Breadcrumb

2.7 Receptor impact modelling for the Galilee subregion

Executive summary

Artesian Spring Wetland at  Doongmabulla Nature Refuge, QLD, 2013 Credit: Jeremy Drimer, University of Queensland

This product details the development of qualitative mathematical models and receptor impact models for the Galilee 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 landscape classes allows a better understanding of how changes in hydrology may result in ecosystem changes.

A receptor impact model describes the relationship between:

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

The relationship between these variables and subsequent responses will identify where further local-level studies of ecosystems and their response to coal resource development are needed.

Coal resource developments

Receptor impact modelling for the Galilee subregion applies the two potential coal resource development futures considered in bioregional assessments (BAs):

  • 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 Galilee subregion the absence of commercially producing coal mines and CSG fields as of December 2012 means that there are no coal resource developments being modelled in the baseline for the purposes of BA
  • 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 Galilee subregion there are 17 proposed new developments, of which 14 are potential new coal mines and 3 are potential new CSG fields. There is enough publicly available information to include seven of these developments in the numerical modelling: open-cut coal mines Alpha and Hyde Park, and combined open-cut and underground coal mines Carmichael, China First, China Stone, Kevin’s Corner and South Galilee.

The difference in results between CRDP and baseline is the change that is primarily reported in a BA. This change is due to additional coal resource development. As there are no coal mines or CSG operations in the Galilee subregion baseline, this effectively means that the CRDP future is simply due to the additional coal resource developments.

Methods

Receptor impact model development is both qualitative and quantitative due to the complexity and uncertainty associated with describing relationships between hydrological change and ecological components of the system. The absence of direct relevant theory and ecological response data of potential impacts due to additional coal resource development that occur in the future requires expert judgement or elicitation to be used in: (i) mapping ecological processes and key components (as signed digraphs); (ii) constructing qualitative models that record – as an increase, decrease or no change – potential landscape class response to sustained hydrological change; and (iii) selecting ecological indicators (receptor impact variables) from the ecological components or processes and the hydrological regimes (hydrological response variables) that support them. The resulting statistical models quantify how changes in hydrological response variables due to coal resource development may potentially impact the receptor impact variables in the short-term (2013 to 2042) and long-term (2073 to 2102) within a landscape class.

Ecosystems

The Galilee subregion occupies diverse environments, from the Great Dividing Range through to vast expanses of semi-arid and arid inland Australia. It includes rivers that flow into Kati Thanda – Lake Eyre, the Gulf of Carpentaria, the Pacific Ocean and the Murray–Darling Basin. In the conceptual modelling product, the Galilee assessment extent was classified into 11 broad landscape groups which comprise 31 landscape classes. The classification was based on five elements derived from the Australian National Aquatic Ecosystem (ANAE) classification framework involving topography, landform, water source, water type and water availability. In addition, each area was identified as either remnant or non-remnant vegetation based on the Queensland remnant regional ecosystem (RE) mapping.

Descriptions for the qualitative mathematical models and receptor impact models are provided under landscape group headings. Of the 11 landscape groups in the Galilee subregion, most (73%) of the zone of potential hydrological change supports the ‘Dryland’ or ‘Floodplain, non-wetland’ landscape groups and these were not modelled as they rely on incident rainfall and localised runoff. Four of the nine remaining landscape groups cover very small areas within the zone of potential hydrological change and were not further assessed. The remaining five landscape groups represent the ecosystems that were the focus for receptor impact modelling. These potentially impacted landscape groups include ‘Springs’, ‘Floodplain, terrestrial GDE’, and ‘Non-floodplain, terrestrial GDE’. In addition, the landscape groups ‘Streams, GDE’ and ‘Streams, non-GDE’ were combined for the purposes of the receptor impact modelling and are considered further in this product as the ‘Streams landscape group’.

‘Springs’ landscape group

The ‘Springs’ landscape group contains springs of two types: recharge (also referred to as ‘outcrop’) springs and discharge springs. There are three clusters of springs that occur within the zone of potential hydrological change of the Galilee subregion. From north to south these are: (i) the Doongmabulla Springs complex, (ii) a series of springs that overlie the Colinlea Sandstone which is of Permian age (hereafter referred to as the ‘Permian springs cluster’) and (iii) a series of springs associated with Triassic geological units (hereafter referred to as the ‘Triassic springs cluster’). Qualitative mathematical models were built for the ‘Springs’ landscape group that described the general dynamics of the aquatic community associated with springs. A receptor impact model was not developed for ‘Springs’ because applying a regional-scale groundwater model to the micro scale of ‘Springs’, was deemed to be inappropriate. Potential hydrological changes at a regional scale cannot be scaled down to apply to ‘Springs’. Instead, a qualitative assessment of potential impacts of hydrological change was made using available hydrological and ecological information obtained from the Lake Eyre Basin Springs Assessment Project and other available sources.

