Scale decisions and good practices in socio-environmental systems modelling: Guidance and documentation during problem scoping and model formulation

Wang, Hsiao-Hsuan et al. Article — Published Version Scale decisions and good practices in socio-environmental systems modelling: Guidance and documentation during problem scoping and model formulation Socio-Environmental Systems Modelling Provided in Cooperation with: Leibniz Institute of Agricultural Development in Transition Economies (IAMO), Halle (Saale) Suggested Citation: Wang, Hsiao-Hsuan et al. (2023) : Scale decisions and good practices in socioenvironmental systems modelling: Guidance and documentation during problem scoping and model formulation, Socio-Environmental Systems Modelling, ISSN 2663-3027, International Environmental Modelling and Software Society, Lugano-Viganello, Vol. 5, pp. --, https://doi.org/10.18174/sesmo.18563 This Version is available at: https://hdl.handle.net/10419/270642 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ Correspondence: Contact H.-H. Wang at hsiaohsua[email protected] Cite this article as: Wang, H.-H., van Voorn, G., Grant, W.E., Zare, F., Giupponi, C., Steinmann, P., Müller, B., Elsawah, S., van Delden, H., Athanasiadis, I.N., Sun, Z., Jager, W., Little, J.C., & Jakeman, A.J. Scale decisions and good practices in socio-environmental systems modelling: guidance and documentation during problem scoping and model formulation Socio-Environmental Systems Modelling, vol. 5, 18563, 2023, doi:10.18174/sesmo.18563 This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. Socio-Environmental Systems Modelling An Open-Access Scholarly Journal http://www.sesmo.org Scale decisions and good practices in socio-environmental systems modelling: guidance and documentation during problem scoping and model formulation Hsiao-Hsuan Wang1*, George van Voorn2, William E. Grant1, Fateme Zare3, Carlo Giupponi4, Patrick Steinmann2, Birgit Müller5,6, Sondoss Elsawah3, Hedwig van Delden7,8, Ioannis N. Athanasiadis9, Zhanli Sun10, Wander Jager11, John C. Little12, and Anthony J. Jakeman13 1 Ecological Systems Laboratory, Department of Ecology and Conservation Biology, Texas A&M University, USA 2 Biometris, Wageningen University & Research, The Netherlands 3 Capability Systems Centre, School of Engineering and Information Technology, University of New South Wales, Canberra, Australia 4 Department of Economics, Ca’ Foscari University of Venice and Fondazione Eni Enrico Mattei, Italy 5 Department of Ecological Modelling, Helmholtz Centre for Environmental Research - UFZ, Leipzig, Germany 6 Chair of Modelling of Human-Environment-Systems, Brandenburg University of Cottbus-Senftenberg, Cottbus, Germany. 7 Research Institute for Knowledge Systems, Maastricht, The Netherlands 8 School of Civil, Environmental and Mining Engineering, University of Adelaide, Adelaide, Australia 9 Data Competence Center, Wageningen University and Research, Wageningen, The Netherlands 10 Leibniz Institute of Agricultural Development in Transition Economies (IAMO), Halle (Saale), Germany 11 Groningen Center for Social Complexity Studies, University of Groningen, Groningen, The Netherlands 12 Department of Civil and Environmental Engineering, Virginia Tech, USA 13 Institute for Water Futures, Fenner School of Environment and Society, Australian National University, Canberra, Australia Abstract Models of socio-environmental or social-ecological systems (SES) commonly address problems requiring interdisciplinary scientific expertise and input from a heterogeneous group of stakeholders. In SES modelling multiple interactions occur on different scales among various phenomena. These scale phenomena include the technical, such as system variables, process detail, inputs and outputs, which most often require spatial, temporal, thematic and organisational choices. From a good practice and project efficiency perspective, the problem scoping and conceptual model formulation phase of modelling is the one to address well from the outset. During this phase, intense and substantive discussions should arise regarding appropriate scales at which to represent the different phenomena. Although the details of these discussions influence the path of model development, they are seldom documented and as a result often forgotten. We draw upon personal experience with existing protocols and communications in recent literature to propose preliminary guidelines for documenting these early discussions about the scale(s) of the studied phenomena. Our guidelines aim to aid modelling group members in building and capturing the richness of their rationale for scoping and scale decisions. The resulting transcripts are intended to promote transparency of modelling decisions and provide essential support for the justification of the final model for its intended use. They also facilitate adaptive H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 2 modifications of the pathway of model development via retracing decisions and iterative reflection upon alternative scale options. Keywords Organisational scale; spatial scale; temporal scale; thematic scale; uncertainty 1. Introduction Models are indispensable tools in the analysis and any subsequent decision making related to socioenvironmental or social-ecological systems (SES) problems and issues. They help describe, represent and analyse complex human-environment interactions for the system of concern, and thereby assist in understanding the system and potential ramifications of specific management decision options and uncontrollable system drivers. Treatment of scale is a crucial consideration in such integrated modelling, along with the identification of issues of concern, management options and governance arrangements, models, and sources and types of uncertainty (Hamilton et al., 2015). Much recent literature on modelling SES has focused on difficulties associated with identifying the appropriate scales at which to represent different system structures and processes, as well as different types of model output (Elsawah et al., 2020; Iwanaga et al., 2021a; Lippe et al., 2019). 1.1 The wide nature of scales Scale decisions in SES modelling include not only those associated with defining temporal and spatial scales per se, but also those associated with defining thematic and organisational scales, especially in the social dimension (e.g., individual, community, and/or national level) (Elsawah et al., 2020; Gibson et al., 2000; Müller et al., 2013; van Delden et al., 2011). Here, the thematic resolution refers to the granular component of scale, regarding the thematic detail at which categorical maps of system components are made. The higher the thematic detail or resolution, the larger the amount of information we have and the more complex the analysis will be (GarcíaÁlvarez et al., 2019). Difficulties in identifying which components of the real system to include in the integrated model, and at what level of detail to represent them (intrinsic scale), are also related directly to issues of scale (Cumming et al., 2006; Gotts et al., 2019; Iwanaga et al., 2022). The focus of this article is on the decision choices with respect to ‘model scale’, including not only spatial and temporal scales but also ‘observational scale’ (scale of measurement and sampling) and ‘policy scale’ (the scale for policy making) (for more details see Section 3.2.). More concretely, we follow Gibson et al. (2000) and Iwanaga et al. (2021a) in using the term “scale” in an expansive sense to refer to the spatial, temporal, thematic and organisational dimensions used to cover the scope of work conducted during the treatment and representation of system processes. In particular, organisational scales, although no less intuitive than temporal, spatial or thematic scales, are usually difficult to define precisely. Even though we agree that ontology and epistemology insights would be essential to reach a more solid and consistent scale type, we provide brief definitions in Box 1 of the various aspects of scale as we intend for them to be understood within the context of the present paper. Box 1: Brief definitions of the various aspects of scale as we intend for them to be understood within the context of the present paper. Scale encompasses the spatial, temporal, thematic and organisational dimensions used to describe system processes represented in the model (Gibson et al., 2000; Iwanaga et al., 2021a). Model scale, in addition to spatial, temporal, thematic, and organisational scales, encompasses observational scale, policy scale and intrinsic scale (Wu & Li, 2006). Thematic scale (= thematic resolution) refers to the amount of detail with which thematic categories are defined (GarcíaÁlvarez et al., 2019). Organisational scale refers to aggregates of individuals in different social networks (e.g., individual, community and/or national level) (Elsawah et al., 2020; Gibson et al., 2000; Müller et al., 2013; van Delden et al., 2011). Observational scale refers to the scale of measurement and sampling (Wu & Li, 2006). Policy scale refers to the scale relevant for policy making (van Delden et al., 2011). Intrinsic scale (= process scale) refers to the scale at which a process operates (Cumming et al., 2006; Gotts et al., 2019; Iwanaga et al., 2022) . H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 3 Organisational scales are often associated with human components, like aggregates of individuals in different informal and formal social networks, for example, households, jobs, farmer networks, internet fora and organisations. While some networks are considered to be “scale-free”, there are no universal units of measure when referring to human components of a system. Moreover, there usually are many interactions between agents, other units, and their environment. Hence, preferences by modellers and stakeholders on organisational scales can easily differ. For example, a model may be at the individual level or consider a larger unit (e.g., a village). An important aspect to consider is the scale of decision making by human components, which is equally challenging to specify. For instance, it is commonly assumed that different types of decisions made by agents are on the same timescale, which is not necessarily valid (Wernz & Deshmukh, 2012). 