Modelling the impact of Phytophthora austrocedri on UK populations of native juniper (Juniperus communis s. l.)

Introductions of non-native plant pests and pathogens are increasing, negatively impacting natural environments. Mitigation requires knowledge of the drivers of pathogen introduction, establishment and spread, rarely available when pathogens first emerge in novel settings. This thesis uses multi-scale ecological modelling to understand environmental and land management factors driving patterns in infection and impact of the newly discovered oomycete pathogen, Phytophthora austrocedri, on juniper (Juniperus communis). I first surveyed potential abiotic and biotic drivers of disease severity across three, geographically separate juniper populations with different infection histories. In all populations, disease severity increased with increasing soil moisture. Associated plant species that could be used to locate microsites at higher risk of infection were also identified. Change in infection intensity during a four-year period was then mapped across a single juniper population and related to environmental factors underpinning the presence and density of juniper and driving P. austrocedri spread. Colonisations usually occurred within a ~500m radius of previously symptomatic trees, with infrequent dispersal beyond 1km, potentially mediated by livestock and deer. By compiling a novel dataset, I revealed larger, more frequent supplementary juniper planting events increased the likelihood of P. austrocedri presence. Stakeholders managing, monitoring, and growing juniper then participated in a survey investigating how practitioners consider disease risks and whether these processes could be better supported by decision tools. Lastly, a machine learning model and risk map was developed that predicted juniper populations in northern England and central Scotland are at highest risk of infection due to acidic soil pH and increased roe deer density. My research demonstrates how incorporating a wider range of abiotic and biotic drivers, exploring scale dependence, and integrating stakeholder knowledge can improve the predictive accuracy of host-pathogen-environment models. The results are used to recommend strategies (e.g. reductions in grazing pressure, natural juniper regeneration and heightened on-site biosecurity) to mitigate the serious threat posed by the pathogen to UK biodiversity and habitat restoration goals.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2021

Author dc:creator

• Donald, Flora

Advisors dc:contributor.advisor

• Purse, Bethan V
• Cunniffe, Nik J
• Green, Sarah
• Searle, Kate

• Statistical modelling
• Plant pathology
• Phytophthora

Licence

• http://purl.org/NET/rdflicense/allrightsreserved

Language dc:language

eng

Author Identifier

0000-0001-6266-9189

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/337425