The use of modelling in an empirical policy context often divides stakeholders into technically trained groups and stakeholders less acquainted with the technicalities of modelling. On presentation of a new model, feedback from the technically trained cohort is often about technical details (e.g., assumption, data input, algorithms) while regularly implying competitive arguments regarding the chosen method. Responses from non-technical stakeholders are mainly focused on aspects of trust and validation, rarely defining the means of accepted validation methods (Fagiolo et al., 2007; Smajgl et al., 2011). Transdisciplinary modelling faces more challenges (Jakeman and Letcher, 2003; Smajgl, 2018). The transdisciplinary approach divides the technical cohort even further and is consistently challenged by disciplinary mandates. Considering that complexity in transdisciplinary models increases substantially if compared to disciplinary models, validation becomes even more problematic, which increases the difficulty to effectively support decision making (Smajgl, 2010). 

Despite these challenges, it has become essential for sustainability to develop transdisciplinary models as traditional disciplinary modelling neglect trade-offs between policy-relevant indicators (Lang et al., 2012). In the context of the UN Sustainable Development Goals (SDGs), scientists have highlighted the potential for such trade-offs between SDGs, particular between SDG 2 (food security), SDGs 6 and 14 (water security), and SDG 7 (energy security), often referred to as the water, food, and energy Nexus (Smajgl et al., 2016). Consequently, based on exasperating experiences, development partners aim to think more holistically, which required an integration of trade-offs and synergies. 

Clearly, such holistic thinking introduces substantially higher levels of complexity, typically fusing assessments of previously isolated system components; in the development context mainly physical, ecological, agricultural, social, and economic processes. New methods needed to be developed (e.g. agent-based modelling) to allow the scientific modelling of complex processes and provide reliable assessments (Matthews et al., 2007; Smajgl and Barreteau, 2013). However, the more complex the modelling, the more challenging validation. Transdisciplinary modelling moves also away from the numerically predictive presentation of results, emphasizes uncertainties and focuses on impact patterns, scenario comparisons, or directional results (David, 2009; Smajgl et al., 2011). 

A foundational activity for using complex, transdisciplinary modelling and still being able to effectively influence policy and planning is the implementation of a participatory process, which allows technical and non-technical stakeholders to avoid a ‘black box’ situation and understand modelling assumptions (Smajgl and Ward, 2013a). Without successful stakeholder engagement processes, most transdisciplinary modelling remains without any traction. This is particularly relevant in situations where high complexity meets competing sector interests, in which policy debates rarely lead to the most sustainable solution by design but instead to the solution that serves the most influential sector(Smajgl et al., 2015; Smajgl and Ward, 2013a). For decision support systems to influence such contested policy and planning it demands the research side to articulate trade-offs effectively and the policy-science interface to establish a high level of trust.  

In a transboundary context, transdisciplinary modelling is particularly difficult to achieve policy impact as many trade-offs occur across borders (Smajgl et al., 2015; Smajgl and Ward, 2013b). This might involve hydrological losses downstream (e.g. due to water diversion upstream) but socio-economic losses upstream (e.g. due to migration), or vice versa (e.g. groundwater context). Aforementioned sector interests are therefore combined with national interests, further complicating the policy dialogue. Establishing a productive policy-science interface that results in mutual trust is highly challenging. Yet, without the combination of transdisciplinary modelling, trade-offs are likely to remain unrecognized and likely to render many development interventions meaningless. Hence, the combination of context-tailored transdisciplinary modelling and participatory stakeholder engagement processes define the most promising approach for achieving in transboundary contexts more sustainable natural resource management solutions. But this requires disciplinary specialists and non-technical stakeholders to embrace innovative engagement processes and modelling techniques. 

References 

David, N., 2009. Validation and Verification in Social Simulation: Patterns and Clarification of Terminology. Lecture Notes in Artificial Intelligence 5466, 117–129.

Fagiolo, G., Moneta, A., Windrum, P., 2007. A Critical Guide to Empirical Validation of Agent-Based Models in Economics: Methodologies, Procedures, and Open Problems. Computational Economics 30, 195–226.

Jakeman, A.J., Letcher, R.A., 2003. Integrated Assessment and Modelling: features, principles, and examples for catchment management. Environmental Modelling & Software 18, 491–501.

Lang, D., Wiek, A., Bergmann, M., Stauffacher, M., Martens, P., Moll, P., Swilling, M., Thomas, C., 2012. Transdisciplinary research in sustainability science: practice, principles, and challenges. Sustainability Science 7, 25–43.

Lang, D., Wiek, A., Bergmann, M., Stauffacher, M., Martens, P., Moll, P., Swilling, M., Thomas, C., 2012. Transdisciplinary research in sustainability science: practice, principles, and challenges. Sustainability Science 7, 25–43.

Matthews, R., Gilbert, N., Roach, A., Polhill, J., Gotts, N., 2007. Agent-based land-use models: a review of applications. Landscape Ecology 22, 1447–1459.

Smajgl, A., 2018. Participatory Processes and Integrated Modelling Supporting Nexus Implementations, in: Hülsmann, S., Ardakanian, R. (Eds.), Managing Water, Soil and Waste Resources to Achieve Sustainable Development Goals: Monitoring and Implementation of Integrated Resources Management. Springer International Publishing, Cham, pp. 71–92.

Smajgl, A., 2010. Challenging beliefs through multi-level participatory modelling in Indonesia. Environmental Modelling and Software 25, 1470–1476.

Smajgl, A., Barreteau, O., 2013. Empiricism and Agent-based modelling, in: Smajgl, A., Barreteau, O. (Eds.), The Characterisation & Parameterisation of Empirical Agent-Based Models, Empirical Agent-Based Modelling – Challenges and Solutions. Springer, New York, pp. 5–27.

Smajgl, A., House, A., Butler, J., 2011. Implications of ecological data constraints for integrated policy and livelihoods modelling: an example from East Kalimantan, Indonesia. Ecological Modelling 222, 888–896.

Smajgl, A., Ward, J., 2013a. A framework for bridging Science and Decision making. Futures 52, 52–58.

Smajgl, A., Ward, J., Foran, T., Dore, J., Larson, S., 2015. Visions, beliefs and transformation: Exploring cross-sector and trans-boundary dynamics in the wider Mekong region. Ecology and Society 20, 15.

Smajgl, A., Ward, J., Pluschke, L., 2016. The Water-Food-Energy Nexus – Realising a New Paradigm. Journal of Hydrology 533, 533–540.

_{“The postings on this site are my own and do not necessarily represent FAO’s views, positions, strategies or opinions.”}