The predictions of hydrological change associated with additional coal resource development are shown in the boxplots in Figure 19 to Figure 27. In these figures, the model nodes are grouped for different river sections and tributaries. Within these groupings, small tributaries have a darker background shading. Nodes are ordered from downstream to upstream. The tributary grouping labelled ‘u/s Greta’ (u/s = upstream of) includes tributaries that join the Hunter River between nodes 1 and 20. The tributary grouping labelled ‘u/s Glennies Creek’ includes tributaries that join the Hunter River between nodes 25 and 51. The tributary grouping labelled ‘u/s Denman’ includes tributaries that join the Hunter River above node 51. Refer to Figure 5 in Section 2.6.1.3 for a schematic depiction of the model nodes and the network topology. The implications of these hydrological changes on landscape classes and assets are not discussed here, but are the focus of the impact and risk analysis reported in companion product 3-4 for the Hunter subregion (as listed in Table 2).
Figure 19 shows the changes to the annual flow (AF) at the 65 model nodes. The biggest percentage reductions occur in some of the small tributaries of the Hunter River, and range up to a median pmax of 81% at node 8 (Doctors Creek). Nine model nodes have reductions in median pmax that exceed 20%. Seven of these have catchment areas of less than 25 km2 and all have areas of less than 80 km2. There are tightly constrained distributions of pmax values around these median values at all the heavily affected nodes except for node 11, where the 5th to 95th percentile range in pmax is –51% to –30%. Apart from reductions in median pmax of about 8% at nodes 46 (Wollar Creek) and 47 (Wilpinjong Creek) there is little effect on AF in the nodes in the Goulburn and Wyong river basins or in those along Wollombi Brook, where the median pmax reductions are less than 2% of baseline flow. The largest reductions in median amax are located in the lower Hunter River and increase with distance downstream. The biggest effects are at nodes 1 and 6 and result in losses of around 29 GL/year, which represent about 3% of the baseline flow. A large proportion of this median reduction (13 GL/year) originates in the Loders Creek tributary (node 7), with the remainder propagating from upstream (node 10).
For the heavily affected nodes, the median year at which maximum hydrological changes occurs is either 2022 or 2037 for the tributaries of the lower Hunter River (nodes 7, 8, 9 and 11), but 2028 for tributaries upstream of Glennies Creek. There is relatively little uncertainty in these dates. The maximum hydrological changes in the Hunter River itself tend to occur in 2037.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 20 shows the changes to the daily streamflow rate at the 99th percentile (P99) at each model node. The biggest changes in pmax occur at the same nine locations with the biggest effect on AF. For P99 the median pmax values for these nine nodes exceed 20% and range up to 68% for node 52 (Dry Creek). At most of these sites there is a greater spread of pmax values than there is for AF. At most of the affected nodes, the percentage reduction in P99 is greater than the percentage reduction in AF.
The year of maximum change in P99 tends to correspond with the year of maximum change in AF, with the exception that it occurs later in the heavily affected nodes 7 and 8 (2052 and 2049, respectively).
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 21 shows the changes to the interquartile range (IQR) at each model node. The changes in IQR associated with the additional coal resource development are always reductions. This implies that the difference in flow rates between high flows (the 75th percentile of daily streamflow) and low flows (the 25th percentile of daily streamflow) is reduced, most likely through a decrease in the 75th percentile. The patterns of change are similar to those of AF (Figure 19) and P99 (Figure 20). The biggest reductions in median pmax occur at nodes on small tributaries of the Hunter River, and include reductions of more than 70% at nodes 8 (Doctors Creek), 52 (Dry Creek) and 55 (unnamed creek west of Muswellbrook). Each of these nodes have catchment areas of less than 25 km2.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 22 shows the changes to high-flow days (FD) at each model node. Once again, the largest reductions in the number of FD occur in the small tributaries of the Hunter River, but there are also largish reductions at two tributary nodes in the Goulburn river basin. The biggest change is a maximum reduction in the median number of FD by 46 days/year at node 52 (Dry Creek). Twelve nodes have median reductions in amax of more than 5 days/year. However, there is much greater uncertainty around changes in the number of high-flow days (and in the timing of the maximum changes) than there is for changes in AF. Along the Hunter River, the frequency of FD is reduced by 4 days/year at node 6 (Singleton) and by lesser amounts elsewhere.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 23 shows the changes to the daily streamflow rate at the 1st percentile (P01) at each model node. Reductions in median pmax of more than 10% are predicted in many of the tributary nodes with large reductions in high-flow characteristics, especially nodes 7, 8, 9, 11 and 55, including 100% reductions in median pmax at node 8 (Doctors Creek) and node 55 (unnamed). These cases are representative of a particular year in a replicate for which there is a nonzero 1st percentile of baseline flow, but the 1st percentile of CRDP flow is zero. Of particular note is that, despite showing little change to the high-flow characteristics, there are substantial predicted changes to 1st percentile flow in the two nodes in the Wyong river basin, where the pmax values show decreases of 51% (node 64) and 57% (node 65).