1.2 Why emphasise the communication and documentation of scale decisions? Here, we emphasise the need for improving communication in modelling exercises or projects as a first step in resolving scale and associated issues in models. Disagreements during problem scoping and conceptual model formulation of SES models are both inevitable and desirable. Without such disagreements, assumptions might be accepted tacitly without sufficient thought given to justifying their acceptance. Thus, disagreements promote transparency, given they are openly communicated within the involved group of stakeholders. Productive debates are more likely when participants agree to clarify their own expectations regarding the outcomes of the model, to allow each other space to disagree and provide alternative viewpoints, and also have a clear understanding of end-user expectations. A useful lens here for such discussions is the principle of fitness-forpurpose. In contending that fitness-for-purpose must go beyond the functional use of a model to include its management, problem and project contexts, it has recently been honed and illustrated in the terms of model usability, reliability and feasibility, along with providing considerations and potential evaluation criteria for those terms (Hamilton et al., 2022). Detailed explanations supporting decisions that define the scope and objectives of the model - such as (1) confining or extending the scope of questions the model will address, (2) including or excluding the explicit representation of particular system components and processes, and (3) using coarser or finer resolutions to represent constituent parts of the model - reveal the richness of thought embodied in this initial stage of model development (Jakeman et al., 2006). The inability to communicate clearly how the modelling process led to the model structure and subsequently to the model results undermines model credibility and limits model usefulness (Grimm et al., 2020). Lack of documentation during problem scoping and conceptual model formulation also prevents modelling group members from recalling their rationale for choosing the current, as opposed to an alternative, pathway of model development (Zare et al., 2020). And the adequacy of the documentation process will very likely affect the comprehensiveness and seriousness of the range of scale choices considered. Hence, we also argue here that communication, in turn, can be greatly improved by better documentation of the modelling process. When properly documented, the content of ensuing debates can provide both guidance and justification for the pathway of model development (Zare et al., 2020). Awareness of the need to improve communication is growing and has resulted in the development of a variety of protocols and guidelines that are applicable to SES modelling (Ayllón et al., 2021; Badham et al., 2019; Grimm et al., 2014; Grimm et al., 2006; Grimm et al., 2010; Jakeman et al., 2006; Müller et al., 2013; van Voorn et al., 2016). All prescribe formal descriptions of each step in the modelling process (e.g., statement of model objectives, conceptual model formulation, quantitative model specification, model evaluation, and model application). Although different authors divide the modelling process into more or fewer steps and name the steps differently, the description of the process itself is consistent among the various schemes. Most schemes focus on documenting the details of a single model development path leading from a statement of objectives to the application of the model; few explicitly consider the need for discussing potential alternative paths that have not been taken in the modelling process. Even though resource limitations can inhibit the exploration of alternative pathways of model development, recently there has been an increasing recognition of the importance of documenting the discussions at each important decision point during model development, in other words, the discussions which led to choosing a particular path forward over others (Ayllón et al., 2021; Grimm et al., 2014; Grimm et al., 2020; Zare et al., 2020). Grimm et al. (2014) introduced the idea of keeping a modelling notebook, analogous to a lab journal in H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 4 laboratory research, which documents key assumptions and decisions along the path of model development to be traced back if needed. Ayllón et al. (2021) further elaborated on the concept of a modelling notebook by suggesting that it should also document the outcomes of relevant discussions among modelling group members and how these outcomes influenced the future directions of the modelling project. Zare et al. (2020) suggested using a pathway diagramming approach to document more formally the alternative pathways of model development that are discussed and to facilitate iterative reflection upon them. Such diagrams would indicate pathways that led to dead ends, both conceptually and in practice, and alternative potentially successful pathways not taken, as well as the actual pathway of model development. Zare et al. (2020) also proposed a set of symbols representing key pathway diagramming components, such as decision forks, selected options, other possible options, documentation points, and reflection points. 1.3 Why focus on the problem scoping and conceptual model formulation phases? In this paper, we draw upon the above-mentioned literature and personal experience with existing protocols to propose improvements that facilitate documentation of, and reflection on, the discussions that direct the path of model development. But our main focus is on the discussions of scale issues by modelling group members and with stakeholders during problem scoping and conceptual model formulation, as this is an essential phase in SES model development to address well from the outset, notwithstanding that elements of it may be revisited in later phases. Problem scoping and conceptual model formulation provide the foundation for all subsequent model development, and virtually all choices encountered during problem scoping and conceptual model formulation involve issues of scale. The decisions made in that phase will transfer to support the selection of the integrated model type (Kelly et al., 2013) and its computational model implementation, and enable the potential application of subsequent methods that may sharpen choices further. Quantitative methods that enable discrimination among technical scale choices principally include assessing the identifiability of the computational model such as via sensitivity and uncertainty analyses (Guillaume et al., 2017; Guillaume et al., 2019; Razavi et al., 2021; Saltelli et al., 2021). We hypothesise that more formal guidance for documenting the flow of these early discussions of scale will aid modelling group members in capturing the richness of their rationale for scoping and scale decisions, and thus help resolve scale issues. The additional time and money required to document these discussions are likely negligible compared to issues encountered later in the modelling process when scale issues have not been resolved, and the resulting transcripts will provide essential support for the justification of the final model for its intended purpose. The transcribing process will also facilitate adaptive modifications of the pathway of model development via iterative reflection upon the alternative scale options discussed. 