Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Understanding and Managing Frogeye Leaf Spot through Network-Based Modeling in Soybean

Chinthaka Weerarathna, Thien-Minh Le, Jin Wang

Abstract

Frogeye Leaf Spot (FLS), caused by Cercospora sojina, poses a significant threat to soybean production, with yield losses of 30-60%. Traditional mass-action models assume homogeneous mixing, which rarely holds in real fields and limits their ability to inform FLS management. To address this, we developed a network-based model that incorporates real-field structure to improve FLS management in soybeans. Using approximate Bayesian computation, we estimated key epidemiological parameters and found that infection origin can shift the balance between transmission routes. Data analyses indicated that tillage and non-tillage plots did not differ significantly in fungal spread, decay, or disease severity. Finally, we show that early, targeted roguing is more effective than delayed or random removal. Together, these findings offer science-based guidance for FLS management and highlight the value of network-based models to inform agricultural disease control.

Understanding and Managing Frogeye Leaf Spot through Network-Based Modeling in Soybean

Abstract

Frogeye Leaf Spot (FLS), caused by Cercospora sojina, poses a significant threat to soybean production, with yield losses of 30-60%. Traditional mass-action models assume homogeneous mixing, which rarely holds in real fields and limits their ability to inform FLS management. To address this, we developed a network-based model that incorporates real-field structure to improve FLS management in soybeans. Using approximate Bayesian computation, we estimated key epidemiological parameters and found that infection origin can shift the balance between transmission routes. Data analyses indicated that tillage and non-tillage plots did not differ significantly in fungal spread, decay, or disease severity. Finally, we show that early, targeted roguing is more effective than delayed or random removal. Together, these findings offer science-based guidance for FLS management and highlight the value of network-based models to inform agricultural disease control.
Paper Structure (15 sections, 1 theorem, 19 equations, 7 figures, 3 tables, 1 algorithm)

This paper contains 15 sections, 1 theorem, 19 equations, 7 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Methods
  3. The SEIRB Framework for Frogeye Leaf Spot by MBE2024FLS
  4. Proposed Network-Based SEIRB Model
  5. Real Field network based on distance threshold
  6. Proposed SEIRB Network Model
  7. Results
  8. Data Analysis
  9. Parameter Estimation via approximate Bayesian computation (ABC)
  10. Sensitivity Analysis Across Initial Infection Scenarios
  11. Assessing the Effectiveness of Tillage Practices
  12. Tillage effects on fungus spread and decay
  13. Tillage effects on disease severity
  14. Roguing Intervention
  15. Discussion and Conclusion

Key Result

Lemma 1

Let $r_{0}>0$, $k_{0}>0$, $\xi\ge 0$, $\tau_{P(p)}\ge 0$ and $I_p\ge 0$ be constants for a given plot $p$. Consider the soil reservoir dynamics where $\dot B_p(t)$ denotes the time derivative of $B_p(t)$, representing the instantaneous rate of change of the soil inoculum in plot $p$. Then there is a unique nonnegative equilibrium $B_p^\star$ given by Moreover,

Figures (7)

  • Figure 1: Illustration of the SEIRB modeling framework and network representations. (a) Diagram of the SEIRB transmission structure, including direct and soil-mediated infection pathways. (b) Effect of the distance threshold $d$ on the plant contact network: (i) hypothetical soybean field layout; ($\text{i}_1$) the network structure induced by a distance threshold of $d=1$, corresponds to nearest-neighbor connectivity on the lattice; and ($\text{i}_2$) the network structure induced by a distance threshold of $d=2$, resulting in a denser pattern of connectivity.
  • Figure 2: Real field layout used to construct the spatial network Mengistu2014FLS.
  • Figure 3: Observed infected-plant counts and posterior predictive trajectories. Solid line: mean of the best-100 trajectories; the shaded ribbon denotes the 95% credible band. The best-fit model parameter vector is $\vartheta = \left( \theta,\, \beta_{\text{non}},\, \rho_\beta,\, \xi,\, \tau_{\text{non}},\, \rho_\tau,\, d \right) = \left( 3.90\times 10^{-5},\, 4.24\times 10^{-8},\, 1.45,\, 113.5,\, 0.263,\, 0.881,\, 83.1 \right).$
  • Figure 4: Spatial configurations of the initially infected plants for the four seeding scenarios: Random, Cluster 1, Cluster 2, and Polycluster.
  • Figure 5: Posterior sensitivity to initial seeding, shown as boxplots of marginal posterior samples for $\vartheta=(\theta,\beta_{\mathrm{non}},\xi,\tau_{\mathrm{non}},\rho_\beta,\rho_\tau,d)$ under the four seeding scenarios.
  • ...and 2 more figures

Theorems & Definitions (2)

  • Lemma 1
  • proof