This section contains a comparison of the bird and butterfly faunas of mainland SE Sulawesi, Buton Island and Kabaena Island to determine whether diversity of the two groups in a range of habitats is correlated (17.2), a studies of niche separation in the sunbirds and flowerpeckers (17.3 & 17.4), habitat selection and distribution in the Sulawesi and Common Cicadabirds (17.5) and factors affecting the species richness of Nymphalidae butterflies in SE Asia (17.6).

From early July to the end of September, 2000, bird and butterfly species diversity and population density were recorded in a variety of natural and disturbed habitats in several areas of southeastern Sulawesi. These areas include mainland southeastern Sulawesi and the islands of Buton and Kabaena. The principal aim was to investigate factors influencing diversity in these two important groups of organisms and the resultant implications for conservation. Standardised sampling methods were employed, including visual and auditory observations of birds, and hand-netting and baited traps for butterflies, to enable comparisons between areas. Small amounts of anthropogenic habitat disturbance appeared to increase butterfly species richness but decrease bird species richness. However, preliminary results suggest that bird and butterfly species richness may show a closer correlation between different natural habitats. Island area appeared to have less influence on species richness than expected, although a significant number of bird species are not known from Kabaena island. Mistnetting of the Zoothera thrush species on Kabaena provided important data in assessing the taxonomic status of this population. Other bird species targeted for more detailed study included White-bellied Swiftlet (Collocalia esculenta) and migratory shorebirds. Initial conclusions of conservation significance are that relatively small islands such as Buton may contain higher species richness than expected, bird and butterfly diversity may be correlated under certain conditions, thus permitting their use as indicator taxa, and that many endemic species are relatively resilient to habitat disturbance.

To examine patterns of bird and butterfly species diversity in different habitats and areas and the implications for conservation.

• measure bird and butterfly species richness and turnover in different habitats and regions;
• obtain quantitative measures of population density for all species, in addition to data on microhabitat preference and behaviour, to test for general patterns in both birds and butterflies;
• test under what conditions, if any, butterfly and bird diversity are correlated, as expected if either are to be a useful biodiversity indicator taxon;
• test whether island area has a predictable effect on community species richness

Given Indonesia’s numerous island faunas, with complex biogeographic histories, understanding regional patterns of diversity, and how diversity is affected by human activities, are important for conservation of the country’s biodiversity. Birds and butterflies are two highly visible, popular, relatively well studied and diverse taxa, and both have been suggested as ideal indicator taxa for biodiversity. Although the indicator concept is now frequently criticised, few studies have been performed in tropical regions, or attempted to understand under what conditions the concept may be valid. In addition, although there are many studies on the effect of island area on total species numbers, how area influences community species richness is less well understood. Both of these subjects are of conservation relevance.

Rawa Aopa National Park: two areas were surveyed, near the southeastern entrance to the park at Lanowulu, and Aopa swamp.

Andowengga: road west of Andowengga, 35km south of Raterate.

Kendari-Wolasi road: approximately 25km from Kendari.

Moramo: 75km south of Kendari.

Labundobundo: a variety of sites within a 5km radius of Labundobundo village.

Lawele-Lagunturu: several sites around Lawele.

Watambo: swiftlet caves.

Bau-Bau area: Taman Wisata 5km north of Bau-Bau, and Hill House, Keraton.

Kaisabu: caves and trail to river.

Matanauwe: several coastal sites near Matanauwe.

Lake Togomotanu.

Lasalimu: mangrove.

Tangkeno: two sites were surveyed, the trail southeast of the village, and the river 4km south of the village.

Gunung Wumbuntandasa: the ridge (1100m).

Lengora: ridge east of village.

Dongkala: Sungai Kalumpa, 2km south of village.

Raha harbour.

Habitats surveyed include forest, grassland, freshwater, brackish and shore habitats, and open sea. Existing trails were used to sample terrestrial habitats, including riversides, ridgetops, hillsides and flat terrain. Attempts were made to locate as undisturbed habitat as possible within each study region, in addition to disturbed forest, secondary regrowth, edge and scrub.

Birds were identified by sight using Coates & Bishop (1997) and personal knowledge. Bird calls were initially assigned to species through prior knowledge, assistance of local guides, and all later confirmed by sightings. Mistnetting was occasionally used to obtain more detailed descriptions of certain taxa from specimens in the hand.

