Evaluation of action thresholds for chronic rice insect pests in the Philippines. III. Leaffolders

Action thresholds (AT) as insecticide application decision tools were developed and tested against rice leaffolders Cnaphalocrocis medinalis (Guenée) and Marasmia patnalis Bradley (Lepidoptera: Pyralidae) in four sites and 68 crops over a 13-year period in the Philippines. Leaffolder damage levels were generally low with a mean over all sites and crops of 2 – 3% damaged leaves (DL) and 6 – 9% in a given site as a per-crop average based on weekly samplings, with highest incidence per field on a given sampling date of 48%DL. The damage incidence over a typical crop cycle formed a bell-shaped curve with maximum numbers in the late reproductive and early ripening stages. Continuous population buildup over the planting cycle did not occur in any one site over the season, thus a strategy of monitoring earlier planted fields to forewarn against impending economic damage had no basis. ATs were based on different levels of percentage DL, number of larvae per hill (L), and density of flushed moths. Collectively ATs were surpassed in 1, 12, and 4% of fields in each of the three main growth stages. A scoring system was based on benchmark criteria including both leaffolder damage and yield loss. The 15%DL threshold performed best in all growth stages, only equaled by 1L in the vegetative stage, scoring 93 – 99% correct decisions in each of the growth stages. Flushed moths produced the most false positive decisions and the lowest performance scores. The best insecticide response gave only a disappointing 53% control and therefore was a significant weakness in using AT technology. BPMC, endosulfan, and monocrotophos performed the best and azinphos-ethyl and Bt the least. Sites with the highest densities had the most rapid rise in damage levels, thus requiring more frequent monitoring. Low leaffolder damage levels were most likely due to the activity of natural enemies. Given the ability of high tillering varieties to compensate from pest damage, farmers would be better off embracing integrated crop management as a preventative measure to bolster inherent tolerance levels combined with crop monitoring based on historical population levels. In sites with a record of high damage levels, monitoring should start 4 weeks after transplanting (WT) on a weekly basis and more risk averse farmers may increase the interval to twice a week during flag leaf stage. In sites with lower observed levels, monitoring can begin 6 WT on a weekly basis and continue through the flag leaf stage. Significant yield increases were recorded with the best AT characters despite evidence of modern rices to tolerate high damage levels. Reasons for this apparent paradox are discussed.