3-4 Impact and risk analysis for the Namoi subregion

Executive summary

Gulligal Lagoon, which is located about halfway between Gunnedah and Boggabri on the western side of the Namoi River, NSW, 2005 Credit: Courtesy of Neal Foster

The impact and risk analysis for the Namoi subregion is a regional overview of potential impacts on, and risks to, water resources and water-dependent ecological, economic and sociocultural assets due to coal resource development. Hydrological models estimate the changes in water-associated parameters and their uncertainties. These changes provide the input for identifying areas where impacts to ecosystems are likely. Where possible, a quantification of ecosystem impacts allows the ruling out of areas that are very unlikely (less than 5% chance) to experience change.

Results from regional-scale hydrological modelling indicate potential risks to about 1400 km2 of ecosystems, 5500 km of streams, 2 springs and 624 water dependent assets. More detailed local information is required to determine the level of risk and potential impacts.

The Namoi subregion covers an area of 29,300 km2; however, the total area investigated in this assessment, the assessment extent, is 35,660 km2 as hydrological effects due to additional coal resource development may extended past the boundary of the subregion. The subregion is located in the Murray–Darling Basin in central NSW and the landscape is characterised by highlands in the east and south and a broad floodplain in the west. Drainage is dominated by the Namoi River and its tributaries and distributaries, the Mooki River, Coxs Creek, Pian Creek and Turragulla Creek. There are many ecologically important small lagoons, natural and artificial wetlands, and floodplain woodlands in the subregion.

Coal resources

Bioregional assessments (BAs) consider two potential coal resource development futures:

  • baseline coal resource development (baseline): a future that includes all coal mines and coal seam gas (CSG) fields that were commercially producing as of December 2012
    • in the Namoi subregion there are five open-cut coal mines: Boggabri Coal Mine, Rocglen Mine, Sunnyside Mine, Tarrawonga Mine and Werris Creek Mine; and one longwall mine: Narrabri North
  • coal resource development pathway (CRDP): a future that includes all modelled coal mines and CSG fields that are in the baseline as well as the additional coal resource development (those that were expected to begin commercial production after December 2012)
    • in the Namoi subregion ten additional coal resource developments represent the most likely future as of December 2015; eight are modelled: Boggabri Coal Expansion Project, Caroona Coal Project, Maules Creek Project, Narrabri South, Tarrawonga Coal Expansion Project, Vickery Coal Project, Watermark Coal Project and the Narrabri Gas Project. The remaining two mines, Vickery South Coal Project and the Gunnedah Precinct, did not have sufficient information for inclusion in the modelling. Analysis of the impacts of these two developments is restricted to commentary in Section 3.6 of this analysis. The eight mines that are included in the modelling form the basis for relating impacts to additional coal resource developments. The NSW Government bought back BHP’s Caroona coal exploration licences on the Liverpool Plains in August 2016. This occurred after the finalisation and modelling of the CRDP, thus the Caroona Coal Mine was included in the modelling even though it is no longer proceeding. As of July 2017, the Shenhua exploration licence for the Watermark Coal Project was reduced by 51.4% to exclude the Liverpool Plains. However, the company will continue to progress its Watermark Coal Project plans on the remainder of the licence and this is incorporated into the hydrological modelling.

The difference in results between modelled CRDP and baseline is the change that is primarily reported in a BA. This change is due to the additional coal resource development. Potential hydrological changes due to these coal resource developments are reported in companion products 2.6.1 (surface water) (Aryal et al., 2018) and 2.6.2 (groundwater) (Janardhanan et al., 2018) for the Namoi subregion; this product summarises the impacts on, and risks to, water resources and water-dependent ecological, economic and sociocultural assets.

