Knowledge Based Climate Mitigation Systems for a Low Carbon Economy

Final Report Summary - COMPLEX (Knowledge Based Climate Mitigation Systems for a Low Carbon Economy)

Executive Summary:
The COMPLEX team has been working on pathways to a low carbon economy by 2050. An integrative approach was required that would bring natural scientists, human scientists and external stakeholders together to focus on key challenges. We have devoted substantial resources to developing and modifying computer models and, in addition, in responding to the theoretical dimension of this work by focussing on the problem of building and validating models capable of representing technological developments, positive feedback, non-linearity and societal innovation.
All these concepts can be found in the wider literature of policy-relevant science, though their meanings are not easy to express abstractly and scientists often fall back on analogies. An insect like a cricket, for example, passes through successive moults each of which creates a rapid burst of growth while its morphology (form) and physiology (dynamic processes) are conserved and enhanced. A butterfly, in contrast, ends its larval phase with a critical period of morphological and physiological turbulence and emerges as a qualitatively different type of system. The physiology of the butterfly, then, can be used as an example of non-linear dynamics, while the cricket’s development is more linear.
Both types of dynamic are found in human activity systems, but with one critical difference. Insect life-cycles are equifinal - we can predict the end-state, ex ante. The ‘morphology’ and ‘physiology’ of human activity systems, however, is determined by habits, beliefs, patterns of social exclusion, institutional constraints and geo-political power-struggles. As scientists create new knowledge and engage with broad stakeholder communities, the knowledge they create and disseminate can lead to innovations that change the system’s end-state in ways we cannot predict because it is contingent on knowledge we have not yet acquired and actions we have not yet considered. COMPLEX has devoted substantial resources to understanding, facilitating and managing innovative dynamics.
Scientists are not the only group capable of triggering innovations that, in periods of systemic turbulence, can drive rapid systemic re-organisation; all human activity systems can do this, but the potential for innovation and the existence of geo-political turbulence create a range of ethical, theoretical and methodological challenges. Whereas conventional research teams try to maintain the distinction of science from politics, COMPLEX, with the generous backing of DG Research, has been pioneering a high-risk, high-gain strategy that has required us to selectively weaken this boundary and develop scientific tools that provide leverage in periods of systemic turbulence.
Our project’s name, COMPLEX, is not an acronym, but an adjective. The climate / energy / economy nexus is a complex system.
COMPLEX has carried out five, regionally focussed case-studies under 3 workpackages.
WP2: In Norway and Italy, we have had teams working on ‘Climate Related Energies’ - energy sources whose availability and intermittency patterns can be modified by climatic factors. We have developed new models for handling these intermittency problems, explored patterns of resource availability through contemporary space and across time and developed protocols for solving those problems in out case-study regions.
WP3: In Spain and the Netherlands, we have been working with local and municipal stakeholders to characterise the nexus of opportunities and obstacles to renewable energy adoption. Here, as in the other regional case-studies, we have developed new modelling and stakeholder engagement tools.

COMPLEX has developed a repository of models and modules that can be integrated in real-time across the internet and explored a range of qualitative approaches for work with powerful institutions, small local interest groups and multi-national companies.
WP4: We have undertaken extensive work with external stakeholder communities in central Sweden, working with pre-existing groups at all levels in the administrative hierarchy from the regional to supra-national. We have developed a range of modelling tools and elicitation methods for exploring land-use and carbon emissions.
We have also carried out cross-cutting and integrative initiatives, gathering large datasets, explored a wide range of policy options, participated actively in the policy development process and undertaken a wide range of dissemination actions and stakeholder elicitation studies. In this way, COMPLEX has been able to work with and for external stakeholders at every phase in the project cycle from problem formulation to solution.
As a result of this work, COMPLEX has produced nine policy briefs, each targeted on a different stakeholder community and developed with the active co-operation of those stakeholder communities and two edited volumes of applied modelling case-studies available from our web-page.
COMPLEX also developed new methods and protocols for integrating models. These include methods for using the outputs of one model as inputs for another, a system based on wrapper technology that allows users to select alternative sub-models and integrate them in their browsers, a range of attractive modelling tools that can be parameterised and used by external stakeholders. We have developed new, open-access modelling tools that can be used for participatory modelling and developed mathematical protocols for integrating the outputs of different models of the same system, each based on different structural assumptions.
We have also explored the theoretical linkages between recent developments in evolutionary biology (notably the ‘extended synthesis’, anthropology and ecology and been researching the applied anthropology of science-based innovation. Using these insights we have developed managerial tools for facilitating science-based innovation in complex problem-domains.

Project Context and Objectives:
The call to which our team responded explained that currently existing model tools had manifest deficiencies, including the difficulty of representing “pervasive technological developments, positive feedbacks, the difficulty to represent non-linearities, thresholds and irreversibility”.
It was written while the crash of 2008 was fresh in everyone's mind and it was inevitable that we would take this catastrophe as the 'type specimen' of an irreversible, non-linear system-flip, and consider the vexed question of why so many economic models failed to predict it. There were two obvious explanations, namely deficiencies in the modeller's toolkit - perhaps we needed new modelling infrastructure and databases - and problems at the science / society interface - perhaps we needed new ways to think about institutional receptivity, stakeholder engagement and social cohesion. These two explanations were not mutually exclusive, of course, and the call strongly suggested that both lines of research be pursued in parallel. We were to develop new modelling tools, methods and databases, and to engage closely with relevant stakeholders.
The word ‘stakeholder’ is often used in an unfocussed and uncritical way and arguments about who, or what is a legitimate stakeholder are deeply political. Every politically contentious issue in natural and cultural resource management can be re-framed in terms of arguments about stakeholder legitimacy. Many of those research contexts can also be described in terms of ‘unacknowledged stakeholders’ - stakeholders whose legitimacy all parties to the argument would dispute. It is an old joke, in policy-relevant science that, when the argument is about how best to drain the swamp, nobody wants to consult the frogs and fishes. To do so would be to go against ‘common sense’ and would obviously be ‘unscientific’.
COMPLEX made a conscious effort to engage with unacknowledged stakeholders, but there were levels of conflict beyond which integration may be impossible. Political arguments often reduce to a dichotomy between political insiders and outsiders. Both factions have invested heavily in that dichotomy and are often reluctant to allow unacknowledged stakeholders to complicate the issue. They may even join forces to drive the unacknowledged stakeholders out. After a prolonged period of cultural and social exclusion, unacknowledged stakeholders may become so disaffected that they would rather destroy institutional structures than become part of what they see as a corrupt system.
COMPLEX also made an explicit decision to avoid ill-structured or ‘wicked’ problem-domains, where policy-relevant science provides almost no leverage, and to focus on moderate levels of complexity and relatively peaceful research domains. We have used the word ‘stakeholder’ to describe a system whose dynamics the project has influenced. A ‘system’ in this context, could be an institution, a community, an ecosystem, an organism (human or not), or a population of organisms. Every system has a living component and is engaged in some sort of dynamic process.
Systems whose processes the project has not tried to influence or which the project has tried to influence, but failed would not be stakeholders by definition. The climate system, the biosphere and the lithosphere, by this definition, are not stakeholders in the COMPLEX project. An interesting consequence of this definition is that we, the scientists working on the project and our sponsor, DG Research are obviously stakeholders and so too are the communities of farmers, the neighbourhoods and the ecosystems our project has influenced together with our scientific peers, our employers and the journals we have published in.
The corollary of this definition is that COMPLEX, like many other projects, had a reflexive structure; the relationship between the research project and the external stakeholders it works with reflects the relationship between the project and the scientists, partner institutions and sponsors who comprise its internal stakeholder community. The conventional scientific distinction between the objects of study and the subjective scientists studying those objects cannot be maintained. Many of the complexities we have grappled with arise from this reflexive property, as conflicts of interest among internal and external stakeholder communities must be resolved.
We understood that the transition to a low carbon economy by 2050 would involve irreversible step-changes in the cultural, economic and natural domains, with qualitatively different socio-economic configurations before and after. COMPLEX developed new modelling tools for managing step-change dynamics by working across a wide range of spatio-temporal scales, and integrating the knowledge of many stakeholder communities, for example in respect of land-use change driven by carbon-related technologies. By the time the low carbon economy had emerged, many vested interests and culture-clashes would have been resolved and socio-natural systems would have changed irreversibly.
It was imperative that these transformations be managed in a way that maintained social cohesion, prosperity and good governance. Our task as scientists was to help policy makers facilitate qualitative change without compromising cultural and natural life-support systems. The COMPLEX consortium was fully committed to that work, which it supported with a range of research activities that explored the relationship between culture, models, human behaviour, space-time scales and the economics of carbon emission in a complex world. This research was intended to contribute to the management of complex system flips that we expected to occur as we moved to a low carbon economy.
A key difference between classic and complex systems is that, in the former, causal relationships are space-time invariant, while in the latter, the relationship between cause and effect is linked to space-time perspective. Each perspective brings some phenomena into the foreground and backgrounds others. The relationship between cause and effect is often clear within these quasi- classic domains, but may be reversed or confounded as space-time perspectives change. This complex causality is not a philosophical construct. Consider, for example, a limestone catchment. Viewed from the deep time perspective of geology, it is clear that water movement has cut through rocks to create limestone gorges, fissures and underground rivers. Surface water redistributed sediments to create new landscape features. Water movement causes landscape structures. When the same landscape is viewed on the meso-scale of hydrology, however, it is equally clear that persistent landscape structures - riverbanks, underground cave systems and flood plains, say - constrain the passage of water through the catchment. The shift of perspective seems to transform cause into effect and effect into cause. Coming down a step further to the micro-scale of human agency, we see that the system can flip from one set of causal structures to another in response to environmental perturbations and human action. Indeed, humans often manipulate causal structures by modifying a system’s space-time signature, for example by building dams and harnessing hydro-power, by clearing forest or abstracting water from the aquifer.
Complex causality, with its scale-dependent causal inversion, irreversible change and stick-slip dynamics, can be valorised wherever cultural and natural systems interact and our Description of Work (DoW) required us to consider it carefully. We hoped to integrate established economic models with real-world complexity. In practice this meant that the problems of upscaling and downscaling be addressed in an intelligent way. Typical downscaling questions include: How are we to downscale national and supra-national policies to regional and local levels without collateral damage to cultural and natural life-support systems? How are we to disaggregate data on precipitation, say, or economic indicators in a way that reflects real-world complexity? Upscaling questions are similarly complex: How could we represent the interests and needs of local stakeholders without frustrating much-needed policy interventions at national and supra-national scale?
COMPLEX was funded to develop a suite of modelling tools and decision-support systems to inform national and supra-national policy and support communities across Europe working to make the transition to a low-carbon economy. We would approach this from two perspectives. On the one hand, WPs 2, 3 and 4 all dealt with real-world complexity, with stakeholder engagement and upscaling and downscaling problems at regional level. These differences in perspective became manifest as links with different stakeholder communities. These WPs were to fine-tune problem specifications in a way that facilitated system flips and innovation cascades consistent with the shift to a low carbon economy. On the basis of case studies in Norway, Sweden, Netherlands, Spain and Italy and a process of broad stakeholders' engagement, COMPLEX would investigate:
1) better understanding of the space-time dependence between different energy sources driven by climatic variables and to assess positive/negative effects at the regional scale (WP2). A number of external institutional stakeholders, such as utilities and regional environmental protection agencies, have shown interest in the project and would be closely involved in the work related to the two case studies in Norway and Italy that would be undertaken under this work package.
2) Acceptance and planning of climate mitigation options at the landscape level, through improved modelling of spatial and temporal impacts of climate mitigation policies and selected low-carbon technologies and the development a testing of a land-use decision support model. WP3 would explore acceptance, implementation and realisation of climate mitigation (CM) policy options at the scale of the landscape from the perspective of those stakeholders who are most important for their successful implementation - key actors and communities, for example, and knowledge gatekeepers. To do this, we integrated non-mathematical modelling approaches developed for policy research with quantitative decision support tools. Green technologies, such as wind, solar, biofuel production and other forms of biomass use, together with energy efficiency measures such as innovative water treatment or better electricity demand management are needed to move towards a low carbon economy. Large scale implementation of these technologies affects land use dynamics in a complex, unpredictable way and we were interested in unintended consequences (e.g. replacement of food crops for biofuel). This WP, like WPs 2 and 4, accepted that the impacts of policies intended to reduce carbon emissions would have characteristic space-time impacts. These impacts were expected to influence economy, patterns of land-use, social cohesion and compliance. .
3) Land use planning, uncertainty and risk management at a regional (sub-national) level. WP 4 provided process understanding and instruments for support of the transition to a low carbon society by 2050 - with application to the Stockholm-Mälar region of Sweden. This region consists of two NUTS 2 regions and is one of the highly dynamic regions in northern Europe. The aim was to analyze strategic societal choices and their consequences. This included the design of a toolkit for analysis 1) of emerging and optimally selected land use patterns, 2) of economic development and the impact of policy instruments, and 3) of the processes connecting the scientific support to the decision making functions at various levels, including policy processes at shorter and longer time scales. Focus was on finding integrative forms of support to guide the path to a low carbon society under varying climate scenarios and world situations.
Workpackages 5 and 6 dealt with economics and model integration respectively. As the regional case-studies tried to narrow their focus, these two were intended to broaden their perceptions and develop new modelling tools. Our expectation was that this integrative approach would lead to a meeting of minds and methods as the project develops.
Models operate at different geographical scales and spatio-temporal resolutions. Forming vertical linkages between these scales was a significant challenge. To address these problems, we propose a process of engagement between modellers and other stakeholders, so that models are not linked and plugged together mechanically but assessed and interpreted through stakeholder participatory mechanisms to make sure that the temporal, spatial and structural coherence is maintained and that the results do serve the particular needs of stakeholders involved. To expedite this, each regional case-study was to undertake an initial stakeholder engagement exercise and produce a case-study scoping statement. These scoping statements were to feed into a process of ‘model-stakeholder fusion’ that would lead to targeted policy briefs.