Based on the ‘Springs’ landscape group qualitative mathematical model, experts considered that the critical factor in preserving the aquatic community is the rate of groundwater flow that maintains a damp or submerged state in the spring, such that this surface does not become dry. An increase in water depth above this threshold supports a wetted-area regime around the perimeter and downstream of the spring. Surface water and groundwater modelling predict potential impacts from coal mining on groundwater levels and subsurface water availability. Based on all possible combinations of these potential impacts, three cumulative impact scenarios were developed for qualitative analysis of response predictions.

Streams landscape groups

The two streams landscape groups in the Galilee assessment extent contain 10 landscape classes. Four landscape classes are groundwater-dependent ecosystem (GDE) streams (comprising the ‘Streams, GDE’ landscape group) and six are non-GDE streams (comprising the ‘Streams, non- GDE’ landscape group). Both GDE streams and non-GDE streams are widespread in the zone of potential hydrological change. Of the 6285 km of streams in the zone of potential hydrological change, 3484 km (55%) are non-GDE streams and are most prominent in the western and southern part of the zone; GDE streams account for 2801 km (45%) of total streams in the zone. Within the zone of potential hydrological change, annual streamflow shows a high degree of interannual variability. Flows in any given year can vary from almost no flow to major floods. Mean monthly flow is also highly variable with minimal to no flow from July to October, while most surface water flows occur between December and April. The streamflow regime within the zone is thus characterised as one of ‘dry seasonal flows’. A large portion of the Belyando river basin is within the zone of potential hydrological change, where dry seasonal flows result in a boom–bust ecology that follows an annual hydrological cycle unlike more arid rivers further west in the Galilee subregion.

A combined qualitative mathematical model was developed for the two streams landscape groups. The central feature of the qualitative model was the existence and connectivity of refuge habitats with detritus and algae the principal resources that support populations of aquatic invertebrates and fishes. Surface water also recharges stores of deep groundwater in confined aquifers which can contribute to near-surface groundwater.

Two receptor impact models were developed to examine the potential impact of the additional coal resource development on water-dependent ecosystems within the zone of potential hydrological change. The first receptor impact model focused on the response of woody riparian vegetation to changes in flow regime and groundwater. The second model examined the response of a high-flow macroinvertebrate (mayfly nymphs in the genus Offadens, family Baetidae) to changes in flow regime. The density of this species was deemed a suitable indicator of stream health because the species’ responds to streamflow and the species density is impacted by water drying up.

Using the woody riparian vegetation receptor impact model, experts were of the opinion that: (i) mean percent foliage cover would decrease as the depth to groundwater increases, (ii) mean percent foliage cover would decrease as the number of low-flow days increased, and (iii) mean percent foliage cover would increase as the number of flood events with peak daily flow exceeding the 1983 to 2012 2-year return period increased.

Results from the high-flow macroinvertebrates receptor impact model indicate that the experts’ opinion provides no strong evidence that either the number of low-flow days or the mean maximum spell of low-flow days have a significant effect on mean baetid density. This model also predicts that baetid density under reference conditions does not affect outcomes under different low-flow conditions in future assessment years.

‘Floodplain, terrestrial groundwater-dependent ecosystem’ landscape group

The ‘Floodplain, terrestrial GDE’ landscape group vegetation depends on the subsurface expression of groundwater on a permanent or intermittent basis to maintain growth or avoid water stress and adverse impacts on condition. The landscape group contains two landscape classes within the Galilee subregion zone of potential hydrological change: ‘Terrestrial GDE, remnant vegetation’ (about 2358 km2) and ‘Terrestrial GDE’ (about 75 km2).

The ‘Floodplain, terrestrial GDE’ landscape group provides potential habitat for a range of nationally listed threatened plant and animal species. Groundwater and surface water are the key hydrological elements that support woodland communities. Four alternative signed digraphs were developed to represent this landscape group because there was uncertainty of the links associated with trees drawing groundwater to the surface, and the potential for groundwater to contribute to soils depleted of oxygen (anoxic). Surface water and groundwater modelling indicate substantial potential impacts of coal mining to groundwater depth and drawdown, and decreased flood events.