The timing of the maximum changes tends to be later for P01 than for the high-streamflow hydrological response variables. The median tmax for P01 occurs in or later than 2044 at all of the seven most heavily affected nodes.
By comparison to the three flux-based high-streamflow hydrological response variables (AF, IQR and P99), P01 tends to have greater uncertainty – as shown by a large interquartile range relative to the median response – for both pmax and tmax.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 24 shows the increases in the annual number of zero-flow days (ZFD) due to the additional coal resource development. Most of the nodes along the Hunter and Goulburn rivers, together with Glennies Creek and Wollombi Brook, are perennial in both the baseline and CRDP simulations and therefore show no effect on ZFD. Along the regulated part of the Hunter River, this reflects the fact that the river is managed to ensure that minimum environmental flows are met at key points along the river. This requirement is represented in the model, thus under the baseline and the CRDP, there are no zero-flow days. The only nodes with potentially large changes are on some of the small tributary nodes of the Hunter River and the nodes of the Wyong River. The largest predicted changes in ZFD are increases of 200 and 225 days/year at nodes 64 and 65 along the Wyong River. However, there is considerable uncertainty in these projections at all nine nodes where amax exceeds 15 days/year.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 25 shows the increases in the annual number of low-flow days (LFD) due to the additional coal resource development. There are substantial predicted increases in amax in many tributary nodes and in the Wyong river basin. The biggest effect is a median amax value of 295 days/year at node 55. The biggest change along the Hunter River is an increase of 5 days/year at node 6. There is considerable uncertainty in the corresponding tmax values for the heavily affected nodes, but the median tmax values typically occur between 2024 and 2036. Exceptions are node 50 (Goulburn River) where the median tmax is 2065 and the two nodes in the Wyong river basin where the median tmax values are 2051.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 26 shows the changes in low-flow spells (LFS) due to the additional coal resource development. Increases in median amax of more than three spells per year occur at tributary nodes 7, 8, 9, 52 and 55 as well as at nodes 64 and 65 in the Wyong river basin. The biggest changes are increases of 21 spells at nodes 8 and 55. However, there are also substantial increases in LFS at nodes 10, 20, 25 and 31 along the Hunter River, including a median change of 12 spells at node 31 (Liddell). Interestingly, there are two small tributary nodes (26 and 29) where median amax shows a decrease by two spells per year. These reductions in amax result when multiple spells coalesce into a single large spell.
There is considerable uncertainty in the projections of both amax and tmax in Figure 26.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
Data: Bioregional Assessment Programme (Dataset 1)
Figure 27 shows the maximum changes to the length of the longest low-flow spell (LLFS). The changes in the LLFS are very similar to those in LFD. The LLFS is projected to increase by about 130 days at node 8 (Doctors Creek) and by more than 15 days at several other small tributary nodes. The LLFS is projected to increase by 32 and 37 days at the two nodes in the Wyong river basin.
amax = maximum raw change; pmax = maximum percent change; tmax = year of maximum change; u/s = upstream of; NI = negligible impact
Numbers above the top panel are the median of the best 300 replicates under the baseline for the year corresponding to the median tmax. In each boxplot, the bottom, middle and top of the box are the 25th, 50th and 75th percentiles, and the bottom and top whiskers are the 5th and 95th percentiles. Within the groupings, small tributaries have a darker background shading.
Refer to Figure 4 in Section 2.6.1.3 for location of model nodes.
