Back to index of herpetofauna studies

 MICROHABITAT PREFERENCES OF ANURAN LARVAE IN THE FOREST STREAMS OF BUTON

Aim:  

To investigate the microhabitat preferences of the larvae of five species of stream-breeding Anurans on Buton Island.

Objectives: 

1)     To determine the microhabitat niche preferences of individual species of tadpole.

2)     To determine the general microhabitat preferences of stream-dwelling tadpoles on Buton.

Introduction: 

Anurans have a complex life cycle (Wilbur 1983) involving both aquatic and terrestrial/aboreal stages and are thus of interest as an indicator group for the effects of disturbance. 

Stream-breeding frogs and toads are relatively uncommon worldwide and poorly studied, the largest concentration of which is in Asia (Dugeon 1980). These species are especially susceptible to the effects of disturbance, such as increased siltation of stream beds and premature drying out of ephemeral streams following deforestation and through the introduction of aquatic predator species. 

In order to investigate whether Anurans are being affected by any disturbance it is first necessary to have a broad understanding of the ecology of the groups under investigation. Buton has five known species of stream-breeding Anurans; Rana chalconata, Limonectes inflatus, Limonectes modestus, Bufo celebensis and an undescribed species of Rhacophorus, Rhacophorus sp1. Some of these are endemic to Sulawesi and all are poorly studied especially in their larval stages.  

Suitable oviposition sites for adult Anurans are dependent on the surrounding macro-habitat of an area, but the microhabitat within the streams themselves will ultimately determine where the tadpoles will be found. Very little is known about the microhabitat distribution of most Anuran larvae (Alford 1986). This study aims to look into the microhabitat preferences of the larval aquatic stage of the stream-breeding Anurans on Buton by analysing specific microhabitat variables at a number of random sites and relating these to the use or non-use of the sites by tadpoles.

Methods: 

The investigation was carried out during July and August 2002 along 8 streams on Buton within the Lambusango and Kakenauwe reserves. 

A transect along each stream was marked at 10 metre intervals. It has been suggested (Shaffer et al.) that a distance of greater than 5 metres is sufficient to ensure independence of data for the purposes of larval sampling. At each 10 metre interval a 50cm quadrat was placed in the stream a random distance from the nearside bank, the distance being determined by random number tables. 

The quadrat was divided into 400 cross points which allowed the amount of each of six microhabitat variables (plus water) to be calculated at each site by counting the number of cross hairs covering each variable. Other variables such as gradient, percentage bedrock cover and distance to the nearest riffle (rapidly flowing shallow water constantly boiling over a rocky substrate, Shaffer et al. 1974) were also measured. The site was marked using flagging tape and left overnight for any disturbance caused to settle. 

Sites were revisited on the following day and sampled for the presence of each of the five Anuran species using one minute timed sweeps of the area covered by the quadrat with a dip net. The net was held on the substrate at the bottom of the stream whilst water was forced through the net using a scoop. The substrate was also brushed thoroughly to dislodge any tadpoles. Numbers of each species found at each site were recorded for later analysis. Other observations such as presence of fish or adult Anurans were also noted.

Initial results: 

Although the results are yet to be analysed, initial observations suggest that all the stream-dwelling tadpoles from Buton prefer sites which are calm such as pools or slow flowing water (as indicated by the gradient and distance to riffle.) This observation is borne out by the general morphology of the Buton tadpoles. Lack of specialisation may be partly due to the relative age of Buton in relation to older islands such as Borneo. 

Limonectes inflatus appears to be a generalist habitat user although it is most camouflaged against sandy substrates. Rana chalconata tadpoles were most commonly found in deep leaf litter where their darker colouration is more cryptic. Bufo celebensis tadpoles appear to be water column feeders and are patchily distributed. They are conspicuous tadpoles which do not appear to hide from predators as do the other species, a behaviour possibly due to their being distasteful to predators (Gillespie 2002. personal communication).

Report

A dissertation entitled An investigation into the microhabitat preferences of Anuran larvae in the streams of Buton, Southeast Sulawesi will be produced by Lisa Wright, University of Staffordshire by April 2003