|
Indicator description Assemblages of macroinvertebrates in rivers as assessed by AUSRIVAS (Australian River Assessment Scheme) sampling protocols and computer models. Why is it indicative "River health" is considered important in reporting on the state of the environment as it is a concept that goes beyond suitability of water for particular uses and integrates a range of ecosystem values and functions. It complements more traditional assessments of water quality by introducing an ecosystem perspective. Macroinvertebrate assemblages are used because they integrate many aspects of the 'health' of streams and rivers. What does the data show - As the primary objective of the AUSRIVAS river health assessment was to develop and test river health models, the study focused upon sites that were considered to be of regional concern, and test sites were generally chosen because a problem was known or suspected at the site. Therefore, not all rivers and streams were assessed across the state. The large majority of regions in the State, however, have been sampled with the exception of the South West Wilderness area.
- In many catchments of the State, insufficient numbers of reference sites remaining in pristine condition were available to provide a context for the AUSRIVAS river health assessment. Instead, sites in 'least disturbed' areas of the catchment were selected in relation to known or perceived impacts.
- When both AUSRIVAS reference and test sites are considered (a total of 473 sites sampled between 1994-1999), the macroinvertebrate communities (as indicators of aquatic health) in 30% of the riffle habitat sites and 20% of the edgewater habitat sites were found to be impaired.
- Limitations exist in applying AUSRIVAS models in heavily degraded catchments (e.g. through urbanisation and/or agriculture) where sites of reference condition are not readily available.
- The assessment of test sites that were dominated by a single impact (forestry, agriculture, dam, or mining), found that approximately 30-50% of sites were impaired.
- A large proportion of dam-affected sites were substantially impaired. However, recovery in biotic condition in terms of macroinvertebrate taxa present tended to be rapid moving downstream from the dam.
- Acid mine drainage has a significant impact on the water quality and therefore habitat suitability in affected rivers particularly on the west coast.
- River health scores indicated impairment in rivers adjacent to intensive agriculture and seasonal differences in river health impairment may reflect the seasonal variation in the degree of disturbance and/or changes in hydrology at the sites.
- Although river health is assessed using different season and habitat models, it is suggested that AUSRIVAS model O/E outputs for riffle habitats may give a better representation of the condition of the river than the edgewater habitat. With respect to this, approximately 45-50% of riffle habitat sites assessed were found to be impaired across the State.
- It is also considered that the current collection of environmental data for the habitat assessment component of AUSRIVAS doesn't provide a geomorphic context and lacks linkages between biological responses and larger scale catchment attributes.
Data AUSRIVAS includes a set of predictive models that are relevant to particular regions, seasons (spring, autumn and both combined), and habitats for river macroinvertebrates (riffle and edgewater). The models predict the macroinvertebrate community that should be present in specific streams under reference conditions or when a river is considered healthy. It does this by comparing a test site with a group of reference sites that have similar physical and chemical characteristics to those found at the test site. Two complementary ratio indices, O/E FAMILIES and O/E SIGNAL, are used to summarise outputs from the analysis of macroinvertebrate survey data using the AUSRIVAS predictive models. In reporting on river health, the O/E FAMILIES index is primarily used, while the O/E SIGNAL index assists with the broad interpretation of the causes for any apparent site impairment. The values of each index are divided into categories or bands, which are described in the table below. The width of the bands is based on the distribution of index values for the reference sites in a particular model. AUSRIVAS O/E band divisions
 |
Band Label |
Band Name |
Comments |
|
| |
|
O/E Families |
O/E SIGNAL |
|
X |
Richer than reference |
Very high occurrence of expected families
- Biodiversity 'hot spot'; or
- Mild nutrient enrichment
|
Appreciably greater SIGNAL value than expected
- Site of exceptionally high water quality, or
- Impact unrelated to organic/nutrient/sediment pollution
|
|
A |
Reference |
Families found similar to those expected
- Water and habitat roughly equivalent to those at reference sites, or
- Impact on water or habitat quality of a type that does not result in loss of families
|
SIGNAL value similar to expected value
- Water quality roughly equivalent to those at reference sites, or
- Impact on water quality of a type that does not result in loss of families sensitive to organic/nutrient/sediment pollution
|
|
B |
Below reference |
Several expected families not found
- Impairment of either water quality, habitat or both.
