The Use of Mathematics and Computers to Determine Optimal Strategy and Tactics for a Given Insect Pest Control Problem

Abstract

AbstractA computer program has been developed for use in evaluating various strategies of insect pest control. At its present stage of development, the program simulates the effects of various dosages of insecticide, parasite release, and spraying of virus, or any combination of these, weather and pest density on reproduction, dispersal and mortality in a pest population. Effects of changing pest densities on the parasite population and on tree growth and mortality in a hypothetical, 6.4 million-acre balsam fir forest are simulated. Also, the computer simulates dispersal of parasites and disease incidence. All computations are performed separately for each of 625 4-mile-by-4-milc hypothetical squares of forest area in the hypothetical 10,000 square mile area. All operating costs and losses due to lost tree growth and tree mortality are printed each year. The particular set of strategies to be used in each game is included in the input data for each 35-year computer experiment, along with data on physiological parameters, allowable pest thresholds for each type of control, genetic parameters, behavioristic and dispersal data and costs of control. One play of the game takes 50-120 seconds on an IBM 7090 computer, depending on the array of strategies selected.There were a number of significant conclusions from the simulation study. Biological control agents can not be as effective as insecticides unless they keep the pest at very low levels indefinitely. If they only drop the pest to about 10% of the peak level it would have attained without control, this may not be good enough to save the trees, because lethal effects accumulate. That is, pest densities which are not quite high enough to kill a tree quickly can kill it if the pest persists at these densities for several consecutive years. However, by selecting biological control agents with optimal physiological parameters, biological control can be made to produce totally effective control.For all kinds of control, control is vastly more effective if applied 10 years or more before peak pest densities. Therefore, potential pests should be subjected to preventive control at pre-pest densities, rather than being allowed to reach critical densities. The latter policy ignores the hazard of successive years of sublethal pest densities whose effects can cumulate to become lethal. It should be noted, however, that this conclusion follows from exploration of a hypothetical situation, and might not be applicable when considered relative to the exigencies of an actual field problem.