Introduction

Forecasting future events is a widespread activity that has been practised for millennia. Many methods have been used, including the reading of naturally occurring ‘signs’ (e.g. weather patterns), performing special operations (e.g. the inspection of chicken entrails) and relaying messages received from God. The variety of sometimes startling ways that have been used to make forecasts is perhaps indicative of the importance, in many situations, of being able to predict the future.

While it may be a part of the human psyche to wish to see into the future, there are also more specific reasons why it is advantageous to attempt to predict what the future might hold. In pest management, the object of forecasting must be to improve the decisions associated with controlling pests. In the first part of this chapter, we discuss the decision processes by which migrant pests are controlled, in order to define the types of forecast information that are required. This information, together with knowledge of the biology of the specific pest, determines what events must be forecast. We then examine the practical constraints on the production of these forecasts and methods for their evaluation, before suggesting how forecast services can be improved. Our central theme is that more attention must be given to the details of how forecasts are intended to improve control, if the value of migrant pest forecasts is to be improved.

Much of what we have to say about forecasting of migrant pests is also applicable to forecasting of non-migrant pests.