Use and abuse of wildlife models for determining habitat requirements of forest fauna

Summary

Taylor and Haseler's (1993) attempt at estimating habitat requirements of forest fauna and habitat availability is reviewed. Their method used statistical models to relate forest characteristics to the number of hollows in trees. The models were used to predict the number of potential nest trees that are available to four hollow-nesting bird species and to predict the impact of logging on these species. The method did not account for the potential error of the predictions, and nor did it account for uncertainty associated with impacts of management. Methods that may help to account for these sources of error and uncertainty are described.