1.4 Outline of the paper In the following sections, we first justify the nature and importance of identifying scale issues that may arise during problem scoping and conceptual model formulation (Section 2), and then present a framework of considerations for the associated documentation, a structured process for thinking about and validating scale decisions, and guidance on communication issues (Section 3). We also point readers toward recent review papers. Next, we describe our proposed guidelines (Section 4), relating each step in the guidelines to a particular scale issue. We then offer suggestions for future work on scale concepts and issues in SES and related communities around developing a common understanding, building capacity and training, and widespread application to continually refine the guidelines (Section 5). 2. Scale issues arising during problem scoping and model formulation Scale issues inevitably arise during the initial phases of the modelling process, and it is during these initial phases that we confront the most fundamental choices affecting model development. Virtually all choices encountered during problem scoping and conceptual model formulation (e.g., specifying model purpose, bounding the model, representing processes connecting model components, and specifying model inputs and outputs) involve issues of scale. Problem scoping and conceptual model formulation are widely acknowledged and emphasised as crucial rungs upon which subsequent quantitative model development steps depend (Grimm et al., 2014; Jakeman et al., H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 5 2006; Refsgaard & Henriksen, 2004). The tenet that model objectives should guide all phases of the modelling process has long been accepted, regardless of the research field (Ford, 1999; Forrester, 1961; Grant, 1986; Patten, 1971; Schmolke et al., 2010). Perhaps less widely emphasised is the importance of detailed consideration of model purpose, especially for socio-environmental models. As Edmonds et al. (2019) point out, the purpose of a socio-environmental model may be to predict, explain, describe, theoretically explore, illustrate, serve as an analogy, or facilitate social interaction. Problem scoping and conceptual model formulation ideally are conducted within the context of a well-defined model purpose. For modelling that addresses problems requiring interdisciplinary scientific expertise and input from a heterogeneous group of stakeholders, identifying appropriate model objectives is particularly challenging (Elsawah et al., 2020; Iwanaga et al., 2021b; Jakeman et al., 2006). Most commonly a general statement of project objectives from a funding source is what brings modelling groups together in the first place. However, converting these general objectives into specific objectives that a model project or exercise actually can accomplish to serve a specifically stated purpose, whilst taking into account timeframe, various resource and other constraints, is often a difficult task. Furthermore, the specific objectives of the modelling, as initially stated, often warrant modification as alternative pathways of model development present themselves (Zare et al., 2020). Groups involved in modelling socio-environmental systems (SES) generally include scientists from the physical, biological and social sciences, as well as a multi-vocational assemblage of stakeholders from different professions, occupations and pursuits who view the modelling project from different perspectives (Voinov & Bousquet, 2010; Wang & Grant, 2021). A diversity of perspectives on the problem at hand is a strength of interdisciplinary, multi-vocational groups, but can also be a potential source of strife. Difficulties arise when group members underestimate the overhead of interdisciplinary communication. As scientists, we are largely ill-trained to communicate our perspectives in a way that allows the development of a shared understanding of project objectives among stakeholders, who typically have different values and concerns, and suggests how project objectives might be achieved. The typical research management framework is also ill-suited to facilitate the interdisciplinary communication required to address integrated modelling challenges. Arguably, these difficulties arise in large part because of a lack of transparency in discussions about scale decisions (Nabavi et al., 2017), frequently leading to unproductive, time-consuming arguments among group members, which can affect the trustworthiness and fitness-for-purpose of the final model. Such arguments during the problem scoping and conceptual model formulation phases often are exacerbated by a limited budget and impending deadlines, and limited data availability, as well as the expediency of already existing models (‘models on the shelf’) that may, possibly falsely, seem suitable for the current problem. Thus, a lack of transparency in group discussions can lead to scale choices made without sufficient awareness or reflection upon their impact on subsequent model development (Zare et al., 2020). 3. Disentangling issues around scales and communication about scales The literature on scale issues arising during SES modelling is becoming voluminous, as is the literature on the involvement of stakeholders in the modelling process. Thus, for meeting best practice, it has been established what are some of the major scale decisions that should be documented (spatial, temporal, process detail, thematic, organisational) and for whom they should be documented (modellers, stakeholders, and/or the general public) (Figure 1). As stressed in Section 1, scale issues arise in the early stages of modelling, and so communication on scale issues should take place during those stages of an SES modelling exercise. Modellers include those scientists who are involved in model development and their peers who potentially will review the model. Stakeholders include “non-scientist” stakeholders with an interest in the model and its output (e.g., natural resource managers, policy makers, risk assessors, investors), as well as “final impact” stakeholders who would be affected directly by decisions based on model output (e.g., farmers/fishers, developers, manufacturers, local residents) (Cartwright et al., 2016). Accordingly, the content of the “message” embodied in the conceptual model and style in which that message is transmitted to stakeholders, particularly to “final impact” stakeholders, should be tailored to the decision-making context (see literature review on stakeholder participation for environmental management by Reed (2008)). H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 6 Figure 1: A general framework for the documentation of scale decisions during problem scoping and conceptual model formulation, indicating who should communicate what to whom. Adapted from Cartwright et al. (2016). The content and style of documentation supporting effective communication of scale decisions, not only among those directly involved in the modelling process but also with the various parties interested in the model and its output, are arguably not well understood by modellers. The way scale decisions are documented should include consideration of both the interdisciplinarity of peers who will evaluate the scientific credibility of the model, and the level of experience of stakeholders who will assess the usefulness of model output within a particular sociocultural context. In Figure 2 we have organised important scale issues that arise during problem scoping and conceptual model formulation using an Ishikawa fishbone diagram (Ishikawa, 1990). Below we will discuss each “bone” in detail. Our intent is not to provide a literature review, but rather to organise frequently encountered aspects of scale decisions within a framework that facilitates communication of the general set of guidelines that we will suggest in Section 4. Figure 2. Important scale issues that often arise during SES problem scoping and conceptual model formulation (brown arrows), organised using an Ishikawa fishbone diagram (Ishikawa, 1990). Also shown are proposed guidelines associated with each issue (green arrows, discussed in Section 4). H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 7 3.1 Identifying suitable scales 3.1.1 Define scale concepts clearly Scale and associated issues can originate from many sources, and these sources need to be identified and addressed before scales can be clearly defined. We list here issues arising from: • Ambiguity in use and meaning of the word ‘scale’ and words associated with it. For example, the word ‘scale’ mathematically can refer to a series of marks at regular intervals on a line used for measuring or a rule for determining the distances between these marks. Geographically, ‘scale’ refers to the ratio of size to indicate proportional dimensions when shrinking objects to represent them on a map. Three attributes of scale are extent (the ‘width’ of what is possible when measuring), support (the size of the discretised intervals of measuring), and coverage (the ratio between the intervals and the extent) (Bierkens et al., 2000; Gibson et al., 2000). In