Project Results:
COMPLEX has produced a substantial body of modelling infrastructure in the form of models, integration protocols and databases. It has also been active in stakeholder engagement, working with external stakeholders in all stages of the cycle from problem formulation to policy development.
This document contains a strategic summary of the project's work.
For a complete archive of all the project's activities, the reader is referred to Deliverable 7.9 and the four-volumes of edited scientific report that can be downloaded from our website. D 7.9 also contains a complete set of policy briefs (nine in all) generated by our work with external stakeholders.
Here we focus on our principal S&T results.
1.1 Climate Related Energies (WP2)
Facilitating the decarbonisation of the European electricity system is a complex process. On the one hand, this process is embedded in multi-level legislation, market rules, economic and technological constraints, private initiatives and public interests. On the other hand, it must also consider the multiscale climate and environmental variability governing CRE resources (Engeland et al. 2017 for a review). CRE integration in the energy mix potentially conflicts with regard to land and water use as well as environmental regulation (François et al. 2014a,b). From a European policy perspective, how and where to produce, transport, store, complement, sell and buy these renewable energies are critical questions. Answers must be guided by a robust knowledge of where and how climate-related energy is available in regard of energy demand. WP2, mainly focused on this point.
Projections of past and future CRE were assessed using either Global Climate Model (GCM) outputs or on locally more refined values from downscaling techniques that link GCM variables (called predictors) and local climate variables that serve as inputs for CRE modelling. The confidence in a CRE projection is related to the confidence in the values of the corresponding GCM outputs or predictors. The performance of various CMIP5 models (Coupled Model Intercomparison Project) in simulating variables of CRE interest – the statistical properties of daily mean sea level pressure fields over the 1950-2005 period in a first step – has been assessed by attributing weights to each GCM through a Bayesian framework developed within WP2 (Renard et al. submitted). Taking the 20th Century Reanalysis (20CR) data as a reference, results show that, for any chosen observation uncertainty level, the weights of the various GCMs vary markedly but smoothly across 12 European regions.
We built CRE-Prospector, a support tool for multi-CRE resource analysis. It analyses the balance between energy demand and climatic energy resource by exploration of present and future climate databases. Over local, regional, national or continental areas CRE-Prospector simplifies the elementary functions of an electricity system feeding a significant part of the demand with renewables. The generation converts meteorological and hydrological input variables like wind speed and river discharge into energy by integration over the area at a chosen time step. Inputs are either observations, reanalyses or GCM downscaled data. WP2 developed SCAMP, a statistical tool preserving the correlation among the downscaled input variables (Raynaud et al. 2016). Atmospheric temperature and differences between weekdays and holidays modulate the electricity demand in CRE-Prospector (see e.g. François et al. 2016a,b). Transmission is assumed perfect over the area and a bulk storage is sized as a fraction of the mean demand of the area. The simplification of the electricity system helps with prospecting possible steps of a transition to a renewables-only model. Its main merits are the variety and the transparency of the results. It allows, for instance, the rapid exploration of extreme but plausible equipment scenarios (production, transmission and storage) that can be far from the current system state (François et al.. 2015). We used CRE-Prospector for analysing the natural satisfaction of the demand (penetration) and the volatility of the energy balance. The penetration is a key investment performance factor while the volatility indicates either the need of energy backup and transport or the need of demand flexibility. We used CRE-Prospector over 12 regions across Europe for mixing run-of-the river hydropower to solar and wind power (François et al. 2016-a) and for exploring energy penetration sensitivity at low climate frequency along the energy transition process (François 2016).
1.1.1 Matching production and consumption – toward “Prosumption”?
The smoothing role of electricity transport extending to production variability changes the conventional base-load logic. The present clear distinction between consumers and producers is blurred by the emergence of “energy prosumers” at various levels ranging from households equipped with individual PV panels to larger communities or industries sharing more important CRE production means like wind mills or micro-hydro plants. At these various levels each prosumer will try to manage its own local production-demand balance and will connect to grids with the hope of finding backup supply and export opportunity. For instance, PV prosumers start to shift their load into sunshine hours, dealing with fluctuations locally. The myriad of possible prosumer groups will give many possible stable structures of transmission grids and storage facilities.
The multiscale-multisource availability of climate related energy hence reshapes the socio-economy of production, consumption and “prosumption”.
The transition to a renewables-only model will jump from one structure to the next one depending on production and consumption initiatives at multiple levels that all mobilize climate characteristics considerations and the subsidiarity principle– what is the most appropriate level to balance production and consumption - hence the multiscale investigations on the potential CRE-demand balance reported in the following sections that should help anticipating possible transition jumps. Integrating renewables adds uncontrolled variability in the management of electricity systems.
1.1.2 European regions are not equal in terms of CRE potential
The integration of renewables is not equally easy across European regions if we consider for instance that the percentage of consumption naturally satisfied by the renewable production may vary from 60% to almost 100% (François et al. 2016a).
Our exploration with CRE-Prospector concerns 12 regions of ca. 40000 km2 in Europe and North-Africa. These regions show marked differences in the seasonality of the different sources and in their cross-behaviours – solar and hydro are for instance in phase in Italy and in opposition of phase in England. Wind has the less marked seasonality even with a Europe-wide minimum in summer (François et al. 2016a,b). The results of CRE-Prospector bring a new vision of multi-CRE potential in regard to the literature reviewed by WP2 (Engeland et al. 2017). European regions have unequal ease of CRE integrability with best penetration rates increasing from ca. 80 to 95% from North to South and marked differences in the best way to mix the three technologies. Hydro favours CRE penetration in a significant proportion of regions all across Europe. Transition is not a smooth process with performances substantially changing with mix scenarios beyond a rate of CRE generation of 50% of the demand. Transition conditions are different from region to region. Our results on the system balance can be symmetrically interpreted in terms of demand flexibility.
Scenarios with close to 100% renewable energy in Europe are now typically dated for 2050 (European Climate Foundation). Due to the intermittent nature of wind, river and sun power at all space and time scales, they involve challenging issues (François et al. 2014; Engeland et al. 2017). At 2050 horizon, “extreme but realistic” scenarios of grid development oscillate from local generation to global connection through European Electricity Highways. Deciding at what level to generate, store and backup energy is a subsidiarity problem that must be calls for new climate engineering tools able to explore intelligibly the mass of climate data.
Other than nuclear energy, decarbonisation of the European electricity systems means moving from geological resources to climate resources. Climate variability has different temporalities and spatialities to geological variability. Its range of time and space scales overlaps human activity scales making the access to climate related energy intermittent and ubiquitous while geological energy is, to the contrary, scarce in space and constant in time at human terms. This deep difference is at the root of many concerns about electricity systems expressed by the stakeholders we met.
1.1.3 The present state and the potential development of hydropower
Rivers integrate weather variability through their basin and branching structure. Natural river flows are less variable and more predictable than their meteorological drivers – precipitation and temperature essentially (Francois et al., 2017b). With its ability to produce and store energy (François et al. 2015), hydropower holds a specific status in the context of energy transition and renewable energy targets. Our view is that hydro is to be considered as a CRE like wind or solar (François et al. 2014b). Although models simulate rather well hydropower generation from a given catchment, its interaction and complementarity with others CRE are still poorly represented. Modelling results show performance degrees ranging on time scale and hydrological regime (François et al. 2017c).
In addition to the works covering the 12 Europeans regions, we illustrated the role of hydropower through two regional case studies – Northern Italy with an active development of micro-hydro projects (François et al. 2016b, François et al.2017b,c) and Mid-Norway, a region of energy deficit in the hydro dominated Nordic Pool (François et al.2017a). Considering the natural run-of-the-river energy, i.e. ignoring storage effects, we showed that adding hydropower almost doubled the amount of demand satisfied compared to only solar in Northern-Italy (43% to 72%) (François et al. 2016b). Conversely adding wind power generation to the hydro-electricity system of Mid-Norway proved to be almost 100% efficient and to solve the winter deficit problem (François et al. 2017). Under present climate conditions, the Mid-Norway energy deficit is about -70 GWh showing a reduction of 65 GWh resulting from the projected equipment (i.e. wind power and transmission projects).
We also considered climate change effects in these regions. In the case of Mid-Norway, the system sensitivity to climate variability (natural and forced) proved to be lower than its sensitivity to technical developments like wind power reinforcement (François et al. 2017a). In Northern-Italy we showed that the risk of investment should the climate change largely depends on the CRE mix. The penetration of hydropower may increase in mountains or decrease in plains, depending on the hydrological regime (François et al. 2017b).
In summary, hydropower opens opportunities and challenges. To the obvious advantages seen above in terms of complementing the variability of the other CRE, water management has to consider other factors than just energy production issues such as water consumption and environmental constraints (Francois et al., 2014a,b). We would also draw attention to the high sensitivity of water resources to climates in mountainous areas (40% of Europe).
1.2 Stakeholder Engagement in Norway, France and Italy (WP2)
In addition to its technical work on CRE production the WP2 team worked with external stakeholders in Norway, France and Italy. This work was qualitatively unlike that in Spain, the Netherlands and Sweden. Where the Swedish case study looked at stakeholder diversity from regional to national and supra-national scales and the Spanish and Dutch work looked at regional and local patterns, the Norwegian and Italian studies engaged with powerful commercial stakeholders, particularly energy generators, and focussed on problems arising through intermittency of supply, climate variation on a range of spatial scales and temporal scales from hours to decades.
Our work on CREs (Climate-Related Energy) was oriented around renewable energy, climate variability and regional actions with case studies in Italy and Norway linked to French energy companies. The main characteristics of the two regions with respect to the aims of the project were as follows:
Norway: Local value creation and environmental impacts play a crucial role in the development of renewable energy sources in Mid-Norway. Public acceptance depends on the motivation for increased power production in the region. Willingness to agree to/support projects is higher when regional industry and trade development is involved and lower when generated power will be exported from the region. Conclusions from study analyses will differ strongly, dependent on at which level (local, regional, national, European) the considerations are made. Stakeholders expect the results from our analyses to be useful, e.g. as one source of information the counties consider when making their regional climate and energy plans.
Italy: Hydropower and photovoltaics (PV) represent the main renewable energies in the study area (Southern Tyrol). The local energy policy package termed KLIMALAND plays a crucial role in setting the development of renewable energy sources in the region. PV energy source mainly depends on small-medium plants installed on roofs and/or facades. 44% (28%) of the nominal PV power is due to plants in the range of 20 to 200 kW (200-1000 kW, respectively). Public acceptance for PV plants is generally high. The main problems affecting the further development of PV are related to the connection of the many small and medium scale PV energy producers to the power grid in the region. A high density of power plants in a low-voltage section of the power grid may result in power generation exceeding consumption in this section of the grid on sunny days, leading to high loads on the distribution grid with opposite power flow than what it was designed for. The main problems affecting the further development of hydropower depend on conflicting demands with other water usages and environmental constraints. Depending on the technical and financial structure of the electric market in Italy, the typical scale at which integration between different energy sources can be achieved is represented by the market zone (for the study region, this is represented by the North Italy Zone). Workshops were organised independently, but in the same period and with similar structure, in Norway and in Italy.
Project partners made considerable efforts to engage the stakeholder community. Some private sector stakeholders (SINTEF and EDF) were fully embedded in the project as consortium partners. In addition, close relationships were developed with stakeholders like CNR, Sun’R smart energy, Statkraft and Eurac either through continuous collaboration or involvement in workshops. This gave the CRE team a distinct advantage when it came to understanding the perspectives of private sector stakeholders.
However there were problems arising as a consequence of institutional turbulence and changing market conditions. Our institutional stakeholders hade been engaged in the project from the start of the design phase. We knew we were solving an important problem, because key external stakeholders had worked with us to specify it. However the sad truth is that new policies and new market circumstances meant that those stakeholders became appreciably less interested in the problem as time passed.
The Operations Researcher Russell Ackoff produced a paper in 1979 titled: The future of operational research is past in which he declared that ‘managers do not solve problems, they manage messes’. We have been working with institutional stakeholders in a particularly messy period. We solved the problem our stakeholders asked us to solve, but they were no longer interested in the solution when it came. This disappointing result has informed our research on non-linearities and qualitative, irreversible change, which we discuss in greater detail in Deliverable 7.9 (Final Scientific Report).

