No description available.

No description available.

This study examines, from a modelling point of view, the dynamics of infectious diseases in wildlife caused by macroparasites and by tick-borne infections. The overall aim was to investigate the important role played by parasite aggregation in the dynamics of both systems. For macroparasites we first developed some deterministic models that incorporate explicit mechanisms for generating aggregation in parasite distribution, specifically multiple infections and host heterogeneity. We explored the role of aggregation in host regulation and in determining a threshold value for parasite establishment. A large aggregation makes it more difficult for parasites both to regulate hosts, and to get established in a population at carrying capacity. Furthermore, the stabilization yielded by aggregation strongly depends on the mechanism that produces the aggregation. We then introduced some uncertainties into the host-macroparasite system, presenting an individual-based stochastic model that incorporated the same assumptions as the deterministic model. Stochastic simulations, using parameter values based on some real case studies, preserved many features of the deterministic model, like the average value of the variables and the approximate length of the cycles. An important difference is that, even when deterministic models yield damped oscillations, stochastic simulations yield apparently sustained oscillations. The amplitude of such oscillations may be so large as to threaten the parasites’ persistence. With respect to tick-borne diseases we presented a general model framework that incorporated both viraemic and non-viraemic routes of infections. We compute the threshold for disease persistence and study its dependence on the parameters and on host densities. The effects of tick aggregation and correlation between different tick stages on the host have both an important effect on infection persistence, if non-viraemic transmission occurred. In the case of Lyme Disease and Tick-borne Encephalitis (TBE) in Trentino (northern Italy) we showed some numerical results, using parameter estimates based on a detailed field study, and explored the effects of uncertainty on the endemic equilibrium of both diseases assuming only viraemic transmission for Lyme Disease while for TBE we permitted only non-viraemic transmission through co-feeding ticks. In conclusion we have examined the patterns and changes of aggregation in a number of contrasting systems and believe that these studies highlight both the importance of considering heterogeneities in modelling host-parasite interactions and, more specifically, modelling the biological mechanisms that produce aggregation in parasite distributions.

Society requires better insights into how disturbances will alter the global carbon cycle. Ecosystem models help us understand the carbon cycle and make predictions about how the terrestrial land sink will change under future climate regimes. Disturbances drive ecosystem cycling, but modeling disturbances has unique challenges, particularly in incorporating heterogeneity and parameter uncertainty. In this dissertation, I explore two questions. 1) How can we capture disturbance ecology in models?, which I explore in my first and second chapters, and 2) How can we use those models to make projections for the Southeastern US?, which I explore in my third and fourth chapters. Both my first and second chapters point to the practical trade-offs in model structure and realism. In my first chapter, I found that representing spatially implicit contagious disturbances in terms of shape and frequency accurately captured structural changes over time and separated the disturbance regimes of different regions. Representing spatially implicit disturbances in terms of shape and frequency sacrificed the specificity of a space-based approach but may be more computationally efficient. In my second chapter, I developed a framework for calibrating models based on an iterative cycle between uncertainty analysis and literature synthesis, targeted field campaigns, and statistical constraint. I found that targeted field work and statistical constraint reduced parameter uncertainty until structural uncertainty began to dominate. Models that capture disturbance dynamics can help us anticipate effects of global change factors like climate change and invasive species. In my third chapter, I found that elevated temperatures reduce cogongrass biomass, and that cogongrass facilitates pine dominance over oaks in a mixed pine-oak stand. This suggests that cogongrass mediates inter-species competition at an ecosystem scale. Prescribed burns are a management technique used to suppress cogongrass and has an add-on benefit of reducing tick populations. However, climate change may threaten how frequently prescribed fires can be safely deployed. In my fourth chapter, I found that tick populations are most sensitive to leaf litter and humidity, which allows for management strategies as an alternative to prescribed burns.

