The AOR-index is by necessity a relative index; hence, a base-line abundance and distribution of animals must be provided. In
this case the baseline assumed for all farming comparison sce-narios was extensive conventional cattle farming on 100% of the
landscape. This farm type was chosen as the baseline because it
did not exhibit extreme impacts on any of the species/landscape
combinations used and therefore resulted in a visually balanced
set of relative outputs. The effect of landscape on predicted popu-lation density in this baseline was large, between 123% and 858%
( Table 3 ). In this case it should be noted that hare and partridge
abundance in modern Danish landscapes is very low. This means
that a small change in absolute densities for these species can result
in a large AOR-index change in the simulations considered here. In
Præstø both partridge and hare populations were below the pop-ulation density at which the population would be considered as
being viable in the long term. Partridge populations were similarly
low in Bjerringbro. These differences are absolute differences, and