1.3 Stakeholder's Engagement in Spain and the Netherlands (WP3)
WP3 explored acceptance and implementation of climate mitigation policy options at the scale of the landscape. In Spain and the Netherlands, we explored the implementation of Renewable Energy (RE), a key pillar of EU climate policy, from the perspective of those stakeholders who are most important for successful implementation; policy makers, knowledge gatekeepers and communities. In Spain, a top-down filtering approach was developed, in which rapid appraisal of all Spanish regions was used to select 6 regions for detailed case study through telephone interviews with key actors. From these 6 regions, a single region, Navarre, was investigated in detail through participatory action research approaches (PAR), that deployed a spatial model of future renewable energy implementation (APoLUS). In the Netherlands, a bottom-up approach was employed in which key informants at municipal level were engaged in understanding the clean energy transition at local scales. Engagement began with a small and progressive group of stakeholders in the municipality of Dalfsen, in the province of Overijssel. Following this the project engaged a larger and less progressive municipality where new targets for renewable energy had been established. In the municipality of Enschede researchers partnered with municipal staff to both increase awareness of the local goals (together with the various opportunities and challenges involved) and determine stakeholder motivations, cognitions and resources that would enable plans to be made for renewable energy development that had support within the community. Key to this work was the deployment of an interactive Participatory GIS Tool (COLLAGE),that allows stakeholders to negotiate the location, amount and type of installation to be installed within their municipality. Under COMPLEX, COLLAGE was used to facilitate these kinds of negotiations around Renewable Energy Related Landscape Features (RELFs), but could potentially be adapted for other types of developments, e.g. urban or commercial developments, intensive agriculture, infrastructures etc. Researchers from the University of Twente are currently using the COLLAGE tool as part of a Province-wide program to support other municipalities in developing local renewable energy plans and implementing them (Flacke and de Boer 2016).
We developed and applied a wide range of structured approaches and tools aimed at knowledge co-generation and social learning around RE implementation. Though conventional participatory approaches like workshops, interviews and surveys were all employed, the core of the research activities were designed to support participatory modelling activities aimed at understanding future renewable energy implementation, in which stakeholders were co-developers.
In Spain, an integrated model known as the Actor, Policy and Land Use Simulation model (APoLUS) was developed to simulate future land use change for the Navarre region under a range of renewable energy implementation scenarios. APoLUS links a spatially explicit geographical model with policy implementation theory and sociological approaches aimed at widening participation in environmental decision-making (Hewitt et al 2015; Hewitt in press, Kovalevsky et al in press).
APOLUS and COLLAGE are complementary to one another. Information collected related to stakeholder preferences with COLLAGE at the local level can be fed into the APOLUS model to increase the correctness of its output. The results of APOLUS can be fed back into communities to help them determine the type of policy scenarios they would support with respect to the deployment of renewable energy.
1.3.1 The Engagement Process
We made contact with local and provincial government officials, citizens, ENGOs (environmental non-governmental organizations), small businesses, farmers and local media. In the Netherlands the main contact was made through municipal staff involved in the development of energy and carbon plans. This was because there was an interest from both sides in working directly together as partners. Researchers then had direct access to the stakeholders involved in the renewable energy planning and implementation. In Spain external contact was mostly made with regional planners involved in strategic planning focusing on climate change and new landscape strategies. These stakeholders facilitated the contact with other key stakeholders at both the municipal level and the business level involved in renewable energy implementation.
1.3.2 Principal conclusions
In Spain:
Despite Spain’s position as a global leader in RE development in 2010, no further RE development has taken place in the country since 2011, when the government imposed a moratorium by slashing subsidies, implementing punitive connection charges for householders wishing to connect to the grid and outlawing battery storage for small consumers. We concluded that it seemed unlikely that Spain would meet its RE targets for 2020 (Alonso et al 2016). However, recent figures suggest that the decline in energy consumption may mean that the target is still achievable, nonetheless the drastic withdrawal from RE in Spain is disappointing, and a true clean energy transition is no longer a realistic prospect. .
The explanation seems to be that a combination of a liberalized energy market combined with a very substantial problem of over-capacity has caused lawmakers to take fright and establish a series of counter measures to put a brake on the energy transition.
A true green energy transition in Spain requires a decentralized model of energy production and distribution. Yet privatization of energy production and distribution has left policy makers in a weak position to implement the required changes.
In the past, governments have favoured large scale solar and wind developments, and this tendency, before paralysis took hold, seemed to be increasing. Yet true bottom-up transition is very difficult if smaller energy businesses cannot enter the market.
Large-scale wind farms, in particular, are viewed by many stakeholders in an unfavourable light because of their high environmental impacts and the feeling that the profits from these developments are not equitably shared.
Prior to the RE moratorium, the most successful Spanish regions in RE development were those that had the strongest relationships between key sectors like business, public administration, civil society and science and education. Underperforming regions may have the potential to improve in future by developing these links.
Scenario modelling work suggests that much higher RE capacity is easily attainable without major landscape impacts or social conflict. Unfortunately, until powerful actors at the top can be persuaded to remove the obstructive legislation, the lockdown will likely continue. The fundamental take home message is that the clean energy transition cannot become a reality as long as powerful actors are aligned against it. Our research suggests that implementing actor motivations and relationships should be a key consideration in the design of future climate policies.
In the Netherlands:
Increasing the share of Renewable Energy in the Netherlands is a generally accepted necessity. However when it comes to deciding trade-offs between the value of land uses and extent of substitution we run into problems. Implementation needs to be brought about at the level of communities and the people and has an inherent complexity. It is not a choice between one thing and the other in isolation. In order to reduce the feeling of alienation we can bring the big picture to the stakeholders, make the pros and cons visible, and show how various scenarios are possible to achieve targets.
Transitioning to a low carbon economy is a complex and urgent task; the Netherlands is one of the highest per capita carbon emitters in Western Europe. In the Netherlands, where land is intensively managed and highly valued, the provinces and localities are currently facing challenges in implementing the necessary landscape features. Solar farms and urban solar panels, as well as wind turbines demand a certain amount of space and are thus competing with other important current uses and priorities. Little research has been done in understanding the complex interactions between newly installed renewable energy technologies and the previous land uses and so local actors are experimenting and uncovering the impacts themselves (de Boer et al 2015).
This research set out to understand these dynamics, from the perspectives of local, regional and provincial stakeholders. Our results show that the specific trade-offs related to the local dynamics (influenced by municipal, provincial and national policies) are key to overcoming the various obstacles in the transition. Aligning RE development plans to the local context is thus key to speeding up and increasing the efficiency with which RE will take its place in the urban and rural landscape.
This does not mean taking a hands-off approach (decentralisation without support), but actively supporting the different needs of the communities with flexible yet intense policies and programs.
Building on the current national, provincial and local programs to support renewable energy development, the process can be improved by making the location-specific spatial trade-offs explicit and highlighting the intersection between different goals at various levels and from different stakeholders. There is no one size fits all strategy as context differs by region in terms of available land, energy demand, social acceptance and financial capacity. Local goals to achieve CO2 or energy neutrality proved to be a good starting point for frank discussion about the future of land use in each of the various localities in the Netherlands. Having solved the easy problems, strategic and participatory multi-functionality and land use planning can be used to help achieve the lower carbon goals of the Netherlands and increase the energy security and sustainability of local communities.
1.4 The Swedish Case Study (WP4)
COMPLEX also had an active research team in the Stockholm-Mälar region in East-central Sweden (WP4). This region includes a large city, the capital Stockholm, and the surrounding area around the lake Mälaren. It also has six counties, several smaller towns, municipalities, farmland, and forests.
Politically, the governance structure is multi-layered, from the national level (providing the context to the region), the county level, the municipality level, and down to households and individuals. A particular challenge for this region is the heterogeneity among counties with respect to economic prosperity and environmental performance. This may be perceived as an argument for delegation of decision rights on policy choice and implementation from central to local jurisdiction. One important justification is the gains obtained from local knowledge on economic and environmental performances and formation of local communities pursuing sustainable use of resources. The main challenge in reaching a carbon-neutral Stockholm-Mälar region is then to identify, quantify, and balance the advantages and disadvantages of different policy instruments and jurisdictional delegation levels. A specific consideration is the current lack of a strong jurisdiction in between the national state and local municipalities.
In order to address the various problems and issues raised in the Swedish region, we designed a study process which included understanding different stakeholder values and operational capacities, as the use of analytical modelling tools. The tools we developed and used involved partial understandings of various issues, such as economy, land use, environmental considerations, energy-related technological systems, as well as issues about societal transformation at large.
Our work was organized along two lines: 1) stakeholder interactions through workshops, seminars and embedded researchers and 2) computational models for decision support related to a) economy, b) energy, and c) land use, as well as decision making based on d) a societal gaming model for national/regional decisions, and e) a neuro-cognitive model for individual consumer behaviour. The stakeholder interaction work has been described in our final report and the closing section of this document. Here we summarize the modelling efforts:
Our numerical model for cost-efficient land use dynamics under uncertainty showed that a cost effective solution can be reached and the total abatement costs would then correspond to 1 % of cumulative gross regional product in the region when both technological development and uncertainty are acting. Without technological development the cost would be doubled. All classes of abatement measures are needed, but bioenergy, biodiesel, and electric cars are of significant importance. However, the model suggests that the main financial burdens will be born by 1/5 of the municipalities in the most cost effective solution because of asymmetric allocation of resources in the “business as usual” scenario. Another finding is that only a few counties and municipalities make gains in the overall cost effective solution compared with decision making in isolation. A majority faces lower cost when they implement abatement measures within their own jurisdiction.
The Uppsala energy system and GHG-emissions were modelled using the Long Range Energy Alternative Planning system (LEAP). LEAP is an integrated modelling tool for long term forecasting using an annual time step (www.energycommunity.org). The concept is an end-use driven scenario analysis with a “business as usual” scenario and one or more alternative scenarios. It simulates “what if” energy futures along with environmental emissions under a range of user-defined assumptions. On the demand side of the framework, LEAP supports bottom-up accounting and provides a wide range of accounting methods for modelling energy generation, distribution and capacity expansion planning on the supply side. All sectors within an economy or energy system can be included in the model as well as external pollutants. Modelling is based on a comprehensive accounting of how energy is consumed, converted and produced under assumptions given regarding energy demand, population, technology etc.
Locally produced bioenergy can decrease the dependency on imported fossil fuels in the region, while also being valuable for climate change mitigation. Short-rotation coppice willow is a potentially high-yielding energy crop that can be grown to supply a local energy facility while giving climate benefits by sequestering carbon from the atmosphere to the soil. Our study assessed the energy performance and climate impacts when establishing willow on current fallow land in a part of the Stockholm-Mälar region for the purpose of supplying a bio-based combined heat and power plant. Time-dependent life cycle assessment (LCA) was combined with GIS mapping to include spatial variation in terms of transport distance, initial soil organic carbon content, soil texture and yield. The results showed that when current fallow land in our region was used for willow energy, an average energy ratio of 34.6 MJ was obtained and carbon was also sequestered in the soil (compared with the reference land use). Although the climate change mitigation potential over the landscape was improved by selecting the best performing fields, the results showed that to maximise the climate change mitigation, all fields needed to be utilised to produce as much willow energy as possible, since all fields showed climate benefits compared with the reference land of green fallow. This indicates the importance of reference land use in assessment of bioenergy systems.
In the gaming exercise we conducted at a stakeholder workshop, the task was to focus on a general level process to explore how decision makers in a practical case could make use of a very large decision support model with regard to path decisions of an overriding political nature in order to move towards a low carbon society. The target of this study was the interaction between the decision makers and the model. For this purpose, we adapted an off-the-shelf computer game (DEMOCRACY 3) and “trimmed” it to serve our specific purpose (i.e. Swedish decision making) with regard to tasks related to low carbon societal transitions - especially oriented at centrally positioned political actors, or actors with tasks across sectors. Prior to the workshop, we arranged a sequence of small theme-oriented seminars for experts in various fields in order to identify the required changes in the large DEMOCRACY 3 software package.
Our neuro-cognitive model for individual decision making involved an integration of two levels of modelling, an artificial neural network (ANN) model of various brain structures involved in decision making, and a multi-agent system (MAS), where individual agents interact “intelligently”. We believe this is the first time a computational model of this kind has been developed and applied to a climate/environmental problem. The model addressed consumer behaviour, attitude and trust, applied primarily to travel and eating habits. While our approach was quite unique when we started, this type of approach has recently received a growing interest around the world. A recent example relating to our model results is the tendency of commuters in the Stockholm-Mälar region to change to public transport, as the toll fees for Stockholm increased. An opposite effect was observed when new roads/tunnels were built, facilitating car driving through Stockholm.
Our combined modelling and stakeholder interaction efforts provided input to policy makers in the Stockholm-Mälar region, but we also regard the results as providing an early prototype for European societies at large.
1.4.1 Reflections and lessons learned from the Swedish work
When we started the project the Paris 2015 meeting on the climate change challenge was several years into the future. The same was true for the UN meeting in 2015 on Sustainable Development Goals (SDGs) with topics very close to the ones we had to face. From these and other sources, we have learnt that our choice of investigation level, i.e. the sub-national region, was a very fruitful one – and also a level that did draw constantly stronger interest as time went by. However the widening of regions to match the new challenges turned out – in our Swedish case – to be very contested. Some wanted these larger regional frames; some were very reluctant. There was not always a creative interplay between governance issues – including social goal setting - and the mobilisation of technical-economic change efforts. In fact, some earlier experiments had failed, sometimes due to weak persistence. The politics of societal transition seems yet to be in its infancy. There is a call now for a strong degree of societal bravery, of vision and increased public understanding about what is at stake and what the balancing acts are. New ways of designing research and innovation efforts around these issues are needed.
We were expecting a broad range of stakeholder positions on the forthcoming societal transformation process. However, we soon understood that the spectrum of stakeholders was much larger and more complex than initially believed. In addition, the differences in different parts of our region were also larger than we had imagined. The multi-level organisation and sectorial fragmentation in the public sector and the diversity of goals and scales within civil society required a deeper and more holistic approach. The interplay between the two tracks provided the possibilities to shape such an understanding. The work in Sweden has thus improved the knowledge of the systemic nature of the challenges and laid the ground for a design of policy actions and tailored modelling activities.
1.5 Model development, Model Infrastructure & Model Integration
WPs 2, 3 and 4 have been organised thematically. WPs 5, 6 and 7 were cross-cutting activities. As planned in the Description of Work, the boundary between two of the cross-cutting WPs (5 and 6) is now very hard to maintain, as the two sets of activities have converged. WP5 developed a suite of modelling tools that could facilitate the participatory approaches described in WPs 3 and 4. These models generate scenarios of possible futures at the global, national and regional levels quantifying the economic (including energy sector) trajectories together with the impacts on emissions and temperature. The models are designed in a flexible way so that they can be re-parameterised quickly to respond to new circumstances and be integrated (linked) with other established models (e.g. CGE-based models). WP6 dealt with general challenges related to modelling when it was used for producing policy options for low-carbon economy.
The ultimate goal of the modelling suite developed under WP5 was to quantify impacts of climate change mitigation policies across economic and social sectors, while explicitly tracing feedbacks across scales and between systems, and accounting for non-linearities in socio-economic systems. However as the modelling toolkit evolved, the level of integration increased so that WP6 was able to consolidate a generalized ‘socio-environmental model space’, which included empirical models, conceptual models, complex computer simulations, data sets supplied by WP2-5 and, indeed, by WP6 itself.
The COMPLEX project’s model repository and hierarchy of models was based on inputs from all WPs. We have been guided by the principle that a model is any simplified representation of reality. The repository was aimed to support collaboration within the project and beyond in designing, coding, debugging, testing, documenting, and usage of models and modelling frameworks. Currently, it consists of 23 socio-environmental models, which support research on climate change mitigation actions. These models are very diversified in different aspects:
(1) model domains: climate, hydrology, land use, policy, and economy;
(2) spatial characteristics: vary from global level to regional level and even having no spatial dimension;
(3) temporal characteristics: range from yearly to hourly levels;
(4) model type: most of them are quantitative and few are qualitative;
(5) license type: some are open to the public and some are proprietary;
(6) methodology used: agent based, system dynamics, cellular automata, etc.;
(7) programming language used: Vensim, Netlogo, GAMS, FORTRAN, Matlab, C++, etc.
By the end of the project some of the models have been finalized, while others are available as prototypes. More advanced versions will continue to be developed beyond the project’s lifecycle.
We integrated models/modules into a system of models (i.e. two or more interrelated and independent domain-specific models linked together to create a holistic view of economic-energy-climate system). Both software and ‘human-ware’ solutions were explored. The Distributed Model Integration Framework (DMIF) has been developed to work with models wrapped as web services. DMIF is a web based model integration framework designed to link heterogeneous models developed using different programming languages, hosted on different operating systems, and located anywhere on the Internet. To turn heterogeneous models into interoperable components we used wrappers that made models available as web services. By using this approach we were able to convert models developed using such diverse programming tools as GAMS, NetLogo, C++, etc. into interoperable components.
An important functionality supplied by DMIF is that it allows running models in users' browsers, without any prior installation and setup. Once the model is wrapped as a web service it becomes available for users in their browsers, while DMIF provides the basic interface to run the model and analyse the results. In addition, DMIF provides generic interfaces to link web service models in runtime. We define runtime integration as an integration method in which users can select and integrate properly wrapped models using a graphical user interface, during the time of usage. The user needs to provide a URL of the service description, then the system will fetch the properties of the underlying service (inputs and outputs) and expose them in the browser based GUI to connect to appropriate data flows from other services. By using this interface, we can access different web service based models regardless of their underlying software and hardware platforms, and their location. For runtime integration users can define the workflow and data exchange pattern between the participating models. If the data produced by one web service require complex conversion (say aggregation or disaggregation) before passing it to the next web service then skilled users can develop data conversion web services and can include them in the workflow. Simple conversions can be performed directly in the interface.
In this way we created a powerful modelling tool that can enhance stakeholder participation working with models, using them as standalone components or in connection with other models or data sets.
1.6 The challenges posed by integrative modelling
One of the key challenges in our focus is the multiplicity of the types and the number of models, which are used. Within the COMPLEX project alone, 23 modelling frameworks are developed and applied. Their types range from conceptual models to system dynamics models, to general equilibrium models and to agent-based models – to name a few. Naturally, the model suite of the COMPLEX project represents a very small fraction of models, which have been put forward by researchers working in the field of economic-energy-climate modelling worldwide. Such a variety is not to be seen as a surprise. The economy-energy-climate systems are highly complex, their dynamics is subject to various nonlinearities, and inherent uncertainties are profound. These make it impossible to even think of a single model that would be able to capture all necessary components. In this context, we address two major issues: model linking, including a fusion between qualitative and quantitative modelling and integration within multi-model ensembles.
In addition, WP6 dealt with the cognitive foundations of decision-making; it also provided proof-of-the-concept studies on the role of price and heterogeneity of countries/regions in global integration assessment models; finally it develops exploratory spatially explicit economy-landuse modelling framework
1.6.1 Tools for Integrative Modelling
Integrated climate-energy-economy models are a fundamental tool to assess the socio-economic and environmental impact of climate scenarios and mitigation policies. Currently, there is a great variety of models developed under different approaches, operating at different scales, that are being used to assess different questions related to climate mitigation. All of these approaches have their pros and cons and, in recent years, there is an increasing interest on integrating different models in order to get benefit of their respective advantages and to overcome their limitations. In WP5 we have developed an Integrated System of Models (ISM) combining the strengths of various models by utilizing the state-of-the-art in climate, economics, energy technology, and individual behavior change literature as well as in modelling techniques including computational, integrated and participatory modelling. The IMS set up in WP5 combined climate-energy-economy models operating at different scales. Specifically, we considered a global Integrated Assessment Model (IAM) and global Systems Dynamics (SD) model, a country-level Computable General Equilibrium (CGE) model and an Agent-Based Model (ABM) that links behavioural changes among individual households to aggregated changes in green and grey energy use and emissions at the regional level.
The work implemented in WP5 was carried out through a number of tasks. In particular, we started with a literature review of the state-of-the-art in modelling climate-economy and economy-energy systems at global, EU, regional and individual levels (task 5.1). The Deliverable D5.1 presented a review of existing models in all the four involved modelling paradigms: IAMs, SD, CGE and ABM. Importantly, we focused on the theory and practice of modelling abrupt non-linear changes in complex systems, not only looking at such changes in climate system but also in socio-economic systems (Deliverable D5.2). After the main gaps were identified, WP5 continued with the conceptual design of an integral system of models (task 5.2 and deliverable D 5.3). Next two meetings with stakeholders were organized (task 5.3): one in Brussels and one local in the Netherlands. The workshops had a dual role of presenting and discussing models assumptions in order to integrate stakeholders’ feedback in the modelling exercise, as well as a role of preselecting climate and energy policy scenarios to run with ISM (part of D 5.7). The role of climate change and the use of climate scenarios in each of the models employed in ISM is discussed in the Deliverable D5.4.
With respect to data collection (task 5.4) WP5 went beyond available data sources used primarily in the macro models (IAM, CGE and SD) to run a micro-level data collection by means of a survey carried out in the Netherlands and Spain in 2016 (Deliverable D 5.5). Specifically, the survey elicits the factors and stages of a decision-making process with respect to the three types of energy-related actions households typically make: (1) invest in an energy saving equipment, (2) energy conservation due to a change in energy consumption habits, and (3) switching to another energy source. The survey elicits information on the three main steps preceding any of these actions: knowledge activation, motivation, and consideration.
At each step, several psychological factors (e.g. awareness, personal norms, feeling guilt), economical (e.g. income), socio-demographic (e.g. educational level, age), social (e.g. subjective and social norms), and structural and physical (e.g. energy label and ownership of dwelling) drivers and barriers are considered and estimated (Niamir and Filatova, 2016). The survey data has a potential to elicit the role of information and awareness barriers that prevent households from different income and educational backgrounds from reducing their energy and CO2 footprint. The detailed statistical analysis of the survey data will continue beyond the COMPLEX lifetime. The embedded figures present an example of the distribution of currently used electricity sources and motivation to take one of the actions (invest, conserve or switch).
Finally, the software upgrade of existing models (IAM, SD and CGE) and design and implementation of a new ABM is carried out (task 5.5). The integration is realized using a specially-developed web-services (above), which provides software wrappers to assure that different models exchange data. You can read more about these in D5.6 and 5.7 on our webpage.
1.6.2 Distributed Model Integration Framework (DMIF)
The foundational base for WP6 is the COMPLEX project’s model repository and hierarchy of models assembled by WP6 (Twente) based on inputs from all WPs. We are guided by the principle that a model is any simplified representation of reality. The COMPLEX model repository aims to support collaboration within the project and beyond in designing, coding, debugging, testing, documenting, and usage of models and modelling frameworks. The COMPLEX model space consists of a number of socio-environmental models, which support research on climate change mitigation actions.
We integrated models/modules into a system of models (i.e. two or more interrelated and independent domain-specific models linked together to create holistic view of economic-energy-climate system). Both software and ‘human-ware’ solutions are explored. The Distributed Model Integration Framework (DMIF) has been developed by WP6 (Twente) to work with models wrapped as web services. DMIF is a web based model integration framework designed to link heterogeneous models developed using different programming languages, hosted on different operating systems, and located anywhere on the Internet. To turn heterogeneous models into interoperable components we use web service wrappers. By using this approach we are able to convert models developed using such diverse programming tools as GAMS, NetLogo, C++, etc. into interoperable components.
The other important functionality supplied by DMIF stems from the fact that it allows running models in users' browsers, without any prior installation and setup. Once the model is wrapped as a web service it becomes available for users in their browsers, while DMIF provides the basic interface to run the model and analyse the results. As a result we create a powerful modelling tool that can enhance stakeholder participation is working with models, using them as standalone components or in connection with other models or data sets.
1.6.3 Linking EXIOMOD to a model of energy consumption
Energy consumption agent-based model developed in WP5 by Twente includes behavior constraints and bounded rationality; it is linked with the EXIOMOD model to better inform future climate scenarios. This is a proof-of-the-concept example of integration of models of different geographical scales done within COMPLEX.
The integration of the two models is aimed to assure direct feedbacks between potential behavioral change with consequent changes in market shares of low carbon energy vs. fossil fuel based energy and impacts of these on other sectors of economy (ABM=>CGE), and as well as accounting for non-residential electricity demand and changes in households incomes as economy evolves (CGE=>ABM). While EXIOMOD simulates the connections across economic sectors equilibrating annually over many markets of various goods and services within an economy, the ABM will zoom specifically into the energy market, where preferences and energy consumptions choices driven by individual behaviors of households play a decisive role. This more detailed and conceivably more realistic representation of the energy sector promises to yield new budget shares a households spend on (i) energy vs other goods, and (ii) LCE vs. fossil fuel energy sources which then impact the performance of the entire economy.
1.6.4 Linking the EXIOMOD model and a climate-economy GCAM
WP6 undertakes an innovative effort on creating soft linkages between large models so as to create a system of models. The exploratory work on coupling models is done with the EXIOMOD and GCAM models from WP5 by Twente, TNO and BC3. The EXIOMOD model is a CGE model that evaluates the economic impacts of different environmental policies given scenarios of technological change. GCAM model is an integrated assessment model of climate change that evaluates the effects of climate change on the economy as a whole and on the energy sector in particular. The soft integration of these two models is organized in a sequential way: the EXIOMOD results on the labor productivity are used as inputs into the GCAM to generate scenarios of the future energy mix, penetration of new energy technologies, energy prices, emissions and temperature (more details can be found in WP5 report). These outputs are again then used as inputs into the EXIOMOD.
By linking these models, we developed a fully integrated system of climate and economy, using which we simulated two scenarios: (1) business-as-usual scenario that simulates the situation when there is no climate change policy intervention and (2) policy-based scenario, in which the UNFCCC policy targets set by different regions and countries are fed into GCAM and their effect on the different sectors of the economy is captured in EXIOMOD. Preliminary results indicate that by linking the models we are able to simulate complex climate and economy system feedbacks, which could be hardly accomplished using stand-alone model components. Sensitivity analysis is still to be done to derive robust scenarios.
1.6.5 Contextual Interaction Theory, systems dynamics and land use
Beyond the project DoW, OCT, NIERSC, MPG and Twente teams undertook an exploratory exercise on interpreting the (semi-qualitative) Contextual Interaction Theory (CIT) in a quantitative language of system dynamics (SD), and further incorporated this SD module of actor dynamics into the APoLUS land use model developed in OCT.
In the earlier versions of APoLUS developed by WP3, the actor state variables partially adopted from CIT, such as motivation, cognition, resources, power, and affinity were time-independent (static parameters). An approach to model the dynamics of these variables in the SD language was elaborated, and several alternative specifications of SD models of actor dynamics (from quasi-linear to strongly nonlinear) were developed. These SD models now allow simulating the dynamics of the above listed actor state variables, presenting the simulation results as time series Kovalevsky et al., 2017, in press).
Respectively, now time-dependent actor dynamics variables generated by SD model(s) are incorporated in the code of land-use cellular automata APoLUS model (where the transition rules between the successive states previously depended on corresponding static actor parameters). Essentially, the major added value of this exercise is seen in updating the APoLUS from ‘land-use cellular automata model with actor statics’ to ‘land-use cellular automata model with (explicit) actor dynamics’.