Species were identified using Fleming (1983), with notes taken from Butterflies of the South East Asian Islands. After an initial period of familiarisation most species could be identified in flight, or in the hand, but specimens of species that could not be identified, or that might represent new taxa, were collected for further study.

The walk/count method was used at virtually all sites: walking at a constant pace along a trail, all birds seen or heard were recorded, with distance from the trail estimated, and notes on number of individuals, microhabitat and behaviour recorded. Quantitative sampling was conducted from 5.30-8.00am, the period of maximum bird activity, with additional searches for specific groups (e.g., raptors) throughout the remainder of the day. Where possible, night birds were recorded while camping at sites.

A similar walk/count method was used for butterflies, but only individuals flying within an imaginary 5m x 10m box around and in front of the observer were recorded. A 250m section of trail was used as a standard sampling unit. In addition to simple observation, baited traps were used to sample seldom-seen species. At 50m intervals along a trail, one trap was hung from 12-20m up in the midstorey-subcanopy, and one hung directly below in the understorey, 0.5-1m from the ground. Up to 15 pairs of traps were placed at each site. Traps were baited with a 2 day-old fermenting banana and sugar cane mix, and, where possible, placed by 5pm of the evening before sampling to record crepuscular species. Traps were checked several times a day, and individuals removed, marked with a Sharpie marker pen on the wing underside, and released.

Studies of several specific bird taxa were conducted by Operation Wallacea:

Zoothera thrush species: notes were made on call, behaviour and habitat preference of Zoothera thrushes on Buton island, mainland Sulawesi and Kabaena island. Mistnetting was used to capture specimens of the Buton and Kabaena populations to provide detailed morphological and plumage descriptions, as well as blood samples for DNA analysis. These morphological, ecological and molecular data will be used to assess the taxonomic status of the Kabaena population.

White-bellied Swiftlet (Collocalia esculenta): visits were made to three cave nesting sites on Buton island. In each, the cave area was informally surveyed, nests were counted, and notes were taken on nest construction and stage in the breeding cycle. Informal interviews were conducted with local nest collectors, to determine the frequency of collecting visits, extent of nest collection, line of trade and economic value of the harvesting for individuals and the community.

Migratory shorebirds: following up research conducted in Ceram, observations were made at several coastal mudflat sites (Kendari, Raha, Matanauwe) to monitor migration of shorebirds. Individuals were identified and build-up in numbers noted.

Various measures will be used to compare species richness between sites and regions. Total number of species within a site will be estimated from species accumulation curves, based on number of hours, number of individuals recorded, sampling units (i..e, trail sections), trap hours or other measures of effort, fitted to several models. Accumulation curves also helps control for unequal sampling periods between sites. Indices of alpha diversity, based on number of species and individuals recorded during walk/counts and in traps, will also be used. Should the number of species recorded within a day prove a useful measure of overall species richness, this will also be used.

Approximately 210-220 species of butterflies, and approximately 200 species of birds, were recorded during the study period. In a preliminary examination of the bird species, 48 are endemic, 3 are threatened , 12 are near-threatened and c20 are migratory.

Numbers of species for each land-based study area are: Buton c. 128 bird species and c. 175 butterfly species; Kabaena c. 84 bird species and c. 120 butterfly species; and mainland Sulawesi c. 149 bird species and c. 170 butterfly species.

Although the data have yet to be analysed, some patterns are apparent on initial inspection.

