The aerosphere as a network connector of organisms and their diseases

Abstract

Aeroecological processes, especially powered flight of animals, can rapidly connect biological communities across the globe. This can have profound consequences for evolutionary diversification, energy and nutrient transfers, and the spread of infectious diseases. The latter is of particular consequence for human populations, since migratory birds are known to host diseases which have a history of transmission into domestic poultry or even jumping to human hosts. In this chapter, we present a scenario under which a highly pathogenic avian influenza (HPAI) strain enters North America from East Asia via post-molting waterfowl migration. We use an agent-based model (ABM) to simulate the movement and disease transmission among 10 6 generalized waterfowl agents originating from ten molting locations in eastern Siberia, with the HPAI seeded in only ~10 2 agents at one of these locations. Our ABM tracked the disease dynamics across a very large grid of sites as well as individual agents, allowing us to examine the spatiotemporal patterns of change in virulence of the HPAI infection as well as waterfowl host susceptibility to the disease. We concurrently simulated a 12-station disease monitoring network in the northwest USA and Canada in order to assess the potential efficacy of these sites to detect and confirm the arrival of HPAI. Our findings indicated that HPAI spread was initially facilitated but eventually subdued by the migration of host agents. Yet, during the 90-day simulation, selective pressures appeared to have distilled the HPAI strain to its most virulent form (i.e., through natural selection), which was counterbalanced by the host susceptibility being conversely reduced (i.e., through genetic predisposition and acquired immunity). The monitoring network demonstrated wide variation in the utility of sites; some were clearly better at providing early warnings of HPAI arrival, while sites further from the disease origin exposed the selective dynamics which slowed the spread of the disease albeit with the result of passing highly virulent strains into southern wintering locales (where human impacts are more likely). Though the ABM presented had generalized waterfowl migration and HPAI disease dynamics, this exercise demonstrates the power of such simulations to examine the extremely large and complex processes which comprise aeroecology. We offer insights into how such models could be further parameterized to represent HPAI transmission risks as well as how ABMs could be applied to other aeroecological questions pertaining to individual-based connectivity.