Early warning signals in plant disease outbreaks

TL;DR

This paper presents a lattice-based epidemic model for plant diseases, demonstrating that early-warning signals can predict critical transitions from localized outbreaks to widespread epidemics, aiding in forest disease management.

Contribution

Introduces a simplified stochastic lattice model combined with early-warning indicators to predict critical transitions in plant disease outbreaks.

Findings

Early-warning signals successfully forecast the shift to widespread outbreaks.

Critical transition depends on host density and pathogen spread velocity.

Model demonstrates potential for predicting forest disease epidemics.

Abstract

Summary 1. Infectious disease outbreaks in plants threaten ecosystems, agricultural crops and food trade. Currently, several fungal diseases are affecting forests worldwide, posing a major risk to tree species, habitats and consequently ecosystem decay. Prediction and control of disease spread are difficult, mainly due to the complexity of the interaction between individual components involved. 2. In this work, we introduce a lattice-based epidemic model coupled with a stochastic process that mimics, in a very simplified way, the interaction between the hosts and pathogen. We studied the disease spread by measuring the propagation velocity of the pathogen on the susceptible hosts. Quantitative results indicate the occurrence of a critical transition between two stable phases: local confinement and an extended epiphytotic outbreak that depends on the density of the susceptible individuals. 3. Quantitative predictions of epiphytotics are performed using the framework early-warning indicators for impending regime shifts, widely applied on dynamical systems. These signals forecast successfully the outcome of the critical shift between the two stable phases before the system enters the epiphytotic regime. 4. Synthesis: Our study demonstrates that early-warning indicators could be useful for the prediction of forest disease epidemics through mathematical and computational models suited to more specific pathogen-host-environmental interactions.