Zone of potential hydrological change

The zone of potential hydrological change covers an area of 7014 km2 (19.7% of the Namoi assessment extent). The zone is the union of the groundwater zone of potential hydrological change and the surface water zone of potential hydrological change:

  • The groundwater zone of potential hydrological change is defined as the area with at least a 5% chance of exceeding 0.2 m of drawdown due to the modelled additional coal resource development in the regional watertable, which comprises the alluvial aquifer as well as weathered and fractured rock aquifers and covers an area of 2299 km2 (about 6.5% of the Namoi assessment extent).
  • The surface water zone of potential hydrological change contains those river reaches where a change in any one of nine surface water hydrological response variables exceeds a specified threshold due to modelled additional coal resource development. The thresholds incorporate at least a 5% chance of a 1% or greater change in a flow volume, or a 3 day or greater change in frequency. The surface water zone covers an area of 6430 km2 (18% of the Namoi assessment extent) and represents 5521 km of stream length.

The zone was used to rule out potential impacts on ecosystems and water-dependent assets within the Namoi assessment extent. Water resources and water-dependent assets outside the zone are very unlikely to be impacted.

Note that the drawdown in the confined parts of the Pilliga Sandstone was modelled and does not exceed 0.2 m more than 2 km outside of the zone of potential hydrological change. There are no extraction bores or springs within this area so defining the zone based upon the drawdown at the regional watertable is appropriate.

Potential hydrological changes


Results from regional groundwater modelling show drawdown due to modelled additional coal resource development of greater than 0.2 m is very likely (greater than 95% chance) for an area of 156 km2. It is very unlikely (less than 5% chance) that more than 2299 km2 will experience drawdowns of this magnitude due to modelled additional coal resource development. It is very unlikely that drawdown due to coal mining extends more than about 10 km from any coal mine. Results for 2 m and 5 m drawdown extents suggest it is:

  • very likely that an area of at least 117 km2 exceeds 2 m of drawdown and very unlikely that more than 853 km2 exceeds 2 m of drawdown
  • very likely that an area of at least 99 km2 exceeds 5 m of drawdown and very unlikely that more than 520 km2 exceeds 5 m of drawdown.

The range of potential drawdown outcomes reflects the uncertainty in the key input parameters for the various aquifers (e.g. hydraulic conductivity and storage); different parameters and assumptions that are considered plausible based on the current state of knowledge can lead to quite different predicted outcomes.

These numbers are additional to the 116 km2 in the mine pit exclusion zone, where the modelled drawdowns are considered unreliable due to steep hydraulic gradients at the pit face. The modelled estimates of drawdown within the mine pits, while significant, are considered unreliable for use in the receptor impact modelling and are not included when evaluating potential impacts on landscape classes and ecological assets.

Surface water

Within the zone, potential changes to surface water due to additional coal resource development were assessed using three hydrological response variables that represent zero-flow, high-flow and annual flow characteristics of streamflow. Changes in these variables represent the dominant hydrological drivers.

Changes in stream flow are very likely in Back Creek, Merrygowen Creek, Bollol Creek, Maules Creek, Driggle Draggle Creek and two unnamed creeks near Lake Goran. Most of the creeks have catchment areas much less than 100 km2 and effects are localised. The much larger Namoi River is largely insensitive to these changes in inflows because of the volume of flow.

Generally, the predicted changes in streamflow are small relative to the rainfall-related interannual variability, especially for annual flow and high-flow days. The streams likely to see the largest increases in the number of zero-flow days are Back, Merrygowen and Bollol creeks, as well as an unnamed creek. These creeks drain the Maules Creek, Boggabri expansion, Tarrawonga expansion and Watermark coal developments, respectively. The increase in zero-flow days in these creeks may represent a change that is greater than the interannual variability under the baseline, which serves to indicate the degree of hydrological change.

Water quality

The risk to regional stream water salinity due to additional coal resource development will depend on the magnitude of the hydrological changes and the salinity of the groundwater relative to the salinity of the stream into which the water is discharged. Modelling predicts a possible reduction in baseflow and this may lead to a reduction in stream salinity.

In all the streams identified from the regional-scale modelling with potentially large changes in flow regime, the impact on local stream salinity will depend on the relative reductions in catchment runoff and baseflow over time.

