Date of Award


Degree Type


Degree Name

Master of Science (Environmental Management)


School of Natural Sciences


Faculty of Health, Engineering and Science

First Advisor

Associate Professor Mark Lund

Second Advisor

Dr Clint McCullough


The objective of this project was to determine if Galaxiella nigrostriata populations could belong to a metapopulation. Metapopulation theory describes how multiple populations with occasional connectivity are a ‘population of populations’. Some populations’ habitats have optimal conditions (source habitats), others experience regular extinctions (sink habitats). Connectivity allows repopulation of extinct or uninhabited habitats. Galaxiella nigrostriata occurred randomly in 11 seasonal wetlands in the Kemerton wetland complex in south-west Western Australia over a 16 year period. The wetlands did not appear to be connected.

Around 70% of wetlands on the Swan Coastal Plain in south-west WA have been filled or degraded since European settlement around 180 years ago. Of those, seasonal wetlands are at most risk from degradation. Galaxiella nigrostriata mainly live in seasonal wetlands between Augusta and Albany and in three remnant populations on the Swan Coastal Plain. They are small freshwater fish (length), aestivate in moist wetland sediments when wetlands dry and live for about one year. Seasonal wetlands and G. nigrostriata are threatened by nutrient enrichment, salinity, introduced fish, landscape modification and changes to hydroperiod by groundwater abstraction and declining rainfall.

Inundated wetlands that previously contained G. nigrostriata, and wetlands where they had not been recorded, were sampled throughout south-west WA. Fish and crayfish abundance was surveyed and water samples analysed on site and in a laboratory. Physical characteristics of each wetland and surrounding landscape were also recorded. Information about wetlands was analysed to determine if physico-chemical characteristics accounted for G. nigrostriata abundance or distribution between wetlands. Lentocorrals were then established in two Kemerton wetlands prior to inundation. They were sampled following inundation to determine how and where within a wetland G. nigrostriata entered the sediment to aestivate. Aestivation was examined to determine whether any physical features may be lacking which could inhibit population persistence. Galaxiella nigrostriata specimens from each population had morphological measurements and counts taken prior to tissue being removed for genetic analyses. Two mitochondrial DNA markers were used to investigate divergence and connectivity within and between populations and catchments.

Most wetlands were small (mean 0.6 ha), had tannin-stained water and 41% vegetation cover. All wetlands exceeded guideline values for Fe and Zn and those near agricultural land exceeded guideline values for TN and TP. However, no physico-chemical water properties or habitat features impeded G. nigrostriata abundance or distribution between wetlands. Additionally, it was thought there may be a commensal relationship between G. nigrostriata and burrowing crayfish, with G. nigrostriata using burrows to enter the sediment. No relationship was found between G. nigrostriata, crayfish or their burrows, indicating an alternative way for them to enter the sediment. Genetic research and examination of wetland positions in the landscape confirmed G. nigrostriata populations (particularly Kemerton) are part of metapopulation. This research showed populations between catchments had not connected for thousands of years but populations in wetland complexes had recent connectivity.

Management of wetlands requires investigation and monitoring of nearby wetlands which may be part of a metapopulation, and may affect population longevity of all wetlands.


Paper Location