1.6.6 EXIOMOD model and a global climate-energy-economy SDEM
Both models attempt to represent a global coupled climate-economy system. In order to carry out this inter-comparison exercise, the original complex multi-sector EXIOMOD model was simplified into an aggregated one-sector model (the model machinery remained to be based on the CGE modelling paradigm). This simplified EXIOMOD model was then compared with the out-of-equilibrium (system dynamics based) SDEM model in two versions: (i) a single-region, one-sector version and (ii) a few-region, one-sector version. These SDEM versions were specified so that to reflect the main assumptions of EXIOMOD with an essential difference between them being the pricing mechanism. While EXIOMOD relies either on the instantaneous price adjustment in Walrasian price adjustment mechanism (the so-called "Walrasian closure") or, alternatively, on the demand-driven economic dynamics with rigid prices and quantities (the so-called "neo-Keynesian closure"), SDEM is essentially based on an assumption of a finite price adjustment speed in Walrasian price adjustment mechanism. Thus, the undertaken exercise attempts to study the role of the instantaneous market clearing on capital, labor, and consumer goods markets. In the latter, dynamic regimes of idle capital, idle labor [unemployment], and stocks of unsold consumer goods are allowed. Simulations show that the out-of-equilibrium SDEM model yields qualitatively different economic dynamics, with substantial deviations from economic pathways generated by the simplified version of the EXIOMOD model. Particularly, out-of-equilibrium version of the model tends to produce high-frequency oscillations superimposed on overall steady economic growth, that are naturally interpreted as business cycles.
1.6.7 Comparing one-region SDEM model with the multi-region version
This task is built on the SDEM model developed by NIERSC and MPG in WP5. SDEM model is a dynamic model of the coupled climate-socioeconomic system focusing on the strategies of key aggregated economic actors making decisions often pursuing conflicting goals. Jointly they govern the dynamic evolution of the socio-economic system. In the current version of the model, both fossil-fuel-based capital and renewable-energy-based capital determine the production function. We compare a business-as-usual scenario (no mitigation policy) with various mitigation scenarios defined by different level of the global carbon tax rate. The revenues from the carbon tax are recirculated into the economy in the form of investments in renewable-energy-based capital. We explore both the case of constant productivity of renewable-energy-based capital and the case with endogenous improvement of renewable-energy productivity through learning-by-doing effects. The model simulations demonstrate that efficient mitigation policies are feasible with readily affordable costs. From this, we develop a regionalized IAM along the same methodological lines. We consider a large country composed of two regions characterized by different climates and levels of economic development. This is coupled to large residual "country" representing the "rest of the world". It is assumed that a harmonized carbon tax is imposed in both regions of the country and also in the rest of the world. We explore to which extent the transfer of money from carbon tax revenues between the two regions undertaken by a national government can moderate regional disparities in economic development and climate change impacts (Kovalevsky & Hasselmann 2014).
1.6.8 “Multiplication” of models
There have been developed multiple alternative global and regional economy-climate models evaluating plausible future scenarios of economy-climate system development. Often scenarios produced by different models give rather different outcomes; by integrating scenarios based on multiple models, scientists hope to decrease uncertainty and eliminate the bias of a particular model.
Several approaches have been suggested in literature for integration of models from multi-model ensembles. Within the COMPLEX project, we have developed a novel Bayesian-type methodology for posterior integration (reconciliation) of independent probabilistic models describing uncertain systems from different perspectives resting on selection of the models’ posteriorly compatible outcomes. We consider two (or more) independent alternative stochastic model outcomes as priors and, conditional to the event, that both models generate the same (but unspecified) outcome, we suggest a Bayesian formula to define the posterior probabilistic distribution function. Due to its properties, this integration method can be referred to as “multiplication” of models. Note that in this approach, the quality of the model’s performance in the past and present does not play a role.
We have developed three case studies here. The first case study, developed by IIASA, dealt with the integration of two alternative estimates of the net primary production of carbon by Russian forests. While one estimate comes from the combination of all available empirical and semi-empirical methods, including the ground based observations and the remote sensing data, another one relies on the results produced by the available dynamic global vegetation models (DGVMs). The two prior estimates differ by up to 23% across the considered climatic zones. Elimination of these gaps via the multiplication of models helps better quantify the terrestrial ecosystems' input to the global carbon cycle.
The second case study developed by IIASA jointly with the WP2 team focused on the ensemble of sea level pressure models used in WP2. We considered four alternative models of the sea level pressure; integration is performed in three cross-validation runs per each season. We performed a comparison of two alternative integration procedures: 'multiplication' of models and the integration procedure based on the information criterion used by WP2. No one integration method appeared to be consistently “better” across all experiments in reproducing both the mean and the variance of the original distribution.
The third case study focuses on the uncertainty in climate sensitivity – see its description below in the next section.
1.6.9 Uncertainty using SDEM model and multi-model integration
In the third case study on multi-model integration developed by the IIASA team in collaboration with partners from NIERSC and MPG employed the global version of SDEM model, developed in WP5, and amended it with alternative climate sensitivity functions, thus obtaining alternative models producing projections of the future GDP, emissions and temperatures. We estimate future economic losses in GDP due to climate change in each of the five models. Multiplication of models enables to obtain more robust estimates across all models’ pair-wise combinations with a rather small variance. We investigate a mitigation scenario of 30 USD per ton of CO2 and, after the model integration, are able to report a more reliable estimate than prior models, which suggests that in this case the climate change losses can be reduced by appr. 4 times.
1.6.10 An economy-landuse model at different geographical scales
The IIASA team developed statistical methodology to summarize the knowledge about spatial urbanization patterns in a region. The developed modeling procedure was applied to the case study of the Province of Seville, Spain from the WP3 research. We examined association between urbanization processes over space using resampling methods in regression analysis. In general, this work aims to complement existing models of land use change by analyzing the cumulative output of urbanization processes over a single economic phase. The stage of data collection consisted of gathering available data on potential explanatory variables related to land use, population and economy from national and regional government sources. We conducted experimental significance testing and parameter estimation using methods of permutations and bootstrapping, and tested approximation accuracy of the model on a GIS lattice. The developed statistical methodology shows that the land use variable and spatially explicit proxy measurements on economic activity contribute to population densification in case of Seville Province, but the process of regional urbanization cannot be entirely explained by the selected drivers. Thus, a land use modeler should necessarily incorporate uncertainty associated with economic drivers in the model of land use change and use the quantified interdependence between population densification and urban land distribution to refine the probability of change in different areas of the land use map.
1.6.11 Modelling cognitive foundations of decision making
In this work is shared between WP4 and WP6. Computational methods include a combination of cognitive and spatiotemporal modelling approaches with energy/life cycle and socio-economic models, which capture various perceptions, attitudes, and interests regarding regional land use under different scenarios on the path to a low carbon society. In particular, the SLU team has developed a model for decision-making, where several levels of complexity are integrated. Our neuro-cognitive model of the decision making process (DM) of an individual is applied to the choice of transport in a social context, thus integrating ANN and ABM related techniques. The objective was to contribute to an understanding of the relation between individual decisions of citizens and the decisions to be taken by policy makers. Based on the developed neuro-cognitive model, we also model interaction of several individuals for social decision making, exemplified by choice of transport and with consequences for climate change. Our model was intended to give insights on the emotional and cognitive processes involved in DM under various internal and external contexts. We also addressed the relation between short and long term decisions, where individual preferences and attitudes play a crucial role. Knowledge and experience of the outcome of our decisions and actions can eventually result in changes in our neural structures, attitudes and behaviour. In such a feedback loop between individuals and environment/society trust is an important parameter that we explore further.

