The hierarchical effects of anthropogenic impacts and natural filters on the distribution of freshwater mussels in the Hawkesbury-Nepean River, NSW, Australia

Abstract

In comparison with North America, there is a poor level of taxonomic diversity in the Australian freshwater mussel fauna. Seventeen endemic hyriid species have been recorded in this country, while over 30 species have been recorded from one reach of a US river. While there have been no extinctions recorded among Australian mussel species, there are several species under threat, and mussel populations in many river systems are in serious decline. One of these catchments, the Hawkesbury-Nepean River, is considered to be among the most highly impacted of the eastern coastal rivers. Mussel populations in this river are believed to have declined in density and distribution, but to date this observation has been based on anecdotal, rather than scientific, evidence. This study aims to address the dearth of knowledge regarding population ecology for hyriid species by compiling a detailed assessment of mussel species, densities and sizes for a series of 100 sites throughout the greater Hawkesbury-Nepean River. Mussel populations were sampled concurrently with a range of physical habitat variables to identify natural and anthropogenic factors that may be important for defining the observed mussel assemblage structure. The results from these assessments were used to develop a hierarchical model that ranked the relative importance of these physical habitat variables in determining mussel distribution. An initial survey of sites in the Nepean River indicated that a high level of riparian disturbance was strongly related with decreased mussel density, and often with the apparent absence of mussels from many sites. Distribution of species and density of mussel assemblages were strongly associated with geomorphic reach type, so that different reaches supported different suites of mussel species. Presence of mussels was strongly associated with annual maximum dissolved oxygen, minimum total nitrogen, and minimum conductivity. Each species was associated with different annual patterns of water quality. Velesunio ambiguus were most frequently associated with maximum dissolved oxygen, and minimum conductivity and total phosphorus. In contrast, Hyridella depressa were not linked with dissolved oxygen patterns, but were negatively affected by turbidity and total phosphorus. Hyridella australis were associated with annual maximum dissolved oxygen, and not apparently affected by other aspects of water quality. Densities of mussels were compared with single samples for water quality at a second set of sites. Data from this study suggested that at higher densities, mussel populations impacted water quality, rather than water quality impacting mussel populations. The results indicated that at higher densities, mussels were able to favourably impact on water quality, reducing levels of phosphorus in the water column. Both small and large impoundments were seen to influence the composition of mussel assemblages. The presence of a large impoundment, mid catchment, was strongly associated with the absence of mussels upstream, and the presence of lower density populations. For small impoundments, weir height and geomorphic reach type were linked with differences in species distribution and densities. Velesunio ambiguus were present in greater densities in pools above weirs, while H. depressa and H. australis were present in higher numbers below weirs. Differences in distribution were also observed among different size classes of mussels. Small mussels were virtually absent above the large impoundment. Below the large impoundment, they were mostly found below small weirs located in less impacted sandstone gorge reaches. Within reaches, mussel distribution was most frequently related with coarse sand and the presence of a roughness element characteristic for the geomorphic reach type. Small mussels were associated with boulder stabilised habitats, and medium mussels with cobble based habitats. Large mussels were rarely associated with any type of roughness element. While each species showed some microhabitat preference, this was generally associated with the geomorphic reach type, rather than a particular aspect of substrate conditions. Within a hierarchy, catchment scale variables, including position in relation to a major impoundment, tended to dominate the delineation of mussel presence. Some aspects of water quality were associated with mussel density at a reach scale. The level of riparian disturbance, a reach scale factor, was strongly associated with mussel density. In general, factors associated with larger scales for the river were more influential in determining mussel presence and mussel density. However, the exact nature of these factors varied among species. When compared with northern hemisphere mussel fauna, Australian hyriid species were affected by a similar range of impacting factors. While the distribution of mussel species throughout the catchment was originally defined by natural filters, much of their present distribution patterns can be directly linked with anthropogenic changes within this highly impacted river system. Many of these anthropogenic impacts operate at catchment and reach scales, with the result that suitable habitats may not be colonised in some reaches of the river.

Date of Award	2006
Original language	English