everyday language, ‘scale’ often refers to the size of something, e.g., “the large scale of the COVID-19 pandemic”. (Note that ‘resolution’ refers to the level of detail at which an entity or process is represented.) • Ontological differences. It is critical to clarify the ontology (Gotts et al., 2019). For example, when talking about ‘farms’, scale could concern the spatial boundaries of farms (measured in spatial units), the judicial farm units (possibly expressed in monetary units), or the position of farmers in a social network. • Discrepancies among what is desired by the exercise or from a stakeholder perspective, the process understanding of the modeller, and the support available through data (van Delden et al., 2011). These do not necessarily overlap, as available data and the scale at which process(es) can be modelled may not provide the support desired by the exercise (similar to balancing model complexity between model requirements and the available support from data) (Wagener et al., 2001). • Differences in the objective functions that modellers and stakeholders may have in mind, for which different scales may be applicable, including different appropriate or preferential scales for different processes and related model components. Any effort to simulate the behaviour of an SES will have to make deliberate decisions on scales. There may not be one ‘perfect’ scale for the SES model under consideration. • Mismatches between the real-life scale of attributes and their discrete classification and/or numerical representation in models (e.g., the classification of soil types and the implementation of layers in soil hydrological models). Interpolation and extrapolation may (partially) solve issues associated with scale mismatches between model and data, but this usually comes at the expense of increased uncertainty about model predictions. • The (implicit) assumption that attributes are scaled homogeneously, whereas they may scale heterogeneously. For example, in climate studies it is not uncommon to select only the dominant soil type and cropping system in a fixed radius around individual weather stations and assume these occupy the whole area, disregarding other soil types and cropping systems in that area that may still play a substantial role (Schils et al., 2022). For some areas this homogeneity may be roughly accurate, while in other areas soil types or cropping systems may be very diverse. A particular feature with modelling SES is that ‘scale’ can encompass multiple aspects. Many ecological studies consider only time and space (Turner et al., 2001). Although many ‘natural’ boundaries are related to time and space because they are defined by dominant biophysical, geological or chemical processes (e.g., in the case of a catchment area for groundwater modelling), identification of the scale of these boundaries may not be trivial. Boundaries may also be the result of underlying processes. For instance, several patterns observed in ecosystems can be explained by smaller-scale mechanisms of resource concentration (Kéfi et al., 2010). Moreover, SES also typically involve ‘artificial’ boundaries. Even though some of these artificial boundaries concern time or space, like country borders or the administrative start or finish of political cycles, many concern other attributes relating to thematic or organisational scales. The choice for organisational scales may be rather straightforward, for instance, many judicial or governmental units enforce a ‘pyramid-like’ structure in which one agent controls multiple agents ‘downstream’. Ecological food webs may follow a similar structure, in particular when functional groups are considered (Allen & Hoekstra, 1994). Often, however, the organisational scales are debatable. For example, scales may result from social processes, such as networking among actors. Thus, agents can have different roles (people are workers, team H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 8 members, and parents, for instance), where each role can be accompanied by different action modes, and agents interact with other agents from the same group with the same role (working together, playing as a team, and being with the family) – the so-called ‘agent-group-role’ paradigm (Brugière et al., 2022). Further entanglement of scales may occur when agents are part of different social and organisational networks. Decisions need to be made about the number and type of actors and the level of differentiation (e.g., number of actor groups, household/farm types, type of irrigation licence, types of technology, types of behaviour), as well as decisions about the thematic resolution in all other parts of the model (e.g., land use classes, crop types, economic sectors). 3.1.2 Towards a structured process to inform thinking about relevant scales Good modelling practices should include explicit discussion and documentation of the choices of scale that were made, why they were made, and by whom. We argue it is best to not aim for agreement about scale at the start, but rather to allow room for discussion and to explore different scale options, including the possibility of having multiple scales. For each different viewpoint provided by stakeholders, at least the following questions/items should be considered: • Are there natural or artificial boundaries to time, space, or otherwise? • Are there agents in the system? What limitations or demands on the scale do they impose? What are the options for agents to act that are relevant for inclusion in the model? What is the best level of detail at which to represent the different (groups of) agents? • Are there any emerging scales, such as spatial patterning or social clustering? • Regarding space, is the best approach a point model, or a one-, two-, or three-dimensional model? • If maps are used, are they grid-based or vector-based? • Are there any transformations used, like (dis)aggregations, interpolation or extrapolation? • What heterogeneities are there, and what is their origin? • What is the extent, support and coverage of the data? And of the model? • What is desired from the application side (this may be stakeholder-specific)? 3.2 Identifying appropriate dimensions for scale-matching Once the sources of scale issues are clarified, and discussions about the desired scale(s) are underway, several new questions must be addressed about the matching of scales. As mentioned above, what is desired as scale and what is supported by data and process knowledge regarding the scale may not overlap. First, what are potential scale-matching methods and what is meant by it? What is being matched to what? Possible answers involve: (1) a match to what the different disciplines involved in the modelling process find important for answering the research question, (2) a match to the resolution and extent of the available data, or (3) a match to the resolution and extent of the envisioned application scale. It is important to realise that selection of model scale commonly involves a trade-off, or tension, between the scales desired for model application and the scales supported by the available data. As mentioned briefly in the introduction, we should further recognise that there are different types of model scales. Such scales are intrinsic scale (the scale at which a process operates), observational scale (the scale of measurement and sampling), and policy scale (the scale for policy making). The relation among these scales is discussed by Wu & Li (2006) and Bierkens et al. (2000) and has been taken into consideration regarding the ‘model scale’ decisions that we focus on here. Sitas et al. (2021) provide a chapter on system scoping in SES research, including a table of methods containing policy scoping and social-ecological inventories, and links to introductory texts describing associated methods. Furthermore, they acknowledge that system scoping is commonly constrained by institutional and biophysical boundaries. Existing conceptual frameworks directed at system scoping include the Intergovernmental Science Policy Platform for Biodiversity and Ecosystem Services (IPBES) framework (Díaz et al., 2015), and Ostrom’s framework for analysing the sustainability of SES (Ostrom, 2009). These conceptual frameworks can help to set the boundaries, conditions and variables for a deeper exploratory or analytical exercise based on addressing specific questions or problems. H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 15 • C2: Re-validate scale decisions if the context changes. Scale choices are assessed within the domain of model applicability (sensu Rykiel, 1996) and are validated under the specified conditions that the model is designed to simulate. • C3: Expect and embrace change in the selection of scale choices as inevitable due to the complexity of the system being studied and the speed at which new information may become available. During problem scoping and conceptual model formulation, it is common to go through several iterations of a set of concurrent steps/activities. The flexibility to formally assess alternative scale choices must be maintained, as this is critical to problem scoping and conceptual model formulation. Scale choices should not be invalidated prematurely, and justifications should be provided for decisions to declare a scale choice (in)valid. Such justifications add to the possibilities of exploring alternative pathways. • C4: Clarify which scale and model validation criteria and approaches are used, and why. Consider the distinction between conceptual, operational and data validity (see Subsection 3.3.2 for suggestions). • C5: Realise that data do not represent an infallible standard in the sense that their scale should not solely dictate choice. Explain the origin of the involved data and how they are used in the scale validation. This especially includes data that have been interpolated or extrapolated. • C6: Try to include both qualitative and quantitative measures of system performance. Project-specific tests are important. Identify questions with a known range of acceptable answers at the scales chosen and scale-dependent patterns of system behaviour that the model is expected to produce. 