Student information

MSc thesis topic: Spatial characterisation of synergies in Socio-Ecological Systems for resilience

Cities are complex and dynamic, shaped by the intricate interactions between human and nature. These interactions form what we refer to as socio-ecological systems (SES)—complex networks where social, economic, and ecological components are interwoven. Understanding the spatial patterns and dynamics within SES is crucial for creating cities that withstand shocks and disturbances while also adapt and improve in the face of adversity—resilience. By examining SES through this lens, we can identify strategies that enhance urban resilience, fostering cities that adapt, learn, and flourish amidst dynamic challenges.

In the study of social-ecological systems (SES), researchers have employed various approaches to explore and characterize synergies—the interactions between social and ecological components. These approaches include correlation analysis, qualitative frameworks, network analysis, and multi-criteria decision analysis (MCDA). While each method offers unique insights, one approach stands out for its relevance to our research: Spatial Bayesian Network (SBN) modelling. SBNs explicitly capture probabilistic relationships between SES components. They allow us to quantify uncertainties and dependencies. For example, if green space decreases, how does it impact air quality, social well-being, and biodiversity? Moreover, SBNs seamlessly integrate spatial data (e.g., land cover, population density) into the network. By considering spatial dependencies, we can explore how changes propagate across the urban landscape. It also facilitate scenario testing. We can simulate disturbances (e.g., urban expansion, extreme weather events) and observe their effects on SES. This informs strategies for enhancing resilience.

In this thesis, we aim to characterize spatial synergies within SES using SBNs, contributing to evidence-based developments.

Objectives and Research questions

Characterize spatial synergies: Investigate how SES components interact spatially. Identify areas where positive synergies prevail (e.g., co-location of green spaces and community centers).
Quantify trade-offs: Assess trade-offs between different SES components. For instance, if investing in housing density improves economic growth, how does it affect green space availability?
Evaluate resilience: Use SBNs to evaluate the resilience of urban systems. How do different SES configurations respond to disturbances? Which nodes contribute most to overall resilience?

Requirements

GIS proficiency: Familiarity with GIS is essential. You’ll work with spatial data layers, perform analyses, and create maps.
Python basics: Basic knowledge of Python is required. Some SBN models and algorithms have already been developed, and you’ll need to adapt and extend them.
Willingness to learn Bayesian Network: While prior knowledge of Bayesian models is beneficial, a willingness to learn and explore Bayesian concepts is essential.

Literature and information

Santillán-Carvantes, P., Balvanera, P., Thomsen, S. et al. Spatial characterization of social-ecological systems units for management in Tropical Dry Forests. Landsc Ecol 38, 4303–4323, 2023.
Ana Stritih, Sven-Erik Rabe, Orencio Robaina, Adrienne Grêt-Regamey, Enrico Celio, An online platform for spatial and iterative modelling with Bayesian Networks, Environmental Modelling & Software, 2020.
Das, M., Ghosh, S.K. (2020). Spatial Bayesian Network. In: Enhanced Bayesian Network Models for Spatial Time Series Prediction. Studies in Computational Intelligence, vol 858. Springer, Cham.

Theme(s): Modelling & visualisation, Human – space interaction