The fixed-width strip transect method is increasingly becoming an important bird sampling
technique in Australia. In this survey, different sampling durations and strip widths were evaluated for
the sampling of birds in Yellow Box woodland remnants in the ACT.

The survey showed that the choice of sampling duration and strip width had a significant effect on
the estimation of bird abundance and species diversity using the fixed-width strip transect method in
Yellow Box woodland remnants. Different measures of bird abundance and species diversity were also
found to vary in sensitivity with changes in strip width. Use of the logarithm link function to analyse
bird counts further demonstrated that the appropriate strip width will depend upon the flocking
behaviour of the birds. It is evident from this survey that there is a need for more intensive surveys to
develop and validate the fixed-width strip transect method for the sampling of birds in savannah
eucalypt woodlands. Until that is done, it is inappropriate to compare results between studies which
employ the fixed-width strip transect method, but use different sampling durations and strip widths.

Bird species richness and abundance were recorded along 28 fixed-width strip transects located
within 10 study sites dominated by Yellow Box Eucalyptus melliodora and Blakely's Red Gum
E. blakelyi. This was conducted across four seasons in 1994. This paper reports the seasonal variation
in bird species composition and abundance observed in these study sites. The results suggest that
Yellow Box-Blakely's Red Gum woodland remnants serve as (1) important bird refuges within the
landscape matrix in the lowland of the Australian Capital Territory; (2) wintering sites critical to partial
migrants and some resident species; (3) breeding sites for both residents and summer migrants; and
(4) refuges for nomadic species during periods of extreme drought. In addition, they demonstrated that
birds respond differently to their habitats at different times of the year and the importance of basing
management decisions upon information from more than one season.

Data on bird species diversity, bird density, species diversity of various foraging groups and of
honeyeaters were analysed in relation to mean dieback scores of woodland remnants in the Australian
Capital Territory. Bird species diversity and bird density were found to be significantly reduced as
dieback increased. The reduction in bird species diversity can be accounted for by the declines in
subcanopy bird species diversity and insectivorous bird species diversity, the latter being restricted
most to the foliage insectivores, ground insectivores and bark insectivores. Honeyeaters were also
shown to decline significantly with an increase in dieback. Reasons for the decline in bird species
diversity with increased dieback, the dieback-insects-birds cycle and the implications of dieback for
the conservation of bird species are discussed in relation_ to these observations.