Reductions in catchment runoff are more likely to affect runoff peaks, while baseflow reductions have a more noticeable effect on low flows. In streams, such as Back Creek, Merrygowen Creek and Ballol Creek, Tulla Mullen Creek and Mooki River near Maules Creek, located near Boggabri, Tarrawonga, Vickery and Watermark coal mines respectively, where modelling results suggest increasing numbers of zero-flow days, it is likely that channel pools will be subject to longer periods of salt concentration by evaporation and less efficient flushing. These are conditions that favour increasing salinity in these water bodies. Increases in baseflow, potentially leading to increases in alluvial aquifer and stream salinity, cannot be ruled out, however, this is not an outcome that has been reported in the literature and remains an area for further investigation. The magnitude and extent of water quality changes cannot be determined without specifically representing water quality parameters in the modelling. This remains a knowledge gap.

Regulatory requirements are in place in NSW that aim to minimise potential salinity impacts due to coal resource development.

Impacts on, and risks to, ecosystems

The impact and risk analysis investigates how hydrological changes due to additional coal resource development may affect ecosystems at a landscape scale. Twenty-nine landscape classes, aggregated to six landscape groups, represent the ecosystems in the Namoi subregion. Landscape groups and some classes within other landscape groups ‘ruled out’ of the ecological modelling include:

  • the ‘Dryland remnant vegetation’ landscape group, as it comprises non-water-dependent vegetation communities for the purposes of the BAs
  • the ‘Human-modified’ landscape group, which comprises highly modified agricultural and urban landscapes; some impacts and risks are considered under economic assets
  • The 'Non-GAB springs' landscape class as the 15 springs in this landscape class are assumed to draw their water from the regional watertable and none of the springs are located within the zone of potential hydrological change.

Potential impacts of hydrological changes on ecosystems in the zone of potential hydrological change are assessed using qualitative mathematical models and receptor impact models. These models use indicators of the ecosystem condition, such as the probability of presence of a particular species, or projected foliage cover of the canopy vegetation, to assess the impacts of hydrological changes. Eight receptor impact models quantify potential ecological changes in the ‘Floodplain or lowland riverine’ and ‘Non-floodplain or upland riverine’ landscape groups and the Pilliga riverine (upland and lowland) ecosystems within these two landscape groups (Table 16). The results identify assessment units (parts of ecosystems and landscape classes) that are potentially ‘at minimal risk of ecological and hydrological changes’, ‘at some risk of ecological and hydrological changes’, or ‘more at risk of ecological and hydrological changes’ (thresholds used are described in Section

‘Floodplain or lowland riverine’ landscape group

The lowland riverine landscape classes in this group include ecosystems adjacent to the Namoi River and its major tributaries. Receptor impact modelling considered modelled hydrological changes and the responses of corresponding receptor impact variables, which are projected foliage cover, presence of tadpoles and assemblages of macroinvertebrates in the edge habitat.

Potential ecosystem impacts estimated from the receptor impact modelling showed changes in one or more of the receptor impact variables at a confined set of locations across the distribution of the associated landscape class. For example, modelling predicted the largest declines in the average number of families of aquatic macroinvertebrates due to additional coal resource development (ranging from 16 to 17 families at the 5th percentile and 4 to 3 families at the 50th percentile) in the Maules Creek and Bollol Creek.

Thresholds indicative of a relative measure of risk across a given landscape class from a combination of the receptor impact model results provided an assessment of potential ecosystem impacts. The greatest concentration of ‘more at risk of ecological and hydrological changes’ and ‘at some risk of ecological and hydrological changes’ assessment units are located along the Namoi River and its tributaries, Maules Creek, Back Creek and Bollol Creek. Of the 1425 assessment units included in one or more of the receptor impact models, 51 show ‘at minimal risk of ecological and hydrological changes’ and 29 ‘more at risk of ecological and hydrological changes’, with most of these risk categories being related to the potential impacts on lowland riverine and floodplain wetland landscape classes. A more detailed and local consideration of risk needs to consider the specific values at the location that the community are seeking to protect (e.g. particular assets), and bring in other lines of evidence that include the magnitude of the hydrological change and the qualitative mathematical models.

