Application of Survival Ratio for Monocyclic Process ofHemileia vastatrixin Predicting Coffee Rust Infection Rates

Kushalappa, A. C., Akutsu, M., and Ludwig, A. 1983. Application of survival ratio for monocyclic process of Hemileia vastatrix in predicting coffee rust infection rates. Phytopathology 73: 96-103. Apparent infection rates corrected for leaf formation (p") were based on the proportion of rainy and windy days (for 28 days before DP) determined based on proportion of leaves (PLR) and leaf area rusted multiplied by the proportion of leaf density. The monocyclic process (PRA) on 60 branches marked among 15 plants on 1 ha of cultivar Mundo equivalents for environment (MPEE) related to the whole monocyclic Novo coffee in Vifosa, Minas Gerais, Brazil, at 14-day intervals from process was derived from MPEE = DISEE X INFEE. The inoculum was August 1978 to July 1980. Hourly temperature and leaf wetness were quantified as the proportion of leaf area occupied by visible spores and recorded and the data were transformed to infection equivalents for proportion of spore area index (PSAI). The calculated proportion of spores environment (INFEE) and to dissemination equivalents for environment surviving dissemination and able to cause infection was designated as the (DISEE). The INFEE was determined by laboratory experiments to be the proportion of infective spore area index, PISAI = PSAI X DISEE X productofinfectionequivalentsforhoursoffreewaterandfortemperature, INFEE. The following regression equations were developed to predict which were calculated from the function of hours of free water and of coffee rust infection rates: Y, = 2.0804 + 0.1533Xi + 0.0852X 2 (r 2 = 0.72), temperature, respectively, for infection, as determined by laboratory2 and Y2 = 1.393 + 0.1149X1 + 0.0708X 2 (r 2 = 0.49), in which Yis p" for 28 experiments. The functions are Y= 1 1.996 exp (0. 1089t) and Y = sin days after the date of prediction ( Y, is based on PLR, Y2 is based on PRA), (188.lx 41.6x 2 151.3x ), respectively, in which Y and Y' are the X, islogit PISAI and X2is host available for infection,-logit xy when xy is proportions of maximum infection observed (= INFEE), x is the PLR or PRA on the date of prediction corrected for leaf formation during Schri~dter's temperature equivalent, and t is hours of free water. DISEE was the 28 days after the date of prediction. Additional key words: Coffea arabica, epidemiology, prediction model. Multiple regression analysis has been used to predict functions of environment for subprocesses developed by development of coffee rust (1,8,9,17) and other plant diseases laboratory experiments. Meteorological field data have been (4,6,7). The disadvantages of using multiple regression to predict transformed to subprocess equivalents for environment and plant disease development also have been reported (5). In the survival ratios for monocyclic process using biological field data, regression models developed to predict coffee and wheat rusts (1,4, which were used in combination with data representing the amount 6,7,9,17), the combinations of independent variables have been of host tissue available for infection to develop a multiple based on meteorological rather than biological functions. regression model for predicting coffee rust infection rates. The Furthermore, variables relating to a single function have often been cause and effect of independent variables for the best-fitting model repeated, which makes the model less stable. These models are have been quantified by using path analysis (13). unlikely to describe paths of causation in biological systems because the relationship among variables in multiple regression MATERIALS AND METHODS models is a sum of linear functions and because the independent variables themselves are principally meteorological and thus are Laboratory experiments. The influence of temperature and not directly related to disease development. However, certain hours of free water on infection were studied under laboratory combinations or sequences of meteorological factors at known conditions. Young, completely expanded leaves of 6-mo-old time intervals may control the progress of various epidemiological seedlings of Coffea arabica L. 'Mundo Novo' were inoculated with processes and thus should relate directly to disease development, a suspension (2 mg/ml) of freshly collected urediniospores of especially when paths of pathogen actions are synchronized. Hemileia vastatrix race II in 0.12% Tween-80. The spore Many simulation models define the functional relationship suspension was atomized onto the seedlings, and they were placed involved among various subprocesses of a monocyclic process. in a humidity chamber without lights at 21 ± 2 C. The seedlings Zadoks and Schein (20) used principles of ecology to evaluate were removed from the humidity chamber at various time intervals, resistance. They quantified horizontal resistance as survival ratios dried with a fan for 60 min, and then kept in the dark until the for successful completion of various subprocesses of a monocyclic longest moisture treatment (48 hr) was completed. Each treatment process. consisted of five pairs of leaves, one in each of five seedlings or In the present study, the environment related to three replicates. The experiment was repeated twice. The seedlings were subprocesses (sporulation, dissemination, and infection) of the then transferred to a greenhouse and the numbers of lesions formed monocyclic process of Hemileia vastatrix Berk & Br. have been were counted after 40 days. The mean number of lesions per leaf for quantified as subprocess equivalents for environment based on the each period of free water on the leaves was represented as a proportion of the maximum number of lesions observed, and an The publication costs of this article were defrayed in part by page charge payment. This exponential fit to the data was estimated by using a computer article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. § program SPSS (16). The equations were designated here as 1734 solely to indicate this fact. functions of environment (for temperature and for hours of free 0031-949X/83/01009608/$03.00/0 water) for infection, FEINF (acronyms not defined in the text are 01983 The American Phytopathological Society defined in a footnote to Fig. 1).