Breadcrumb

2.7 Receptor impact modelling for the Hunter subregion

Executive summary

Google Earth image of the Hunter River west of Muswellbrook

This product details the development of qualitative mathematical models and receptor impact models for the Hunter subregion. Receptor impact models enable the Bioregional Assessment Programme (the Programme) to quantify the potential impacts and risks that coal resource developments pose to water-dependent landscape classes and ecological assets. Applying receptor impact models across landscapes classes allows a better understanding of how changes in hydrology may result in changes in ecosystems.

A receptor impact model describes the relationship between:

  • one or more hydrological response variables, which represent hydrological characteristics of the system that potentially change due to coal resource development (for example, drawdown or the annual flow volume) and
  • a receptor impact variable, which is a characteristic of the system (for example, probability of presence of riffle-breeding frogs) that, according to the conceptual modelling, is potentially sensitive to changes in the hydrological response variables.

Receptor impact modelling for the Hunter subregion applies the two potential coal resource development futures considered in the bioregional assessments:

  • baseline coal resource development (baseline): a future that includes all coal mines and coal seam gas (CSG) fields that are commercially producing as at December 2012. In the Hunter subregion the baseline includes 42 mining operations, comprising 22 open-cut mines and 20 underground mines
  • coal resource development pathway (CRDP): a future that includes all coal mines and CSG fields that are in the baseline as well as those that are expected to begin commercial production after December 2012. In the Hunter subregion as of September 2015, the additional coal resource development includes a further 22 proposals for coal resource developments, including 3 new open-cut coal mines, 3 new underground coal mines and 16 expansions to baseline mining operations. As of May 2015, there is no CSG production in the Hunter subregion, nor any proposals for CSG development in the future.

The difference in results between CRDP and baseline is the change that is primarily reported in a bioregional assessment. This change is due to additional coal resource development. Potential hydrological changes have been presented in companion products 2.6.1 (surface water modelling) and 2.6.2 (groundwater modelling); the process of developing qualitative mathematical models and receptor impact models is summarised in this product.

To simplify the impact and risk analysis for this additional coal resource development, the Hunter subregion is broken into five landscape groups – ‘Riverine’, ‘Groundwater-dependent ecosystem (GDE)’, ‘Coastal lakes and estuaries’, ‘Non-GDE vegetation’ and ‘Economic land use’ – which comprise 26 landscape classes, broadly differentiated by ecohydrological characteristics. Only landscape classes that intersect the zone of potential hydrological change were candidates for qualitative models.

Two modelling workshops were held to build the qualitative mathematical models and receptor impact models and required input from experts in these landscapes and/or the Hunter subregion. Receptor impact models were developed for landscape classes that experts considered more likely to be at risk from hydrological changes due to additional coal resource development, and for which they had the expertise to inform model development.

Six receptor impact models were developed: ‘Perennial streams – riffle-breeding frogs’, ‘Perennial streams – Hydropsychidae larvae’, ‘Intermittent streams – riffle-breeding frogs’, ‘Intermittent streams – hyporheic invertebrate taxa’, ‘Forested wetlands (riverine forest) – projected foliage cover’ and ‘Wet and dry sclerophyll forests – projected foliage cover’. They quantify relationships between components of four qualitative models, representing five landscape classes.

Qualitative models were developed for a further seven landscape classes: ‘Highly intermittent or ephemeral’ streams, ‘Rainforest’, ‘Freshwater wetland’, ‘Lakes’, ‘Lagoons’, ‘Saline wetlands’, and ‘Seagrass’. Quantitative models of ecosystem response were not progressed for these landscape classes because the project team and experts at the qualitative modelling workshops considered it unlikely that these landscape classes would be impacted by hydrological changes due to additional coal resource development, or were too uncertain about ecosystem responses to potential hydrological changes to provide meaningful quantitative estimates (e.g. in the ‘Freshwater wetland’ landscape class). The qualitative model for the ‘Rainforest’ landscape class, was premised on rainforests in sheltered gullies and slopes in hilly-to-steep terrain, which were considered unlikely to be affected by drawdown of the regional watertable. This does not take into account the 10 km2 of rainforests that overlaps with the alluvium in the Hunter subregion suggesting that, despite their topographic position, these rainforest communities may have some dependence on alluvial groundwater in some circumstances. This rainforest community represents a gap in the receptor impact modelling.

