Modelling fungicide resistance management in an uncertain world

Chemical disease control exerts strong selection pressures on plant pathogens. This promotes the evolution of resistant pathogens, thereby threatening crop yields and food security. Resistance management – optimising fungicide deployment to delay the emergence or spread of resistant pathogen strains – has therefore been studied for decades. A particular focus has often been how chemicals should be sprayed in combination, most often comparing applying fungicides with different modes of action as a mixture (simultaneously) vs. in alternation (sequentially). Until relatively recently few general messages have emerged (Cunniffe et al., 2015).

However, the recent introduction by van den Bosch et al. (2014) of a simple set of governing principles to allow rates of selection for resistance to be predicted has begun to change this. A recent PhD student in my laboratory used these governing principles to extend earlier modelling work by Hobbelen et al. (2011) and to conclusively show that yields are almost always optimised by spraying fungicides as a mixture (Elderfield et al., 2017). The result is invariant to model structure and parameterisation, as well as the particular disease in question. However, the modelling shows the effective lifetime is also optimised by spraying a mixture containing as little fungicide at high-risk of resistance as possible. This permissible minimum amount of chemical was identified via a threshold on crop yield, below which disease was considered to be economically-inviable for the grower. However, in reality disease pressures – as well as prices obtained by growers for their crop and the costs of chemicals – vary from season to season, making such a threshold very difficult to identify. Farmers and their agronomists are also notoriously risk-averse, although they do sometimes rely on decision support systems to predict likely disease pressures and so necessary levels of chemical input.

The potentially confounding factors of stochasticity, economics and risk aversion were omitted from the initial work. Decision support systems were also not considered, although this could potentially allow less fungicide to be sprayed. Including each of these factors would make the recommendations from the modelling more practically-useful. The proposed project will therefore extend the earlier modelling work, focusing on a stochastic extension to the deterministic modelling of Hobbelen et al. (2011) and Elderfield et al. (2017), initially by accounting for temporal variability in the spread of disease. The model will be based on the existing model for septoria leaf blotch on UK winter wheat, which has been parameterised and tested on field data. However, a particular focus of the project will be developing robust recommendations that are not tied to any one particular pathosystem.

Eligible background

The project would be particularly suitable for a student with a strong interest in mathematical modelling, and ideally a background in computer programming. Candidates with undergraduate training in mathematics, physics, computer science or engineering are therefore very welcome to apply. However, training is potentially available for any student who is well-motivated but who has not had the opportunity to develop modelling and computing skills during their first degree, for example students with undergraduate training in biology or ecology.

Supervisors

The primary supervisor will be Dr Nik Cunniffe at the University of Cambridge, Department of Plant Sciences https://www.plantsci.cam.ac.uk/directory/cunniffe-nik. The project will be co-supervised by Dr Frank van den Bosch at Rothamsted Research https://www.rothamsted.ac.uk/our-people/frank-van-den-bosch

Please contact the primary supervisor on [Email Address Removed] for further details of how to apply.

BBSRC DTP Programme

The studentship is part of the BBSRC DTP Programme. The first six months will be spent completing tailored training courses and two laboratory rotations, before progressing to a PhD. An important element of the Programme is a three-month internship to gain experience in a non-academic environment. The Programme organises a number of events, training courses and workshops to foster cohort development, skills enhancement and networking opportunities.

References

Cunniffe, N.J., Koskella, B., Metcalf, C.J.E., Parnell, S., Gottwald, T.R. and Gilligan, C.A. (2015) Thirteen challenges in modelling plant diseases. Epidemics. 10:6-10

Elderfield, J.A.D., Lopez-Ruiz, F.J., van den Bosch, F. and Cunniffe, N.J. (2017) Using epidemiological principles to explain fungicide resistance management strategies: why do mixtures outperform alternations? BioRxiv. 168831

Hobbelen, P.H., Paveley, N.D. and van den Bosch, F. (2011) Delaying selection for fungicide insensitivity by mixing fungicides at a low and high risk of resistance development: a modelling analysis. Phytopathology. 101:1224-1233.

van den Bosch, F., Oliver, R., van den Berg, F. and Paveley, N.D. (2014) Governing principles can guide fungicide resistance management tactics. Annual Review of Phytopathology. 52:175-195