|
Appreciably lower SIGNAL value than expected
- Impairment of water quality
|
|
C |
Well below reference |
Many expected families not found
- Substantial impairment of water and/or habitat quality
|
Much lower SIGNAL value than expected
- Substantial impairment to water quality
|
|
D |
Impoverished |
Very few of the expected families found
- Severe impairment of water and/or habitat quality
|
Very low SIGNAL value
- Severe impairment to water quality
|
|
The names of the bands refer to the relationship of the index value at a test site to the reference condition (Band A). Under comments for each index, an explanation of the band is stated first, followed by possible interpretations as dot-points. Source: Barmuta, et. al. (1998)
Six different AUSRIVAS models have been applied for reporting on river health in Tasmania. Each model has a different range of O/E FAMILIES index values representing the separate bands, which are described in the table below. Bands and O/E value ranges for AUSRIVAS models
|
| |
O/E Taxa Bands |
|
Model |
Band X |
Band A |
Band B |
Band C |
Band D |
|
TAS Autumn Edge |
>1.19 |
1.19-0.82 |
0.81-0.45 |
0.44-0.08 |
0.07-0 |
|
TAS Autumn Riffle |
>1.14 |
1.14-0.86 |
0.85-0.58 |
0.57-0.30 |
0.29-0 |
|
TAS Spring Edge |
>1.18 |
1.18-0.83 |
0.82-0.48 |
0.47-0.13 |
0.12-0 |
|
TAS Spring Riffle |
>1.15 |
1.15-0.86 |
0.85-0.57 |
0.56-0.28 |
0.27-0 |
|
Combined Edge |
>1.15 |
1.15-0.85 |
0.84-0.55 |
0.54-0.25 |
0.24-0 |
|
Combined Riffle |
>1.13 |
1.13-0.88 |
0.87-0.63 |
0.62-0.38 |
0.37-0 |
|
Source: Krasnicki et. al. 2001
The most recent river health rating available for each site across the state ranges between 1994-99, with few sites having been assessed every year. An overview of river health for all sites (reference and test) assessed across the state is presented in the riffle and edgewater river health ratings figures below. A summary of river health rating percentages for the reference and test sites is available.
Specific test sites were chosen for assessment between 1994-1999, because they were considered to be of regional management concern. Test sites that were dominated by a single impact (forestry, agriculture, dam, or mining) were assessed and, overall, approximately 30-50% of the sites were found to be impaired. - The combined season edgewater and riffle habitat models found that 33.8% and 45.2% of the sites were impaired, respectively.
- The autumn edgewater and riffle habitat models deemed 34.6% and 41.7% of the sites to be impaired respectively.
- The spring edgewater and riffle habitat models found that 33.5% and 53.1% of the total test sites assessed were impaired respectively.
The variation in ratings between riffle and edgewater habitats at test sites is possibly because a greater structural diversity exists within edgewater macroinvertebrate communities than in riffle communities resulting in a higher variation of edgewater fauna sampled; more pollution tolerant fauna may be harboured in edgewaters compared to riffles; and conversely, riffles may be effected more by flow related impacts (Krasnicki et al. 2001). Different impact types appeared to have varied effects upon the macroinvertebrate communities of the test sites assessed, as depicted by the O/E values in the figure below. 
- Sites located in mature forestry pine plantations tended to be assessed as unimpaired probably because they were located where little or no significant recent history of forestry operations had occurred.
- A large proportion of dam-affected sites were substantially impaired with losses of up to 80% taxa. Riffle habitat indicated greater impairment than edgewater habitat, possibly because riffle taxa are more susceptible to daily or seasonal fluctuations in flow than edgewater taxa. Recovery in biotic condition in terms of macroinvertebrate diversity tended to be rapid moving downstream from the impacting dam, probably influenced by the degree to which the natural hydrological regime is restored from unregulated tributary flows (Krasnicki et al. 2001).
- Sites impacted by mining were assessed as impaired. Riffle habitats were more degraded than edgewater habitats possibly because of a better food source and microhabitat availability from higher organic material inputs into edgewaters than riffles. The majority of sites on the west coast were impacted by acid mine drainage.
- Sites indicated impairment from intensive agriculture and seasonal differences in the level of impairment may reflect the seasonal variation in the degree of disturbance and/or changes in hydrology at the sites.
- More detailed information is available in the Tasmanian Australia Wide Assessment of River Health report (Krasnicki et al. 2001).
Information on river health in specific river catchments using AUSRIVAS can be obtained in the State of Rivers Reports that have been produced for select catchments across the State. Related Indicators
Acknowledgment Core Inland Waters and Wetlands Indicator IW 13 (ANZECC 2000), Tom Krasnicki and Martin Read |
