RESUMEN
Reed (Phragmites australis (Cav.) Trin. ex Steud.) beds are important habitat for marsh birds, but are declining throughout Europe. Increasing numbers of the native marsh bird, the Greylag goose (Anser anser L.), are hypothesized to cause reed bed decline and inhibit restoration of reed beds, but data are largely lacking. In this study, we experimentally tested the effect of grazing by Greylag geese on the growth and expansion of reed growing in belts along lake shorelines. After 5 years of protecting reed from grazing with exclosures, reed stems were over 4-fold denser and taller than in the grazed plots. Grazing pressure was intense with 50-100% of the stems being grazed among years in the control plots open to grazing. After 5 years of protection we opened half of the exclosures and the geese immediately grazed almost 100% of the reed stems. Whereas this did not affect the reed stem density, the stem height was strongly reduced and similar to permanently grazed reed. The next year geese were actively chased away by management from mid-March to mid-June, which changed the maximum amount of geese from over 2300 to less than 50. As a result, reed stem density and height increased and the reed belt had recovered over the full 6 m length of the experimental plots. Lastly, we introduced reed plants in an adjacent lake where no reed was growing and geese did visit this area. After two years, the density of the planted reed was six to nine-fold higher and significantly taller in exclosures compared to control plots where geese had access to the reed plants. We conclude that there is a conservation dilemma regarding how to preserve and restore reed belts in the presence of high densities of Greylag geese as conservation of both reed belts and high goose numbers seems infeasible. We suggest that there are three possible solutions for this dilemma: (1) effects of the geese can be mediated by goose population management, (2) the robustness of the reed marshes can be increased, and (3) at the landscape level, spatial planning can be used to configure landscapes with large reed bed reserves surrounded by unmown, unfertilized meadows.
RESUMEN
The extent to which animal migrations shape parasite transmission networks is critically dependent on a migrant's ability to tolerate infection and migrate successfully. Yet, sub-lethal effects of parasites can be intensified through periods of increased physiological stress. Long-distance migrants may, therefore, be especially susceptible to negative effects of parasitic infection. Although a handful of studies have investigated the short-term, transmission-relevant behaviors of wild birds infected with low-pathogenic avian influenza viruses (LPAIV), the ecological consequences of LPAIV for the hosts themselves remain largely unknown. Here, we assessed the potential effects of naturally-acquired LPAIV infections in Bewick's swans, a long-distance migratory species that experiences relatively low incidence of LPAIV infection during early winter. We monitored both foraging and movement behavior in the winter of infection, as well as subsequent breeding behavior and inter-annual resighting probability over 3 years. Incorporating data on infection history we hypothesized that any effects would be most apparent in naïve individuals experiencing their first LPAIV infection. Indeed, significant effects of infection were only seen in birds that were infected but lacked antibodies indicative of prior infection. Swans that were infected but had survived a previous infection were indistinguishable from uninfected birds in each of the ecological performance metrics. Despite showing reduced foraging rates, individuals in the naïve-infected category had similar accumulated body stores to re-infected and uninfected individuals prior to departure on spring migration, possibly as a result of having higher scaled mass at the time of infection. And yet individuals in the naïve-infected category were unlikely to be resighted 1 year after infection, with 6 out of 7 individuals that never resighted again compared to 20 out of 63 uninfected individuals and 5 out of 12 individuals in the re-infected category. Collectively, our findings indicate that acute and superficially harmless infection with LPAIV may have indirect effects on individual performance and recruitment in migratory Bewick's swans. Our results also highlight the potential for infection history to play an important role in shaping ecological constraints throughout the annual cycle.