‘Non-floodplain or upland riverine’ landscape group

The receptor impact model for the upland riverine landscape classes modelled the relationship between cease-to-flow hydrological response variables (zero-flow days and maximum zero-flow spells) and two receptor impact variables: average number of families of aquatic macroinvertebrates in edge habitat and the probability of presence of tadpoles from the Limnodynastes genus. There were no detectable differences in mean changes in either average number of families of aquatic macroinvertebrates or the probability of presence of tadpoles across the upland riverine landscape classes between the baseline and modelled CRDP futures across the different percentile simulation periods (2042 and 2102).

The receptor impact model for the ‘Upland riparian forest GDE’ landscape class was based on the relationship between the effect of changes in groundwater drawdown and the frequency of overbank flows on projected foliage cover in the riparian trees. There were only a small number of assessment units where projected foliage cover predictions indicated a decline at the 5th percentile, and no assessment units at the 50th percentile for either simulation period. The limited change in this receptor impact variable is consistent with the associated hydrological response variables, where very small parts of the ‘Upland riparian forest GDE’ landscape class were exposed to changes in additional groundwater drawdown or change in frequency of overbank flows.

Pilliga riverine (upland and lowland)

The experts considered the Pilliga region, which encompasses both the Pilliga and Pilliga Outwash Interim Biogeographic Regionalisation for Australia (IBRA) subregions, as a separate entity for the purposes of the ecological modelling due to its distinctive biophysical attributes. Potential ecological impacts were a reflection of the modelled changes in groundwater and surface water, and the corresponding receptor impact variables, which were average number of families of aquatic macroinvertebrates and projected foliage cover of riparian vegetation. Predicted declines in both of these ecosystem indicators showed that potential changes were confined to assessment units along Bohena Creek and were equivalent to the ‘at some risk of ecological and hydrological changes’ category across the Pilliga riverine landscape class based on the risk thresholds defined for other receptor impact models.

There were 8 water-dependent landscape classes intersecting the Pilliga region that did not have an associated receptor impact model and hence ecosystem impacts remain undefined. These include the two Great Artesian Basin (GAB) springs that intersect the Pilliga region. It is unclear whether these springs source their water from the regional watertable used to define the zone, so it is not known whether they are potentially impacted. The classification as GAB springs is based on their association with underlying sandstone formations; their connection to the GAB requires further investigation.

Impacts on, and risks to, water-dependent assets

Ecological assets

Assessment of the potential impacts on ecological assets includes multiple lines of evidence: overlay analysis, qualitative mathematical models derived from expert elicitation and predictions of receptor impact variables as ecosystem indicators.

The Namoi subregion has 1690 water-dependent ecological assets in the assessment extent. Of these, 624 are in the zone of potential hydrological change and are subject to potential hydrological changes due to additional coal resource development. Water-dependent ecological assets in the zone associated with ‘Habitat (potential species distribution)’ asset class in the ‘Vegetation’ subgroup include (Table 34):

  • 15 species listed under the Commonwealth’s Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act) such as:
    • koala (Phascolarctos cinereus)
    • swift parrot (Lathamus discolor)
    • regent honeyeater (Anthochaera phyrgia)
  • 7 assets listed on the Collaborative Australian Protected Area Database (CAPAD)
  • the Pilliga Important Bird Area
  • 6 EPBC Act-listed threatened ecological communities.

Out of the 624 ecological assets, 135 are ‘more at risk of hydrological changes’ because all or part of the area where these assets occur is within one or more of the potentially impacted landscape groups and there is a greater than 50% chance of the modelled hydrological change exceeding the defined threshold (Section These ‘more at risk of hydrological changes’ assets include:

  • 76 assets in the ‘Surface water feature’ subgroup, none of which are listed in A directory of important wetlands in Australia (DIWA)
  • 12 assets in the ‘Groundwater feature (subsurface)’ group including
    • Cadna-owie Hooray Equivalent Great Artesian Basin recharge area
    • Gunnedah Basin Groundwater Management Zone
    • Lower Namoi Alluvium Groundwater Management Zone
    • Great Artesian Basin Groundwater Management Zone
  • 47 assets in the ‘Vegetation’ subgroup including
    • the Pilliga Important Bird Area, contiguous with the Pilliga Nature Reserve and forming the largest intact native forests west of the Great Dividing Range
    • the habitat of 5 threatened ecological communities and of 11 listed species in the EPBC Act.

