Providing evidence to improve practice

Individual study: The use of constructed and natural ponds by frogs in agricultural land of the Southern Tablelands, New South Wales, Australia

Published source details

Hazell D., Hero J., Lindenmayer D. & Cunningham R. (2004) A comparison of constructed and natural habitat for frog conservation in an Australian agricultural landscape. Biological Conservation, 119, 61-71


With the advent of agriculture and other land-use changes in Australia, many natural wetlands have been destroyed of severely degraded. In response, many artificial ponds (farm dams) have been created as a conservation intervention to provide habitat for aquatic species. However, farm dams are suggested to be less complex in structure and biological diversity. Consequently, in this study the diversity of frogs was compared between natural wetlands and constructed ponds.


Study ponds: In agricultural land within the Upper Shoalhaven catchments in the South Tablelands of New South Wales (Australia) the diversity of anurans (frogs and toads) was compared between natural wetlands and man-made ponds. Twenty-two natural pond systems were located, including ox-bow lakes (billabongs), chains of intermittently connected deep ponds (chain-of-ponds), and other wetlands. In cases where there were multiple ponds, one was selected randomly. Each natural pond was matched with a farm dam pond (>10 years old) that had comparable management (i.e. level of stock access and stock type) and landscape features, giving a total of twenty-two farm ponds. Natural and constructed pond pairs were separated by between 1 and 3 km.

Frog surveys: Sites were surveyed in the austral spring/summer of 1999/2000, once spring breeding species became active. Surveying was ceased on nights when the temperature dropped below 5°C, the mean minimum daily spring temperature. Each site was surveyed on two nights, with natural and constructed pond pairs being censused on the same night. Audio strip transect and visual transect sampling was used. Species recorded by either of these methods (sse below) were recorded as present, those not detected were recorded as absent.

Audio strip transects: A 50 m transects was conducted for 20 min along the drainage line with the waterbody as the central point, therefore encompassing terrestrial, semi-aquatic, and aquatic microhabitats. The number of frogs calling were categorised as: 0 = 0 frogs; 1 = 1-5 frogs; 2 = 6-10 frogs; 3 = 11-20 frogs; 4 = 21-50 frogs; and 5 = 50+ frogs.

Visual transects: Four 5 x 2 m transects were censused. These were separated between four microhabitats: areas with bare ground; areas with emergent vegetation in shallow water; areas with tussock cover; and areas with shrub/tree canopy cover. If all four were present, one transect was conducted per microhabitat. If there were less than four, transects were divided depending on how common each microhabitat was.


A total of thirteen anuran species were recorded, with each study pond having at least one species present. There was a total of 12 species in natural ponds and of 11 species in farm dams. An average of 4.5 (±0.4 standard error) species were recorded in natural ponds and an average of 5.0 (± 0.3 standard error) species were recorded in farm ponds. Table 1 (attached) shows the number of natural and constructed ponds at which each species was recorded.

Overall, the results suggest that the man-made ponds are equally attractive as natural waterbodies as breeding sites for the species present in the study area. Only one species, the striped marsh frog Limnodynastes peronii went against this trend with only six man-made waterbodies occupied in comparison with 14 natural ponds (of similar dimensions) occupied.

Note: If using or referring to this published study, please read and quote the original paper. This is available from Please do not quote as a case as this is for previously unpublished work only.