Louping ill virus (LIV) is a tick borne disease that causes mortality in red grouse, an economically important game bird of British uplands. The aim of this thesis is to extend previously published models of LIV , to consider the potential impact of different management strategies. In addition a new route of infection and the seasonal biology of both grouse and ticks will be explored. Grouse chicks are known to eat ticks as part of their diet in the first three weeks of life which may contribute to virus persistence if chicks consume infected ticks. This novel route of infection is incorporated in to the model which predicts that ingestion increases the range of host densities for which the virus is able to persist. The ingestion of ticks by grouse also reduces the tick population so that for low host densities the ingestion of ticks by grouse reduces the tick population so virus cannot persist. The model is adapted to take account of the seasonal biology of grouse and ticks. Although the temporal predictions of the seasonal models show some differences the addition of seasonality does not alter the model predictions of when LIV is likely to persist at different grouse and deer densities. Consequently seasonality is felt to be unimportant when considering management strategies. The treatment of sheep with acaricide in an attempt to reduce the tick population on a grouse moor is currently being trialled in Scotland. We use a model to predict the likely effect of this strategy at different deer densities. The number of ticks found attached to sheep varies so we consider the effect of tick attachment rates as well as acaricide efficacy. Although we predict that acaricide treated sheep can reduce the tick population and therefore LIV in grouse in some circumstances the treatment is less effective in the presence of deer. Consequently we use a model to make theoretical predictions of the effectiveness of acaricide treated deer as a control strategy for reducing LIV in red grouse. The effect of culling deer on LIV in grouse is also modelled and contrasted with the effect of acaricide use. It is predicted that acaricide treatment of deer could be highly effective, particularly if the deer density is first reduced by culling. Finally we considered the direct treatment of red grouse with acaricide. Female grouse can be given an acaricidal leg band which protects her directly and indirectly protects her chicks as they acquire some acaricide whilst brooding. Trials have suggested this can reduce tick burdens for individuals. We use the model to determine the potential effect that treating individual broods may have on the whole grouse population. The model predictions suggest that unless acaricide efficacy on chicks is high and long lasting treating individual broods is unlikely to reduce LIV in the whole population but will still provide some benefit for the individuals. The effectiveness of treatment is reduced by higher deer densities. The success of the management strategies considered in this thesis appear to be restricted by the presence of deer. It may therefore be that a combination of treatments including the treatment of deer may be of the greatest benefit to the grouse population.

The field of vector-borne disease research uses multidisciplinary approaches to help understand complicated interactions. This dissertation, covers three different aspects of tick-borne disease research which all focus on exploring tick-borne diseases in the non-endemic areas of Denton, County Texas and the state of Texas with a focus on Rickettsia spp. These aspects include tick sampling, testing ticks for the presence of Rickettsia spp., and creating species distribution maps of the Rickettsia spp. Rickettsia amblyommatis and tick species Amblyomma americanum.

Tick populations have been immigrating into northeast Tennessee from east Tennessee, Kentucky, Virginia, and North Carolina. Counties in states bordering northeast Tennessee harbor tick species associated with human illness. Human diseases transmitted by ticks include ehrlichiosis, spotted fever rickettsial group diseases, tularemia, anaplasmosis, babesiosis, Lyme disease, alpha-gal syndrome, Heartland virus, Powassan virus, and southern tick-associated rash illness (STARI). These diseases cause morbidity and mortality in human populations and may pose a high risk to individuals, wildlife, and livestock. The Cherokee National Forest covering the east Tennessee border provides a permissible environment for ticks to immigrate and thrive. Residents of northeast Tennessee frequently use the natural environment for a variety of purposes, creating exposure risk at the human-animal-environment interface. This study performed a scoping review and meta-analysis addressing topics informing epidemiological investigation of tick populations. The meta-analysis identified geography, climate, and Shannon-Wiener Diversity Index as the most significant variables associated with northeast Tennessee tick populations. Additionally, tick surveillance in northeast Tennessee counties was performed. These counties included Carter, Greene, Hancock, Hawkins, Johnson, Washington, and Unicoi. Primary tick species present in the summer included the American Dog tick (Dermacentor variabilis); the winter included the Blacklegged/Deer tick (Ixodes scapularis). Canonical correlation analysis was used to identify which environmental variables had the most influence, to what degree, and in a positive or negative direction. Altitude, total forest land, forest canopy, and fraction of surface water area were statistically significant. More altitude was correlated with more clinical cases; less total forest land, canopy, and fraction of surface water area was correlated with less clinical cases. Lastly, species distribution modeling of the invasive Asian longhorned tick was conducted. Study results indicate a low to moderate risk for tick-borne illness exposures among human populations, which is poised to increase. Species distribution modeling and clinical case data reports suggested an increasing exposure risk from improved habitat suitability. Increased risk is related to climate change and tick population growth in metropolitan areas. Finally, surveillance and control methods are summarized for integration into public health interventions.

