Modeling cascading drought effects in a national food supply network using minimum cost flow

Droughts increasingly threaten food security, yet assessments often overlook how local agricultural losses cascade through complex supply chains. The authors developed a spatially explicit network optimization model employing a minimum cost flow algorithm to simulate food distribution within a synthetic national-scale Swedish food supply network (a model network constructed from open geospatial data) of more than 830,000 nodes. To test the system’s response to extreme hydro-climatic stress, they applied a severe drought scenario analogous to the conditions experienced in 2018. The model revealed a significant spatial disconnect: while agricultural yield losses were concentrated in southern farming regions, unmet consumer demand emerged predominantly in northern municipalities. Furthermore, network analysis showed that flow reconfiguration under drought stress created new critical bottlenecks, effectively shifting the system’s overall vulnerabilities. While these specific geographic results are illustrative of the framework’s diagnostic capability rather than empirically verified operational failures, they demonstrate that drought impacts propagate nonlinearly through supply networks. This underscores the critical importance of considering network topology in food security assessments. Ultimately, that approach offers a scalable framework for identifying hidden vulnerabilities and informing resilience strategies in national food supply infrastructures.