Potential Impact:
This section of our report deals describes the efforts made by all partners under WP7 (Dissemination and Exploitation). COMPLEX has well-developed plans for future work with external stakeholders in Spain, Netherlands, Sweden, Italy and Norway. We have developed a substantial body of modelling infrastructure and field-tested it in a wide range of research contexts, producing a two-volume report on the work. All our project Foreground is available in the public domain so there are no intellectual property issues to resolve. The project has been extremely active in dissemination actions; organising a wide range of stakeholder engagement exercises, publishing high-impact refereed periodicals and producing no less than 7 monograph volumes of policy briefs and dissemination actions, all available in the public domain. Our research has also advanced the careers of project staff - many, but not all, early career researchers and specialists in stakeholder engagement. Our scientists and stakeholder-partners are our ambassadors; it is they who will carry the project forward over the coming decades.
Our future options are unbounded, but space to report on project impacts is limited. Therefore we confine this statement to lists and counts of our dissemination actions and impact indicators.
3.1 Media and Press
Although we have produced a number of press-releases and distributed them through university press offices, we have found that the scatter-gun approach is something of a lottery. Our press release on the paper in Nature Geoscience, for example, was upstaged by breaking news. Our press releases on the co-evolutionary ecology of co-operation were not widely understood by science journalists more accustomed to stories about finding the genes for autism. We had a steep learning curve to climb and discovered that personal interest stories, public lectures and stories focussed on regional or topical issues worked best.
• Niamir, L. Interview, Twente Graduate School and Institute for Innovation and Governance Studies, University of Twente, 2015.
• T Filatova Film about the ‘Prof De Winter Prize for the Best article of the year’ video http://www.utwente.nl/en/newsevents/2013/11/166791/tatiana-filatova-wins-professor-de-winterprijs and TYPO?
• T Filatova Inaugural speech video for the installation as a Member the Young Academy of the Royal Netherlands Academy of Arts and Sciences (DJA/KNAW), in Dutch.
• Winder N. presentation on innovation to graduate students at CEMUS. https://vimeo.com/63773349
• Video Contribution to the Färgfabriken seminar series (interview with Svedin) TYPO?
• Liljenström et al. (2012, 2014, and 2016) Publication of three debate articles on climate and policy in the Swedish newspaper Dagens Nyheter
• Film produced as record of stakeholder engagement exercise in Sigtuna 21st Jan 2016: What shall we do about carbon?
http://owsgip.itc.utwente.nl/projects/complex/images/uploaded_files/VTS_01_1_VOB.mp4
University of Twente News:
https://www.utwente.nl/en/news/!/2016/2/37463/participatory-models-can-induce-change-and-action

Video: Kovalevsky, D.V. Addressing positive feedbacks and impacts of abrupt climate change in actor-based system dynamics Integrated Assessment models. KlimaCampus Colloquium. 09 April 2015, Max Planck Institute for Meteorology, Hamburg, Germany.
Video of the lecture: https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17477

