- Title
- The application of AUSRIVAS and SIGNAL2 for assessing the impacts of wastewater treatment plants on water quality in the lower Hunter Valley, New South Wales
- Creator
- Franks, Catherine Ann
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2016
- Description
- Masters Research - Master of Philosophy (MPhil)
- Description
- Two commonly used biological water quality assessment methods in Australia are SIGNAL2 and AUSRIVAS. Both rely on taxonomic data from one macroinvertebrate sample per site to be assessed to classify water quality, and it is thought that one sample is unlikely to be adequate to reliably assess water quality in all types of fluvial system. This study set out to investigate the variability associated with these single-sample methods when applied to streams in the lower Hunter River catchment, NSW, when monitoring the impacts on water quality of discharge from sewage treatment plants. The investigation is somewhat different to previous studies using these biotic indices since no previous research has been undertaken to assess the variability of these single-sample methods for reliably assessing water quality in a particular fluvial system within a robust statistical framework. In this case, the creeks of the lower Hunter region are seasonally variable with respect to flow, have been degraded by past land use practices, have the additional load of treated effluent from wastewater treatment plants and have not previously been the subject of investigation using these methods. Hunter Water Corporation operates several inland wastewater treatment plants which discharge treated effluent into the creeks of the lower Hunter Valley, New South Wales, Australia. Currently, the water quality monitoring program involves the periodic measurement of various water quality parameters; however incorporating biological monitoring into this program would give a more holistic assessment of water quality and ecosystem health. The within-site variability of SIGNAL2 scores and AUSRIVAS OE50 values was investigated by collecting ten macroinvertebrate samples from sites upstream and downstream of the wastewater treatment plants at Cessnock, Kurri Kurri, Branxton and Farley during Spring 2002, Autumn 2003, Spring 2003 and Autumn 2004. The percentage chances of misclassifying the water quality at a particular site when using a single sample compared to a ten sample mean classification were calculated, and the consequences of this variability for impact assessments was investigated. This study demonstrates that one sample is insufficient to reliably assess water quality, particularly in the waterways such as those of the lower Hunter region where macroinvertebrate taxa may be less abundant and not uniformly distributed as a result of degraded water quality. Applying the multiple sample strategies using ten macroinvertebrate samples, no evidence of ongoing impacts on water quality as a result of wastewater treatment plant effluent discharge was detected using SIGNAL2 and AUSRIVAS at Cessnock, Branxton and Farley. However, at Kurri Kurri there was an observable deterioration in water quality between the upstream and downstream sites. The optimal numbers of macroinvertebrate samples required to produce more reliable assessments of water quality and creek health were determined for SIGNAL2 and AUSRIVAS, so that these methods could be usefully incorporated into Hunter Water Corporation’s ongoing water quality monitoring program. It was determined that four samples per site allowed for the detection of a difference in mean SIGNAL2 scores of one SIGNAL2 unit between sites (power = 0.9) for the SIGNAL2 method, and six samples per site were recommended to detect a difference in mean OE50 value of 0.2 OE50 units (power = 0.9) between sites using AUSRIVAS. Therefore, in the creeks of the lower Hunter Valley it is recommended that multiple macroinvertebrate samples are necessary to produce a statistically valid and reliable biotic index, and that the single sample methods of SIGNAL2 and AUSRIVAS are ineffective. These results also demonstrate that it is necessary to derive the optimum sample numbers to obtain statistically significant outcomes using these biotic indices. The importance of these rapid biological assessment techniques is unquestioned, but the implementation of them must produce scientifically robust and defendable outcomes. As an outcome from this investigation it is recommended at the use of SIGNAL2 and AUSRIVAS should include a multi-sampling strategy, not a single sample, particularly for waterways that have been disturbed by industrial, urban or agricultural impacts. Furthermore, where waterways represent different systems, it is necessary to conduct a preliminary study to statistically determine the validity of the number of samples necessary to characterise a site. The definition of different systems is based on a number of landscape and riverine characteristics; for example the streams (rivers) in the current investigation have relatively restricted catchments, but are connected to a larger river system.
- Subject
- AUSRIVAS models; Australia; effulent discharge; impact assessment; macroinvertebrates; output variability; rapid biological assessment
- Identifier
- http://hdl.handle.net/1959.13/1316819
- Identifier
- uon:23274
- Rights
- Copyright 2016 Catherine Ann Franks
- Language
- eng
- Full Text
- Hits: 884
- Visitors: 1758
- Downloads: 567
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Abstract | 201 KB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Thesis | 3 MB | Adobe Acrobat PDF | View Details Download |