Ticks are of global interest as the pathogens they spread can cause diseases that are of importance to both human health and economies. In Scotland, the most populous tick species is the sheep tick Ixodes ricinus, which is the vector of pathogens causing diseases such as Lyme borreliosis and Louping-ill. Recently, both the density and spread of I. ricinus ticks have grown across much of Europe, including Scotland, increasing disease risk. Due to the nature of the tick lifecycle they are particularly dependent on environmental factors, including temperature and habitat type. Because of this, the recent increase in tick-borne disease risk is believed to be linked to climate change. Many mathematical models have been used to explore the interactions between ticks and factors within their environments; this thesis begins by presenting a thorough review of previous modelling of tick and tick-borne pathogen dynamics, identifying current knowledge gaps. The main body of this thesis introduces an original mathematical modelling framework with the aim to further our understanding of the impact of climate change on tick-borne disease risk. This modelling framework takes into account how key environmental factors influence the I. ricinus lifecycle, and is used to create predictions of how I. ricinus density and disease risk will change across Scotland under future climate warming scenarios. These predictions are mapped using Geographical Information System software to give a clear spatial representation of the model predictions. It was found that as temperatures increase, so to do I. ricinus densities, as well as Louping-ill and Lyme borreliosis risk. These results give a strong indication of the disease risk implications of any changes to the Scottish environment, and so have the potential to inform policy-making. Additionally, the models identify areas of possible future research.

Ehrlichia ruminantium est l'agent causal de la cowdriose, une maladie tropicale mortelle des ruminantstransmis par les tiques Amblyomma. Jusqu'à présent, il n'existe pas de vaccin efficace dû à la faible protection croisée des souches vaccinales vis-à-vis des isolats de terrain. Ceci est principalement lié àdiversité génétique d'E. ruminantium au sein les zones géographiques. Par conséquent, la caractérisation de lastructure génétique de la population d'E. ruminantium à l'échelle mondiale et régionale est importante pour définir les meilleures stratégies de contrôle et améliorer les stratégies de surveillance de la cowdriose. / Ehrlichia ruminantium is the causal agent of heartwater, a ruminant tropical fatal diseasetransmitted by Amblyomma ticks. Up to now, no effective vaccine is available due to a limitedcross protection of vaccinal strains on field isolates mainly associated to a high geneticdiversity of E. ruminantium within geographical locations. Thus, both characterization of E.ruminantium genetic population structure at worldwide and regional scale and estimation of E.ruminantium tick prevalence are important to delimitate better control strategies and improveheartwater monitoring strategies

In this paper, I will show how applying a spatially explicit context to an existing environmental health surveillance framework is vital for more complete surveillance of disease, and for disease prevention and intervention strategies. As a case study to test the viability of a spatial approach to this existing framework, the risk of human exposure to Lyme disease will be estimated. This spatially explicit framework divides the surveillance process into three components: hazard surveillance, exposure surveillance, and outcome surveillance. The components will be used both collectively and individually, to assess exposure risk to infected ticks. By utilizing all surveillance components, I will identify different areas of risk which would not have been identified otherwise. Hazard surveillance uses maximum entropy modeling and geographically weighted regression analysis to create spatial models that predict the geographic distribution of ticks in Texas. Exposure surveillance uses GIS methods to estimate the risk of human exposures to infected ticks, resulting in a map that predicts the likelihood of human-tick interactions across Texas, using LandScan 2008TM population data. Lastly, outcome surveillance uses kernel density estimation-based methods to describe and analyze the spatial patterns of tick-borne diseases, which results in a continuous map that reflects disease rates based on population location. Data for this study was obtained from the Texas Department of Health Services and the University of North Texas Health Science Center. The data includes disease data on Lyme disease from 2004-2008, and the tick distribution estimates are based on field collections across Texas from 2004-2008.