4.4 Communicating scale decisions Communication and justification are the central issues around scale choices. Enhancing the quality of communication and documentation is crucial for improving the transparency of the modelling process and stakeholder trust in the model. Subsection 4.1 discussed considerations and principles to help identify suitable scales, Subsection 4.2 presented questions to consider in choosing a scale, and Subsection 4.3 dealt with scale validation. This Subsection considers some good communication practices during all three steps and is intended to aid model reproducibility and communication effectiveness and thereby help both modellers and stakeholders. We propose the following guidelines for effective communication between modellers and stakeholders regarding the decision making related to scales: • D1: Get clarity on the decision objective(s) or research question(s). This precision is needed for justification of the scale decisions adopted. A project will seldom be directly aimed at scales, so a ‘translation’ of the societal and research objectives to scale decisions is usually needed. • D2: Analyse the audience of stakeholders for possible inclusion or exclusion in the communication process (see Box 3 below). Be aware of differences in the level of interest in, and understanding of, the role and importance of scale among stakeholders and modellers. Do not consider the ‘scientific view’ to automatically be the right one. Recognise that behavioural traits can affect group interactions and effective communication, and that they may lead to the possibility of procedural mistakes and cognitive biases; for instance, focusing on the time scale and ignoring important organisational scale aspects, or emphasising fast practical model use while overlooking concerns regarding scale validation. Box 3: A list of questions for project leaders and teams to consider in analysing an audience of stakeholders for possible inclusion or exclusion. 1) Who are the decision makers and stakeholders? 2) What are their interests, desires and motivations for participating in project planning? 3) How do audience members understand the system of interest? (i) How do they interact with, or experience, the system? (ii) What features of the system do they pay attention to? (iii) How do they monitor or observe the system? (iv) How do practitioners make sense of changing conditions? (v) What local vocabularies of terms, images and concepts do they use to explain various system aspects and characterise problems within the system? 4) What other driving factors, significant events, and past conflicts with the decision-making context should be considered? 5) What prior experiences have audiences had with models? H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 16 • D3: Pay attention to communication between team members as well as stakeholders. In interdisciplinary research exercises such as occur in SES modelling, communication among team members is as important as communication between stakeholders and the research team. Improved communication among team members enhances the transfer of knowledge within the team and helps to achieve a strong connection among different disciplines, especially regarding critical decisions such as scale. • D4: Organise communication in a suitable way; in the Subsection below, we discuss some approaches that can be followed. Important principles are: 1) Involve sufficient two-way communication between modellers and stakeholders. 2) Consider the earlier-mentioned challenges with communicating scale concepts to nontechnical audiences, including mitigating (i) confusion regarding scale concepts, (ii) conflicts with pre-existing scales in other studies, (iii) the use of inappropriate communication methods, and (iv) the existence of cognitive boundaries and behavioural phenomena that inhibit effective communication. 3) Avoid technical language that is unnecessary or without proper explanation. • D5: Provide concise and comprehensible documentation with respect to the important information regarding decisions and the rationale supporting those decisions. When properly performed and completed, this documentation makes it possible for people from within and outside the project to review the decisions and, if applicable, potentially provide input to alter the decisions. It should contain essential information about the following: 1) The actual decision, why it was made, who made the decision, and how it was made. For example, were dedicated exercises, workshops, or even serious games organised to explore scale issues and possible outcomes of decisions? Was the decision the result of a group discussion, or did a particular person (for example, the project leader or a specific stakeholder) make the final decision? If it were the results of a group discussion, who was involved in the process (Moallemi et al., 2022)? 2) Documented considerations regarding whether we use one or more scales for the study depending on the different phenomena, and why (see A1), what is desired for a particular simulation or project (see A2), and the possible trade-offs between desired scales and support provided by data (see A5). Also include options for scales that were not followed up. 3) A clear description of the joint definition of objectives and system boundaries (see A3) and the used data typologies (see A4). 4) A description of the validation (see the guidelines under C), and a critical reflection of the expected consequences of the decisions regarding scale, including what scale means in the project (e.g., is it temporal or spatial?) (see B1), whether it should be the same for all disciplines or not (see B2), and what the desired temporal, spatial, thematic and organisational scales are for the social and for the environmental view (B3). Consider what could be the result of not selecting particular scale options? The ramifications of certain decisions need not only be in terms of model output and uncertainty; there can also be social ramifications, for example, that the minority who was not in favour of the chosen scale will show a reduced willingness to continue participating in the remainder of the project. • D6: Use proper language. In their communication with stakeholders and fellow scientists, people should opt for the use of non-technical language and means of communication that are broadly supported and traceable. Avoid the use of scientific terms as well as popular ‘slang’ and establish a common ontology to avoid confusion. For example, a ‘matrix’ can be a mathematical object from linear algebra, the physical attribute of the soil, or the landscape over which a species is dispersed. This is confusing, and many stakeholders will not even know what this word means. Technical terms and implicit assumptions must be clarified. • D7: Communication and documentation should be an ongoing, regularly occurring process, with sufficient information about the possible scale options and scale decisions presented in such a way as to be as transparent as possible. • D8: Activities aimed at communication should be planned from the start of the project to ensure incentivised communication. The occurrence of many, short interactions among a few team members H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 17 has been found to improve the transfer of relevant project information between scientists and stakeholders (Podestá et al., 2013). Hence, ample communication moments should be planned and followed. 