2.6.1.6.3 Summary and discussion

The prediction results show that the additional coal resource development in the Hunter subregion can affect all of the hydrological response variables. Comparisons across the 65 model nodes show that the relative hydrological changes are larger for the nodes where the maximum footprint for the additional coal resource development affects a larger proportion of the catchment. For instance, the mine footprints in the catchments contributing to nodes 52 and 55 comprise 66% and 58% of the total contributing area, respectively (these areas are reported in Table 4 in Section 2.6.1.3). The resulting median pmax values for the high-flow flux-based variables, AF and P99, are around –68% and –59%, respectively. In general the median change in these variables is highly correlated and commensurate with the maximum proportion of the catchment that is included in the footprint for the additional coal resource development. Large changes are also predicted for node 8 for many hydrological response variables (–81% for AF and –65% for P99). Here the footprint for the additional coal resource development comprises 55% of the contributing area to this node, but since the node is also heavily affected by mining under the baseline, even larger hydrological changes are predicted relative to its proportional area.

The prediction results show that the additional coal resource development in the Hunter subregion has more noticeable effects on hydrological response variables in the small tributaries of the Hunter River, where the proportion of contributing area affected by mining is more likely to be high, than at the nodes along the Hunter River itself. This is particularly apparent in streamflow at nodes 7 to 9 (Loders Creek, including Doctors Creek), which enter the Hunter River just upstream of Singleton, and at nodes 52 (Dry Creek) and 55 (unnamed creek) in the vicinity of Muswellbrook. The catchments of nodes 7 to 9 include the Bulga and Mount Thorley–Warkworth additional coal resource developments, which affect 35–55% of the contributing areas (Table 4 in Section 2.6.1.3); the catchments of nodes 52 and 55 include the Bengalla and Mount Pleasant mines. Other nodes with substantial percentage changes in the high-streamflow hydrological response variables are nodes 26, 27, 29 and 35. The first three of these nodes are all located in the vicinity of the Glendell, Integra, Liddell and Mount Owen mines, while the catchment of node 35 includes parts of the Drayton South and Mount Arthur mines. All these nodes have relatively small contributing areas. While there are bigger predicted changes in amax at nodes further downstream, the proportional effects of these changes are diluted by relatively unaffected inflows. The prediction that the biggest effects occur downstream of multiple mine developments highlights the cumulative nature of potential hydrological changes, particularly on low-flow characteristics.

The biggest changes (in terms of amax) on the low-streamflow hydrological response variables are also predicted to occur at these small tributary nodes. However, unlike for the high-streamflow hydrological response variables, there is also a substantial change in the low-streamflow hydrological response variables in the two nodes of the Wyong river basin. These nodes are located near the proposed Wallarah 2 and Mandalong underground mines. The effect of these reductions in baseflow is to turn what was previously a perennial stream into one that flows on only about 40% of days in the most heavily affected year. Although this is a large reduction, it must be remembered that the projections presented in Figure 19 to Figure 27 are for the worst-case year during the entire simulation period (2013 to 2102). There is no implication, particularly for the low-flow variables, that the changes will be this severe in every year.

Substantial increases in ZFD would be expected where the river connection with groundwater is broken – that is, if drawdown causes the watertable in the alluvium to drop below the channel bed. In such a situation, a perennial or intermittent stream would switch to an ephemeral system, which by definition only flows in response to rainfall-runoff events. As part of the Hunter subregion groundwater modelling (companion product 2.6.2 for the Hunter subregion (Herron et al., 2018b)) hydraulic property information from the Wallarah 2 environmental assessment (Mackie Environmental Research, 2013) was used to illustrate the process by which local information can be used to constrain the regional-scale predictions. The parameters from the environmental assessment indicated relatively poor connectivity through the overburden above the mine. The simulations informed by lower hydraulic conductivity parameters resulted in drawdowns towards the lower end of the range of predictions from the regional groundwater model (see Section 2.6.2.8.4 in Herron et al., 2018b). This suggests that the larger changes in baseflows from the set of regional groundwater parameters are over-estimates and that the Wyong River is unlikely to switch to an intermittent flow regime (i.e. a variable connection to groundwater). This is an area where more detailed investigation is needed to reduce the uncertainty around these predictions.

The changes due to the additional coal resource development on the low-streamflow hydrological response variables (P01, LFD, ZFD, LFS and LLFS) appear to be slightly more noticeable than those on the high-streamflow hydrological response variables (AF, P99 and FD). The flux-based variables (AF, IQR, P99 and P01) have similar median pmax values at the most heavily affected nodes. However, of the two frequency-based variables that are most directly comparable – FD and LFD – the increases in median amax values in the latter are typically larger than the decreases in the former. However, the uncertainties in predicted pmax (for the flux-based variables), amax (for the frequency-based variables) and tmax are greater for the low-flow variables.

For high-streamflow hydrological response variables, the tmax at nodes with noticeable changes occurs approximately when the maximum footprint for the additional coal resource development occurs. This indicates that the instantaneous streamflow reduction caused by the mine footprint for the additional coal resource development dominates amax and pmax in these hydrological response variables while the changes from the cumulative effect on baseflow over time caused by watertable drawdown are negligible. This conclusion is supported by the tightly constrained changes in pmax at most nodes which suggest that the biggest effect on the high-streamflow hydrological response variables is caused by interception and retention of surface runoff at the mine sites, rather than by reduced baseflow associated with groundwater drawdown.

For low-streamflow hydrological response variables, the tmax at nodes with noticeable changes does not occur consistently with the time when the maximum footprint for the additional coal resource development occurs. At many of the most heavily affected nodes, the predicted median tmax values tend to be a little later for two of the low-streamflow hydrological response variables – P01 and LLFS. This indicates that the causes of the changes on the low-flow variables are controlled by a combination of the instantaneous change from the additional mine footprints and the cumulative effect on baseflow over time caused by watertable drawdown. Therefore, it is expected that uncertainty in predicting the changes on low-streamflow hydrological response variables is much larger than that on high-streamflow response variables.

Last updated:

18 January 2019

Product Finalisation date

Numbered headings

ASSESSMENT

PRODUCT DATA