Mouse Plagues in South Australian Cereal-Growing Areas. Ii. An Empirical Model for Prediction of Plagues.

Abstract

A model for predicting the occurrence of mouse plagues is presented based on rainfall and crop yield. The model shows that the probability of a plague occurring in the following autumn is raised if: grain yield in the current season is high but was low 2 years previously; autumn rainfall in the current year is high; and Nov. rainfall is high and Oct. rainfall is low in the current year. The model was developed using logistic regression analysis of data from 6 sites and tested using independent data from a further 6 sites. Using all data, the model accounts for 41% of the deviance in plague occurrences. Hypotheses proposed by previous authors attributing the occurrence of mouse plagues to good cropping seasons, unusually high autumn rainfall and prior drought are mostly supported. Spring rain is a previously unrecognized contributing factor. Possible application of the model for preventive control of mouse damage is discussed.