4.4.1. Methods for communicating about scale decisions Different ways of communication should be used to maximise the probability of getting the information across. Communicators should use simple exercises and tools to reduce confusion about the scale concept, identify the available options, and demonstrate the process. We list some available tools and suggestions for selecting tools: • E1: Using ID cards (Zare et al., 2019). An ID card is an organised collection of explicit statements about each step of the modelling process which can be used to communicate a common story about the scale selection step and a better understanding of the scale selection process between modellers and stakeholders. Such a card shows a quick summary of what has been said about this step across multiple guidelines using a common language and provides information about different aspects of the step such as aim, actors involved, and main decisions. Team members should fill out such cards. • E2: The use of narratives with examples and a scientific storytelling approach has been found to be useful in the communication of complex information to non-specialists (Howarth & Anderson, 2019). This could be, for example, defining the spatial and temporal scales in narratives about the case study in a different time and physical boundary and conveying the differences within the story. • E3: The elucidating explanation of the concept (Hall et al., 2014). This uses an example of the concept, here scale, that lists the concept’s essential features and also offers a list of examples and nonexamples. Therefore, the audience has the opportunity to understand the difference between examples and non-examples by looking for their essential features. This tool is helpful if the concept of scale is mistaken with other concepts such as resolution or if there are several similar options for scale. • E4: The practice of regular reflection on the modelling pathway during a project, particularly with respect to scale choices, can increase transparency and trust among stakeholders and help them reach better decisions as well as promote efficiency and effectiveness of the modelling process (Zare et al., 2021). • E5: Documentations should include the following information, noting that stakeholder reflections on the information should be solicited to avoid one-way communication: 1) the scale decisions that were made, 2) the other scale options that were discussed but not chosen, 3) what each alternative option might mean in terms of possibly affecting project results, costs and next steps in the project, 4) what each group of stakeholders thought about the scale decisions, and 5) the final decisions and selected scales. • E6: Tools for documentation should be more than just deliverables and reports at the final stage of the project. Metadata, summaries and easy-to-understand fact sheets for each document could convey the meaning and aim of each document to people who will not read the details in a final report. • E7: Reflective communication: Reflections could be solicited within a workshop or by sending a list of reflective questions with any document to facilitate engagement and draw attention to the points of concern. Reflection to validate, confirm, or debunk the result of a decision (such as scale selection) should happen frequently, and thereby assist in efficiently reaching objectives by identifying the need for iteration as early as possible (Zare et al., 2020). Steps in the reflective cycle include: 1) describing the event and situation, 2) expressing the sentiments and thinking about them, 3) assessing and analysing the discussion, and finally, 4) concluding and setting an action plan (Gibbs, 1988). 5. Summary and suggestions for future work The choice of scales is an inevitable and critical decision in SES modelling. But the interdisciplinary nature of SES modelling may easily create confusion and frustration when discussing and communicating scale issues during problem scoping and conceptual model formulation. Compromises or sometimes simply convenient choices may H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 18 be made to accommodate appeals from researchers, stakeholders, and policy makers with specific interests in a particular discipline, question, field or scale. Thus, the final scale choice for any phenomenon is not necessarily the optimal or most scientifically solid and justifiable for all envisioned applications. Accordingly, there are significant benefits in developing a better understanding of issues surrounding the selection of scales in SES modelling. This paper is intended to stimulate further discussions and inspire future work surrounding the issue of documenting scale choice in SES modelling. We foresee future work including, but not limited to: • The development of an improved understanding and alignment of scale and associated concepts within the SES and other relevant modelling communities. Despite the fact that many articles focus on the scale issues, confusion regarding scale concepts and lack of understanding of the choice of scale and its implications are still prevalent among modellers and stakeholders. • The development of education and training materials with easy-to-understand (non-technical, discipline-neutral) terminology in order to advance the handling of scale issues and build associated capacity within the scientific community. These are vital to establishing a common understanding of the concept of scale and relevant terminologies between researchers and non-scientific actors (stakeholders, policymakers and citizens) as well as modellers from different disciplines (the natural sciences and social sciences). We suggest that special workshops could be organised on topics related to scale in SES modelling. For example, topics could include the importance of scale issues, the implications of scale choices, the limitations of models, and the ramifications for modelling results following a particular choice of scales. • The application of scale guidelines in SES projects and their subsequent refinement to make the discussions and decisions around scale issues transparent and traceable. The potential benefits can only materialise when suggestions for communication and documentation are adopted and applied in SES modelling practice. In the present work, we have presented a preliminary set of guidelines, which now need to be tested ‘in the field’. Through real case applications, we hope to further improve the guidelines and make them operational. A clear protocol or template illustrated with examples would be extremely helpful to facilitate the usage of the guidelines. Eventually, an automatic protocol like a web-based tool could facilitate the scale decision process with step-by-step questions. In this position paper, we have attempted to identify the potential issues, traps and pitfalls in the process of decision making on scale issues and have presented a set of prescriptive guidelines on scale issues in four categories. We argue that following these guidelines can facilitate discussions between researchers and stakeholders, support the justification of scale choice, and help assess the strengths and limitations of models and their outputs. Acknowledgments This paper builds on discussions from the Lorentz workshop "Participatory and Cross-scale Modelling of Social Ecological System" at Lorentz Centre, Leiden, the Netherlands in June 2022. The authors acknowledge Lorentz Center, Leiden University and NWO for hosting and providing financial support. References Allen, T., & Hoekstra, T. (1994). Toward a definition of sustainability. In Sustainable ecological ecosystems: Implementing an ecological approach to land management, General Technical Report, RM-247 (pp. 98–108). USDA Forest Service. Ayllón, D., Railsback, S. F., Gallagher, C., Augusiak, J., Baveco, H., Berger, U., Charles, S., Martin, R., Focks, A., & Galic, N. (2021). Keeping modelling notebooks with TRACE: Good for you and good for environmental research and management support. Environmental Modelling & Software, 136, 104932. https://doi.org/10.1016/j.envsoft.2020.104932 Badham, J., Elsawah, S., Guillaume, J. H. A., Hamilton, S. H., Hunt, R. J., Jakeman, A. J., Pierce, S. A., Snow, V. O., BabbarSebens, M., Fu, B., Gober, P., Hill, M. C., Iwanaga, T., Loucks, D. P., Merritt, W. S., Peckham, S. D., Richmond, A. K., Zare, F., Ames, D., & Bammer, G. (2019). Effective modeling for Integrated Water Resource Management: A guide to contextual practices by phases and steps and future opportunities. Environmental Modelling & Software, 116, 4056. https://doi.org/10.1016/j.envsoft.2019.02.013 H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 19 Bankes, S. (1993). Exploratory modeling for policy analysis. Operations Research, 41(3), 435-449. https://doi.org/10.1287/opre.41.3.435 Bierkens, M. F. P., Finke, P. A., & de Willigen, P. (2000). Upscaling and downscaling methods for environmental research (Vol. 88). Springer. Brugière, A., Nguyen-Ngoc, D., & Drogoul, A. (2022, 2022-December-01). Handling multiple levels in agent-based models of complex socio-environmental systems: A comprehensive review [Review]. Frontiers in Applied Mathematics and Statistics, 8. https://doi.org/10.3389/fams.2022.1020353 