Speaker: Viola Priesmann, Max Planck Institute for Dynamics and Self-Organization & Georg-August-University Göttingen

Date | Time: March 20, 2026 | from 14:00 to 14:50

Place: Room 3 – Building Hangar II.

Abstract: Many living systems, from virus spread in societies to information spread in neural networks, are characterized by stochastic spreading of discrete events on complex networks. This spread then does not occur on a static network, but on an adaptive one, where the spreading proper influences the network’s coupling strength. This feedback loop between spreading activity and coupling strength can generate either stabilize and optimize information flow, or it can generate catastrophic resonance effects. We will investigate how these different phases emerge, and how they shape disease spread and information flow in complex networks.

##########################################################

Title: Near-Threshold Stochastic Dynamics and Arbovirus Emergence in Non-Endemic Regions

Speaker: Maíra Aguiar, Basque Center for Applied Mathematics (BCAM)

Date | Time: March 20, 2026 | from 14:50 to 15:40

Place: Room 3 – Building Hangar II

Abstract: Arboviruses such as dengue and chikungunya are increasingly reported in temperate and non-endemic regions, driven by climate variability, global mobility, and the expansion of competent mosquito vectors. Most risk assessments, however, rely on deterministic metrics like the basic reproduction number (R₀), assuming sustained transmission is unlikely when R₀ < 1. Under this framework, areas with low mosquito abundance, short transmission windows, and sporadic viral importation are considered low-risk.

We show that this approach can underestimate outbreak potential. Many non-endemic settings operate near the transmission threshold, where stochastic effects dominate and rare introductions can trigger substantial outbreaks despite subcritical average conditions. Using the 2024 dengue outbreak in Fano, Italy, we demonstrate that stochastic transmission models reproduce observed outbreak timing and magnitude, whereas deterministic models predict rapid extinction.
This mechanism extends beyond specific case studies: near-threshold stochasticity can generate outbreak patterns resembling supercritical dynamics wherever competent vectors and episodic viral introductions occur. Recognizing these dynamics is crucial for improving epidemic intelligence and public health preparedness. Integrating stochastic models with high-resolution mosquito surveillance, climate data, and human mobility enables more realistic outbreak risk assessments, informing early warning systems and targeted interventions in emerging epidemic settings.

More information is available here.