AN EVALUATION OF POTENTIAL ECONOMIC RETURNS FROM THE STRATEGIC CONTROL APPROACH TO THE MANAGEMENT OF AFRICAN ARMYWORM SPODOPTERA-EXEMPTA (LEPIDOPTERA, NOCTUIDAE) POPULATIONS IN EASTERN AFRICA

The costs of armyworm control in candidate outbreak source areas in Et hiopia, Kenya and Tanzania were compared with estimates of the costs o f potential damage if these source outbreaks were not controlled. The latter costs were estimated using a modelling approach, taking account of known moth migration trajectories, their frequency of occurrence, and of the agricultural productivity of potential destinations of the migrant moths. Two parallel studies were conducted: one for the outbre aks in the period January to June, and the other during October to Dec ember. Strategic control of armyworm using ground spraying, whereby in festations are controlled even if they are of no immediate economic im portance but when their future progeny pose a potential threat, will b e worthwhile during the period October to December on the basis of: (a ) available estimates of control costs; and (b) a sensitivity analysis of potential losses in relation to moth reproduction, For January to June the potential outbreak sources worthy of control are fewer, but t he two most important sources could lead to more devastation than any of those in October to December. The maximum estimated control cost (U S$ 33 per hectare, 1987 prices) involved the use of helicopters during an operation in Tanzania. The next highest were for ground-spraying o perations in Ethiopia in 1985 (US$ 17) and for an aerial operation in Kenya (US$ 16). However, even if the latter estimates are used, strate gic control will still be worthwhile in one fifth of the study area re gardless of season, given 20% crop damage and a moth reproductive rate per generation of at least five. Comparison of the results with those based on a hypothetical situation of no migrations showed only minor differences in many cases but the effects of moth movements were more marked for the period October to December than during January to June. This exercise also identified the Meru district in Kenya and the sout h Morogoro region of Tanzania, in particular, as zones which have unre markable agricultural value of their own but which can be sources of m oth populations capable of causing economically important damage elsew here. It is proposed that strategic control operations should be restr icted to those areas likely to be responsible for the most potential d amage. Such target areas have been identified in the study and are mos tly in east central and southwestern Kenya and northern and central Ta nzania.