• Butterfly and bird diversity do not appear to correlate across a disturbance gradient. Butterfly abundance and species diversity, at least per unit sampling effort, are both higher in selectively logged forest and along forest edge. Light gaps and forest edge often contain flowering bushes, on which many butterfly species feed, as well as providing suitable sites for perching males. However, a sharp decline in species richness in secondary growth habitats suggests that apparently elevated diversity at disturbed and edge habitats may be an artifact of increased ease of sampling, which might be detected through species accumulation curves. For birds, there appears to be a significant decline in species richness and abundance with disturbance. Although birds are more visible at the forest edge, recording species largely from calls during the dawn chorus eliminates this sampling artifact. Across natural habitat gradients, such as elevation or forest-grassland, bird and butterfly species richness appears to show a better correlation, as required if either taxon is to be used as an indicator.
• Buton island lies sufficiently far off mainland Sulawesi that certain butterfly and bird populations are subspecifically distinct, suggesting relatively little migration between the two islands. Assumed to have been connected to the mainland during previous ice ages, Buton therefore represents a forest fragment that has been isolated for a substantial period of time. Abundant research on the effect of area on species diversity suggests that diversity on Buton should be substantially lower than the adjacent mainland. However, for birds and butterflies, community species richness shows little difference between these two areas. The decreasing number of habitats with declining area may be the most important factor affecting species diversity, which suggests that small areas of habitat may be more valuable for species conservation than expected. These results, if supported by additional study, also have important implications for understanding regional patterns of species richness.
• Several other recent studies have found positive correlations between abundance, range size and number of sites occupied, and tolerance of disturbed habitats. In conservation terms, therefore, endemic species would be expected to occur in the fewest sites, at low abundance, and in more undisturbed habitats. Initial impressions from this study are that many Sulawesi endemics are tolerant of disturbed habitats, widespread and abundant. It seems possible that the factors influencing range size and rarity in continental species may be different to those in island regions; an ability to tolerate habitat disturbance may be a prerequisite for long-term survival within an island area.
• It is expected that this study will uncover a number of undescribed taxa in the butterfly fauna, since many species appear to be new records for Buton and Kabaena islands, already both known to have endemic subspecies. However, much additional museum research is required. Among the birds, an undescribed taxon showing plumage differences from the Buton and mainland Zoothera erythronota (Red-backed Thrush) was repeatedly observed on Kabaena island. Two adult males and one possible female were mistnetted to obtain detailed morphological measurements, plumage descriptions and blood samples for DNA analysis.
• Buton island’s bird fauna has been thoroughly surveyed by Operation Wallacea teams over the last four years, but one bird species was recorded for the first time, Green Pygmy Goose (Nettapus pulchellus). A single male was observed on Lake Togomotanu, representing a rare record for the Wallacean region.
• Three White-bellied Swiftlet (Collocalia esculenta) caves were visited on Buton island to investigate the impact of nest collection on bird populations. One cave had been visited before (1999) by Operation Wallacea, and numbers showed a considerable decline. The two other caves were totally cleared of nests, some of which contained eggs. This occurred quite late in the breeding season and it seems unlikely that this method of harvesting is sustainable.
• Each year Operation Wallacea carries out an informal survey of a site near Bau-Bau where a small population of the threatened Yellow-crested Cockatoo (Cacatua sulphurea) is known to breed. A single pair was observed on three different visits to the site, as in the preceding two years, although two pairs were previously documented.
• Mistnetting for particular bird species resulted in morphological and plumage data for Ashy Woodpecker (Mulleripicus fulvus), Indonesian Rusty-breasted Cuckoo (Cacomantis sepulcralis), Collared Kingfisher (Halcyon chloris), Brown-throated Sunbird (Anthreptes malacensis), Olive-backed Sunbird (Nectarinia jugularis), Black-naped Monarch (Hypothymis azurea), Pacific Swallow (Hirundo tahitica), Flyeater (Gerygone sulphurea), Sulawesi Babbler (Trichastoma celebense), Scaly-breasted Munia (Lonchura punctulata), Buff-banded and Barred Rails (Gallirallus philippensis, G. torquatus), and White-breasted Waterhen (Amaurornis phoenicurus) on Buton, Sulwesi Babbler on Kabaena, and Vinous-breasted Sparrowhawk (Accipiter rhodogaster) and Sulawesi Pygmy Kingfisher (Ceyx fallax) on mainland Sulawesi.

A dissertation entitled Species Richness and Community Composition of Nymphalidae (Lepidoptera) in Different Habitats in Sulawesi was produced by Karen Lander, University of Coventry  in March 2001.  A copy of this dissertation which was awarded a First is available at the Labundo field centre and from the Op Wall UK offices.

