MORPHOLOGY AND HABITAT PARTITIONING IN FRESHWATER GOBIES

Aim-

To investigate freshwater river systems and determine habitat partitioning and its relationship to the morphology of common species of freshwater fish.

Objectives- 1) To identify habitat preferences of a number of the more common species of freshwater fish.

2) To investigate the relationship between the morphology of the selected species and their preferred habitat.

Introduction

Research conducted last year identified more than 60 species freshwater and estuarine species of fish around the Labundobundo area of Buton Island. The survey was continued this year with the overall aim of compiling a species checklist for the entire island. In addition, this project was developed to gain an understanding of the ecology of the freshwater fish communities.

Rivers provide an array of different habitats for fish populations and adaptations to these different habitats provides the basis for complex fish communities to develop. Very little is known about the structure of these communities in the rivers in this region so it is important that exploration has begun into the factors that influence the species distribution.

Habitat partitioning in rivers is often determined by physical factors such as substrate types, flow rates and oxygen concentrations. The graded ability of different fish species to deal with local environmental conditions gives selective advantages to certain fish ‘types’ in particular environments. This reduces resource competition and is often responsible for diversification and speciation. An understanding of the principles of hydrodynamics linked with ecomorphology can be used to explain why certain species tend to exist in particular habitats. Body size, shape and fin morphology can be used to catagorise fish and hypothesize which habitat and lifestyle they are adapted to.

Aspect ratios calculated from fin span and area are a good indication of how particular species move around their environment, in a similar way to birds and bats. Control methods of buoyancy provides another indicator of habitat specialization.

By compiling information on morphological characteristics and body proportions and comparing these to the habitat locations in which particular species are most abundant, an indication of the partitioning in the rivers habitats can be developed that could lead to a greater understanding of the freshwater ecosystems.

Methodology

The study was conducted over 4 weeks during July and August 2002 around the Kakenauwe and Lambasango forest reserve sites. A wide range of rivers types were visited to provide contrasting habitats.

Fish samples were collected by electrofishing using portable equipment that generates a potential of up to 400 volts between the fishing electrodes. The fish shocker was powered by a 24 volt battery and set to generate pulsed DC to the electrodes. To standardise fishing effort between each site, one operator fished continuously with two others either side with nets. Fishing occurred at a rate of about 1-2metres per minute. At each location the river was divided into sections with characteristic flow patterns: pools, riffles, glides. As each site was fished the fish catch was kept in individually numbered pots or bags.

When possible the position of each location was recorded using GPS. Temperature, pH, water conductivity and dissolved O² were measured using portable meters. A general description of each location was made that included depths, widths and lengths of the watercourse, the degree of disturbance/land use of the surroundings, canopy cover and macro invertebrate species observed. The substrate was characterised using the Wentworth Classification of substrate particle substratum size and was used to determine flow rates.

Fish were anaesthetised in benzocaine and the Standard Length of all the fish in each sample recorded. Fish were identified to species where possible (Kottelat et al., 1993).

Five common species were chosen for further study, these were Rhyachcithys aspro, Eleotris melanosoma, Ophieleotris aporos, Glossogobius celebes and Sicyoterius micrusus.

These five species were some of the most abundant species in the samples and were visually rather different in appearance. Approximately 50 individuals of each species were collected and fixed in 10% formalin.

The following morphological measurements were made on each specimen (mm):

Total length,
Standard length,
Body depth,
Body width,
Depth of caudal peduncle,
Anterior dorsal fin depth and base length,
Posterior dorsal fin depth ad bas length,
Anal fin depth and base length,
Pectoral fin length and width,
Pelvic fin length,
Sucker diameter (when present),
Eye diameter,
Snout length,
Jaw width.

Additional data was collected on the preserved specimens:

Wet weight (g),
Transverse body section,
Caudal fin type
Number of dorsal fins,
Dorsal fin structure,
Presence of suckers,
Mouth position.

One of each species was dissected to confirm or deny the presence of a swim bladder. The area of the pectoral fins and the sucker were appropriate was recorded in 20 specimens of each species. This was done by drawing around the fin/sucker onto paper and later a computer scanning program to determine the area. This will allow aspect ratios to be calculated and comparisons to be made between species.

Initial Results

In the first week of the 4 week survey, 5 species of fish were identified as being reasonably common in the watercourses around the Labundobundo area and suitable for study. Approximately 35 sites were fished, including 14 pools, 4 glides and 17 riffles. 50 Eleotris melanosoma and Glossogobius celebius, 30-35 Ophieleotris aporos Sicyopterius micrusus and 20 Rhyachickthys aspro were individually measured.

Early observations suggest that fish caught in pools were larger than from riffles even within species, signifying potential life-cycle movements that would be interesting in the final analysis. The fact that the Superfamily Gobioidei all possess a ventral sucker formed from the fusion of the paired pelvic fins provides members of this group the ability to adhere to the substratum and withstand strong currents. Initial findings suggest that Glossogobius celebius and Sicyopterius micrusus, which are both gobies were found more abundantly in riffles areas. In contrast other species without suckers such as Eleotris melanosoma and Ophieleotris aporos seem to be more frequent in the open pools. The fifth species Rhyachickthys aspro that does not possess a sucker was nevertheless commonly found in riffle areas. Its large pectoral fins and streamlined body shape seems to provide a different solution to the problem of remaining in strong currents. Statistical analysis of the data will confirm the significance of these observations.

Final report title

A dissertation entitled Relationships between morphology and habitat partitioning in freshwater gobies and related species will be produced by Tom Wakefield, University of Liverpool by May 2003.

References

A J. Whitten, M. Kottelat, S. Wirjoatmodjo, S. Nurani Kartikasari. (1993). Freshwater fishes of Western Indonesia and Sulawesi. Periplus Editions (HK) LTD.