3.2 Books
COMPLEX is strongly committed to the Open Access model of publication. Sadly the cost of open access publication through journals is often prohibitive and many of our target journals are not yet open access. We have therefore made extensive use of internet repositories and have been actively supporting an open-access model of publication for books, monographs. Our partner institutions in Spain and Sweden have been especially active in this work and we have published 7 book-length monographs to date, coincidentally creating the nucleus of a new book series on Human-Nature Interaction:
1. The Behavioural Ecology of Project-Based Science Sweden, Sigtunastiftelsen (Human Nature Series)
2. Final Scientific Report, Volume 1: The Quest for a Model-Stakeholder Fusion. Sigtunastiftelsen (Human Nature Series)
3. Final Scientific Report, Volume 2: Non-linearities and System-Flips. Sigtunastiftelsen (Human Nature Series )
4. Final Scientific Report, Volume 3a: Establishing Policy-Relevance: Human-Environment Interaction Sigtunastiftelsen (Human Nature Series )
5. Final Scientific Report, Volume 3b: Establishing Policy-Relevance: Developing and Evaluating Policy Options Sigtunastiftelsen (Human Nature Series)
6. Liljenström, H. and Svedin, U. Eds. (2016) Towards a Fossil-free Society – In the Stockholm-Mälar Region. Sigtuna. COMPLEX WP4 Final Scientific Report, Human Nature Series. Sigtuna: Sigtunastiftelsen, ISBN 978-91-976048-4-0.
7. Hernández Jiménez, V., Encinas Escribano M. A., Hewitt, R., Ocón Martín, B., Román Bermejo, L.P. and Zazo Moratalla, A. (2016), ¿Qué territorio queremos? Estrategias participativas para un futuro común. [What kind of territory do we want? Participatory strategies for a common future]. Observatorio para una Cultura del Territorio, Madrid, Spain.
3.3 Books In Preparation
Final Scientific Report, Volume 4: COMPLEX: what we did, what we learned and why it matters Sigtunastiftelsen (Human Nature Series) (forthcoming)
Hewitt, R, Hernández Jiménez, V., Encinas Escribano M. A., Ocón Martín, B., Román Bermejo, L.P. and Zazo Moratalla, A. (in preparation), Participatory Modeling for Resilient Futures: Action for Managing Our Environment from the Bottom-Up. Elsevier series “Developments in Environmental Modelling”. Series editor: Brian D. Fath. Book proposal accepted, book in progress, publication scheduled for 2017.
3.4 International conferences
Conferences have been an important part of our dissemination policy, they provide a chance to test ideas on our scientific peers and receive feedback, they build competence among early career researchers, and provide senior scientists with an opportunity to give keynote addresses. Strategically important conferences have sometimes been targeted with multiple presentations because they give us direct access to strategically significant peers and often generate spin-off publications. The following were international conferences, most of which produced proceedings or papers:
1. François, B., Hingray, B., Hendrickx, F. and Creutin, J.D. ‘The value of storage water: a climatological signature for global change impact studies’. AIC conference (In French), Grenoble, France, 2012
2. François, B., Hingray, B., Hendrickx, F. and Creutin, J.D. ‘Estimating performance of a multi- purpose Alpine water reservoir under climate change’. ICEM, Toulouse, 2013
3. Argent, Robert M, Richard S Sojda, Carlo Guipponi, Brian Mcintosh, and Alexey A Voinov. 2014. “Best Practice in Conceptual Modelling for Environmental Software Development” 7th Intl. Congress on Env. Modelling and Software, San Diego, CA, USA.
4. Ramos, M.H. Creutin, J.D. Engeland, K., François, B. and Renard, B. ‘System’s flips in climate- related energy systems’. EGU2014-9087, Vienna, Austria, May 2014
5. François, B, Borga, M., Anquetin, S., Creutin, J.D. England, K., Favre, A.C. Hingray, B., Ramos, M.H. Raynaud, D., Renard, B., Sauquet, E., Sauterleute, J.F. Vidal, J.P and Warland, G. (2014) ‘Integrating hydropower and intermittent climate-related energies: a call for hydrology’, EGU2014-5761, Vienna, Austria, May 2014
6. Engeland K, M Borga, J-D Creutin, MH Ramos, L Tofte, J-P Vidal, et al. Space-time dependence between energy sources and climate related energy production. EGU General Assembly 2014. EGU, Vienna, Austria, 2014.
7. François, B., Creutin, J.D. Hingray, B. and Zoccatelli D. ‘Integration of small run-of-river and solar power’. EGU2014-5845, Vienna, Austria, May 2014
8. Kolbjorn Engeland, Marco Borga, Jean-Dominique Creutin, Maria-Helena Ramos, Lena Tøfte, and Geir Warland: Space-time dependence between energy sources and climate related energy production. EGU2014-4840, Vienna, Austria 2014.
9. Lena S. Tøfte, Julian Sauterleute, Geir Warland, Sjur Kolberg: Modeling CRE in a changing climate and energy system (mid-Norway). EGU2014-7051, Vienna, Austria 2014
10. Resilience 2014: Resilience and Development. Montpellier, France. May 4-8, 2014
11. ESEIA-IGS Conference Smart and Green Transitions in Cities and Regions: Enschede, the Netherlands. April 24-25 2014
12. 7th International conference on Land planning, FUNDICOT, Madrid Complutense University. Title: Planificación Participativa para un Planeta Resiliente. 27-29/11/2014
13. 7th Intl. Congress on Environmental Modelling and Software, San Diego, CA, USA. June 15-19, 2014. Getachew F. Belete presented 2 papers: “An architecture for integration of multidisciplinary models” and “Integration of Models for Low Carbon Economy”.
14. CSDMS Annual Meeting, “Data meet models, models meet data.” Boulder, Colorado, USA. May 26-28, 2015. Getachew F. Belete: Real-time integration of models
15. Niamir, L., and T. Filatova, 2015. Linking Agent-based energy market with Computable General Equilibrium Model. 20th WEHIA. Sophia Antipolis, France.
16. Niamir, L., and T. Filatova, 2015. Simulating Nonlinearities in the Electricity Market, Navarre Region-Spain. 11th Social Simulation Conference (SSC2015), Groningen, The Netherlands, 2015b. Springer Proceedings in Complexity.
17. IALE 9th World Congress “Crossing Scales, Crossing Borders: Global Approaches to Complex Challenges” Portland, Oregon, EEUU, 5-10 July 2015.
18. François , B., Hingray, B., Creutin, J.D. Borga, M., Raynaud, D. and Vautard, R: ‘Climate related energy sources: sensitivity study to climate characteristics across Europe’. EGU2015-482, Vienna, Austria, April 2015
19. François , B., Borga, M., Creutin, J.D. Hingray, B., Raynaud, D. and Sauterleute, J.F. ‘Complementarity between solar and hydro power: Sensitivity to climate characteristics in Northern-Italy’, EGU2015-3580, Vienna, Austria, 2015
20. A. Shchiptsova, D. Kovalevsky, E. Rovenskaya: Reconciling Information from Alternative Climate-economic Models: A Posterior Integration Approach, Systems Analysis 2015 - Celebration of Howard Raiffa, 2015, IIASA, Laxenburg, Austria
21. R. Hewitt, A. Shchiptsova, E. Rovenskaya: Understanding the Drivers of Urban Expansion: Case Study of Seville Province, Systems Analysis 2015 - A Conference in Celebration of Howard Raiffa, 11-13 November 2015, IIASA, Laxenburg, Austria
22. Borga, M., François, B., Creutin, J.D. Hingray, B., Zoccatelli, D. and Tardivo, G. ‘Assessment of potential for small hydro/solar power integration in a mountainous, data-sparse region’. EGU2015-10401, Vienna, Austria, April 2015
23. François, B., Hingray, B., Creutin, J.D. and Hendrickx, F. ‘Influence of the management strategy model on estimating water system performance under climate change’. EGU2015- 3612, Vienna, Austria, April 2015
24. Tøfte L. S.: Regional estimation of response routine parameters. EGU2015-12432, Vienna, Austria, 2015.
25. Raynaud, D., Hingray, B., Chardon, J., Anquetin, S., Favre, A.C. François, B. and Vautard, R., Tobin, I. ‘Multivariate weather prediction with atmospheric analogs for different European regions’. EGU2015-834, Vienna, Austria, April 2015
26. Puspitarini, H.D. H., François, B., Hingray, B., Raynaud, D., and Creutin, J.D. ‘Fluctuation Analysis of Climate-Related energies in Europe’. Renewable Energy & Green Technology Conference, Kuta, Indonesia, 2015.
27. 13th European Week of Regions and Cities. October 12-15, 2015, Brussels. Cheryl de Boer presented WP3 as part of a workshop given by AESOP (Association of European Schools of Planning). Energy issues in regional and urban development.
28. The International Society for Ecological Modelling Global Conference 2016, 8-12 May 2016 Towson University, MD, USA. Richard Hewitt presented research at this and also chaired the session entitled “Ecological Landscape and Land Use Change Modelling”
29. International Environmental Modelling and Software (IEMS) International Congress. July 10-14, 2016, Toulouse. Cheryl de Boer presented WP3 research as part of a session on Modelling for Low Carbon Economies.
30. Borga, M., B. François, B. Hingray, D. Zoccatelli, J.D. Creutin, C. Brown, Linking top-down and bottom-up approaches for assessing the vulnerability of a 100 % renewable energy system in Northern Italy, Poster, EGU conference, Vienne, Autriche, Avril, 2016.
31. Tøfte, L.S. Martino S., Mo, B.: Using climate response functions in analysing electricity production variables. A case study from Mid-Norway EGU 2016 -12170, Vienna Austria 2016.
32. Tøfte L. S., Martino S: Analysing electricity production in today's and tomorrow's climate. Presentation at seminar and Kick-off meeting in the EEA Grants project Intelligent energetic system in protected areas. Iasi, Romania, June 2016.
33. EGU Conference , Vienne, Autriche, Avril, 2016. 5 separate presentations from WP2
34. Renard B., Vidal J.-P. A performance weighting procedure for GCMs... 13th International Meeting on Statistical Climatology, Canmore, Alberta, CAN, Juin 2016.
35. Economic Development in Africa Conference, CSAE, University of Oxford, March 29-31, 2016.
36. Latin American and Caribbean Economic Association Conference, EAFIT, Medellin, Colombia, November 2016
37. 2016 8th International Congress on Environmental Modelling and Software, Toulouse, France. 4 Presentations from WPs 5 and 6: Niamir and Filatova on climate change awareness, Belete et al on linking heterogenous models, Belete on designing integration systems and Shchiptsova et al on Reconciling Information from Climate-Economic Model Ensembles.
38. A. Shchiptsova, R. Hewitt, E. Rovenskaya: Measuring Spatial Feedbacks in Urban Systems, European Meetings on Cybernetics and Systems Research (emcsr avantgarde), 30 March – 1 April 2016, Bertalanffy Center (BCSSS), Vienna, Austria
39. E. Rovenskaya: Reconciling information from climate-economic model ensembles, Data Intensive System Analysis for Geohazard Studies, 18-21 July 2016, Sochi region, Russia
40. E. Rovenskaya, A. Shchiptsova, D. Kovalevsky, Reconciling information from climate-economic model ensembles, International Conference in Memory of Academician Arkady Kryazhimskiy on Systems Analysis: Modeling and Control, 03-08 October 2016, Institute of Mathematics and Mechanics, Ural Branch of the Russian Academy of Sciences, Ekaterinburg, Russia
41. Kovalevsky, D.V. Hasselmann, K. Modelling the impacts of a national carbon tax in a country with inhomogeneous regional development: an actor-based system-dynamic approach. ERSA 54th Congress “Regional Development & Globalisation: Best Practices”, 26-29 August 2014, St. Petersburg State University (SPbU), St. Petersburg, Russia. [Conference paper at IDEAS, URL: https://ideas.repec.org/p/wiw/wiwrsa/ersa14p743.html]
42. Kovalevsky, D.V. Hasselmann, K. Assessing the transition to a low-carbon economy using actor-based system-dynamic models. The 7th International Congress on Environmental Modelling and Software (iEMSs 2014), 15-19 June 2014, San Diego, California, USA [Conference paper in Brigham Young University ScholarsArchive, URL: http://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=1211&context=iemssconference ]
43. Kovalevsky, D.V. Hasselmann, K. Actor-based system dynamics modelling of win-win climate mitigation options. The 8th International Congress on Environmental Modelling and Software (iEMSs 2016), 10-14 July 2016, Toulouse, France. [Conference paper in Brigham Young University ScholarsArchive, URL: http://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=1588&context=iemssconference ]
44. Kovalevsky, D.V. Modeling herding behavior on financial markets affected by exogenous climate-related shocks. The 8th International Congress on Environmental Modelling and Software (iEMSs 2016), 10-14 July 2016, Toulouse, France. [Conference paper in Brigham Young University ScholarsArchive, URL: http://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=1586&context=iemssconference ]
45. Shchiptsova, A., Kovalevsky, D., Rovenskaya, E. Reconciling information from climate-economic model ensembles. The 8th International Congress on Environmental Modelling and Software (iEMSs 2016), 10-14 July 2016, Toulouse, France. [Conference abstract in Brigham Young University ScholarsArchive, URL: http://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=1587&context=iemssconference ]
46. Kovalevsky, D.V. Modelling the win-win opportunities of climate mitigation policies. The Third International Conference “Sustainable Development: Society and Economy”, 20-23 April 2016, St. Petersburg State University (SPbU), St. Petersburg, Russia. NEW ITEM
47. Flacke, J., & de Boer, C. (2016) An Interactive GIS-Tool for Collaborative Local Renewable Energy Planning. Agile conference 2016, 14-17 June 2016, Helsinki, Finland. available at: https://agile-online.org/conference_paper/cds/agile_2016/shortpapers/130_Paper_in_PDF.pdf
48. Hewitt, R., Kovalevsky, D.V. de Boer, C., Hasselmann, K. Modelling actors’ influence on land use change: a dynamic systems approach. Submitted as a conference paper to 20th AGILE International Conference on Geographic Information Science, Wageningen University, The Netherlands, 10-12 May 2017

3.5 Other Conferences, Working Papers and Book Chapters
In addition to the international conferences listed above, COMPLEX team members have attended at least 30 National Conferences and workshops and produced a large number of discussion papers and reviews, often as invited participants. Invited talks and keynotes are an important dissemination mechanism because they guarantee the presenter an interested and receptive audience and often lead to publication. We have also been invited to contribute to a number of edited books:

1. T.Filatova (2013). Annual conference of the Groningen Center for Social Complexity Studies, University of Groningen, The Netherlands. Invited talk “Changing climate – changing behavior: flood hazards, risk perception and markets.”
2. T.Filatova (2013) Leibniz Institute for Agricultural Development in Central and Eastern Europe (IAMO), Halle, Germany. Invited lecture "Agent-based land markets: methods, challenges and perspectives"
3. T.Filatova (2013). Wageningen University & Research centre, the Netherlands. Invited talk “Changing climate – changing behavior :Modeling abrupt structural shifts in complex socio-environmental systems from the bottom-up”
4. T.Filatova (2013). Institute of Environmental Systems Research, University of Osnabrueck, Germany. “Behavioral changes and abrupt structural shifts in complex socio-environmental systems”
5. Voinov A. USGS, Reston, USA - Invited lecture: “Values in modeling: can science be really value neutral?”, December 2014.
6. Voinov A. ESS Summer-School, Scientific policy-advice under (deep) uncertainty – The case of energy scenarios, Invited lecture: “Values in modeling: can science be really value neutral?”, October 2014.
7. Voinov A. CSDMS - 4th annual meeting: Uncertainty & Sensitivity in Surface Dynamics Modeling - University of Colorado — Boulder - Plenary keynote: “Exploring climate mitigation and low-carbon transition: new challenges for model integration”, May 2014
8. Voinov A. James Hutton Institute, Aberdeen, UK, Invited talk “Exploring low-carbon transitions by means of model integration”, March 2014.
9. Voinov A. IQ SCENE, UCL Energy Institute, UK, Invited talk “Exploring low-carbon transitions by means of model integration”, March 2014.
10. T.Filatova (2014) Stockholm Resilience Center, Stockholm, Sweden. Invited talk ‘Structural changes in coupled socio-ecological systems’
11. Sundberg, C (2014). Challenges of adding a life cycle perspective to municipal-level decision support for transition to a climate-neutral society. Proceedings SETAC conference Basel May 2014
12. Voinov A. The Economics of Climate Change, International Conference, National Chengchi University, Taipei, Taiwan - Keynote “Exploring climate mitigation and low-carbon transitions: new challenges for modeling”, August, 2015.
13. T.Filatova (2015) University of Hamburg, Research Unit Sustainability and Global Change, Hamburg, Germany. Invited talk ‘Changing climate – changing behavior.
14. Leslie Harris Centre of Regional Policy and Development, Memorial University of Newfoundland, Newfoundland, Canada, May 2nd 2016
15. Richard Hewitt. Invited presentation at James Hutton Institute, Aberdeen, UK, on 5th August 2016.
16. T.Filatova (2016) EAWAG/ETH, Workshop, Zurich, Switzerland. Invited talk ‘Combining ABM, surveys and lab experiments’
17. T.Filatova (2016) Future Earth Cluster "Linking Earth-System and Socio-Economic Models", Aix-en-Provence, France. Invited talk ‘Economic Actors in Social-Ecological Systems: What can ABM offer?’
18. Voinov A. ICM - Innovations in Collaborative Modeling, Michigan State University, Plenary speaker “Biases, Beliefs and Values in Participatory Modeling and Citizen Science”, June, 2016
19. Voinov A. Science of Future - Kazan, Russia, Keynote speaker “Participatory modeling: integrating models and stakeholders”, Sept., 2016.
20. E Rovenskaya, A Shchiptsova, D Kovalevsky (2016): Reconciling information from climate-economic model ensembles. Geophysical Center RAS, Moscow Russia, 4: 4002-4002
21. A. Shchiptsova, R. Hewitt, E. Rovenskaya: Exploring Drivers of Urban Expansion, 4th International Conference on Computer Science Applied Mathematics and Applications (ICCSAMA2016), 2-3 May 2016, IIASA, Laxenburg, Austria.
22. Kovalevsky, D.V. Hasselmann, K. (2013): Out-of-equilibrium actor-based system-dynamic modelling of the economics of climate change. Workshop paper. GSS Preparatory Workshop for the 3rd Open Global Systems Science Conference (2014), 29-30 October 2013, Beijing, China. URL: http://blog.global-systems-science.eu/?p=2063
23. Kovalevsky, D.V. Hasselmann, K. Integrated Assessment modelling of global impacts of shrinking Arctic sea ice. All-Russian Conference with International Participation “State of Arctic Seas and Territories under Conditions of Climate Change”, 18-19 September 2014, Northern (Arctic) Federal University (NArFU), Arkhangelsk, Russia.
24. Kovalevsky, D.V. Actor-based system-dynamic Integrated Assessment modelling: addressing out-of-equilibrium dynamics, positive feedbacks and impacts of abrupt climate change. Climate & Economics Workshop, 13 October 2014, Nansen Environmental and Remote Sensing Center (NERSC), Bergen, Norway.
25. Kovalevsky, D.V. “The Arctic feedback” model. EU FP7 EuRuCAS 3rd Workshop, 27-28 May 2015, Nansen International Environmental and Remote Sensing Centre (NIERSC), St. Petersburg, Russia.
26. Kovalevsky, D.V. Abrupt climate change and climate policy modeling. EU FP7 EuRuCAS Climate Policy Modeling Workshop, 24-26 June 2015, Nansen International Environmental and Remote Sensing Centre (NIERSC), St. Petersburg, Russia.
27. Kovalevsky, D.V. Prasolov, A.V. Modelling the coupled climate–economic dynamics within time-to-build approach. The Fourth International Workshop on Natural Resources, Environment and Economic Growth, 01-02 October 2015, European University at St. Petersburg, St Petersburg, Russia.
28. Kovalevsky, D.V. Rovenskaya, E.A. Posterior integration of climate–economic models. The Second Readings in Memory of Prof. B.L. Ovsievich “Economic-Mathematical Studies: Mathematical Models and Information Technologies” – All-Russian Conference, 26-28 October 2015, St. Petersburg Institute of Mathematical Economics of the Russian Academy of Sciences, St. Petersburg, Russia.
3.5.1 Book chapters
1. Shchiptsova, A., Hewitt, R., & Rovenskaya, E. (2016). Exploring Drivers of Urban Expansion. In Advanced Computational Methods for Knowledge Engineering (pp. 153-165). Springer International Publishing.
2. Hewitt, R. (accepted October 2017): The Actor, Policy, and Land Use Simulator (APoLUS). In Camacho Olmedo, M.T. Paegelow, M, Mas, J.F. and Escobar, F. (eds). Geomatic simulations and scenarios for modelling LUCC: A review and comparison of modelling techniques. (in press): Lecture Notes in Geoinformation and Cartography' LNGC series (http://www.springer.com/series/7418) Springer.
3. in press. Pacheco, J.D. van Delden, H., & Hewitt, R. (accepted). The importance of scale in land use models: experiments in data conversion, data resampling, resolution and neighbourhood extent. In Camacho Olmedo, M.T. Paegelow, M, Mas, J.F. and Escobar, F. (eds). Geomatic simulations and scenarios for modelling LUCC: A review and comparison of modelling techniques. (in press): Lecture Notes in Geoinformation and Cartography' LNGC series (http://www.springer.com/series/7418) Springer.
4. in press. Hewitt, R., Hernández Jiménez, V, Román Bermejo, L, and Escobar, F. (accepted): Who knows best? The role of stakeholder knowledge in land use models- an example from Doñana, SW Spain. In Camacho Olmedo, M.T. Paegelow,M, Mas, J.F. and Escobar, F. (eds). Geomatic simulations and scenarios for modelling LUCC: A review and comparison of modelling techniques. (in press) Lecture Notes in Geoinformation and Cartography' LNGC series (http://www.springer.com/series/7418) Springer.
5. Liljenström, H. and Hassannejad Nazir, A. (2016): Decisions and Downward Causation in Neural Systems. In: (R. Wang & X. Pn, Eds.) Advances in Cognitive Neurodynamics (V). pp. 161-167. Singapore: Springer. DOI 10.10007/9778-981-10-0207-6_7
6. Hassannejad Nazir, A. and Liljenström, H. (2015b) Neurodynamics of Decision Making – A Computational Approach. In: (R. Wang & X. Pn, Eds.) Advances in Cognitive Neurodynamics (V). Singapore: Springer. DOI 10.10007/9778-981-10-0207-6_7.
7. Svedin, U. and Liljenström, H. (2016) Paths to a low carbon society by 2050 – The Stockholm-Mälar case. In: (L. Ekenberg, K. Hansson, M. Danielson, G. Cars, et al, Eds.) Deliberation, Representation and Equity: Research Approaches, Tools and Algorithms for Participatory Processes, Open Book Publishers, (in print.) ISBN Paperback: 9781783743032
8. Svedin, U. (2015). Urban Development and the Environmental Challenges – “Green” Systems Considerations for the EU, In: W. Leal Filho et al. (eds.), Sustainable Development, Knowledge society and Smart Future Manufacturing Technologies, page 81-112, World Sustainability Series, Springer International Publishing DOI 10.1007/978-3-319-14883-0_7
9. Svedin, U. (2015). Urban Development and the Environmental Challenges – “Green” Systems Considerations for the EU, In: W. Leal Filho et al. (eds.), Sustainable Development, Knowledge society and Smart Future Manufacturing Technologies, page 81-112, World Sustainability Series, DOI 10.1007/978-3-319-14883-0_7
3.5.2 In preparation - Book Chapters
Niamir L. and T.Filatova ‘Transition to Low-carbon Economy: Simulating Nonlinearities in the Electricity Market, Navarre Region-Spain’ in ‘Advances in Social Simulation’, W.Jager (Eds), Springer: Heidelberg.
3.6 Journal articles
As scientists, our most important dissemination action are lodged in refereed journals. We have used wide range of journals, ranging in impact from Nature down to relatively obscure open-access periodicals. We do not despise lower impact journals any more than we would any other dissemination action. However, a glance at the list below would show that COMPLEX has a set of core journals that we target for hydrology, natural resource management, environmental modelling and the circular economy.
1. Winder, I.C. and Winder, N.P. 2013. An agnostic approach to ancient landscapes. Journal of Archaeology and Ancient History. 9. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-210069
2. Filatova T., P.H. Verburg, D.C. Parker, C.A. Stannard (2013). “Spatial agent-based models for socio-ecological systems: challenges and prospects”, Environmental Modelling & Software (IF=4.420) Volume 45, p. 1-7 http://www.sciencedirect.com/science/article/pii/S1364815213000807
3. Hewitt, R., Van Delden, H., & Escobar, F. (2014). Participatory land use modelling, pathways to an integrated approach. Environmental Modelling & Software, 52, 149-165.
4. Voinov, Alexey, Ralf Seppelt, Stephan Reis, Julia E.M.S. Nabel, and Samaneh Shokravi. 2014. “Values in Socio-Environmental Modelling: Persuasion for Action or Excuse for Inaction.” Environmental Modelling & Software 53 (March): 207–12. doi:10.1016/j.envsoft.2013.12.005.
5. Voinov, Alexey, and Tatiana Filatova. 2014. “Pricing Strategies in Inelastic Energy Markets: Can We Use Less If We Can’t Extract More?” Frontiers of Earth Science 8 (1): 3–17. doi:10.1007/s11707-013-0410-y.
6. Arodudu, Oludunsin, Esther Ibrahim, Alexey Voinov, and Iris van Duren. 2014. “Exploring Bioenergy Potentials of Built-up Areas Based on NEG-EROEI Indicators.” Ecological Indicators 47 (May). Elsevier Ltd: 67–79. doi:10.1016/j.ecolind.2014.04.042.
7. Rojo, M. S., Moratalla, A. Z., Alonso, N. M., & Jimenez, V. H. (2014). Pathways towards the integration of periurban agrarian ecosystems into the spatial planning system. Ecological Processes, 3(1), 1.
8. Winder, I.C. and Winder, N.P. 2014. Reticulation and the human past. Annals of Human Biology 41: 300-311.
9. Voinov A., T. Filatova (2014) “Pricing strategies in inelastic energy markets: can we use less if we can’t extract more?” Frontiers in Earth Science (IF=0.883) March 2014, Volume 8, Issue 1, pp 3-17, http://link.springer.com/article/10.1007%2Fs11707-013-0410-y
10. François, B., M. Borga, S. Anquetin, J.-D. Creutin, K. Engeland, A.-C. Favre, B. Hingray, M. H. Ramos, D. Raynaud, E. Sauquet, J. Sauterlete, C. Vidal, and G. Warland, 2014: Integrating hydropower and intermittent climate-related renewable energies: a call for hydrology. Hydrology Today, 28, 5465–5468. doi:10..1002/hyp.10274.
11. François, B., B. Hingray, F. Hendrickx, and J. D. Creutin, 2014: Seasonal patterns of water storage as signatures of the climatological equilibrium between resource and demand. Hydrol. Earth Syst. Sci., 18, 3787-3800. doi:10.5194/hess-18-3787-2014.
12. Tol, Richard S J (2014) Ambiguity reduction by objective model selection, with an application to the costs of the EU 2030 climate targets. Energies, 7 (11). pp. 6886-6896. ISSN 1996-1073
13. Tol, Richard S J (2014) Bootstraps for meta-analysis with an application to the total economic impact of climate change. Computational Economics, 46 (2). pp. 287-303. ISSN 0927-7099
14. de Boer, C., Hewitt, R., Bressers, H., Alonso, P. M., Jiménez, V. H., Pacheco, J. D., & Bermejo, L. R. (2015). Local power and land use: spatial implications for local energy development. Energy, sustainability and society, 5(1), 1.
15. Raghothama, J & Meijer, S (2015). What do policy makers talk about when talking about simulations? ISAGA Proceedings, Kyoto, Japan July 2015.
16. Raghothama, J. & Meijer, S. (2015) ‘Gaming, Urban Planning and Transportation Design Process ‘in: Geertman et al. (eds.), Planning Support Systems and Smart Cities, p. 297-312, Springer.
17. Filatova, T. (2015) "Empirical agent-based land market: Integrating adaptive economic behavior in urban land-use models", Computers, Environment and Urban Systems (IF=1.537) 54, p.397–413, http://www.sciencedirect.com/science/article/pii/S0198971514000714
18. Gren, I-M. (2015) Estimation of values of forest carbon sequestration and nutrient recycling: An application to the Stockholm-Mälar region’. Forests 6(10): 3594-3613. http://www.mdpi.com/1999-4907/6/10/3594
19. Hassannejad Nazir, A. & Liljenström, H. (2015) A Cortical Network Model for Cognitive and Emotional Influences in Human Decision Making. BioSystems 136:128-141 doi:10.1016/j.biosystems.2015.07.004.
20. A Kryazhimskiy, E Rovenskaya, A Shvidenko, M Gusti, D Shchepashchenko, V Veshchinskaya (2015): Towards harmonizing competing models: Russian forests' net primary production case study. Technological Forecasting and Social Change, 98: 245-254
21. Hasselmann, K., Cremades, R., Filatova, T., Hewitt, R., Jaeger, C., Kovalevsky, D., ... & Winder, N. (2015). Free-riders to forerunners. Nature Geoscience, 8(12), 895-898.
22. Winder, N.P. and Winder, I.C. 2015. Compassion, complexity and self-organisation: human evolution and the vulnerable ape hypothesis. Internet Archaeology 40. http://intarch.ac.uk/journal/issue40/3/
23. Puspitarini, H. D., B. François, B. Hingray, D. Raynaud, and J.-D. Creutin, 2015: Fluctuation analysis of climate related energies in Europe. ARPN Journal of Engineering and Applied Sciences, 10.
24. François, B., B. Hingray, J.-D. Creutin, and F. Hendrickx, 2015: Estimating Water System Performance Under Climate Change: Influence of the Management Strategy Modeling. Water Resources Management, 29,4903-4918. doi:10.1007/s11269-015-1097-5.
25. François, B., B. Hingray, D. Raynaud, M. Borga, and J.-D. Creutin, 2015: Increasing Climate-Related-Energy penetration by integrating run-of-the river hydropower to wind/solar mix. Renewable Energy Journal, 87, 686-696. doi:10.1016/j.renene.2015.10.064.
26. Alonso, P. M., Hewitt, R., Pacheco, J. D., Bermejo, L. R., Jiménez, V. H., Guillén, J. V., ... & de Boer, C. (2016). Losing the roadmap: Renewable energy paralysis in Spain and its implications for the EU low carbon economy. Renewable Energy, 89, 680-694.
27. Hammar, T., Hansson, P.-A. & Sundberg, C. (2016). Climate impact assessment of willow energy from a landscape perspective: a Swedish case study. GCB Bioenergy. Doi: 10.1111/gcbb.12399
28. Kovalevsky, D.V. Hewitt, R., de Boer, C., Hasselmann, K. (2017): A dynamic systems approach to the representation of policy implementation processes in a multi-actor world.
Discontinuity, Nonlinearity, and Complexity, 6(3) (in press)
29. A Kryazhimskiy (2016): A Posteriori Integration of Probabilities. Elementary Theory. Theory of Probability & Its Applications, 60 (1): 62-87
30. François, B., M. Borga, J. D. Creutin, B. Hingray, D. Raynaud, and J. F. Sauterleute, 2016: Complementarity between solar and hydro power: Sensitivity study to climate characteristics in Northern-Italy. Renewable Energy, 86, 543-553. doi:10.1016/j.renene.2015.08.044.
31. François, B. 2016: Influence of the North-Atlantic Oscillation on Climate-Related-Energy Penetration in Europe, Renew. Energy, 99, 602-613. doi: 10.1016/j.renene.2016.07.010.
32. Hingray, B., J. Blanchet, 2016: Uncertainty components estimates in transient climate projections. 1. Bias of empirical estimators in the single time and time series approaches. Submitted to J. of Climate.
33. Hingray, B., J. Blanchet, J.P. Vidal, 2016: Uncertainty components estimates in transient climate projections. 2. Robustness of unbiased estimators in the single time and time series approaches. Submitted to J. of Climate.
34. Raynaud, D., Hingray, B., Zin, I., Anquetin, S., Debionne, S., Vautard, R. 2016-a: Atmospheric analogues for physically consistent scenarios of surface weather in Europe and Maghreb. International Journal of Climatology. In print. doi: 10.1002/joc.4844.
35. Filatova, T., J.G. Polhill, S. van Ewijk (2016) Regime shifts in coupled socio-environmental systems: Review of modelling challenges and approaches. Environmental Modelling & Software (IF=4.420) 75, p. 333–347, http://www.sciencedirect.com/science/article/pii/S1364815215001127
36. Polhill, J.G. T.Filatova M.Schlüter, A.Voinov (2016) ‘Modelling systemic change in coupled socio-environmental systems’, Environmental Modelling & Software (IF=4.420) 75, p. 318–332, http://www.sciencedirect.com/science/article/pii/S136481521530075X
37. Belete, G.F. Voinov, A., 2016. Exploring temporal and functional synchronization in integrating models: A sensitivity analysis. Computers & Geosciences 90, 162-171.
38. Belete, G.F. Voinov, A., Laniak, G.F 2017. An overview of the model integration process: from pre-integration assessment to testing. Environmental Modelling and Software 87, 49-63.
39. Voinov, A. et al., 2016. Modelling with stakeholders - Next generation. Environmental Modelling and Software, 77, pp.196–220.
40. Argent, R.M. Voinov A., et al., 2016. Best practices for conceptual modelling in environmental planning and management. Environmental Modelling & Software, 80, pp.113–121. Available at: http://www.scopus.com/inward/record.url?eid=2-s2.0-84959341722&partnerID=tZOtx3y1 [Accessed March 16, 2016].
41. Voinov, A. et al., 2015. Estimating the potential of roadside vegetation for bioenergy production. Journal of Cleaner Production, 102(March 2007), pp.213–225. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0959652615003959.
42. van Duren, Iris, Alexey Voinov, Oludunsin Arodudu, and Melese Tesfaye Firrisa. 2015. “Where to Produce Rapeseed Biodiesel and Why? Mapping European Rapeseed Energy Efficiency.” Renewable Energy 74 (February). Elsevier Ltd: 49–59. doi:10.1016/j.renene.2014.07.016.
43. Polhill, J. Gary, Tatiana Filatova, Maja Schlüter, and Alexey Voinov. 2016. “Modelling Systemic Change in Coupled Socio-Environmental Systems.” Environmental Modelling & Software 75 (January). Elsevier Ltd: 318–32. doi:10.1016/j.envsoft.2015.10.017.
44. Farley, Joshua, and Alexey Voinov. 2016. “Economics, Socio-Ecological Resilience and Ecosystem Services.” Journal of Environmental Management 183. Elsevier Ltd: 389–98. doi:10.1016/j.jenvman.2016.07.065.
45. Arodudu, O., Voinov A. et al., 2017. Towards a more holistic sustainability assessment framework for agro-bioenergy systems — A review. Environmental Impact Assessment Review, 62, pp.61–75. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0195925516300944.
46. Hewitt, R., & Díaz-Pacheco, J. (2017). Stable models for metastable systems? Lessons from sensitivity analysis of a Cellular Automata urban land use model. Computers, Environment and Urban Systems, 62, 113-124.
47. Hasselmann, K. (2013): Detecting and responding to climate change. Tellus, Series B: Chemical and Physical Meteorology, 65, 20088
48. Kovalevsky, D.V. Hasselmann, K. (2014): A hierarchy of out-of-equilibrium actor-based system dynamic nonlinear economic models. Discontinuity, Nonlinearity, and Complexity, 3(3), 303-318. DOI:10.5890/DNC.2014.09.007
49. Kovalevsky, D.V. (2016): On the sensitivity of ecological economics models of lake water resource management to the welfare function parameters. Transactions of the Karelian Research Centre of the Russian Academy of Sciences. Limnology, No. 9, 102-108. DOI:10.17076/lim455. URL: http://journals.krc.karelia.ru/index.php/limnology/article/download/455/311
3.6.1 In Preparation / In Press
François, B., S. Martino, L. Tøfte, B. Hingray, B. Mo, J.-D. Creutin, 2016: Effects of increased wind power generation on Mid-Norway’s energy balance under climate change: A market based approach. Submitted to Energies.
Creutin, J.D. Hewitt, R., Tøfte, L., Ramos, M.H. and Borga, M. (2016). Energy companies as key stakeholders. Chapter x of complex report
François, B., Hingray, B., Borga, M., Zoccatelli, D., Creutin, J-D., Brown, C.: Linking top-down and bottom-up approaches for assessing the energy penetration of a 100 % solar and run-of-the river power system in Northern Italy. In preparation.
Raynaud, D., B. Hingray, B. François, 2016: Long term occurrence of low renewable electricity production periods in Europe. In preparation.
Renard, B., J-P. Vidal, 2016. Performance weighting of GCMs. Part 1: A method based on explicit probabilistic models and accounting for observation uncertainty. In preparation.
Renard, B., J-P. Vidal, 2016. Performance weighting of GCMs. Part 2: Daily atmospheric variability across Europe. In preparation.
François, B., Borga, , B., Zoccatelli, D., 2016: Assessing small hydro/solar power complementarity in ungauged, mountainous areas: a crash test study for hydrological prediction methods. In preparation.
François, B., Raynaud, D., Hingray, B., Creutin, J.-D. Influence of winter NAO pattern on variable renewable energies potential in Europe over the 20th century. In preparation
Bhattacharyya, S., M. Intartaglia, and A. McKay, 2016. “Does Climate Aid Affect Emissions? Evidence from a Global Dataset,” World Development, [Revise and Resubmit].
Belete, G.F. Voinov, A , Bulavskaya, T, Niamir, L, Dhavala, K, Arto, I, Moghayer, S, and Filatova, T, (in review). Web service based approach to linking heterogeneous climate-energy-economy models for climate change mitigation analysis. International Journal of Energy.
Belete, G.F. Voinov, A., Morales, J. (in review). Designing the Distributed Model Integration Framework – DMIF. Environmental Modelling and Software.
Engeland, K., M. Borga, J.-D. Creutin,, B., François, M. H. Ramos, J.-P. Vidal 2016: Space-time variability of climate and intermittent renewable electricity production - a review. Renewable and Sustainable Energy Reviews, submitted.
3.7 Journal Special Issues Edited by COMPLEX personnel
Although we have made extensive use of pre-existing publication opportunities, there were occasions when COMPLEX approached journal editors with plans for special journal issues. COMPLEX personnel edited four special issues in the life of the project:
1. A special issue on using simulation models to study systemic shocks / regime shifts in coupled social-ecological systems: Polhill, J.G. T.Filatova M.Schlüter, A.Voinov (2016) ‘Modelling systemic change in coupled socio-environmental systems’, Environmental Modelling & Software (IF=4.4207) 75, p. 318–332, http://www.sciencedirect.com/science/article/pii/S136481521530075X
2. A special issue on the current state of the art in the field of spatial agent-based models of social-ecological systems: Filatova T., P.H. Verburg, D.C. Parker, C.A. Stannard (2013).
3. Voinov, A., Kolagani, N., McCall, M. 2016. Modelling with stakeholders – Next generation. Virtual Thematic Issue of Environmental Modelling & Software.
4. “Spatial agent-based models for socio-ecological systems: challenges and prospects”, Environmental Modelling & Software (IF=4.4207) Volume 45, p. 1-7 http://www.sciencedirect.com/science/article/pii/S1364815213000807