Carrada, G. (2006). Communicating science: Infrastructure, hman resources, international cooperation, research and development, environment and health, societal Issues, industrial innovation - A scientist's survival kit. Office for Official Publications of the European Communities. Cartwright, S. J., Bowgen, K. M., Collop, C., Hyder, K., Nabe-Nielsen, J., Stafford, R., Stillman, R. A., Thorpe, R. B., & Sibly, R. M. (2016). Communicating complex ecological models to non-scientist end users. Ecological Modelling, 338, 51-59. https://doi.org/10.1016/j.ecolmodel.2016.07.012 Cumming, G. S., Cumming, D. H. M., & Redman, C. L. (2006). Scale mismatches in social-ecological systems: Causes, consequences, and solutions. Ecology and Society, 11(1), 14. http://www.jstor.org/stable/26267802 Díaz, S., Demissew, S., Joly, C., Lonsdale, W. M., & Larigauderie, A. (2015). A Rosetta Stone for nature’s benefits to people. PLOS Biology, 13(1), e1002040. https://doi.org/10.1371/journal.pbio.1002040 Edmonds, B., Le Page, C., Bithell, M., Chattoe-Brown, E., Grimm, V., Meyer, R., Montañola-Sales, C., Ormerod, P., Root, H., & Squazzoni, F. (2019). Different modelling purposes. Journal of Artificial Societies and Social Simulation, 22(3), 6. https://doi.org/10.18564/jasss.3993 Elsawah, S., Filatova, T., Jakeman, A. J., Kettner, A. J., Zellner, M. L., Athanasiadis, I. N., Hamilton, S. H., Axtell, R. L., Brown, D. G., Gilligan, J. M., Janssen, M. A., Robinson, D. T., Rozenberg, J., Ullah, I. I. T., & Lade, S. J. (2020). Eight grand challenges in socio-environmental systems modeling. Socio-Environmental Systems Modelling, 2, 16226. https://doi.org/10.18174/sesmo.2020a16226 Elsawah, S., Guillaume, J. H. A., Filatova, T., Rook, J., & Jakeman, A. J. (2015). A methodology for eliciting, representing, and analysing stakeholder knowledge for decision making on complex socio-ecological systems: From cognitive maps to agent-based models. Journal of Environmental Management, 151, 500-516. https://doi.org/10.1016/j.jenvman.2014.11.028 Ford, A. (1999). Modeling the environment: An introduction to systems dynamics modeling of environmental systems. Island Press. Forrester, J. W. (1961). Industrial dynamics. Cambridge: MIT Press. García-Álvarez, D., Lloyd, C. D., Van Delden, H., & Camacho Olmedo, M. T. (2019). Thematic resolution influence in spatial analysis. An application to Land Use Cover Change (LUCC) modelling calibration. Computers, Environment and Urban Systems, 78, 101375. https://doi.org/10.1016/j.compenvurbsys.2019.101375 Gibbs, G. (1988). Learning by doing: A guide to teaching and learning methods. Oxford: Further Education Unit. Gibson, C. C., Ostrom, E., & Ahn, T. K. (2000). The concept of scale and the human dimensions of global change: A survey. Ecological Economics, 32(2), 217-239. https://doi.org/10.1016/S0921-8009(99)00092-0 Glynn, P. D., Voinov, A. A., Shapiro, C. D., & White, P. A. (2017). From data to decisions: Processing information, biases, and beliefs for improved management of natural resources and environments. Earth's Future, 5(4), 356-378. https://doi.org/10.1002/2016EF000487 Gotts, N. M., van Voorn, G. A. K., Polhill, J. G., Jong, E. d., Edmonds, B., Hofstede, G. J., & Meyer, R. (2019). Agent-based modelling of socio-ecological systems: Models, projects and ontologies. Ecological Complexity, 40, 100728. https://doi.org/10.1016/j.ecocom.2018.07.007 Grant, W. E. (1986). Systems analysis and simulation in wildlife and fisheries sciences. New York: Wiley. Grimm, V., & Railsback, S. F. (2012). Pattern-oriented modelling: a ‘multi-scope’ for predictive systems ecology. Philosophical Transactions of the Royal Society B: Biological Sciences, 367(1586), 298-310. https://doi.org/10.1098/rstb.2011.0180 Grimm, V., Augusiak, J., Focks, A., Frank, B. M., Gabsi, F., Johnston, A. S. A., Liu, C., Martin, B. T., Meli, M., Radchuk, V., Thorbek, P., & Railsback, S. F. (2014). Towards better modelling and decision support: Documenting model development, testing, and analysis using TRACE. Ecological Modelling, 280, 129-139. https://doi.org/10.1016/j.ecolmodel.2014.01.018 Grimm, V., Berger, U., Bastiansen, F., Eliassen, S., Ginot, V., Giske, J., Goss-Custard, J., Grand, T., Heinz, S. K., Huse, G., Huth, A., Jepsen, J. U., Jørgensen, C., Mooij, W. M., Müller, B., Pe’er, G., Piou, C., Railsback, S. F., Robbins, A. M., Robbins, M. M., Rossmanith, E., Rüger, N., Strand, E., Souissi, S., Stillman, R. A., Vabø, R., Visser, U., & DeAngelis, D. L. (2006). A standard protocol for describing individual-based and agent-based models. Ecological Modelling, 198(1), 115-126. https://doi.org/10.1016/j.ecolmodel.2006.04.023 Grimm, V., Berger, U., DeAngelis, D. L., Polhill, J. G., Giske, J., & Railsback, S. F. (2010). The ODD protocol: A review and first update. Ecological Modelling, 221(23), 2760-2768. https://doi.org/10.1016/j.ecolmodel.2010.08.019 Grimm, V., Johnston, A. S. A., Thulke, H. H., Forbes, V. E., & Thorbek, P. (2020). Three questions to ask before using model outputs for decision support. Nature Communications, 11(1), 4959. https://doi.org/10.1038/s41467-020-17785-2 Grimm, V., Revilla, E., Berger, U., Jeltsch, F., Mooij, W. M., Railsback, S. F., Thulke, H.-H., Weiner, J., Wiegand, T., & DeAngelis, D. L. (2005). Pattern-oriented modeling of agent-based complex systems: Lessons from ecology. Science, 310(5750), 987-991. https://doi.org/10.1126/science.1116681 H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 20 Guillaume, J. H. A., Helgeson, C., Elsawah, S., Jakeman, A. J., & Kummu, M. (2017). Toward best practice framing of uncertainty in scientific publications: A review of Water Resources Research abstracts. Water Resources Research, 53(8), 6744-6762. https://doi.org/10.1002/2017WR020609 Guillaume, J.H.A., Jakeman, J.D., Marsili-Libelli, S., Asher, M., Brunner, P., Croke, B., Hill, M.C., Jakeman, A.J., Keesman, K.J., Razavi, S., & Stigter, J.D. (2019) Introductory overview of identifiability analysis: A guide to evaluating whether you have the right type of data for your modeling purpose. Environmental Modelling and Software, 119: 418-432. https://doi.org/10.1016/j.envsoft.2019.07.007 Hall, D. M., Lazarus, E. D., & Swannack, T. M. (2014). Strategies for communicating systems models. Environmental Modelling & Software, 55, 70-76. https://doi.org/10.1016/j.envsoft.2014.01.007 Hämäläinen, R. P., Luoma, J., & Saarinen, E. (2013). On the importance of behavioral operational research: The case of understanding and communicating about dynamic systems. European Journal of Operational Research, 228(3), 623634. https://doi.org/10.1016/j.ejor.2013.02.001 Hamilton, S. H., ElSawah, S., Guillaume, J. H. A., Jakeman, A. J., & Pierce, S. A. (2015). Integrated assessment and modelling: Overview and synthesis of salient dimensions. Environmental Modelling & Software, 64, 215-229. https://doi.org/10.1016/j.envsoft.2014.12.005 Hamilton, S. H., Pollino, C. A., Stratford, D. S., Fu, B., & Jakeman, A. J. (2022). Fit-for-purpose environmental modeling: Targeting the intersection of usability, reliability and feasibility. Environmental Modelling & Software, 148, 105278. https://doi.org/10.1016/j.envsoft.2021.105278 Howarth, C., & Anderson, A. (2019). Increasing local salience of climate change: The un-tapped impact of the media-science interface. Environmental Communication, 13(6), 713-722. https://doi.org/10.1080/17524032.2019.1611615 Ishikawa, K. (1990). Introduction to Quality Control (Translator: J.H. Loftus). Productivity Press, New York, USA. Iwanaga, T., Steinmann, P., Sadoddin, A., Robinson, D., Snow, V., Grimm, V., & Wang, H.-H. (2022). Perspectives on confronting issues of scale in systems modeling. Socio-Environmental Systems Modelling, 4, 18156. https://doi.org/10.18174/sesmo.18156 Iwanaga, T., Wang, H.-H., Hamilton, S. H., Grimm, V., Koralewski, T. E., Salado, A., Elsawah, S., Razavi, S., Yang, J., Glynn, P., Badham, J., Voinov, A., Chen, M., Grant, W. E., Peterson, T. R., Frank, K., Shenk, G., Barton, C. M., Jakeman, A. J., & Little, J. C. (2021a). Socio-technical scales in socio-environmental modeling: Managing a system-of-systems modeling approach. Environmental Modelling & Software, 135, 104885. https://doi.org/10.1016/j.envsoft.2020.104885 Iwanaga, T., Wang, H.