Figure 3. The relationships between bat activity (number of
call files per detector per night) and A) distance to nearest forest (in
meters) with high relative pest abundance (RPA; dark grey) and low RPA
(light grey); B) distance to nearest forest for high-frequency bats
(dotted purple line), mid-frequency bats (dashed red line), and
low-frequency bats (solid blue line); and C) RPA for high-, mid-, and
low-frequency bats. Ticks on the x-axis indicate spread of data points
on the x-axis. Lines indicate predicted mean; shaded areas represent
95%confidence intervals.
Bat diversity
Bat diversity (i.e., number of species per detector-night) declined
significantly with increasing distance from forests
(χ2 = 29.80, df = 1, p < 0.001), such that
bat diversity declined by 1.4 species
from
the forest edge (4.1 ± 0.2 species/night) to 4000m away from the forest
edge (2.7 ± 0.2 species/night; Fig. 4). Bat diversity was not affected
by relative pest abundance (χ2 = 0.16, df = 1, p =
0.692) and the distance to forest effect did not vary with relative pest
abundance (χ2 = 2.05, df = 1, p = 0.152). Modelled bat
diversity increased through the summer season, peaking on 25 July
(χ2 = 70.64, df = 2, p < 0.001). Bat
diversity declined with increasing precipitation (χ2 =
8.89, df = 1, p = 0.003) , but was not affected by minimum daily
temperature (χ2 = 0.05, df = 1, p = 0.831).