Economic assets

There are 168 economic assets in the Namoi assessment extent. Of these, there are 47 assets in the ‘Groundwater management zone or area’ subgroup and 39 in the ‘Surface water management zone or area’ subgroup that are potentially impacted by hydrological changes due to additional coal resource development.

Of the 8953 bores in the assessment extent, 2555 bores were identified within the zone of potential hydrological change, and 2051 of these are very unlikely to be impacted due to additional coal resource development as they are in the surface water zone. There are 133 bores (excluding the 25 bores in the mine pit exclusion zone) where there is a greater than 5% chance of more than 2 m additional drawdown.

In relation to surface water availability, maximum reductions in annual flow in the Namoi Regulated River water source area are less than 1% and unlikely to lead to reductions in water availability, although total reductions can be as high as 4.2 GL/year.

There are reductions in water availability in the Mooki River, Maules Creek, Driggle Draggle Creek, Bollol Creek, Merrygowan Creek, Tulla Mullen Creek and one unnamed creek near the Lake Goran, but these are all less than 1% of the total water availability in each unregulated water source under the baseline.

Cease-to-pump rules apply to some water sourced in NSW to ensure sufficient water is retained in unregulated rivers to meet environmental requirements. For example, in the ‘Lower Coxs Creek Management Zone’, users must cease to pump when flow is equal to or below 15 ML/day at Tourable Gauge and 11 ML/day at Boggabri Gauge.

There is a 5% chance of additional 12 cease-to-pump days per year for Bundock Creek from the additional coal resource developments in the medium term. Additional cease-to-pump days for Bohena Creek, and the Mooki River at Breeza are generally less and limited to a maximum of 9 days with a 5% chance at the Bohena Creek and a maximum of 6 days (5% chance) for the Mooki River at Breeza.

Sociocultural assets

There are 31 water-dependent sociocultural assets in the assessment extent, including 22 heritage sites and 9 Indigenous assets. Fourteen of these are in the zone of potential hydrological change, and therefore are potentially impacted. The impact of potential hydrological changes on these assets requires a quantitative understanding of the nature of their water dependency. Some examples of these assets include the Boggabri Lagoon and the Burburgate Carved Tree, and built infrastructure such as the Wee Waa and Gunnedah courthouses, and heritage-listed buildings, cemeteries and graves (see Table 49 and Figure 50 of Section 3.5). The Bioregional Assessment Programme does not have the expertise to comment on potential impacts of changes in hydrological regimes on the value of Indigenous assets and built infrastructure. Evaluating potential impacts on these sites would require further local-scale assessment.

Future monitoring

Post-assessment monitoring is important to test and validate (or not) the risk predictions of the assessment. At the highest level, monitoring effort should reflect the risk predictions and focus the effort where the changes are the largest and incorporate those areas where modelling limitations did not allow a risk quantification. However, it is important to place some monitoring effort at locations with no predictions and lower risk predictions to act as a reference and confirm the range of potential impacts and identify unexpected outcomes.

The BA for the Namoi subregion has identified that potential hydrological or ecosystem impacts due to additional coal resource development are likely in areas concentrated around the main proposed coal resource development locations. Groundwater monitoring would be expected to be a focus for the following water sources as they have the largest number of bores with a greater than 5% chance of more than 2 m drawdown, In order they are: Gunnedah-Oxley Basin, Murray Darling Basin, Upper Namoi Zone 4 Namoi Valley, Upper Namoi Zone 7 Yarraman Creek, Upper Namoi Zone 8 Mooki Valley, Southern Recharge and Upper Namoi Zone 3 Mooki Valley.

