Cunniffe Group: Modelling pest and pathogen entry and spread pathways to enhance the UK Plant Health Risk Register
Supervisors
Professor Nik Cunniffe
Sam Grant (DEFRA)
Brief Summary
The UK Plant Health Risk Register (UK-PHRR) is central to UK plant biosecurity, tracking over 1,400 pests and pathogens that threaten UK crops, trees, gardens, or ecosystems. It provides an agreed, evidence-based framework for prioritising government action by ranking the relative risk posed by different pests and pathogens. This PhD project will evaluate past performance of the UK-PHRR and explore possible improvements, focusing on methods that remain practical in the context of limited data.
Importance of Research
Increasing rates of global trade and travel, and changing climatic patterns, have led to more frequent outbreaks of plant disease epidemics worldwide. Epidemics in plants grown as crops can lead to lead to devastating impacts, including on food security and rural livelihoods. Globally plant pests and pathogens are estimated to destroy up to 40% of yield, leading to hundreds of billions of dollars of economic losses. Epidemics in natural environments can have significant effects on the full range of ecosystem services. For example, tens of millions of elm trees were lost from the UK landscape due to the Dutch Elm disease epidemic in the 1970s; we are probably living through a similarly impactful outbreak of ash dieback. In the light of these potential impacts, there is a clear need for mathematical and statistical approaches to inform plant health policy. However, to date, only relatively simple approaches are used by the UK government. This project aims to assess the impact of - and potentially to change - this.
Project Summary
UK-PHRR rankings are based on scores, ranging from 1 (very low risk) to 125 (exceptionally high risk). A key component in these scores is the “likelihood”, calculated differently depending on whether a pest/pathogen is (thought to be) absent from the UK or known to already be present. If the pest/pathogen is absent, the likelihood is a measure of the risk of it being introduced, accounting for both entry and establishment. However, if the pest/pathogen is already present, the likelihood assesses the risk of it spreading to its maximum extent.
In both cases, the likelihood score is produced according to a set of simple rules and/or expert judgement. Given the number of pests/pathogens that must be accounted for, and the lack of data for many of them, this is pragmatic and scalable. However, it raises the question of whether and how the likelihood could be improved.
Pests/pathogens often enter via the trade in plants and/or plant products, and network-based models of this pathway are now available. Risks of establishment can be informed by species distribution models based on climatic conditions, as well as information on the density and location of susceptible host plant species. Similarly, predictions of future spread can be made using models, which range in complexity from simple projections of a fixed spread rate over time to detailed process-based simulations.
Additionally, given the UK-PHRR has now been in use for over a decade. Over this time an average of around 5 pests/pathogens has entered annually. This permits us to ask questions around the accuracy of the UK-PHRR as a tool, by comparing likelihood scores from historic versions of the UK-PHRR with what came to pass over time.
Assessing the accuracy of the UK-PHRR, and whether and how its predictions can be improved is the focus of this PhD project. A key challenge is that any new methodology developed must be practically useful, in a setting in which there is a large and ever-increasing number of pests and pathogens to rank, and often very little concrete information on their biology and current distribution.
What will the successful applicant do?
The project has four objectives:
- Evaluate the predictive performance of the UK-PHRR. Compile a dataset of pest/pathogen introductions since the UK-PHRR was first introduced and use it to assess the accuracy of UK-PHRR likelihood scores at the time of entry. Repeat for pests/pathogens which had already entered the UK, but which have markedly changed in extent over the lifetime of the UK-PHRR. Identify classes of pests/pathogens that had risks which were systematically over-/under-predicted, to attempt to improve the simple rule-based system currently used.
- Develop trade-driven pest/pathogen entry models. Building on existing network based models (e.g. Alonzo Chávez et al., 2025; Douma et al., 2016; Montgomery et al., 2022), create a flexible pathway model which calculates the probability of entry in terms of the trade flows between different countries, the degree of confidence we have in whether a given pest/pathogen is absent/present/widespread in a trading partner, the degree of symptoms/ease of detection at the point of entry and further flows of commodities within the UK. Models will be validated for a selection of case study pests/pathogens using interception data from Defra. For potential extension to all pests/pathogens on the UK-PHRR, consider systematically how models respond to lack of information, focusing closely on how uncertainties in key inputs/parameters can be propagated through to final assessments of relative risks.
- Understand risks outside of (regulated) trade. Although trade is the predominant pathway, pests/pathogens can also be transported by travellers. Again, building on models which have recently appeared (Gottwald et al., 2019), we will build models using global databases on patterns of travel, testing whether simplified and parameter-sparse versions could be used to improve the likelihood scores used in the UK-PHRR. An intriguing possibility is to attempt to extend this to include risks via e-commerce, focusing on using web-scraped data from online marketplaces capturing live plant and seed sales, as well as published literature on pest transport via mail.
- Couple models to estimate risk of establishment and/or spread. Even if a pest/pathogen enters the country, establishment and invasion requires plant host species and suitable environment conditions (and perhaps a vector species, too). We will use host maps (e.g., CEH landcover maps) and species occurrence data from global pest databases (e.g., CABI, EPPO) to develop simple methods to estimate risk of establishment, again concentrating on methods that can be used quickly, and which have small requirements in terms of data (e.g., using Köppen Geiger climate classifications rather than full species distribution models). We will also test the extent to which spatio-temporal spread models - which are now increasingly well-refined (see, e.g., Cunniffe et al. (2016), Mastin et al. (2020), Ellis et al. (2025)) - can be used to estimate impacts of any pest/pathogen invasion, should establishment occur.
The project would be well suited to a student with a background in mathematics, engineering, physics or theoretical ecology, ideally with prior knowledge of computer programming and/or statistics, motivated to transition to work on biological problems. However, students with a background in wet-lab biology have enjoyed and been very successful in my laboratory in the past. Any such candidates with a strong interest in making a transition to mathematical modelling are very much encouraged to get in touch to discuss.
References
Alonso Chávez, V. et al. (2025) Early detection strategies for invading tree pests: Targeted surveillance and stakeholder perspectives. Journal of Applied Ecology. 62, 857-871 https://doi.org/10.1111/1365-2664.70009
Cunniffe et al. (2016) Modeling when, where, and how to manage a forest epidemic, motivated by sudden oak death in California. PNAS. 113:5640-5645 https://doi.org/10.1073/pnas.1602153113
Defra (2025) The UK Plant Health Risk Register. Available online at https://planthealthportal.defra.gov.uk/pests-and-diseases/uk-plant-health-risk-register/
Douma, J.C. et al. (2016) Pathway models for analyzing and managing the introduction of alien plant pests – an overview and categorization. Ecological Modeling. 339, 58–67 https://doi.org/10.1016/j.ecolmodel.2016.08.009
Ellis et al. (2025) Developing epidemiological preparedness for a plant disease invasion: Modelling citrus huánglóngbìng in the European Union. Plants, People, Planet. 7:1403-1423. https://doi.org/10.1002/ppp3.10643
Gottwald, T. et al. (2019) A probabilistic census-travel model to predict introduction sites of exotic plant, animal and human pathogens. Philosophical Transactions of the Royal Society, B. 374, 20180260 https://doi.org/10.1098/rstb.2018.0260
Mastin et al. (2020) Optimising risk-based surveillance for early detection of invasive plant pathogens. PLOS Biology. 18:3000863 https://doi.org/10.1371/journal.pbio.3000863
Montgomery, K. et al. (2022). Forecasting global spread of invasive pests and pathogens through international trade. Ecosphere. 14, e4720 https://doi.org/10.1002/ecs2.474