Supervising Scientist Environmental Monitoring Program

Search:

Explanatory notes on ecosystem level responses

Monitoring using macroinvertebrate community structure

Macroinvertebrate communities have been sampled from a number of sites in Magela Creek at the end of significant wet season flows, each year from 1988 to the present. The design and methodology have been gradually refined over this period (changes are described in the Supervising Scientist Annual Report 2003-04 (pp 32-34), section 2.2.3. The design is now a balanced one comprising upstream and downstream sites at two 'exposed' streams (Gulungul and Magela Creeks) and two control streams (Burdulba and Nourlangie Creeks).

Samples were collected from each site at the end of each wet season, i.e. between April and May. For each sampling occasion and for each pair of sites for a particular stream, a dissimilarity index is calculated. This index is a measure of the extent to which macroinvertebrate communities of the two sites differ from one another. A value of 'zero' indicates macroinvertebrate communities identical in structure while a value of 'one' indicates totally dissimilar communities, sharing no common taxa.

Disturbed sites may be associated with significantly 'higher' dissimilarity values compared with undisturbed sites. Analysis of the macroinvertebrate data set from 1988 to 2007 is shown in figure 1. This figure plots the paired-site dissimilarity values using family-level (log-transformed) data, for the two 'exposed' streams and the two 'control' streams.

Inferences that may be drawn from the data shown in Figure 1 are weakened because there are no pre-mining (pre-1980) data upon which to assess whether or not significant changes have occurred as a consequence of mining. Notwithstanding, the plots show that the mean dissimilarity value for each stream across all years is approximately the same (~0.3) and that the values are reasonably constant over time. Confirming this, single-factor ANOVA (a statistical comparison) shows no significant difference in the mean dissimilarities between the two treatment groups, 'control' versus 'potentially disturbed' streams.

Dissimilarity indices such as those used in Figure 1 may also be 'mapped' using multivariate ordination techniques to depict the relationship of the community sampled at any one site and sampling occasion with all other possible samples. Samples close to one another in the ordination indicate a similar community structure. Figure 2 depicts the ordination derived using the same macroinvertebrate data that were used to construct the dissimilarity plot from Figure 1. Data points are displayed in terms of the sites sampled in Magela and Gulungul Creeks downstream of Ranger for each year of study (to 2007), together with all other control sites sampled for the same period. Because the data-points associated with these two sites are interspersed among the points representing the control sites, this indicates that these 'exposed' sites have macroinvertebrate communities that are similar to those occurring at control sites.

Collectively, these results provide good evidence that changes to water quality downstream of Ranger as a consequence of mining in the period 1994 to 2007 have not adversely affected macroinvertebrate communities.

A related study of macroinvertebrate communities, sampled from shallow lowland billabongs in May 2006, is aimed at providing a biological basis for developing water quality closure criteria for the billabongs immediately adjacent to Ranger (see research highlight in Supervising Scientist Annual report 2006-2007 (pp 63-68), section 3.5).

Monitoring using fish community structure

Assessment of fish communities in billabongs is conducted between late April to the end of June of each year. Data are gathered, using non-destructive sampling methods, from 'exposed' and 'control' sites in deep channel billabongs and shallow weedy lowland billabongs. Details of the sampling methods and sites were provided in the 2003-04 Supervising Scientist Annual Report.

For both deep channel and shallow lowland billabongs, comparisons are made between a directly exposed billabong in Magela Creek catchment downstream of the mine versus a control billabong from an independent catchment (Nourlangie Creek and Wirnmuyurr Creek). The similarity of fish communities in exposed sites to those in control sites is determined using multivariate dissimilarity indices, calculated for each sampling occasion. Dissimilarity indices are described and defined above ('Monitoring using macroinvertebrate community structure'). A significant change or trend in the dissimilarity values over time could imply mining impact.

Channel billabongs

The similarity of fish communities in Mudginberri Billabong (directly exposed site downstream of Ranger in Magela Creek catchment) and Sandy Billabong (control site in the Nourlangie Creek catchment) was determined using multivariate dissimilarity indices calculated for each annual sampling occasion. A plot of the dissimilarity values from 1994 to the present is shown in Figure 3.

In the Supervising Scientist Annual Report for 2003-2004 (pp 35-38), section 2.2.3, an apparent decline over time was noted in the paired-site dissimilarity measures. The value for 2007 is the second lowest on record (Figure 2.16), although higher than for 2006.

