning at these key interfaces through smartphone apps and dynamic data visualisation; and iv) close partnerships with city-based researchers that emphasize capacity development through ‘learning by doing’ augmented with targeted skill-building workshops and an opportunity grants fund for cross-city knowledge sharing. The partners will work closely with CCKE to identify opportunities for mutually beneficial engagement that enhances the capacity development impacts of FRACTAL and that supports CCKE’s efforts across the FCFA programme.
CSAG and UCT have a proven track record of managing large international consortium grants. The project finance officer together with the wider university research contracts office will coordinate payment to sub-contracted partners, and will be responsible for gathering financial reporting from these partners. The Finance Officer, Academic Coordinator and PI will hold overall responsibility for financial reporting back to NERC/DFID.
Complementary activities and leverage opportunities
We note that the consortium brings exceptionally strong leverage of resources, and these will inherently be involved in the project to bring notable direct and indirect added value. We highlight some of these here. In terms of computational resources the LBNL contributes substantial HPC time for the regional partners to use in dedicated model experiments (and in so doing building HPC capacity). Likewise the UKMO, SMHI, and EC JRC all bring substantial computing resources. Likewise CSAG, UKMO, and NASA have substantial observational data resources that are available for the project. CSAG and SMHI are core project leaders in the WCRP CORDEX program and thus have direct access to the CORDEX archive and to emerging new CORDEX RCM data sets, and the CORDEX programme office is based at the SMHI. CSAG also lead the CORDEX statistical downscaling and coordinate the experiments for the projects domain with access to the downscaling output. With regard to major project linkages, CSAG is a core research partner in the SASSCAL program that undertakes complementary research in the region, and the PI is on the steering committee of the emergent Africa-wide Climate Research for Development (CR4D) program operating under the Africa Climate Policy Centre (ACPC), as well as being engaged on a broad range of regional climate services and capacity building projects, and leads of collaborates on a wide range of physical climate research, including seasonal forecasting, all of which have complementary value.
The consortium also brings excellent connections to key science organizations of relevance to Africa (for example the IPCC TGICA and the WCRP WGRC are co-chaired by the PI, and the AgMIP is organized by our NASA partner), and also with other development and funding agencies with interests in southern Africa (e.g. World Bank, US AID, and JAMSTEC). Likewise excellent linkages have been established with the cities, and a strong track record of capacity building through coursework and co-exploration, and a strong network of users of the CSAG Climate Information Portal for a co-exploration approach to climate services. Consortium members are also engaged with the GFCS and Future Earth, as well as the critical development of a dialogue on the ethics of climate services (including the drafting of a white paper for the climate services community).
The consortium is similarly connected in the urban and decision making realm through ACC (African Center for Cities), AAPS (African Association of Planning Schools), ICLEI (Local Governments for Sustainability). Through ACC the consortium has links to Cities Alliances and UN Habitat.
Mindful of the need to reduce the carbon footprint, the project will use remote workshops and web-based meeting tools wherever possible. However, in order to generate the kinds of rich dialogue required for the transformative vision of the project, some face-to-face workshops are essential. The number of workshops and the need for project travel is reduced by the multiple means of engagement between researchers and users, most notably, the long-term embedded researchers and the remote tools to be developed, such as the smartphone app.
Berkes F. & H. Ross (2013) Community Resilience: Toward an Integrated Approach. Soc. Nat. Resour. 26, 5–20.
Berkhout F. (2012) Adaptation to climate change by organizations. Wiley Interdiscip. Rev. Clim. Chang. 3, 91–106.
Bulkeley H. & Castan Broto V. (2013) Government by experiment? Global cities and the governing of climate change, Transactions of the Institute of British Geographers, vol. 38, no. 3, pp. 361-375.
Cumming GS., et al. (2006) Scale Mismatches in Social-Ecological Systems: Causes, Consequences, and Solutions, Ecology and Society, vol. 11, no. 1. [http://www.ecologyandsociety.org/vol11/iss1/art14/]
Daron D., et al. (2014) The role of regional climate projections in managing complex socio-ecological systems, Regional Environmental Change, DOI 10.1007/s10113-014-0631-y
Daron, J.D. 2014. Challenges in using a Robust Decision Making approach to guide climate change adaptation in South Africa. Climatic Change.
Faghmous J.H. & Kumar V. (2014) A Big Data Guide to Understanding Climate Change: The Case for Theory-Guided Data Science, Big Data, vol. 2, no. 3, pp. 155-163. doi:10.1089/big.2014.0026.
Folke C.Å. et al. (1997) Ecosystem appropriation by cities. Ambio vol. 26. no. 3, pp. 167-172. 10.2307/4314576.
Gemenne F. et al. (2014) Climate and security: evidence, emerging risks, and a new agenda, Climatic Change, 10.1007/s10584-014-1074-7.
International Energy Agency (IEA) (2014) Africa Energy Outlook Report: A Focus on Energy Prospects in Sub-Saharan Africa (http://bit.ly/1txi2LG).
Kemmis S. & McTaggart R. (2000) Participatory action research, in: D Denzin & Y Lincoln (eds.), Handbook of Qualitative Research, pp. 567-605.
Klocke, Daniel, Robert Pincus, and Johannes Quaas. “On constraining estimates of climate sensitivity with present-day observations through model weighting.” Journal of Climate 24.23 (2011): 6092-6099.
Koster, Randal D., et al. “Regions of strong coupling between soil moisture and precipitation.” Science 305.5687 (2004): 1138-1140.
Lane S.N. et al. (2011) Doing flood risk science differently: an experiment in radical scientific method, Transactions of the Institute of British Geographers NS, vol. 36, pp. 15–36.
Mowles C. (2009) Consultancy as temporary leadership: negotiating power in everyday practice, International Journal of Learning and Change, vol. 3, no. 3, pp. 281 – 293.
Parnell S. & Pieterse E. (eds.) (2014) Africa’s Urban Revolution, London, Zed Books.
Red Cross Climate Centre, START & UNISDRA-Africa (2013) Using games to experience climate risk: Empowering Africa’s decision makers, Final Report: CDKN.
Senge P. & Scharmer O. (eds.) (2001) Community Action Research: Learning as a Community of Practitioners, Consultants and Researchers, London, Sage.
Shaw A. & Kristjanson P. (2013) Catalysing learning for development and climate change: An exploration of social learning and social differentiation in CGIAR. Copenhagen, CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) .
Shin S.I. & Sardeshmukh P.D. (2011) Critical influence of the pattern of Tropical Ocean warming on remote climate trends. Climate Dynamics, vol.36, pp. 1577-1591, doi: 10.1007/s00382-009-0732-3.
Venturini T. (2012) Building on faults: How to represent controversies with digital methods. Public Understanding of Science, 21(7). 796–812. DOI:10.1177/0963662510387558.
Wise R.M. et al. (2014) Reconceptualising adaptation to climate change as part of pathways of change and responses, Global Environmental Change, vol. 28, pp. 325–336. DOI: 10.1016/j.gloenvcha.2013.12.002
UN Habitat (2013) State of the World’s Cities 2012/2013, New York, Routledge.