Learning across disciplines in socio-environmental problem framing

Modelling complex socio-environmental problems requires integration of knowledge across disparate fields of expertise. A key challenge is understanding how social learning across disciplines occurs in scientific research teams, in order that integrated knowledge is co-created. This article introduces a new framework for training researchers to integrate their knowledge across disciplines, based on current understanding of how inter- and transdisciplinary learning in research teams occurs. The framework was generated from a synthesis of learning, cognitive, and social science theories, and combines facilitated, structured negotiation processes with co-creation of boundary objects. It was used in two, 9 to 10-day intensive training workshops for doctoral students. This article describes the framework, workshop design, analysis of data collected during the workshops related to knowledge integration processes, what has been learned from the results, and the impact on participants. All participants indicated the experience was transformative, provided knowledge and skills unavailable elsewhere, filled gaps in their graduate education programs, and improving confidence in their capacity for inter- and transdisciplinary research. Pre- and post-workshop surveys confirm that the framework changed participants’ knowledge, behaviors, and competencies for engaging across disciplines. Many students have reported they have used the framework in a variety of other research and education settings, indicating they are able to transfer their new competencies to other contexts. Findings contribute to understanding of how to more effectively train researchers to integrate knowledge across disciplines for complex societal problem solving.

An agent-based model to support community forest management and non-timber forest product harvesting in northern Thailand

Agent-based models are popular in common-pool resource management to represent complex systems and stimulate collective action and management, where they are used to evaluate scenarios of stakeholders’ choice in participatory simulations. We developed the “CoComForest” (COllaborative COMmunity FOREST management) model to support community forest management (CFM) and non-timber forest product (NTFP) harvesting in Nan Province, northern Thailand. The model was used as a computer-based role-playing game to support sharing of perceptions and knowledge among stakeholders, and in participatory simulations to explore future CFM scenarios. The Unified Modelling Language was used to build the conceptual model, subsequently implemented under the CORMAS (COmmon-pool Resource and Multi-Agent System) simulation platform. Several tests were conducted in the laboratory for verification and calibration before using this tool with 21 diverse stakeholders during a field workshop. Three different participatory gaming and simulation sessions were organized. The first one focused on the co-validation of the model with participants. They accepted most of the model functionalities and the scheduling of the rounds of play. The model was used in the subsequent two sessions to simulate the scenarios of firebreak establishment and introduction of outsiders intensively harvesting NTFPs, respectively. The results showed that the intensive harvesting practices of outsiders accelerated the depletion of resources, whereas the prevention of wildfire by establishing firebreaks could increase the resource availability in the landscape. The debriefing session at the end of the workshop focused on the analysis of simulation results and the relationships between the players’ decision-making and their actual circumstances. Individual in-depth interviews conducted after the workshop helped to evaluate the use of this model with local stakeholders. Most participants considered the model as a useful common representation of the system they manage collectively. Its use in participatory simulations facilitated communication among the stakeholders searching for an adapted and acceptable collective action plan to improve CFM at the sub-district level in order to prevent the overharvesting of NTFPs by outsiders.

Editorial: SESMO Special Issues

As the journal nears the end of its second official year, we are pleased to start accepting submissions to our first two Special Issues. The first Special Issue is on Resilience of complex coupled Socio-Technical-Environmental systems through the modeling lens with guest editors Tatiana Filatova, Tina Comes (4TU Resilience Engineering Centre), Christoph Hoelscher (ETH Zurich) and Juliet Mian (Resilence Shift). This Special Issue aims to bring together cutting-edge research and international practice to offer insights into the latest scientific modelling methods, gaps, challenges and opportunities and best practice examples relating to operationalising resilience across a range of socio-technical-environmental applications. The second Special Issue is on Large-scale behavioural models of land use change with guest editors Calum Brown (Karlsruhe Institute of Technology), Tatiana Filatova (University of Twente), Birgit Müller (Helmholtz Centre for Environmental Research – UFZ), and Derek Robinson (University of Waterloo). This Special Issue is focussed on better understanding and modelling of temporal or spatial scales in land use dynamics. We invite new proposals for Special Issues that fit within SESMO’s aims and scope. Our Special Issues are cohesive collections of articles focussed on a specific contemporary theme related to socio-environmental systems modelling. The Special Issue can build on previous work and research gaps, but can also explore new and emerging terrain relevant to our aims. Although the conceptualisation of a Special Issue may be initiated in a conference or workshop, it is critical that such a proposal also builds on the original dialogue. Articles should also be canvassed from across the globe. SESMO is an open access journal with no article processing or publication charges for authors. If you have a topic to propose, please contact us to discuss further.

