Modeling spatial variability and pattern of rust and brown eye spot in coffee agroecosystem

Abstract

The advent of Precision Agriculture has made possible the analysis of complex spatial patterns of plant disease epidemiology considering statements of integrated disease management. The objective of this work was to use geostatistics, statistics and geographical information systems to characterize the structure and magnitude of spatial dependency of rust (Hemileia vastatrix) and brown eye spot (Cercospora coffeicola) incidence and severity in coffee agroecosystem cultivated with Catuai Vermelho IAC-99 (Coffea arabica L.). Evaluations of incidence and severity of rust and brown eye spot were accomplished at 67 georeferenced points arranged in 6.6202 ha of coffee crop, in the years of 2005, 2006 and 2007. Exponential models of covariance enabled the characterization of the magnitude and structure of rust and brown eye spot spatial variability in the evaluated dates. Ordinary block kriging presented satisfactory performance to map rust and brown eye spot outbreaks based on kriging error coefficients. Kriged maps enabled the visualization of intensity of rust and brown eye spot in each evaluation date. Assessments of incidence and severity presented highly statistical correlation based on linear regression models, also confirmed by the spatial variability of kriging maps. Kriging maps of rust and brown eye spot enabled to observe that intensity of disease was dispersed in foci patterns along the coffee plantation, indicating that the current strategy of disease control based in total area may be replaced by site specific disease management, with less environmental impact and sustainability of coffee crop, according to statements of integrated disease management and precision agriculture.