The following papers are being prepared for publication by Mark Catterall and/or Julia Robinson-Dean;

• A Checklist of the Birds of the Tukangbesi Islands (to be completed by December 2003)
  A Checklist of the Birds of Kabaena Island (to be completed by February 2003)
• A Checklist of the Birds of Buton Island (to be completed by April 2003)
• Notable New Bird Records for Buton Island (to be completed by November 2003)
• The Sympatric Occurrence of Common and Sulawesi Cicadabirds (to be completed by December 2003)
• The Yellow Crested Cockatoo Population in Buton Island and its’ Conservation Needs (to be completed by November 2003)
• The Owl and Nightjar Fauna of the Islands of SE Sulawesi (to be completed by March 2003)
• A New Species of Red-backed Thrush Zoothera erythronota kabaena subsp. nov (Muscicapidae: Turdidiae) from Kabaena Island, Indonesia
  Robinson-Dean, J. C. Willmott, K.R. Catterall, M.J. Kelly, D.J. Whittington, A. Phalan, B. Marples, N.M. & Boeadi, D.R.S. (2002) Forktail 18 (2002) 1 -10
  

• An Annotated Taxonomic Checklist of the Butterflies of SE Sulawesi (to be completed by May 2003)

In addition to the above, a report entitled A Comparison of Bird and Butterfly Diversity in SE Sulawesi giving the detailed results of the work outlined in the above summary will be prepared by June 2001 by Julia Robinson-Dean and Keith Willmott.

A Field Guide to the Butterflies of SE Sulawesi will be produced by Keith Willmott by May 2003. The guide will contain photographs of each species recorded from the recent survey and those which are expected might also occur together with details of their identification features. Known sites and habitat preferences will also be included. Photographs will be taken primarily from the collection made over the last few months since many are distinct subspecies. The Field Guide will be included on the Operation Wallacea web site so that it is freely available to researchers.

Four species of sunbird and three species of flowerpecker are found on Buton Island. These are the Black (Nectarinia aspasia), Olive-backed (Nectarinia jugularis), Brown-throated (Anthreptes malacensis), and Crimson (Aethopyga siparaja) Sunbirds and the Grey-sided (Dicaeum cebelicum), Yellow-sided (Diaceum aureolimbatum), and Crimson-crowned (Diceum nehrkorni) Flowerpeckers. There is little published data on these species. While the sunbirds are relatively widespread across parts of the Wallacea region and beyond, the flowerpeckers are endemic to the Sulawesi subregion. There is a degree of niche overlap between these species, similar food requirements such as invertebrates and flowers, but little is known about how the ecological niches of these species are separated. The aims of this first roject which was carrried out in June and July was to study the fine-scale habitat usage of the seven species, in order to determine to what extent niche separation occurs and to assess whether or not there are prticular associations with food plants.

Although background research into the species in question revealed a paucity of published data, information given in the field guide to the birds of the region by Coates and Bishop, together with previous research by members of the Operation Wallacea team, indicated preferences for particular habitat types. These included secondary forest and forest edge, scrub, semi-cultivated land, plantations and mangroves. The knowledge of Freddy Pangedja a wildlife guide and tracker proved invaluable in locating potential study sites. Study sites were selected to provide easy access along roads and tracks from base locations and to provide examples of all the various habitat types. The habitat types surveyed during this study were secondary forest edge, semi-cultivated/scrub and scrub/plantation.

After an initial trial period, the study method used was tosurvey 20m either side of 100m transect lines. Within this area, a survey of the vegetation was carried out to assess the proportion of cover in each of four height classes (low = 0-2m; lower-middle = 2-5m; upper-middle = 5-10m; high = over 10m). A broad description of the habitat type and situation was also made and where possible the species of flowering and fruiting trees, shrubs and climbers were also noted. A total of twenty transects were worked in three main areas; the vicinity of Bau-Bau, around Labundo Bundo in central Buton and in Maligano in the north of the island. Each of these transects was sampled up to 7 times depending on available time. Sampling was generally carried out in early morning and late afternoon to coincide with periods of greatest activity, but this was also influenced by weather conditions. As far as possible surveys were carried out during periods of fine weather in order to avoid the bias resulting from reduced activity during rainfall or strong winds. Weather conditions were noted during each survey.