The relationships identified in the qualitative mathematical modelling workshop were formalised into a receptor impact model that described the response of annual mean percent foliage cover in a 0.01 km2 plot to changes in groundwater levels and peak daily streamflow.

The experts’ elicited responses suggest that percent foliage cover may decrease as groundwater drawdown increases due to additional coal resource development. The mean of the average percent foliage cover will decrease from about 10% without any change in groundwater level, to about 7% if groundwater levels decrease by 6 m relative to the reference level in 2102 (holding all other variables at their median values). However, there is considerable uncertainty within these predictions.

The receptor impact modelling indicates that there is evidence in the experts’ responses for overbank flood events (Events R2.0) to have a major quadratic effect on percent foliage cover. The frequency of overbank flood events will generally have a positive effect on average percent foliage cover, unless the frequency of flooding becomes too high. The predicted increase in the average frequency of overbank flood events, from 0.05 in the 30 years preceding the reference year, to an average value of 0.9 over the future period, causes the average percent foliage cover to increase by almost 10%. However, when the frequency of overbank flood events exceeds 1 there is a negative effect on percent foliage cover.

‘Non-floodplain, terrestrial groundwater-dependent ecosystem’ landscape group

The ‘Non-floodplain, terrestrial GDE’ landscape group contains four landscape classes. Two are distributed within the zone of potential hydrological change: ‘Non-floodplain, terrestrial GDE’ (about 6 km2) and ‘Non-floodplain, terrestrial GDE, remnant vegetation’ (1184 km2). The vegetation in this landscape group depends on the subsurface expression of groundwater on a permanent or intermittent basis to maintain growth or avoid water stress and adverse impacts on condition. Much of the vegetation, in particular trees, relies on accessing groundwater from shallow aquifers. Groundwater is needed for trees to persist in areas where evapotranspiration exceeds rainfall.

The geological units of the Rewan Group and Clematis Group are the most likely sources of groundwater for the ‘Non-floodplain, terrestrial GDE’ landscape group. This landscape group provides potential habitat for a range of nationally listed threatened plant and animal species.

A qualitative model was developed that focused on recruitment dynamics associated with groundwater-dependent native tree species, with tree canopies providing a range of (as yet unspecified) ecosystem roles. A receptor impact model was designed that focused on the response of mean annual percent foliage cover of woody vegetation to changes in groundwater. The quantitative modelling, based on expert elicitation of response values, indicated a site with higher foliage cover at the 2012 reference point is more likely to have higher foliage cover in the future than a site with a lower foliage cover value at this reference point. The model also indicated that percent foliage cover may decrease as groundwater drawdown increases due to the additional coal resource development. The model shows that the mean of the average percent foliage cover will decrease from about 48% without any change in groundwater level, to about 47% if the levels decrease by 10 m relative to the reference level in 2012. However, there is uncertainty in these results.

Gaps and limitations

There are a number of gaps and limitations for building qualitative ecosystem models and quantitative receptor impact models. These include:

  • the expert elicitation process reflects the subjectivity and bias inherent in the knowledge base of the experts (e.g. in defining the scope of the model; its components and connections; the ecologically important hydrological variables; representative receptor impact variables; and the magnitude and uncertainty of responses to change)
  • water quality changes due to a shift in the relative contributions of surface water runoff and groundwater to streamflow, or due to enhanced connectivity between aquifers of differing water quality, are not represented
  • there are areas where the hydrological change to the stream network cannot be quantified adjacent to some of the coal mine developments, thus impeding the analysis of potential risks across those parts of the landscape
  • limited understanding of the connectivity between groundwater and surface water systems including: the connectivity within landscape classes and groups of classes and between ecosystems; potential surface water – groundwater interaction; the nature of groundwater interactions between riverine and terrestrial ecosystems
  • an absence of experts with specific knowledge of the local hydrological and ecological systems within the zone of potential hydrological change of the Galilee subregion.

These gaps and limitations place some constraints on the comprehensiveness of the ecosystem analysis undertaken for this BA, and it is important to flag these so that further investigation can be better targeted.

Results from the application of receptor impact models and the spatial distribution of hydrological results are reported in Section 3.4 of the impact and risk analysis in companion product 3-4 for the Galilee subregion.

Last updated:
6 December 2018
Thumbnail of the Galilee subregion
PRODUCT FINALISATION DATE
2018
PRODUCT CONTENTS
ASSESSMENT