DisplayTitle

Analysis of Summer Epidemic Progress of Apple Scab at Different Apple Production Systems in the Netherlands and Hungary

First author: Department of Plant Protection, Centre of Agricultural Sciences, University of Debrecen, P.O. Box 36, H-4015 Debrecen, Hungary; second and third authors: Wageningen University and Research Centre, Applied Plant Research, P.O. Box 200, 6670 AE Zetten, the Netherlands; fourth author: Department of Applied Mathematics and Probability Theory, University of Debrecen, P.O. Box 12, H-4010, Debrecen, Hungary; and fifth author: Department of Agricultural Sciences, Imperial College London, Wye Campus, Wye, Ashford, Kent TN25 5AH, UK

Two, 4-year studies on summer epidemic progress of apple scab were conducted at Randwijk, the Netherlands, from 1998 until 2001 and at Eperjeske, Hungary, from 2000 until 2003. Disease assessments were made on scab-susceptible cv. Jonagold. A range of nonlinear growth functions were fitted to a total of 96 disease progress curves (3 treatment classes × 2 plant parts × 2 disease measures × 4 years × 2 locations) of apple scab incidence and severity. The three-parameter logistic model gave the most consistent fit across three treatment classes in the experiment (integrated, organic-sprayed, and organic-unsprayed). Parameters estimated or calculated from the three-parameter logistic function were used to analyze disease progress. These were disease incidence and severity on the day of the first assessment (Y_s); final disease incidence or upper asymptote for incidence (Y_if) or severity (Y_sf); fruit incidence and severity on day 40, after which no new lesions on fruits appeared (Y₄₀); leaf incidence and severity on day 75, at which shoot growth stopped (Y ₇₅); relative (β) and “absolute” (θ) rates of disease progress; inflection point (M); and area under the disease progress curve (AUDPC_S) standardized by the duration of the total epidemic. Comparisons among disease progress curves were made by correlation and factor analysis followed by Varimax rotation. There were large differences but high positive correlations among the parameters Y_s, Y_f, θ, and AUDPC_S across the three treatment classes. In the factor analysis, two factors accounted for more than 85% of the total variance for both incidence and severity. Factor 1 gave an overall description of epidemic progress of both scab incidence and severity and included the parameters Y_f, Y₄₀, Y₇₅, θ, and AUDPC_S. Factor 2 identified a relationship between the relative rate parameter (β) and the inflection point (M) for severity and a relationship between disease incidence and severity. For an integrated or an organic orchard, θ, AUDPC_S, and one of Y_f or Y₇₅ (because of the link with host phenology) can characterize apple scab epidemics during summer. Based on these findings, improved scab management approaches were provided for integrated and organic apple production systems.

Additional keywords: Akaike information criterion, disease management, disease warning model, epidemiology, fixed effect , nonlinear mixed-effect modeling , random effect , R language .