Four landscape classes that intersect the zone of potential hydrological change do not have qualitative models. They are ‘Heathland’, ‘Grassy woodland’, ‘Semi-arid woodland’ and ‘Creeks’ (part of the ‘coastal lakes and estuaries’ landscape group). These landscape classes generally depend on rainfall and/or local groundwater and are considered very unlikely (less than 5% chance) to be impacted by hydrological changes due to additional coal resource development.

Descriptions of the qualitative mathematical models and receptor impact models are provided under the landscape group headings. To define the results space for the receptor impact variables, elicitation scenarios were chosen to represent the range of modelled hydrological results under baseline and additional coal resource development. The receptor impact variables serve as indicators of potential impact to the ecosystems they represent. Riffle-breeding frogs, for example, are not found along every reach of intermittent and perennial stream in the Hunter subregion, and the models do not assume this. Rather the models predict whether, everything else being equal, a given hydrological change will enhance or diminish the habitat suitability of a reach of stream for riffle-breeding frogs, and by extension other components of the ecosystem that have similar flow dependencies or depend on components that do. Results from application of the receptor impact models to modelled hydrological changes in the Hunter subregion are reported in Section 3.4 of companion product 3-4 (impact and risk analysis) for the Hunter subregion.

‘Riverine’ landscape group

In the qualitative mathematical model for the ‘Permanent or perennial’ streams landscape class, experts identified groundwater levels, riffle flow, overbench flow, overbank flow and zero-flow days as critical determinants of instream and riparian habitat condition. Reductions in groundwater levels and streamflow, resulting in increases in zero-flow days, are generally considered to have negative impacts on riparian and subsurface habitats.

Two instream receptor impact models were developed, reflecting the response of riffle-breeding frogs and flow-dependent macroinvertebrates to changes in zero-flow days. As the number of zero-flow days increases, experts generally considered that:

  • the probability of presence of riffle-breeding frogs would drop quite dramatically, with the model reflecting a chance of no presence under extremely dry conditions
  • the density of Hydropsychidae (net-spinning caddisflies) larvae would drop dramatically, with the model reflecting the possibility of less than 1 per m2 under increasingly intermittent flow regimes.

In the qualitative mathematical model for the ‘Lowly to highly intermittent’ streams landscape, the same hydrological variables as for the perennial streams were identified as critical to habitat condition, with flow reductions and lowering of the watertable generally assumed to result in negative impacts.

The two receptor impact models represent the response of hyporheic invertebrate taxa (organisms found where surface water and groundwater mix below the bed of a stream) and riffle-breeding frogs to changes in zero-flow days and the duration of zero-flow spells. With increasing zero-flow days, experts generally considered that:

  • the probability of presence of riffle-breeding frogs would drop quite dramatically, with the possibility of no presence under extremely dry conditions
  • the hyporheic invertebrate taxa richness would fall, with the chance of fewer than 10 in 6 L of water under extremely dry conditions.

‘Groundwater-dependent ecosystem’ landscape group

Eight of nine GDE landscape classes occur within the zone of potential hydrological change. The ‘Spring’ landscape class does not intersect the zone and can be ruled out as very unlikely to be impacted by additional coal resource development.

The qualitative model for the ‘Forested wetland’ landscape class was based on the biological processes and environmental factors that regulate tree, shrub and herb composition in riparian systems of inland and coastal rivers of the Hunter subregion. The model recognises the possibility for coal resource development to impact the groundwater regimes that support these forest communities. Qualitative analysis generally indicates a negative predicted response on trees, seeds and seedlings to changes in hydrology, with a corresponding decline in shade, habitat structure, bank stability and orchids and fungi.