3.8 Career and Professional Development
Our work with students and early career researchers, the people who will be the project’s ambassadors over the coming decades, has been an important feature of our dissemination actions. The project co-ordinator, for example, has averaged six or seven student seminars and workshops a year, the most recent in North Wales in the final month of the project. We have produced a book and teaching materials about the behavioural ecology of project-based science and made it available free for download.
We have also supported a large group of students and early career researchers and had the pleasure of seeing staff working at all levels in the project advance and consolidate their careers,
Theses and dissertations
1. Hewitt, R. (2014). Integrating Stakeholder Knowledge in Cellular Automata Models of Land Use Change. Unpublished Ph.D. thesis, Universidad de Alcalá. Available at: https://www.researchgate.net/profile/Richard_Hewitt3/publication/281651759_Integrating_stakeholder_knowledge_in_Cellular_Automata_models_of_land_use_change/links/55f3229208ae63926cf2324f.pdf
2. Pera, F.A. (2016) Tendencias en la relación de la agricultura intensiva en Europa con las regiones potencialmente vulnerables al cambio climático. Master's Thesis (using APoLUS model) Departamento de Geología, Geografía y Medio Ambiente, Universidad de Alcalá.
3. Raynaud, D. (2016). Hydroclimatic Variability and the Integration of Renewable Energy in Europe. Ph.D. thesis, Université Grenoble Alpes. Available at: ftp://ftp.lthe.fr/pub/raynaud/
4. L. Niamir “Agent-based Energy Market: modeling non-marginal changes” Ph.D
5. M.B.Tariku (2014) “Household energy demand in the Netherlands: application of an agent based model to assess the potential of carbon emission reduction” M.SC.
6. Mülder (2016) ‘Understanding the impact of barriers on energy saving decisions of households’. B.Sc
7. Belete, G.F. (Submitted). Integrating models on the web: application for socio-environmental studies. Unpublished PhD thesis. University of Twente.

3.8.1 Dissertations in Preparation:
Doctoral Thesis of Torun Hammar planned for May 2017.
Licentiate Thesis of Huayi Lin planned for May 2017
Licentiate Thesis of Azadeh Hassannejad Nazir planned for June 2017
MSc thesis in Energy Systems Engineering by Björn Isaksson, Modelling of a Fossil Fuel Free Energy System in Uppsala 2050, in collaboration between SLU, Uppsala kommun and technical consultancy Semcon, using and further developing the model built by Complex.
3.8.2 Evolution of team members’ careers and exit strategies
1. V. Hernandez-Jimenez - Appointed independent evaluator by Joint Research Centre for FP7 Project BeWater
2. P. Martinez Alonso – Successfully passed civil service exams for position of State Agronomist
3. J.D. Pacheco – Appointed to University of La Laguna, Tenerife, Chair on Natural Risks and Hazards
4. R. Hewitt- Took up new post of Spatial Landscape Planner, James Hutton Institute, Aberdeen
5. C. de Boer - Took up new post of Assistant Professor, ITC
6. B. François – Took up a Post-Doctorate at the University of Massachusetts Amherst
7. I.C. Winder - Took up a post as Lecturer in Biology at the University of Bangor
8. A. Laurie Schwarz – Appointed Teaching Fellow at Department of Economics, University of Sussex, 2016.
9. T Filatova - Promoted to Associate Professor at University of Twente
10. G.F. Belete – Completed his PhD thesis.
11. A.A.Voinov - Inaugurated as full professor at University of Twente
12. Stephan Barthel has moved from post doc to post in industry and Stockholm University and, in 2016 was appointed senior researcher/Professor Gävle Högskola and senior researcher Stockholm Resilience Center/Stockholm University)
13. Sebastiaan Meyer moved from Associate Professor KTH/Transport Science to Full Professor KTH/health systems by 1st January 2016
14. Cecilia Sundberg (postdoc) was appointed Associate Professor at KTH Royal Institute of Technology in October 2015
15. Sara Borgström (postdoc) became Associated professor KTH from 1st December 2016
16. Thereza Webster (Project Operations Manager) completed a Master’s degree at Newcastle University Business School through part-time study and was appointed Project Manager of COHERE (H2020)
17. Dmitry Kovalevsky has been appointed wissenschaftliche Mitarbeiter at the Climate Service Center, Germany (GERICS), where he will work on a Horizon 2020 project (IMPREX) that seeks to improve the prediction and management of hydrological extremes.
3.8.3 Awards for Team Members
François, B. - ’EGU2016 : Early Career Scientist’s Travel Award’ for ’The inﬂuence of the North-Atlantic Oscillation on variable renewable energy penetration rate in Europe’, Vol. 18, EGU2016-482, 2016, session ERE3.1 : Energy Meteorology.
2013 Filatova T. De Winter Prize for paper: “Spatial agent-based models for socio-ecological systems: challenges and prospects”
3.9 International Workshops and Summer-Schools Convened by COMPLEX
1. Water and Society – an interdisciplinary summer school held in Oléron, France in May, 2014 (N. Winder was the director of this second edition of the school)
2. Workshop on Multi-model integration, 29-30 September 2014, IIASA, Laxenburg, Austria
3. Summer School on Economic Growth and Governance of Natural Resources (MSA2015), 20 July – 1 August 2015, Lomonosov Moscow State University, Moscow, Russia
4. Workshop on Multi-model integration, 13-14 June 2016, IIASA, Laxenburg, Austria

3.10 Retrospective and Prospects
COMPLEX has been very active in dissemination and exploitation. Our work with external stakeholders has created lasting networks of engagement and opportunities for career advancement that will allow us to capitalise them. We have matched our research efforts with a substantial investment in public engagement, scientific dissemination and education. WP6, our integration WP, ran into delays in Year 3 and these delays had knock-on consequences for year 4. We are happy and proud to report that all these obstacles have been removed. In accordance with the DoW, all our project foreground has been delivered to external stakeholders, and is accessible through our website. We are confident that the project will have a strong and valuable legacy in the European Research Area.

List of Websites:
http://owsgip.itc.utwente.nl/projects/complex/

complex@ncl.ac.uk

Final Report Summary - COMPLEX (Knowledge Based Climate Mitigation Systems for a Low Carbon Economy)

Compartir esta página

Descargar