University of Tasmania
Kozakiewicz_whole_thesis.pdf (6 MB)

Landscape factors shaping host connectivity and pathogen dynamics in urban bobcats (Lynx rufus)

Download (6 MB)
posted on 2023-05-27, 09:37 authored by Kozakiewicz, CP
As a major factor affecting connectivity in wildlife populations, landscape heterogeneity can have substantial impacts on disease transmission. Urban development is a particularly acute contributor to landscape heterogeneity and declines in wildlife population connectivity, but few studies examine disease transmission and spread in urban environments. In this thesis, I utilise fine-scale genetic inference of how landscape heterogeneity shapes host connectivity and pathogen transmission in an urban environment. I focus on populations of bobcats, which are sensitive to urban development and indicators of connectivity in urban landscapes, from coastal southern California, one of the most urbanised landscapes in North America. Within these populations, I study feline immunodeficiency virus (FIVLru), a bobcat specific, rapidly mutating retrovirus and model for genetic inference of pathogen transmission in heterogenous landscapes. The field of landscape genetics investigates how landscape heterogeneity affects genetic variation and has potential as a framework for studying pathogen transmission and spread. I conducted a comprehensive review of landscape genetic studies of pathogen dynamics (Chapter 2). I found that landscape genetics has been underutilised in disease ecology, partly due to a lack of cross-disciplinary awareness within the field of disease ecology, and a lack of landscape genetic frameworks for pathogen systems. I emphasise the utility of landscape genetics for disease ecology and highlight emerging frontiers, including recent phylogeographic approaches and multi-species analytical frameworks. In Chapter 3, I conducted a landscape genomic study to investigate how urban and non-urban landscape factors are influencing gene flow among bobcats in southern California. I identified five genetically distinct populations, separated by major highways and urban development. Replicating landscape resistance analyses among these populations enabled me to assess the generality of landscape effects on gene flow. I found that urbanisation had a pervasive impact on connectivity, influencing region-wide patterns of gene flow as well as locally within three populations, but that availability of riparian habitat may mitigate these urban impacts. This work demonstrates the value in replicating landscape genetic analyses across populations, showing that factors affecting connectivity in urbanising environments may vary depending on spatial scale and local landscape structure. Having identified host population structure that is driven by major highways, I then (Chapter 4) conducted a phylogeographic analysis of FIVLru to investigate the impact of these barriers on FIVLru. Estimates of FIVLru divergence times dating back 118 years revealed a history of changes in pathogen transmission as urbanisation has increased. I identified strong phylogeographic structure, reflecting host population structure, and low FIVLru genetic diversity northwest of Los Angeles, suggesting relative isolation of FIVLru populations. Southeast of Los Angeles, FIVLru was genetically diverse with deep phylogenetic branches but little phylogeographic structure, suggesting divergence from host population structure. However, when accounting for variation in branch depth, movement rates of FIVLru across highways did not differ among populations. Divergence from host population structure southeast of Los Angeles was potentially a product of incomplete lineage sorting due to greater FIVLru sequence diversity and population size. Lastly (Chapter 5), I implemented ecological phylogenetic tools to identify landscape and host factors influencing patterns of FIVLru phylogenetic differentiation and dispersal rates. Overall, I found that urbanisation plays less of a role in reducing FIVLru connectivity than for host connectivity. However, FIVLru sequences from bobcat capture locations that were more divergent in the amount of vegetation land cover were more distantly related. Specifically, this was the case for forest land cover northwest of Los Angeles and scrub land cover southeast of Los Angeles. My results suggest FIVLru transmission differs between areas of high natural vegetation and areas of low natural vegetation (which are often urban areas). Further supporting the importance of vegetation for FIVLru transmission, I found a weak positive overall effect of vegetation density on FIVLru dispersal velocities. In summary, this thesis: i) identifies and implements a variety of emerging methods for elucidating landscape effects on host and pathogen spatial genetic structure; ii) identifies factors affecting bobcat connectivity in a highly urbanised environment; iii) indicates how host population structure and landscape heterogeneity shape FIVLru phylogenetic structure and transmission dynamics; and iv) demonstrates the utility of replicating analyses of genetic structure across multiple populations and spatial scales to contextualise observed patterns and relationships. Collectively, this work represents a rare example of integrating genetic estimates of both host and pathogen connectivity in a heterogeneous landscape. These insights provide valuable information for managing an urban wildlife host-pathogen system, while showcasing the utility of landscape genetics and emerging ecological phylogenetic tools for studying connectivity in heterogeneous landscapes.


Publication status

  • Unpublished

Rights statement

Copyright 2018 the author Chapter 2 appears to be the equivalent of a post-print version of an article published as: Kozakiewicz, C. P., Burridge, C. P., Funk, W. C., VandeWoude, S., Craft, M. E., Crooks, K. R., Ernest, H. B., Fountain-Jones, N. M., Carver, S., 2018. Pathogens in space: advancing understanding of pathogen dynamics and disease ecology through landscape genetics. Evolutionary applications, 11(10), 1763‚Äö-1778

Repository Status

  • Open

Usage metrics

    Thesis collection


    No categories selected


    Ref. manager