This paper evaluated the risk that Aenolamia postica (Walker) Fennah populations reach the economic threshold in sugar cane fields in Veracruz, México. A risk deductive model was constructed to include the sequence of events leading to damaging populations, considered the top event or critical failure in the crop. Model events were identified and quantified, and model was validated on field conditions. The model components and their state values were identified as: temperature e" 28°C, precipitation e" 45% during June and July, soil clay content e" 40%, infested adjoining fields, deficient weed control, wind dominance, crop phenology and variety, deficient chemical and biological control, and irrigation. Sensitivity analysis showed that the most important events triggering high densities of A. postica were high temperatures and precipitation, previous field infestation, nymph and weed presence. Event probability estimates were combined using Boolean algebra to compute the minimum, mean and maximum probabilities for the top event, yielding values of 0.417, 0.563, y 0.734 respectively. Model was tested in field, by selecting sugar cane fields having the model properties and compared to fields without these features. Fields were sampled in both conditions during 2004 year and high-risk fields had significantly (F = 13, 4, gl = 1, 18, P = 0,0018) higher densities (2.4 adults m-1) than low-risk plots (0.4 adults m-1) thus agreeing with the model forecast.
Sugar cane; fault tree analysis; system analysis; Monte Carlo simulation; preventive pest management