The surveys ranged from 10 minutes up to 30 minutes depending on the density of the vegetation and ease of locating and identifying birds. As a measure of abundance was not being made the varying durations of the surveys should not bias the results. The data collected during each survey were as follows; bird species and sex, position in the vegetation (i.e. low, lower middle, upper middle, or high) and activity engaged in. The activities were categorised as follows; perching, preening, feeding, flying, vocalising, moving and interacting. Information on food plants used during the survey period were also recorded.

Only two of the three flowerpecker species recorded as present on Buton were noted during this survey. The Crimson-crowned Flowerpecker is known to frequent higher elevation secondary forest on mainland Sulawesi and has only been recorded twice over the last 5 years on Buton.

Preliminary analysis of the data utilising Neu’s Selection Index method show the following trends arising; across species, vertical habitat usage does not seem to be linked to the availability of that habitat level. Prefernces for upper-middle and high level vegetation were apparent. Differences in species may be revealed on further analysis. There also appears to be a trend towards utilising different levels of the haitat for different species and is probably related to the presence of food sources and the varying habits of the different species.

Grey-sided and Yellow-sided Flowerpeckers exhibited a marked preference for the fruits of the low growing tree Pipturus argenteus, to the extent that competition for the food source even when in abundant supply, was noted. Not all habitat usage of course is linked to feeding, since different habitats are often selected for perching, preening and vocalising. Further analysis of the results will be carried out to assess these associations.

A dissertation entitled Habitat Usage by Sunbirds and Flowerpeckers on Buton Island was produced in December 2000 by Karen Shelley from the University of Leeds.

Introduction.

Four species of Sunbird, Black (Nectarinia aspasia), Olive Backed (Nectarinia jugularis), Brown Throated (Anthrepetis malocensis), and Crimson (Aethopyga siparaja), three species of Flowerpeckers, Grey sided (Dicaeum celebicum), Yellow sided (Dicaeum aureolimbatum) and Crimson Crowned (Dicaeum nehrkorni) and two species of White-eyes, Pale Bellied (Zosterops consobrinosum) and Lemon Bellied are found on Buton Island. Sunbirds are spread widely across the Wallacea region and further. The Flowerpeckers are all endemic to Sulawesi as is the Pale Bellied White Eye although it is restricted to the South East of the island. The Lemon Bellied is fairly widespread throughout South East Asia.

2 Pale-bellied White-eyes ; Photo Mark Catterall

The aim of this project was to study the abundance and habitat usage of four species of Sunbirds, two species of Flowerpeckers and two species of White-eyes on Buton Island. The comparative abundance between species and the comparative abundance of species within different habitats will also be examined. The Crimson Crowned Flowerpecker was not included in the study because it has only previously been recorded twice in past studies on Buton Island. It is known to frequent areas of higher elevated secondary forest on mainland Sulawesi upwards of 1000m. The highest point in the study area was only 300m above sea level.

Nine belt transects varying between 0.5 and 2km covering a number of different habitat types were surveyed around the Labundo Bundo area. The habitat types covered included mangrove forest, coconut and banana plantation, good and degraded secondary forest, scrubland, paddy field and village environments. As none of the sites contained a continuous stretch of the same habitat each transect was mapped by hand in 10m blocks either side of the transect as far as observation was possible up to a maximum of 50m. The transects were walked at set times, 6am and 3.30pm as these were the times of peak bird activity. The surveying ws completed by walking at approximately 1km per hour. Each bird siting was recorded on a map of the transect indicating the length along and the perpendicular distance from the transect. Additional information of the bird such as microhabitat and behaviour was also recorded. This included the position of the bird in the forest profile, its activity and where possible the tree species that it was sited in. Surveys were carried out in periods of fine weather to avoid bias from adverse weather conditions such as rain and high wind. Transects were located along established communication links due to limited access to other areas.

Preliminary analysis of results prior to statistical analysis have shown that all of the species that were studied are widespread through a variety of habitat types although they do occur in varying degrees of abundance. Within the Sunbird species that were studied the most widespead appeared to be the Brown-throated and Black Sunbirds as they were found in the vast majority of the habitats surveyed and they were generally found in relatively large numbers. Olive-backed Sunbirds were quite common in their preferred habitat type which appeared to be coconut plantations and scrubland. Crimson Sunbirds were the least abundant and were usually found in forest edge habitats. Out of the two species of flowerpeckers that were studied the Grey-sided was found to be more abundant and could be seen in a wide variety of habitats. The Yellow- sided was also found in quite large numbers although not as wide spread as the Grey-sided preferring forest edge habitats although it was found in lower numbers elsewhere. Both species of white-eyes were found to be widespread although they did not appear to be as common as either the flowerpeckers or sunbirds.

