Effects of migration network configuration and migration synchrony on infection prevalence in geese

Highlights

• We study the effects of migration network configuration and synchrony on infection prevalence.

• Migration network configuration and synchrony can affect pathogen transmission dynamics.

• Increasing the number of serial stopover sites reduces the infection prevalence.

• Higher migration synchrony reduces the number of cumulative infections.

• We call for studies exploring the role of stopover sites availability in pathogen transmission.

Abstract

Migration can influence dynamics of pathogen-host interactions. However, it is not clearly known how migration pattern, in terms of the configuration of the migration network and the synchrony of migration, affects infection prevalence. We therefore applied a discrete-time SIR model, integrating environmental transmission and migration, to various migration networks, including networks with serial, parallel, or both serial and parallel stopover sites, and with various levels of migration synchrony. We applied the model to the infection of avian influenza virus in a migratory geese population. In a network with only serial stopover sites, increasing the number of stopover sites reduced infection prevalence, because with every new stopover site, the amount of virus in the environment was lower than that in the previous stopover site, thereby reducing the exposure of the migratory population. In a network with parallel stopover sites, both increasing the number and earlier appearance of the stopover sites led to an earlier peak of infection prevalence in the migratory population, because the migratory population is exposed to larger total amount of virus in the environment, speeding-up the infection accumulation. Furthermore, higher migration synchrony reduced the average number of cumulative infection, because the majority of the population can fly to a new stopover site where the amount of virus is still relatively low and has not been increased due to virus shedding of infected birds. Our simulations indicate that a migration pattern with multiple serial stopover sites and with highly synchronized migration reduces the infection prevalence.

Introduction

Many species migrate between their wintering and breeding grounds in response to seasonal changes in habitat conditions, such as food availability (Altizer et al., 2011, Dingle, 2014). Meanwhile, migration can also facilitate pathogen transmission, as migratory animals can disperse pathogens over long distances (Dingle, 2014, Pulgarín-R et al., 2019), or trigger infection outbreaks by exposing the population to pathogens in novel habitats (Lisovski et al., 2018, van Dijk et al., 2015). For example, migration of passerine birds has contributed to the spread of the West Nile Virus across North America (Owen et al., 2006), and the migration of waterfowl has contributed to the global spread of the highly pathogenic avian influenza (HPAI) H5N1 and H5N8 (Si et al., 2009, Xu et al., 2016).

Migration, however, can also reduce infection in a migratory population by so-called migration escape (Loehle, 1995, Satterfield et al., 2015). Migration allows hosts to ‘escape’ from the accumulated pathogens in the habitat. For example, Lesser black-backed gulls (Larus fuscus) with a relatively long migration distance have a lower seroprevalence of avian influenza virus (AIV) compared to those with a relatively short migration distance (Arriero et al., 2015). Previous studies that examined the interactions between bird migration and infection dynamics of pathogens focused on spatial–temporal and phylogenetic correlations between animal movements and infection outbreaks (Bourouiba et al., 2010, Huang et al., 2019, Tian et al., 2015, Xu et al., 2016), whilst other aspects of migration that might affect infection prevalence have not yet been investigated, such as the configuration of the migration network and the synchrony in timing of migration.

Migratory animals, particularly migratory birds, can use stopover sites in a serial configuration, in which all individuals use the same stopover sites successively. The number of these stopover sites varies among species. For example, Sandpiper (Calidris mauri) and Black turnstone (Arenaria melanocephala) use stopover sites more frequently and spend less time to refuel on each site than Dunlin (Calidris alpine), Red knot (C. canutus) or Bar-tailed godwit (Limosa lapponica) (Iverson et al., 1996, Krementz et al., 2011, O'REILLY and Wingfield, 1995). On the other hand, migratory birds, such as Swan geese (Anser cygnoides), Bar-tailed godwits (Limosa lapponica), Brent geese (Branta bernicla) and Greater white-fronted geese (Anser albifrons) can use stopover sites in a parallel configuration (Batbayar et al., 2013, Battley et al., 2012, Green et al., 2002, Kölzsch et al., 2016), in which all individuals split to use multiple stopover sites at the same time. Hence, there are potentially many distinct network configurations with respect to the use of serial and parallel stopover sites. The configuration of a migration network is expected to influence the aggregation of migratory birds and their exposure to pathogens at these stopover sites (Buehler and Piersma, 2008, Rohani et al., 2009). Moreover, the increase in anthropogenic activities has decreased the availability of many stopover sites (Yamaguchi et al., 2008). For example, suitable stopover sites in the East Asian-Australian flyway have experienced a dramatic loss over the past 20 years, especially the stopover sites located in China (Jia et al., 2018, Zhang et al., 2015, Xu et al., 2019), which changes the configuration of migration networks. The effects of stopover sites loss on pathogen infection prevalence in a migratory population are not yet fully understood. To obtain a better understanding of infection dynamics in migratory populations and the spatio-temporal distribution of infection outbreaks, more studies are required that take into account the changes in network configurations of migratory species.

Apart from different configurations of migration network, migration synchrony (i.e., timing of migration) also varies among migratory species, due to e.g., differences in body condition, competition for limited resources, global warming, and optimization of mating opportunities (Morbey and Ydenberg, 2001, Muraoka et al., 2009). For example, a Swan goose population might only take weeks to leave a habitat, whereas a population of Barnacle geese might take months. Previous studies proposed that highly synchronized migration might be associated with high infection prevalence, as larger flocks lead to increased contact probabilities among individuals (Buehler and Piersma, 2008, Gaidet et al., 2011). However, no study has investigated how migration network configuration and migration synchrony affect the infection prevalence in a migratory goose population.

In this study, we applied a time-discrete SIR (susceptible-infected-recovered) model to various scenarios of spring migration to explore how variations in configuration of migration network and synchrony of migration affect infection prevalence. The model and scenarios were applied to infection of a low pathogenic avian influenza (LPAI) virus in migratory goose species, since the outbreaks of AIV caused concerns, but the relationship between goose migration and virus dispersal is not fully understood (Ren et al., 2016, Takekawa et al., 2010, Yin et al., 2017). We aimed at answering the following questions: (1) How does the configuration of a migration network affect infection prevalence? (2) Does highly synchronized timing of migration increase infection prevalence? (3) Is there a specific migration pattern, regarding the number of stopover sites and migration synchrony that minimizes pathogen infection?