-H., Koralewski, T., Grant, W., Jakeman, A., & Little, J. (2021b). Toward a complete interdisciplinary treatment of scale: Reflexive lessons from socioenvironmental systems modeling. Elementa: Science of the Anthropocene, 9(1), 00182. https://doi.org/10.1525/elementa.2020.00182 Jakeman, A. J., Letcher, R. A., & Norton, J. P. (2006). Ten iterative steps in development and evaluation of environmental models. Environmental Modelling & Software, 21(5), 602-614. https://doi.org/10.1016/j.envsoft.2006.01.004 Kéfi, S., Eppinga, M. B., de Ruiter, P. C., & Rietkerk, M. (2010). Bistability and regular spatial patterns in arid ecosystems. Theoretical Ecology, 3(4), 257-269. https://doi.org/10.1007/s12080-009-0067-z Kelly, R. A., Jakeman, A. J., Barreteau, O., Borsuk, M. E., ElSawah, S., Hamilton, S. H., Henriksen, H. J., Kuikka, S., Maier, H. R., Rizzoli, A. E., van Delden, H., & Voinov, A. A. (2013). Selecting among five common modelling approaches for integrated environmental assessment and management. Environmental Modelling & Software, 47, 159-181. https://doi.org/10.1016/j.envsoft.2013.05.005 Lippe, M., Bithell, M., Gotts, N., Natalini, D., Barbrook-Johnson, P., Giupponi, C., Hallier, M., Hofstede, G. J., Le Page, C., Matthews, R. B., Schlüter, M., Smith, P., Teglio, A., & Thellmann, K. (2019). Using agent-based modelling to simulate social-ecological systems across scales. GeoInformatica, 23(2), 269-298. https://doi.org/10.1007/s10707-018-003378 Louie, M. A., & Carley, K. M. (2008). Balancing the criticisms: Validating multi-agent models of social systems. Simulation Modelling Practice and Theory, 16(2), 242-256. https://doi.org/10.1016/j.simpat.2007.11.011 McEvoy, N., & Whitcombe, A. (2002). Structured risk analysis. Infrastructure Security, InfraSec 2002, Meinen, B. U., & Robinson, D. T. (2021). Agricultural erosion modelling: Evaluating USLE and WEPP field-scale erosion estimates using UAV time-series data. Environmental Modelling & Software, 137, 104962. https://doi.org/10.1016/j.envsoft.2021.104962 Moallemi, E. A., Zare, F., Hebinck, A., Szetey, K., Molina-Perez, E., Zyngier, R., Hadjikakou, M., Kwakkel, J., Haasnoot, M., & Miller, K. K. (2022). Knowledge co-production for decision-making in human-natural systems under uncertainty. https://doi.org/10.5281/zenodo.7426011 Moallemi, E. A., Zare, F., Reed, P. M., Elsawah, S., Ryan, M. J., & Bryan, B. A. (2020). Structuring and evaluating decision support processes to enhance the robustness of complex human–natural systems. Environmental Modelling & Software, 123, 104551. https://doi.org/10.1016/j.envsoft.2019.104551 Modell, S. (2009). Institutional research on performance measurement and management in the public sector accounting literature: A review and assessment. Financial Accountability & Management, 25(3), 277-303. https://doi.org/10.1111/j.1468-0408.2009.00477.x Müller, B., Bohn, F., Dreßler, G., Groeneveld, J., Klassert, C., Martin, R., Schlüter, M., Schulze, J., Weise, H., & Schwarz, N. (2013). Describing human decisions in agent-based models – ODD + D, an extension of the ODD protocol. Environmental Modelling & Software, 48, 37-48. https://doi.org/10.1016/j.envsoft.2013.06.003 Nabavi, E., Daniell, K. A., & Najafi, H. (2017). Boundary matters: the potential of system dynamics to support sustainability? Journal of Cleaner Production, 140, 312-323. https://doi.org/10.1016/j.jclepro.2016.03.032 H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 21 Ostrom, E. (2009). A general framework for analyzing sustainability of social-ecological systems. Science, 325(5939), 419422. https://doi.org/doi:10.1126/science.1172133 Patten, B. C. (1971). A primer for ecological modeling and simulation with analog and digital computers (Vol. 1). Academic Press. Podestá, G. P., Natenzon, C. E., Hidalgo, C., & Ruiz Toranzo, F. (2013). Interdisciplinary production of knowledge with participation of stakeholders: A case study of a collaborative project on climate variability, human decisions and agricultural ecosystems in the Argentine Pampas. Environmental Science & Policy, 26, 40-48. https://doi.org/10.1016/j.envsci.2012.07.008 Razavi, S., Jakeman, A., Saltelli, A., Prieur, C., Iooss, B., Borgonovo, E., Plischke, E., Lo Piano, S., Iwanaga, T., Becker, W., Tarantola, S., Guillaume, J. H. A., Jakeman, J., Gupta, H., Melillo, N., Rabitti, G., Chabridon, V., Duan, Q., Sun, X., Smith, S., Sheikholeslami, R., Hosseini, N., Asadzadeh, M., Puy, A., Kucherenko, S., & Maier, H. R. (2021). The future of sensitivity analysis: An essential discipline for systems modeling and policy support. Environmental Modelling & Software, 137, 104954. https://doi.org/10.1016/j.envsoft.2020.104954 Reed, M. S. (2008). Stakeholder participation for environmental management: A literature review. Biological Conservation, 141, 2417-2431. https://doi.org/10.1016/j.biocon.2008.07.014 Refsgaard, J. C., & Henriksen, H. J. (2004). Modelling guidelines––terminology and guiding principles. Advances in Water Resources, 27(1), 71-82. https://doi.org/10.1016/j.advwatres.2003.08.006 Robinson, S. (1997, 7-10 December 1997). Simulation model verification and validation: Increasing the users' confidence. Proceedings of the 1997 Winter Simulation Conference, Atlanta, GA. Rykiel, E. J. (1996). Testing ecological models: the meaning of validation. Ecological Modelling, 90(3), 229-244. https://doi.org/10.1016/0304-3800(95)00152-2 Saltelli, A., Jakeman, A., Razavi, S., & Wu, Q. (2021). Sensitivity analysis: A discipline coming of age. Environmental Modelling & Software, 146, 105226. https://doi.org/10.1016/j.envsoft.2021.105226 Sargent, R. (2008). Verification and validation of simulation modelling. Proceedings of the 2008 Winter Simulation Conference, New York, NY. Schils, R. L. M., van Voorn, G. A. K., Grassini, P., & van Ittersum, M. K. (2022, 2022/01/01/). Uncertainty is more than a number or colour: Involving experts in uncertainty assessments of yield gaps. Agricultural Systems, 195, 103311. https://doi.org/10.1016/j.agsy.2021.103311 Schmolke, A., Thorbek, P., DeAngelis, D. L., & Grimm, V. (2010). Ecological models supporting environmental decision making: a strategy for the future. Trends in Ecology & Evolution, 25(8), 479-486. https://doi.org/10.1016/j.tree.2010.05.001 Sitas, N., Ryan, P., & Schultz, L. (2021). Systems scoping. In R. Biggs, A. de Vos, R. Preiser, H. Clements, K. Maciejewski, & M. Schlüter (Eds.), The Routledge Handbook of Research Methods for Social-Ecological Systems (pp. 83-94). Routledge. https://doi.org/10.4324/9781003021339 Trébuil, G., Bousquet, F., Ekasingh, B., Baron, C., & Le Page, C. (2005). A multi-agent model linked to a GIS to explore the relationship between crop diversification and the risk of land degradation in northern Thailand highlands. In F. Bousquet, G. Trébuil, & B. Hardy (Eds.), Companion modeling and multi-agent systems for integrated natural resource management in Asia (pp. 167-190). International Rice Research Institute. Turner, M., Gardner, R., O'neill, R., Gardner, R., & O'Neill, R. (2001). Landscape ecology in theory and practice (Vol. 401). Springer. van Delden, H., van Vliet, J., Rutledge, D. T., & Kirkby, M. J. (2011). Comparison of scale and scaling issues in integrated landuse models for policy support. Agriculture, Ecosystems & Environment, 142(1), 18-28. https://doi.org/10.1016/j.agee.2011.03.005 van Voorn, G. A. K., Verburg, R. W., Kunseler, E. M., Vader, J., & Janssen, P. H. M. (2016). A checklist for model credibility, salience, and legitimacy to improve information transfer in environmental policy assessments. Environmental Modelling & Software, 83, 224-236. https://doi.org/10.1016/j.envsoft.2016.06.003 Voinov, A., & Bousquet, F. (2010). Modelling with stakeholders. Environmental Modelling & Software, 25(11), 1268-1281. https://doi.org/10.1016/j.envsoft.2010.03.007 Wagener, T., Boyle, D. P., Lees, M. J., Wheater, H. S., Gupta, H. V., & Sorooshian, S. (2001). A framework for development and application of hydrological models. Hydrology and Earth System Sciences, 5(1), 13-26. https://doi.org/10.5194/hess-5-13-2001 Wang, H.-H., & Grant, W. E. (2021). Reflections of two systems ecologists on modelling coupled human and natural (socioecological, socio-environmental) systems. Ecological Modelling, 440, 109403. https://doi.org/10.1016/j.ecolmodel.2020.109403 Wernz, C., & Deshmukh, A. (2012). Unifying temporal and organizational scales in multiscale decision-making. European Journal of Operational Research, 223(3), 739-751. https://doi.org/10.1016/j.ejor.2012.06.038 Wu, J., & Li, H. (2006). Concepts of scale and scaling. In J. Wu, K. B. Jones, H. Li, & O. L. Loucks (Eds.), Scaling and uncertainty analysis in ecology (pp. 3-15). Springer. Zare, F., Guillaume, J. H. A., & Jakeman, A. J. (2019). Constructing customized modelling guidelines: A participatory integrated assessment and modelling example. 23rd International Congress on Modelling and Simulation, Canberra, Australia. Zare, F., Guillaume, J. H. A., ElSawah, S., Croke, B., Fu, B., Iwanaga, T., Merritt, W., Partington, D., Ticehurst, J., & Jakeman, A. J. (2021). A formative and self-reflective approach to monitoring and evaluation of interdisciplinary team research: An integrated water resource modelling application in Australia. Journal of Hydrology, 596, 126070. https://doi.org/10.1016/j.jhydrol.2021.126070 H.-H. Wang et al. (2023) Socio-Environmental Systems Modelling, 5, 18563, doi:10.18174/sesmo.18563 22 Zare, F., Guillaume, J. H. A., Jakeman, A. J., & Torabi, O. (2020). Reflective communication to improve problem-solving pathways: Key issues illustrated for an integrated environmental modelling case study. Environmental Modelling & Software, 126, 104645. https://doi.org/10.1016/j.envsoft.2020.104645