• Estimated population size per kilometer square for each bird species using the belt transect method.
• The abundance of the different species in the different habitats types.
• Cluster analysis to determine if there is any relationship between the different species and how they interact with one another.

Data collected can also be compared with survey data from previous years carried out by Operation Wallacea.

A dissertation entitled Habitat Usage and Abundance of Sunbirds, Flowerpeckers and White-eyes on Buton Island was produced in March 2001 by Jack Kirby from the University of Durham. A copy of this dissertation is available at the Labundo field centre and in the UK offices.  Lenita Maria Turk from the University of Keele also produced a dissertation of the same title.

17.5 HABITAT USAGE AND POPULATION SIZE OF THE SULAWESI CICADABIRD, CORACINA MORIO

The aim of the project was to study the population size and habitat usage of the Sulawesi endemic species of cicadabird on Buton Island to assist in identifying its conservation status. A comparison with the Sulawesi cicadabirds seen on the mainland of Sulawesi will also be made.

Belt transect were used to sample the population size with the perpendicular distance from the transect to the bird being recorded for each registration. Observations on the bird’s position in the canopy and habitat preferences were made.

Field observations suggested that the Sulawesi cicadabird is territorial with birds being seen in similar locations on subsequent visits and aggressive behaviour between males being observed. The data from the transects which were repeated 8 times each will be analysed using the Common Bird Census technique developed by the British Trust for Ornithology. Data from those transects repeated less times will be analysed for the number of registrations per hectare surveyed.

A dissertation entitled Habitat Usage and Population Size of the Sulawesi Cicadabird, Coracina morio on Buton Island was produced in March 2001 by Ceinwen Stone from the University of Keele.  A copy of this dissertation which received a First is available at the Labundo field centre and at the Op Wall UK offices.

The Nymphalidae family of butterflies are the most ecologically diverse butterfly group in SE Asia and comprise a substantial proportion of the butterfly fauna of SE Sulawesi. They are characterised by having 4 legs instead of the usual 6 and are all nectar or fruit feeding species that can be attracted to baited traps. The ability to sample butterflies in a range of habitats including the forest canopy, enabled the factors affecting species distribution in this family to be examined as part of a larger survey of the butterfly fauna of SE Sulawesi.

Van Sommeren traps baited with a 2 day fermented mix of bananas and sugar cane were used to sample the Nymphalidae in a variety of habitats. Wherever possible traps were paired with one in the canopy and one sampling the lower storey. Pairs of traps were set every 50m along 250m transects. Transects were placed in primary forest, selectively logged forest, secondary growth forest, forest edge, ridge and savannah on Buton Island and Rawa Aopa National Park on the mainland of SE Sulawesi. Traps were normally set in the morning and taken down in the evening. All butterflies captured were counted and identified. At each trap site the light levels were estimated from the percentage of sky that could be seen through the canopy from a photograph. In addition the height of the vegetation as measured from the trap to each of the four corners of a 5m X 5m quadrat beneath the trap was recorded. In addition to the trap sampling, the 250m transects with the traps and a few additional transects were walked. All butterflies seen or captured by sweep netting were recorded together with the time of their sighting or capture.

The data are being analysed using an Access database. For each trap site the following data will be entered; date, trap hours, light levels, understorey vegetation height, height of trap, weather and the numbers of each sex of each species of butterfly caught. The following details for the transect count data will be recorded; date, habitat type, weather, and the time and identity of each butterfly caught.

The data will allow the most significant factors (light levels, habitat type, understorey vegetation, height of trap, weather) affecting the distribution of each species to be examined, how each of these factors affects overall species diversity and the daily flight time patterns for the commoner species.

A dissertation entitled Factors affecting the distribution of Nymphalidae butterflies in SE Sulawesi was produced by Karan Lander from the University of Coventry in March 2001. A copy of this dissertation is available at the Labundo field centre and from the UK Op Wall offices.