The receptor impact model for the ‘Forested wetland’ landscape class quantifies the response of projected foliage cover of riverine forests of central and western Hunter to changes in hydrology. Coastal swamp oak communities are not represented by this model, nor riverine forests along the regulated Hunter River and Glennies Creek. Experts generally thought that:

  • initial foliage cover in the reference period has a positive effect on the likelihood of future foliage cover
  • groundwater extraction has a negative effect on projected foliage cover
  • increased frequency of both overbench and overbank flows has a positive effect on projected foliage cover.

Experts at the Hunter subregion qualitative modelling workshop determined that ‘Wet sclerophyll forest’ and ‘Dry sclerophyll forest’ landscape classes could be represented in a single model. Groundwater was assumed to be the dominant water source. The qualitative model assumes trees, seedlings and shrubs are adversely affected by groundwater depletion, with reductions in shade, habitat structure, nectar, nectar consumers, predators, sap-eating and leaf-eating insects, gliders and koalas. Insects are shown as potentially increasing due to reduced predation.

The ‘Wet and dry sclerophyll forests’ receptor impact model reflects the experts’ view that:

  • initial foliage cover in the reference period has a positive effect on future foliage cover
  • groundwater extraction has a negative effect on projected foliage cover.

A qualitative mathematical model for the ‘Rainforest’ landscape class was developed based on a modified version of the wet and dry sclerophyll forests model. Experts assumed that rainforests generally occupy sheltered gullies and slopes in hilly-to-steep terrain, that they might use groundwater opportunistically, and that the groundwater would be from local sources (e.g. perched watertables), not connected to a regional watertable. Thus, the rainforests were considered unlikely to be affected by coal resource development and a quantitative model was not progressed.

The ‘Freshwater wetland’ landscape class is described as a complex of marsh and pond habitats. These wetlands are reported in the literature as largely dependent on local, perched groundwater systems. They were considered unlikely to be impacted by coal resource developments higher in the catchment and quantitative models were not progressed for this landscape class. However, given some uncertainty about freshwater wetland connections with regional groundwater, it is recommended that an assessment of the potential for impacts on a freshwater wetland be informed by understanding of the local and regional hydrology.

‘Coastal lakes and estuaries’ landscape group

Qualitative models were developed for subtidal benthos and intertidal wetland communities, which pertain to ‘Seagrass’, ‘Lakes’, ‘Lagoons’ and ‘Saline wetlands’ landscape classes. The subtidal benthos qualitative model reflects the view that seagrass beds are sensitive to subsidence from underground mining, leading to an increase in water depth, hence reduced light penetration, to the seagrass beds. As the proposed Wallarah 2 and Mandalong Southern Extension mines are not under the coastal lakes, there is no risk of subsidence of the lake beds from these developments. The proposed Chain Valley extension does involve coal extraction from under Lake Macquarie, but not within the high water mark subsidence barrier, as required under existing regulations to minimise risks from subsidence in these areas. A quantitative model was not developed for subtidal benthos.

The qualitative model for intertidal wetlands identified an interaction between groundwater and saltmarsh, which was very local in scale. There was considerable uncertainty about interactions with regional groundwater, if any. Modelling of groundwater drawdown due to additional coal resource development indicated a small chance of drawdown in areas of mapped saline wetlands on the eastern side of Lake Budgewoi, the Dora Creek inflow to Lake Macquarie and around the Lake Macquarie entrance. A quantitative model was not developed for intertidal wetlands.

The receptor impact modelling described in this product guides how companion product 3-4 (impact and risk analysis) for the Hunter subregion is framed. This product will describe the difference between results for the CRDP and the baseline (due to the additional coal resource development) for only those developments that can be modelled in the Hunter subregion.

Last updated:
18 January 2019
Thumbnail of the Hunter subregion
PRODUCT FINALISATION DATE
2018
ASSESSMENT