Modelling in Support of the Sustainable Development Goals
A model is a simplified version of reality, yet our sustainability challenges are complex. With increasing complexity comes greater uncertainty. So, what is the role of modelling to support sustainable development? In this workshop, different approaches to modelling were presented, and issues of uncertainty, data, and communication were discussed among participants.
The Sustainable Development Goals (SDGs) consist of 17 goals that provide a plan to address some of society's greatest challenges. As they are, the goals are abstract, high-order, and universal. Therefore, translation is required across various contexts at different scales. Likewise, modellers translate complex systems at different scales into simplified versions of reality.
The aim of the workshop was to discuss the role of modelling to support sustainable development. Taking place on Thursday 3 March 2022, presenters and participants joined online to share experiences and learn from each other, across different disciplines, applications, and approaches to modelling for sustainability, for example, global climate modelling, integrated assessment modelling, migration modelling, and life cycle assessment.
The workshop is one of a series of workshops, arranged by the Strategic Research Areas (SRAs) and Sustainability Forum at Lund University, aimed at creating a venue to share experiences, discuss collaboration, and foster deeper reflection on sustainability.
Global Climate Modelling
Global climate modelling helps to understand the scale and intensity of climate impacts across time and space. The climate system includes the atmosphere, the hydrosphere (e.g. oceans), the cryosphere (e.g. snow, ice), the lithosphere (e.g. land surface), the biosphere (e.g. plants, animals), and the interactions between them. Modelling such a complex system requires massive amounts of data, very precise mathematical equations, and immense computing power. Erik Kjellström, Professor of Climatology at SMHI, provided a crash course. In essence, Kjellström suggests global climate modelling divides the Earth’s atmosphere, oceans, and land surface into an interconnected patchwork of cubes. Within each cube, mathematical equations describe our understanding of the dynamics within and between these ‘spheres’ in order to say something about our past, present, or future climate.
Integrated Assessment Modelling
Integrated assessment modelling seeks to combine various models that describe social, economic, and environmental systems. Francesco Gardumi, Researcher at KTH Climate Action Centre, presented the Climate, Land-Use, Energy, Water Systems (CLEWs) Model. This integrated model assesses the interlinkages between primary resources, energy systems, conversion processes, and the delivery of final services in order to understand how the production and use of food, energy, and water affect climate change, and vice versa.
While there is great potential to describe a broader system using integrated assessment modelling, challenges exist in what Gardumi called “squaring a circle”. Integrated assessment modelling sees the combining of data input and modelling methods, with different temporal and spatial resolution, as well as assumptions underpinning each model. Importantly, our social, economic, and technical systems influence this data in different ways, with varying or incomplete knowledge of the mechanisms involved. Therefore, integrated assessment modelling requires immense effort to combine data at various resolutions and to reconcile assumptions across various models.
Orderly, safe, and responsible migration and mobility of people is among the targets of Goal 10 – Reduced Inequality. Migration also has clear synergies with other goals, for example, no poverty (Goal 1), sustainable cities and communities (Goal 11), as well as peace and justice (Goal 16). Lina Eklund, Researcher at the Department of Physical Geography and Ecosystem Science, suggests migration modelling may predict changes in localised population as well as predict future migration flows linked to climate change. Eklund suggests human migration has generalisable effects on the physical environment, which can be measured and modelled to assess migration patterns based on environmental indicators. For example, satellite data can track land abandonment, forest regeneration, or reduced light emissions to quantify the number of people leaving a region, and new construction, land-use change, and increased light emissions to quantify the number of people arriving to a region.
Life Cycle Assessment
Life cycle assessment (LCA) is a modelling method to assess the environmental impact of products or services offered in the economy across their entire life cycle, for example, extraction of raw materials, manufacturing, transportation, use, and end-of-life. The aim is to identify opportunities to improve environmental performance, inform decision-makers, and communicate impacts among industry and citizens. A series of international standards (ISO 14040, 14044) dictate the methods used to ensure rigour and standardisation. Nelly Garcia Gonzalez, Researcher at the Department of Environmental and Energy Systems Studies, presented the process of conducting an LCA. Gonzalez highlighted the importance of defining your system boundaries, collecting primary and secondary data, and interpreting results.
Data Quality, Uncertainty, and Communication
Across the different approaches to modelling, there are common experiences, limitations, and challenges, and these were discussed among workshop participants. Data availability and data quality are long standing challenges for modellers, where incomplete or flawed data used as input into a model produces nonsense output, a concept called garbage in, garbage out (GIGO). Therefore, quality data often includes metadata, which describes the who, what, when, where, why, and how the data was generated. Of course, data must first be available, relevant to the system being modelled, and at a resolution that allows for meaningful analysis.
Complexity and uncertainty are increasingly taken up by modellers. For example, the systems that researchers are wishing to model have many interrelated components, with feedback loops and rebound effects. As such, it is increasingly the case that modelling tools are coupled, where the output(s) of one model are used as input(s) to another. Coupled modelling leads to a more complex modelling system, but with greater uncertainty and potentially unexpected results. Therefore, greater effort and skill is required to interpret modelling results to make meaningful recommendations and avoid negative rebound effects.
Finally, participants highlighted again and again the responsibility and necessity for modellers to communicate – communicate metadata, communicate assumptions, communicate uncertainty. Three specific suggestions related to communication were made to improve the ability of models to support sustainable development.
- Dialogue to support integrated assessment modelling – Participants agree that greater communication is needed among modellers and owners of datasets in order to overcome some of the limitations associated with this modelling approach.
- Involve users of model output – There is potential to involve the user of any model output throughout the process of data acquisition, modelling, and interpretation. Participants suggest that this ensures the goal and scope of the model meets the needs of the user, as well as supports their ability to interpret the results.
- Include modelling curriculum in higher education – With modelling applications increasing, it was suggested to integrate modelling more in the curriculum to support students’ abilities to understand and interpret modelling data and results.
Modelling has a clear role to support sustainable development, with diverse modelling approaches and applications. Most importantly, the workshop highlights opportunities for future-oriented methodologies, research proposals, and collaboration.
This workshop was an initiative by the Strategic Research Areas (SRAs) and Sustainability Forum at Lund University, co-organised by MERGE, SPI, and MECW.