Building trust in SWAT model scenarios through a multi-institutional approach in Uruguay

The inherent complexity of numerical models and the diversity of stakeholders in integrated water resources management (IWRM) create challenges in achieving credibility, salience and legitimacy to develop trust in model-based scenarios. In Uruguay, there has been significant debate on increasing agricultural production while managing agriculture’s environmental impacts (e.g., on water quality and environmental flows). This paper reports on the evolution of a stakeholder process in a case study with a multi-institutional participatory modelling group, supported by researchers. This specific participatory modelling (PM) project is unique in that the active stakeholders are the actual hydrological modellers, and the role of “experts” is mainly in facilitation and capacity building. The results highlight the different bottlenecks and the factors that enabled effective collaboration in this PM project. The main bottlenecks were related to: different views on representation of the watershed, the quality and usability of different input data, the public information for the technical implementation of the model, and the priority of output scenarios. The factors that enhanced collaboration were: a focus on a single basin problem, strong support from upper management, and support from experts in coordination and capacity building. The detailed documentation provided with this project can inspire similar approaches in the future.

Tracing resilience, social dynamics and behavioral change: a review of agent-based flood risk models

Climate change and rapid urbanization exacerbate flood risks worldwide. The recognition of the crucial role that human actors play in altering risks and resilience of flood-prone cities triggers a paradigm shift in climate risks assessments and drives the proliferation of computational models that include societal dynamics. Yet, replacing a representative rational actor dominant in climate policy models with a variety of behaviorally-rich agents that interact, learn, and adapt is not straightforward. Focusing on the costliest climate-exacerbated hazard, flooding, we review computational agent-based models that include behavioral change and societal dynamics. We distinguish between two streams of literature: one stemming from economics & behavioral sciences and another from hydrology. Our findings show that most studies focus on households while representing decisions of other agents (government, insurance, urban developers) simplistically and entirely overlooking firms' choices in the face of risks. The two communities vary in the extent they ground agents' rules in social theories and behavioral data when modeling boundedly-rational decisions. While both aspire to trace feedbacks that agents collectively instigate, they employ different learning and interactions when computing societal dynamics in the face of climate risks. Dynamics of hazard, exposure, and vulnerability components of flood risks driven by incremental adaptation of agents are well represented. We highlight that applying a complex adaptive system perspective to trace the evolution of resilience can lead to a better understanding of transformational adaptation. The methodological advances in computational models with heterogeneous behaviorally-rich adaptive agents are relevant for adaptation to different climate-driven hazards beyond flooding.

Formalising theories of human decision-making for agent-based modelling of social-ecological systems: practical lessons learned and ways forward

Incorporating representations of human decision-making that are based on social science theories into social-ecological models is considered increasingly important – yet choosing and formalising a theory for a particular modelling context remains challenging. Here, we reflect on our experiences of selecting, formalising and documenting psychological and economic theories of human decision-making for inclusion in different agent-based models (ABMs) of natural resource use. We discuss the challenges related to four critical tasks: How to select a theory? How to formalise a theory and how to translate it into code? How to document the formalisation? In this way, we present a systematic overview of the choices researchers face when including theories of human decision-making in their ABMs, reflect on the choices we made in our own modelling projects and provide guidance for those new to the field. Also, we highlight further challenges regarding the parameterisation and analysis of such ABMs and suggest that a systematic overview of how to tackle these challenges contributes to an effective collaboration in interdisciplinary teams addressing socio-ecological dynamics using models.

