Diseases characterized by long distance inoculum dispersal (LDD) are among the fastest spreading epidemics in both natural and managed landscapes. Management of such epidemics is extremely challenging because of asymptomatic infection extending at large spatial scales and frequent escape from the newly established disease sources. We compared the efficacy of area- and timing-based disease management strategies in artificially initiated field epidemics of wheat stripe rust and complemented with simulations from an updated version of the spatially explicit model EPIMUL, using model parameters relevant to field epidemics. The model was further used to expand the number of epidemic mitigations beyond that feasible to incorporate in the field. The field experiment was conducted for two years in two locations having different climatic conditions. Culling and protection treatments were applied at different times after epidemic initiation and to different spatial extents surrounding the outbreaks. In each experiment, treatments were replicated four times in plots 33.5 m long and 1.52 m wide with a 0.76 x 0.76 m inoculated focus centered within each plot. Disease gradients were assessed along the center lines of the plots at 1.52 m intervals both upwind and downwind from the focus. Both field and simulation results indicated that control measures applied over the entire population were highly effective in suppressing the epidemics by more than 99% but may not always be logistically and economically feasible at large spatial scales. Comparison between the variable sized treatment areas and application timings suggested that implementing contiguous premises (CP) cull at 1 day after first sporulation in the outbreak focus reduced rust by 52 and 60% in Corvallis and Madras, respectively. However, altering the cull size did not significantly affect the disease epidemic development, which suggested that early timing had a greater influence in suppressing the epidemics than did increased area of application. However, sufficiently large, treated areas may compensate for a delay in application timing to some extent. Results from these replicated treatments may help to devise appropriate management strategies for other LDD pathogens. 

In this study, we compared the efficacy of different control strategies for management of epidemics caused by LDD pathogens using field experiments and simulation studies. These control strategies included area-based treatments, application timing, and a combination of both tactics. We parameterized the compartment based SEIR model ‘EPIMUL’ using model parameters relevant to the wheat stripe rust pathosystem and compared outputs with results from the field studies. Field experiments were conducted at two different locations in Oregon, one at the Hyslop Crop Science Field Research Laboratory in Corvallis and the other at the Central Oregon Agriculture Research and Extension Center in Madras for two consecutive winter wheat seasons, 2018-2019 and 2019-2020. All experiments in both locations were planted with an older club wheat (Triticum aestivum ssp. compactum) cultivar called ‘Jacmar’. Plots were planted with the long axis oriented parallel with the anticipated predominant wind direction during the time of epidemic spread. Stripe rust was established in the center (0.762 x 0.762 m) area of each plot. At both locations, the experimental design was a randomized complete block with seven treatments and four replications per treatment. Different disease control treatments were applied to either the outbreak focus area or the general population. End of season WSR prevalence (percentage of leaf area covered by stripe rust lesions, assuming a maximum of 100 lesions per tiller) was assessed visually in each treatment plot.

The compartment based SEIR model EPIMUL was used to simulate the effect of different mitigation practices for the control of epidemic outbreaks and to complement the field results. In our field experiment, we tested only a limited number of treatments because of logistical constraints. Without these limitations in EPIMUL we were able to test an extended range of timing and size of the cull and fungicide treatments, including different efficacy levels of the fungicide, to better understand the effect of timing and extent of control practices in the control of disease epidemics.

Please consider README_for_Data.pdf for the instruction on how to use data. All the codes used in the analysis have been included.

If you have any questions, please write to  Professor Christopher C. Mundt (mundtc@science.oregonstate.edu)