The Role of Host and Habitat Spatial Heterogeneity in the Distribution of Ticks and Tick-borne Diseases on Scottish Upland Moorland.

Abstract

The problems associated with ticks and tick-borne diseases cause significant impacts on both human and wildlife populations throughout the world. In the UK, the future of Scottish red grouse shooting, an activity of both cultural and economic importance, is threatened by the apparent increase in ticks and louping-ill virus. The cessation of moorland management for red grouse production would result in the loss, not only of the iconic heather moorland landscape, but also of a habitat important for the conservation of other species such as mountain hares.
Mathematical models of disease transmission provide important tools for the improvement of disease mitigation strategies. However, most disease transmission models assume spatial homogeneity, which in the case of tick-borne diseases, is unlikely to be true. In order to advance our understanding of tick-borne disease transmission, it is necessary to investigate the spatial relationships between ticks, tick hosts, and the habitat. This thesis examines how host (red deer, roe deer, sheep, red grouse and mountain hare) and habitat heterogeneity affects the distribution of ticks on Scottish moorland and investigates the prevalence of two important tick-borne diseases, louping-ill and Lyme disease, in questing, field-collected ticks. The data is used to test the predictions of an influential model of louping-ill virus.
Both hosts and habitat were found to be important predictors of Ixodes ricinus nymph abundance, though there was some site-to-site variation in the precise pattern exhibited. In general, the highest nymph numbers were collected from areas of heather habitat, and lowest numbers from areas of boggy ground. Nymph numbers increased with increasing red deer density, but were negatively associated with both increasing mountain hare and red grouse density.
Polymerase chain reaction techniques were used to detect louping-ill virus and Borrelia burgdorferi sensu lato in individual, field-collected, questing nymphs and adults. A total of 1,063 ticks were tested for louping-ill virus, of which 621 were also tested for Borrelia burgdorferi s.l.. Louping-ill virus RNA was detected in 7.8 % of all ticks tested. The percentage of ticks positive for louping-ill virus did not vary with tick stage, collection site, month, or vegetation type, and was not correlated with host density or host seroprevalence of louping-ill virus. B. burgdorferi was detected in l.4% of the ticks. No simultaneous infections of louping-ill virus and B. burgdorferi were detected.
Field data on tick abundance and pathogen prevalence collected during this study was used to challenge an influential model of louping-ill virus dynamics. The model proved relatively successful at predicting tick abundance at the six study sites, but underestimated the extent of louping-ill virus. The original model was adapted to form a two-patch model allowing red deer movement between the two patches. The results of the two-patch model suggest that red deer movement could account for some of the discrepancies found between the one-patch model predictions and the empirical data. Allowing the movement of red deer between two patches with different host densities shifted the threshold for louping-ill virus persistence, allowing its persistence in almost all red deer-red grouse-mountain hare scenarios.