The decline is primarily attributed to the particularly high abundances of chequered rainbowfish (Melanotaenia splendida inornata) and to a lesser extent glassfish (Ambassis spp) in Mudginberri Billabong in the early years of the study, relative to Sandy Billabong. Chequered rainbowfish have declined in Mudginberri Billabong since sampling commenced in 1989. The decline in rainbowfish numbers and, by association, the paired billabong dissimilarity value, is not related to any change in water quality over time as a consequence of water management practices at Ranger. This issue was examined in greater detail in the Supervising Scientist Annual Report 2004-05 (pp 66-71), section 3.6 where the environmental correlates (1) wet season stream discharge, (2) natural, wet season stream solute concentration, (3) length of previous dry season, and (4) habitat conditions on Magela Creek floodplain, were identified as possible causes of the decline in rainbowfish.

The decline has been less influenced by chequered rainbowfish and glassfish in the latter years, suggesting that more subtle changes in community structure are also occurring.

Shallow lowland billabongs

The scope of the monitoring programme for fish communities in shallow billabongs was reviewed in October 2006. An outcome of the review was a reduction to biennial sampling in six billabongs that comprise three 'control' versus 'exposed' site pairs. The reduction in sampling effort to biennial sampling was justified on the basis that an extensive baseline of data had been collected since 1994 together with evidence that changes to fish communities arising from potential mining impacts would arise over long time spans.

In a similar manner to fish communities in channel billabongs (discussed above), the similarity of fish communities in the directly exposed sites downstream of Ranger on Magela Creek (Georgetown, Coonjimba and Gulungul Billabongs) to those of the control sites (Sandy Swamp and Buba Billabongs on Nourlangie Creek and Wirnmuyurr Billabong - a Magela floodplain tributary) was determined using multivariate dissimilarity indices calculated for each sampling occasion. A plot of the dissimilarity values of the control-exposed site pairings - Coonjimba-Buba, Georgetown-Sandy Swamp and Gulungul- Wirnmuyurr Billabongs - from 1994 to the present, is shown in Figure 4.

The paired-site dissimilarities shown in Figure 4 average between 40 and 60% indicating fish communities in each of the billabongs comprising a site pairing are quite different from one another. Differences between the fish communities are most strongly linked to the type and density of aquatic plant communities in each billabong. For example, Coonjimba Billabong is dominated by an emergent reed (Eleocharis sp). Increasing density of Eleocharis (measured as weight) results in reduced fish abundance (Figure 5) and fish species number (regression R² = 0.12, p = 0.012). In contrast, Buba Billabong is dominated by emergent grasses and in this habitat the number of fish species is greatest at an intermediate grass density (unimodal regression, R² = 0.2, p = 0.0048). Thus fish find increasing densities of Eleocharis unfavourable but are favoured by emergent grasses until these become so dense that they physically prevent fish movement. For either habitat, excessive aquatic plant densities are unfavourable for fish communities.

The particularly high dissimilarity values observed in the Coonjimba-Buba and Gulungul- Wirnmuyurr site pairings in 2002 and the Coonjimba-Buba site pairing in 2007 (Figure 4) are attributable to high plant densities in one or both of the billabongs comprising the site pairing. In 2002, there was a particularly high biomass of aquatic vegetation caused by drier (than normal) conditions at the time of sampling. This sampling was conducted after a wet season of relatively low rainfall, possibly resulting in favourable growing conditions and the rapid drying of billabongs that effectively concentrated and increased the density of aquatic plants in the available sampling areas.

 

The relatively high dissimilarity associated with the Coonjimba-Buba Billabong pairing in 2007 is attributable to unusually high Eleocharis densities in Coonjimba Billabong in this year. This increase in density appears to be related to natural growth cycles, whereby a reduction in density (as experienced in 2003) is followed by a recovery period. Of particular note, one of the sites randomly selected for sampling in Coonjimba Billabong comprised the only area without Eleocharis. As expected, the fish community sampled here differed greatly from the other sites. This caused a large variance in the data set for the fish community structure within Coonjimba, in turn reflecting the high paired-site dissimilarity value for 2007.

The dissimilarity associated with the Coonjimba-Buba site pairing has increased over time (R² = 0.21, p = 0.0008) (Figure 4). This increase is significant irrespective of the previously-discussed high values in 2002 and 2007 (R² = 0.21 p = 0.0026, 2002 and 2007 removed). ERA's water chemistry data and eriss water quality spot checks taken at the time of sampling show that inputs of salts from the mine site (based on electrical conductivity data) to Coonjimba Billabong have increased over time. However, because the levels of contaminants (including Mg and U) are relatively low in relation to the concentrations known to adversely affect fishes, it is unlikely the increasing paired-site dissimilarity values are directly related to minesite inputs. The increase in dissimilarity is more likely the result of changes to the structure of aquatic plant communities in both billabongs. Whether altered water quality or hydrological conditions in Coonjimba Billabong as a consequence of mining at Ranger are the cause of these changes is not known at this stage. Further analysis is required to properly infer the cause of this increase in paired-site dissimilarity over time.

• Further information about ecosystem monitoring at Ranger Mine can be found in the Supervising Scientist Annual Reports.