There are 3629 km of streams that are potentially impacted in the not modelled areas, which means that the ecological impacts in those same locations remains unquantified and future work may need to address these constraints. For the modelled streams, future surface water monitoring should focus on streams that pass near the additional coal resource developments, and particularly for Back, Merrygowan, Bollol and Driggle Draggle creeks, given the changes in flow regime modelled to occur there.

Gaps and opportunities

The BA for the Namoi subregion incorporated the best available information within the constraints and timing of the Programme. For example, at the time of modelling sufficient information was not available for two additional coal resource developments: the Gunnedah Precinct and Vickery South Coal Project. The Gunnedah Precinct has the potential for cumulative impacts with the baseline Sunnyside Mine (currently in care and maintenance). The Vickery South Coal Project has the potential for cumulative impacts with the Vickery Coal Mine (an additional coal resource development), as well as the Rocglen Mine (a baseline development). Therefore, considering alternative coal resource development pathways is an important future consideration.

The assessment is regional and cumulative, and provides an important framework for local-scale environmental impact assessments of new coal resource developments and the local geological, hydrogeological and hydrological modelling that supports them. The results do not replace the need for detailed site-specific studies, nor should they be used to supplant the results of detailed studies that may be required under state legislation. There are opportunities to tailor the regional-scale BA modelling results for more local analyses (e.g. combining detailed local geological information with the groundwater emulators developed through BA, where appropriate).

Some of the broader knowledge gaps identified that can improve the understanding the potential impacts of coal resource development include:

  • Spatially explicit data on the thickness of the stratigraphic layers and a more detailed understanding of fault locations and depths would improve the geological model and may provide an improved precision to the groundwater modelling. Including faults into the current modelling is unlikely to change the spatial extent of the zone of potential hydrological change at regional resolution of 1 km2, however, local-scale accuracy and precision is likely to improve.
  • Improved mapping of depth to groundwater, and its spatial and temporal variation, not only has potential to constrain hydrological change predictions, it provides much needed context for the interpretation of the ecological impacts due to hydrological change. Interactions between changes in groundwater availability and the health and persistence of terrestrial groundwater-dependent vegetation remain uncertain due, in part, to sparse mapping of groundwater depths outside of alluvial layers.
  • A higher density of surface water model nodes and gauging information, located immediately upstream of major stream confluences and upstream and downstream of mine operations, would allow the point-scale information to be interpolated to a greater proportion of the stream network and improve the extent of surface modelling (and consequently some of the receptor impact modelling).
  • An improved ecohydrological understanding of water-dependent vegetation communities and their water requirements.
  • A collection of background data for benchmarking that includes (i) ecological data (e.g. high resolution vegetation mapping) for identifying current conditions and assessing changes in ecosystems and ecosystem indicators, (ii) improved instream water quality monitoring data and separate agricultural, infrastructure and coal resource development impacts on water quality, and (iii) groundwater quality monitoring data.
  • Increased investigation of subsurface ecosystem changes, in particular, the hyporheic zone, is highly relevant for the many non-permanent streams in the Namoi subregion and increased investigation of subterranean ecosystems and their response to changes in groundwater.
  • Identifying water-dependent assets valued by the local Indigenous communities would provide a more comprehensive account of sociocultural assets, even if many of those assets are already in the water-dependent asset register through other sources, for example, a wetland may have both ecological and Indigenous value.
  • Putting future changes due to additional coal resource development in the context of a changing climate and changing demands for water, and particularly from agriculture in the Namoi subregion.

In many areas it was not possible to develop receptor impact models and therefore investigate potential risk to ecosystems. In these areas, a simple spatial overlay of landscape classes corresponding to the zone of potential hydrological change identifies the locations (and parts of ecosystems) where additional work may be needed to quantify the risk. For example, expanding the surface water modelling to further areas would increase the coverage of risk to ecosystems. It would be prudent to clarify if the non-modelled areas do not experience any increased surface water-related risk to ecosystems, before focusing solely on areas identified as higher risk from receptor impact modelling.

The full suite of information, including information for individual assets, is provided at Users can explore detailed results for the Namoi subregion using a map-based interface in the BA Explorer, available at

Last updated:
6 December 2018
Thumbnail of the Namoi subregion