Modelling the evolution of pathogenesis

A reconstruction of the dynamic emergence/re-emergence of pathogenesis in Cyprus in relation to famine breakouts during the CE warmth and earlier periods will help assess the vulnerability of the Cypriot society at different climatic/environmental backgrounds.

Spatio-temporal modeling of socio-environmental dynamics of ancient time periods is still a challenge today, even though the field has been very actively explored by an interdisciplinary scientific community during the last 20 years. Modeling under the double constraints of scarcity of data and knowledge led specialists to favor simple but robust models, derived from both continuous and discrete formalisms. Agent-based modeling – ABM took the lead, as a relevant way to combine formalisms within a common framework and to systematize computerized scenario-based explorations. Since recent pioneering work was achieved, the developed models evolved in complexity, while still being limited both by data availability and induced computational burden. Replicating such models is a challenge and too often a limitation. However, sensitivity analysis protocols have been defined since then and their early application to simple epidemiological models proved to be very useful and successful. In parallel, high performance computing technologies and their wider spread among research communities pushed the boundaries of ABM and their application to socio-environmental issues, allowing experimental designs to be defined and explored. In this perspective, one recent protocol is of particular interest to us, as it allows an incremental building of ABM models, based on a systematic analysis of their behavior but also a precise definition of the interactions at work between processes and also between parameters.

Coupling this protocol with recent developments in spatial epidemic modeling will allow us to develop more versatile and accurate models of epidemic spread in Cyprus, adapted to the expected scarcity of data and knowledge. On the whole, we will exploit the most recent developments in agent-based modeling of socio-environmental processes as baseline methods, technologies and protocols.