Back to index of 2002 research reports
ABUNDANCE AND DIVERSITY OF BIRDS IN FORESTS WITH DIFFERENT LEVELS OF DISTURBANCE
Following
previous studies on the biodiversity and distribution of bird species on Buton
island (Catterall 1999, 2000 and in prep., Robinson Dean in prep.), and
preliminary work on bird species and habitat associations (Reid 2002), a study
was undertaken to investigate the relationship between human disturbance to
forest areas and the bird communities present within those areas.
The
disturbance and removal of forest areas on the neighbouring island of Muna has
had dramatic effects on the biodiversity there (Catterall pers. com.). A closer
investigation of the effects of human disturbance on the Kakenauwe and
Lambusanga forests is desirable, as these areas may form the basis of a new
National Park. It is important to collect baseline data about the species
presence and abundance within different forest areas in order to allow
monitoring during the transition period from Forest Reserve to National Park.
Apparently minor changes in land use could have dramatic effects on the bird
communities. By recording baseline information, it will be possible to identify
such changes more easily.
Operation
Wallacea has a forest base at Labundobundo. There are three satellite study
grids (1km2) in the surrounding area, which have been marked to allow
transects to be conducted in East-West and North-South directions. These study
grids are conveniently located at different distances from population centres,
and are therefore subject to differing levels of human disturbance.
Three
study grids (1km2) were marked in an 11 x 11 matrix of 100m
transects. In order to allow independence of data, point counts were conducted
along alternate transects, at intervals of 200m. This effectively generated a 6
x 6 grid for survey purposes. Point counts were conducted between 05:30 and
08:00 along a 100m transect line during the months of July and August 2002. Each
count lasted for 10 minutes, during which time all species identifiable by sight
or call were recorded. A settling period of two minutes was allowed before
counting began to reduce observer influence on the counts. The distance from the
observer was recorded (if seen) or estimated (if heard), along with the number
of individuals at that location. For some social species, estimates of average
group size were collected independently; this value was substituted when it was
not possible to determine group size otherwise. Data on species, number and
distance from observer were entered into a software package called DISTANCE®,
which generated population density estimates. The dawn chorus is known to have
peaks and troughs of singing activity, so each transect line was conducted in
both directions (e.g. East to West, and West to East). In order to predict more
accurately the peak activities of each species within the forest grids, nine
static counts were undertaken on each grid. These static counts monitored only
the bird song activity between 05:30 and 08:00. The static counts lasted for
only five minutes, but were conducted every 15 mins at the point. Once a peak
activity ‘window’ was identified for a species, only counts within this
window were used for calculation of population densities.
In
order to ensure inter-observer consistency, each observer worked alongside a
more experienced observer for two weeks, before recording alone. If observers
were unable to identify calls at a count, these calls were described and
recorded using a microphone and mini-disc player to compare with a library of
known bird calls.
Initial
data has shown similarities and differences between the different grids. Some
species known to populate disturbed forest appear to have higher populations in
the more disturbed habitat (e.g. Green Imperial Pigeon), whereas others appear
to show relatively little change (e.g. Sulawesi Babbler). Some species that are
thought to be sensitive to disturbance (e.g. White-bellied Imperial Pigeon)
appear to show reduced populations in the more disturbed forest area.
The
picture is not a clear one, but indications are that some species are much more
sensitive to disturbance than others. It is hoped to be able to catalogue these
species more accurately after further analysis.
A
paper entitled The
Effect of Forest Disturbance on the Bird Communities of the Lambusango and
Kakenauwe Forest Reserves on Buton Island, Sulawesi Tenggara will be
produced by Dr Dave Kelly and Dr Nicola Marples by April 2003..
Catterall,
M. (1999). A checklist of the birds of Buton Island with notes on their
identification features. Operation Wallacea report. Available on the Operation
Wallacea web site.
Catterall,
M, (in prep.). The owl and nightjar fauna of the islands of S.E. Sulawesi.
Operation Wallacea report. To be available on the Operation Wallacea web site.
Reid,
N. (2002). Population diversity and densities of bird species and their habitat
utilisation of disturbed primary rainforest, Buton Island, Indonesia. B.Sc.
Honours Project. Medical Biology Centre, Queen’s University of Belfast, 97
Lisburn Road, Belfast, Co. Down, BT9 7BL Northern Ireland.
Robinson
Dean (in prep.). Abundance, range size and habitat preference of the birds and
butterflies of S.E. Sulawesi. Operation Wallacea report. To be available on the
Operation Wallacea web site.