When we think about protecting crops from disease, the spotlight usually falls on high-tech fixes – new pesticides, genetic engineering, advanced fungicides. But there’s a powerful, low-tech ally we often ignore: the landscape itself.
Recent theoretical studies carried out by the Epidemiology and Modelling group at the University of Cambridge indicate that the way we arrange crops in the landscape can dramatically slow the spread of plant diseases – without relying on chemicals or costly interventions. By rethinking the spatial design of agricultural landscapes, we can harness nature’s own defences and reshape how we prepare for and respond to agricultural epidemics.
At the core of this work is the concept of the ‘analytical infection rate’ – a measure of how quickly a disease spreads at the very start of an epidemic. This metric serves as the unifying framework for our three recent papers, which together tackle the essential pillars of agricultural epidemic preparedness: speed, efficiency, and design.
Speed: analytical tools for rapid response
Traditionally, scientists have relied on massive computer simulations to predict disease spread, modelling every field and every possible scenario. While detailed, these simulations can be slow and require extensive data – something rarely available in regions like sub-Saharan Africa, where crop maps are often incomplete.
Our first paper, ‘Predicting the effect of landscape structure on epidemic invasion using an analytical estimate for infection rate’, addresses this challenge head-on. We introduce a new analytical method to estimate infection rates, allowing us quickly to gauge how different landscape arrangements could slow or accelerate disease spread.
Think of this method as an ‘accelerator pedal’ for decision-making: it delivers rapid insights in near real time, rather than waiting for simulation results. This speed matters because it empowers farmers and policymakers to make evidence-based choices about land use and crop placement – reducing epidemic risks and potentially slowing outbreaks before they start.
Efficiency: smarter, targeted surveillance
Speed is critical, but it’s not enough. Models are only as accurate as the data behind them – and in many vulnerable regions crop maps are patchy or outdated. A second paper, ‘Where to refine spatial data to improve accuracy in crop disease modelling: an analytical approach with examples for cassava’, brings mathematical speed to the challenge of imperfect data, focusing on Cassava Brown Streak Virus (CBSV), a major threat to African food security.
Surveying every square kilometre to fix map errors is impractical, so to tackle this our team developed a method to prioritize data collection. Using our analytical tool, we pinpoint ‘top priority areas’ where errors have the greatest impact on epidemic predictions. In computational studies of CBSV, this approach cut prediction errors by up to fourfold – simply by directing limited resources to the most critical locations.
Design: building resilience into the landscape
With fast, reliable tools in hand, we can move beyond predicting outbreaks to actively shaping landscapes that resist them. A third paper, ‘Optimizing crop clustering to minimize pathogen invasion in agriculture’, applies our analytical framework to determine the optimal spatial structure of farmland across landscapes to minimize the risks of disease spread.
Contrary to intuition that crops must always be widely dispersed, we found that infection rates can be minimized even when crops are aggregated, provided clusters do not exceed a specific threshold size. The most significant finding was the ability to analytically identify the optimal cluster size and separation distance that keeps the invasion rate at its theoretical minimum for a given disease.
For agricultural planners, this means actionable design principles. Imagine a farmer who wants to maximize profitability (favouring larger clusters) while minimizing risk. Our work provides an exact formula that links a pathogen’s dispersal range to the distribution of crops across a landscape. Instead of vague advice like “spread out your crops,” we can now specify the maximum size of cluster and the minimum size of buffer zones between clusters for a given pathogen.
This marks a critical shift toward proactive infrastructure design – creating landscapes that are structurally resistant to disease.
The new frontier: theory into practice
This trilogy of research – spanning fast analytical tools, smarter surveillance, and proactive landscape design – marks a major step forward in theoretical epidemiology and offers a holistic framework for agricultural epidemic preparedness:
- Speed: analytical tools that complement slow, data-heavy simulations.
- Efficiency: targeted strategies that make surveillance smarter and more resource efficient.
- Design: landscapes planned for resilience, not just productivity.
These advances aren’t just theoretical. They provide actionable strategies for farmers, policymakers, and scientists. As climate change and global trade reshape agriculture, flexible, data-driven, and spatially aware models will be essential to safeguard food supplies. With smarter models, we can build a more resilient, food-secure future.
References:
Suprunenko, Y.F. et al: ‘Predicting the effect of landscape structure on epidemic invasion using an analytical estimate for infection rate.’ Royal Society Open Science, January 2025, DOI: 10.1098/rsos.240763.
Suprunenko, Y.F., Gilligan, C.A.: ‘Where to refine spatial data to improve accuracy in crop disease modelling: an analytical approach with examples for cassava.’ Royal Society Open Science, May 2025, DOI: 10.1098/rsos.250012.
Suprunenko, Y.F., Gilligan, C.A.: ‘Optimizing crop clustering to minimize pathogen invasion in agriculture.’ Nature Scientific Reports, December 2025, DOI: 10.1038/s41598-025-30635-9.
Image: Aerial photography of farmland. Photo by Tom Fisk.
Text by Yevhen Suprunenko. Dr Yevhen Suprunenko is a visiting researcher in the Epidemiology and Modelling group at the Department of Plant Sciences, University of Cambridge. His research focuses on establishing the role of intrinsic scales of an epidemic, cropping pattern and density in epidemic spread and control.