# Perpetuation of Avian Influenza from Molt to Fall Migration in Wild Swan Geese (Anser cygnoides): An Agent-Based Modeling Approach

## TL;DR

This study uses a computer model to show how swan geese spread avian influenza during their molting stage, highlighting the importance of protecting their habitats.

## Contribution

The study introduces an agent-based model integrating field data to explore HPAI perpetuation during the molting stage of wild swan geese.

## Key findings

- Individual movements between molting wetlands and transmission rates are key predictors of HPAI perpetuation.
- Wetland extent significantly affects the number of infections and deaths but not HPAI perpetuation itself.
- Conserving undisturbed habitats during molting could reduce HPAI transmission risk.

## Abstract

Wild waterfowl are considered to be the reservoir of avian influenza, but their distinct annual life cycle stages and their contribution to disease dynamics are not well understood. Studies of the highly pathogenic avian influenza (HPAI) virus have primarily focused on wintering grounds, where human and poultry densities are high year-round, compared with breeding grounds, where migratory waterfowl are more isolated. Few if any studies of avian influenza have focused on the molting stage where wild waterfowl congregate in a few selected wetlands and undergo the simultaneous molt of wing and tail feathers during a vulnerable flightless period. The molting stage may be one of the most important periods for the perpetuation of the disease in waterfowl, since during this stage, immunologically naïve young birds and adults freely intermix prior to the fall migration. Our study incorporated empirical data from virological field samplings and markings of Swan Geese (Anser cygnoides) on their breeding grounds in Mongolia in an integrated agent-based model (ABM) that included susceptible–exposed–infectious–recovered (SEIR) states. Our ABM results provided unique insights and indicated that individual movements between different molting wetlands and the transmission rate were the key predictors of HPAI perpetuation. While wetland extent was not a significant predictor of HPAI perpetuation, it had a large effect on the number of infections and associated death toll. Our results indicate that conserving undisturbed habitats for wild waterfowl during the molting stage of the breeding season could reduce the risk of HPAI transmission.