Fuzzy cognitive mapping of Baltic Archipelago Sea food webs reveals no cliqued views of the system structure between stakeholder groups

Incorporating stakeholder views is a key element in successful environmental management, particularly if the managed system delivers cultural and provisioning ecosystem services directly to the stakeholders, or if there are conflicting views about the ecosystem functioning or its optimal management. One such system is the Archipelago Sea in the Southwestern coast of Finland. It is an area with high biodiversity, offering a range of ecosystem services, from regulating services to provisioning and cultural services. Furthermore, it is subjected to a variety of human activities ranging from eutrophication and marine transport to fishing. The management of the area is also a topic of debate, including discussions of minimum landing size of fish, seal hunting quotas, and the role of cormorants in the ecosystem. Fuzzy cognitive mapping offers a method to evaluate and quantitatively compare different actors’ views on ecosystem structure. The models can be compared quantitatively and simulated to illustrate how they respond to various pressure scenarios. This may reveal differences in the perceptions about what are the important interactions in the ecosystem, and how the system would respond to management measures, potentially explaining differing opinions about the best management strategy. In this work, 30 stakeholders, including policy makers, scientists, eNGOs, fisheries, and recreational users created fuzzy cognitive maps (FCMs) of the Archipelago Sea food web. We found that despite the debate about the management of the area, the stakeholders’ views about the food web structure were not clustered based on the stakeholder group, i.e. the different stakeholder groups did not have distinct ideas about the ecosystem structure. The FCM complexity did not show a pattern based on the stakeholder group either. While the general pattern of the FCMs indicated a shared view of the food web structure across most respondents, there was one map from the recreational group that stood out. The exact setup of the models varied. Across all maps, cod, perch, fishing, zooplankton, and herring were the variables having most links with the other variables. The simulated ecosystem responses indicated that fishing was seen as a key factor affecting food web components, while increases of salinity and oxygen levels have a positive impact on multiple ecosystem components. The value of the approach is to enable a two-way discussion about the food webs and how management of pressures may impact the components.

A bricolage-style exploratory scenario analysis to manage uncertainty in socio-environmental systems modeling: investigating integrated water management options

Exploratory analysis, while useful in assessing the implications of model assumptions under large uncertainty, is considered at best a semi-structured activity. There is no algorithmic way for performing exploratory analysis and the existing canonical techniques have their own limitations. To overcome this, we advocate a bricolage-style exploratory scenario analysis, which can be crafted by pragmatically and strategically combining different methods and practices. Our argument is illustrated using a case study in integrated water management in the Murray-Darling Basin, Australia. Scenario ensembles are generated to investigate potential policy innovations, climate and crop market conditions, as well as the effects of uncertainties in model components and parameters. Visualizations, regression trees and marginal effect analyses are exploited to make sense of the ensemble of scenarios. The analysis includes identifying patterns within a scenario ensemble, by visualizing initial hypotheses that are informed by prior knowledge, as well as by visualizing new hypotheses based on identified influential variables. Context-specific relationships are explored by analyzing which values of drivers and management options influence outcomes. Synthesis is achieved by identifying context-specific solutions to consider as part of policy design. The process of analysis is cast as a process of finding patterns and formulating questions within the ensemble of scenarios that merit further examination, allowing end-users to make the decision as to what underlying assumptions should be accepted, and whether uncertainties have been sufficiently explored. This approach is especially advantageous when the precise intentions of management are still subject to deliberations. By describing the reasoning and steps behind a bricolage-style exploratory analysis, we hope to raise awareness of the value of sharing this kind of (common but not often documented) analysis process, and motivate further work to improve sharing of know-how about bricolage in practice.

Combining social network analysis and agent-based modelling to explore dynamics of human interaction: A review

Agent-based modelling (ABM) and social network analysis (SNA) are both valuable tools for exploring the impact of human interactions on a broad range of social and ecological patterns. Integrating these approaches offers unique opportunities to gain insights into human behaviour that neither the evaluation of social networks nor agent-based models alone can provide. There are many intriguing examples that demonstrate this potential, for instance in epidemiology, marketing or social dynamics. Based on an extensive literature review, we provide an overview on coupling ABM with SNA and evaluating the integrated approach. Building on this, we identify current shortcomings in the combination of the two methods. The greatest room for improvement is found with regard to (i) the consideration of the concept of social integration through networks, (ii) an increased use of the co-evolutionary character of social networks and embedded agents, and (iii) a systematic and quantitative model analysis focusing on the causal relationship between the agents and the network. Furthermore, we highlight the importance of a comprehensive and clearly structured model conceptualization and documentation. We synthesize our findings in guidelines that contain the main aspects to consider when integrating social networks into agent-based models.