DEVELOPMENT OF A PREDICTION MODEL FOR PAPAYA RINGSPOT IN VERACRUZ, MEXICO

A model to predict incidence of papaya ringspot was developed and vali dated from 5 yr of field observations in central Veracruz, Mexico. The model was developed from 1 yr of data collected from papaya (Carica p apaya) plantations in two different locations in Veracruz during 1985- 1986. Incidence of papaya ringspot was evaluated every 15 days, and vi ral infection was confirmed by ELISA. Aphid vectors (Myzus persicae, A phis gossypii, A. nerii, A. citricola, and Macrosiphum euphorbiae) of papaya ringspot virus were collected every 3 days from Moericke yellow pan traps placed at each location. The prediction model was obtained from an examination of the matrix of Pearson's correlation coefficient s and by simple and multiple regression analysis. Model selection was based on Mallow's C-p statistic, proportion of variance explained, var iance inflation factor, analysis of structure, and predictive capacity . The largest amount of variation in the data was accounted for by mod el (y) over cap = -1.45 + 0.42 AN(5) + 0.00016 PW + 0.116 AG(5) -0.005 8 AN(5)(2) -0.0057 MP(5)(2), in which (y) over cap was the incremental increase of disease ((y) over cap(t) - (y) over cap(t-l)) at any give n time (t); AN(5), AG(5), and MP(5)(2) were the numbers of the alate a phid species A. nerii, A. gossypii, and Myzus persicae, respectively. PW was an interaction variable defined as the product of precipitation (P) and speed and duration of wind from the north (W). Values for ind ependent variables were accumulated during a 4-wk period that ended 3 wk before the calculated incremental increase of disease. The equation accounted for 78% (R(2) greater than or equal to 0.78) of the total v ariation of the change of disease incidence ((y) over cap(t) - (y) ove r cap(t-l)) in the original data set. Validity of this model was teste d with data obtained from 60 epidemics in papaya plantations establish ed from 1987 to I989 to represent different dates, plant densities, an d plantation sites. The model predicted the relative rate of disease i ncrease in 38% of the epidemics (23 of 60, R(2) greater than or equal to 0.60). Three other models that accounted for less variance explaine d in the original data set than the first model (R(2) < 0.78) were als o validated. One model predicted the incremental increase in disease i ncidence of 40% of the papaya ringspot epidemics (24 of 60) with R(2) greater than or equal to 0.60. In this model, the disease incidence ch ange was explained by the independent variables AN(5), AG(5), and PW.