3.1 Causal networks

Potential impacts on water and the environment are assessed using a causal network. The assessment uses spatially explicit and systematic evaluations of the likelihood and consequence of potential environmental harm. It also evaluates how compliance with existing regulatory controls and operational practices can minimise and mitigate potential impacts. The causal network also allows local and regional-scale monitoring objectives to be systematically identified and prioritised. An interactive online tool, GBA Explorer, presents this comprehensive information base.

Causal networks are graphical models that describe the cause-and-effect relationships between development activities and endpoints – the values to be protected – for example, the nationally important Mataranka Thermal Pools. The causal network illustrates the complex and interconnected nature of the natural environment and unconventional gas resource development activities in the Beetaloo GBA region. It provides a comprehensive and clearly identified set of inferred direct and indirect pathways where development activities may impact on environmental values. Systematic evaluation of likelihood, consequence and mitigation strategies allows qualitative and quantitative information to be integrated, even when the available knowledge base is limited.

In this impact assessment for the Beetaloo GBA region, a causal network with detailed node descriptions is central to the environmental impact assessment ( Figure 5 ). The impact assessment method was developed as part of the GBA Program based on existing causal network methods described in Peeters et al. (2021b). The causal network illustrates the complex and interconnected nature of the natural environment and unconventional gas resource development activities in the region. Evaluations in the causal network assess how undertaking these activities may cause material changes (a change that exceeds defined thresholds) in nodes linking from activities. The evaluation also considers the degree to which existing regulatory controls and industry practices can mitigate potential impacts.

Existing mitigation and management practices associated with development activities are typically represented by links from activities to stressors along a causal pathway. Whereas subsequent links in the causal network represent the effect an activity has on a process once it occurs. Changes to natural processes are often difficult, if not impossible, to mitigate. This is why it is important that mitigation strategies are implemented early in the chain of events from activities to endpoints.

Nodes represent the different components of the system that make up a causal pathway – drivers, activities, stressors, processes and endpoints ( Table 1 ). Links – represented by arrows – show the hypothesised cause-and-effect relationships between nodes. Grids associated with each link enable representation of spatial variability when evaluating each pathway (Peeters et al., 2021a). Constructing the causal network ( Figure 6 ) is an exercise that involves multiple disciplines and wide stakeholder engagement. Hazards and causal pathways identified in the baseline synthesis and gap analysis (Huddlestone-Holmes et al., 2020) were refined and incorporated into the causal network. A comprehensive process of stakeholder engagement in the hazard analysis and technical reviews aimed to accurately capture unconventional gas operations and regulation while addressing user needs. Extensive independent technical reviews of evaluations and documentation in the causal network was also undertaken.

Endpoints are represented in the causal network as ‘assessment endpoints’ defined as an explicit expression of the ecological, economic and/or social values to be protected, and as ‘measurement endpoints’ defined as measurable characteristics or indicators related to the valued characteristic chosen as the assessment endpoint (Suter, 1990; US EPA, 2016) (Section 5.1.1 in Stage 2 baseline synthesis and gap analysis (Huddlestone-Holmes et al., 2020)). Endpoints in the causal network for the Beetaloo GBA region represent aquifers such as the Cambrian Limestone Aquifer; landscapes, such as terrestrial vegetation communities; protected areas, such as wetlands; and protected fauna, such as the greater bilby or Gouldian finch listed under territory or national legislation ( Table 1).

TABLE 1 Node types, examples and number of each node type in the causal network for the Beetaloo GBA region

Node type



Number of nodes


Major external driving forces (human or natural) that have large-scale influences on natural systems

Resource development



A planned event associated with unconventional gas resource development

Civil construction

Transport of materials and equipment



Physical, chemical or biological agent, environmental condition or external stimulus caused by activities

Dust generation Vehicle movement



A naturally occurring mechanism that could change a characteristic of an endpoint

Confined aquifer drawdown Habitat degradation, fragmentation and loss



A value pertaining to water and the environment that may be impacted by development of unconventional gas resources

Surface water condition

Persistence of Gouldian finch



Visually, the causal network illustrates the complex and interconnected nature of natural and regulatory environments in the Beetaloo GBA region. In general, links from activities to stressors can be mitigated by existing regulatory and industry management frameworks, whereas links from processes to endpoints associated with changes to natural processes are often difficult, if not impossible, to mitigate. Links to subsurface processes and endpoints evaluated as not possible or not material represent the natural barriers that protect overlying aquifers from drawdown or contamination due to gas production from deeper gas plays.

The causal network for the Beetaloo GBA region ( Figure 6 ) consists of 62 nodes, 197 links and 2,078 pathways. Nodes are organised from left to right, starting with the driver node resource development . Surface and ecology nodes are in the upper half of the figure, and subsurface and hydrology nodes are in the lower half of the figure. Hazards and causal pathways identified in the Stage 2 baseline synthesis and gap analysis (Huddlestone-Holmes et al., 2020) were refined and incorporated into the causal network.

Links are evaluated according to their likelihood (‘possible’, ‘not possible’), consequence (‘material’, ‘not material’) and management (‘avoidable’, ‘unavoidable but can be mitigated’, ‘unavoidable and cannot be mitigated’) ( Table 2 ). Outcomes of the overall assessment categorise potential impacts on endpoints by their level of concern ( Table 2 ).


TABLE 2 Evaluation language used to describe links in the causal network and the corresponding level of concern used to describe the impact pathway through to endpoints


Not possible

Possible but not material

Possible and material but can be avoided

Possible, material and unavoidable but can be mitigated

Possible, material, unavoidable and cannot be mitigated

Level of concern

Very low concern

Low concern

Potential concern

Potentially high concern

Impacts are not physically possible or are extremely unlikely (having a probability of less than 1 in 1,000)

Impacts can be avoided by current legislation or because the impact does not represent a material change

Impacts can be minimised or mitigated by existing management controls

Impacts cannot be avoided or mitigated at the scale of the GBA region


The causal networks for the GBA Program are presented in the GBA Explorer. This web-based tool allows users to interact with a graphical presentation of the causal networks with immediate access to the node descriptions, link evaluations and overall assessment summary. Users can visualise the whole entire causal network or simplify it by selecting specific pathways. Spatial data are also presented via interactive maps that include the spatial information used to inform the assessment and spatial link evaluations, and to make spatial impact maps depicting pathways for activity to endpoint, stressor to endpoint and process to endpoint.

Future studies could extend the causal network to other industries, such as pastoralism or tourism. What is not possible with this method is assessing the cumulative impact of multiple stressors on a baseline. This requires firstly a comprehensive and quantitative baseline assessment and secondly future projections of the magnitude and likelihood for all stressors (Peeters et al., 2021b).

FIGURE 5 The impact assessment for the Beetaloo GBA region explained


View a higher resolution copy of this figure

The assessment is based on a resource development scenario (number of wells, well pads, access roads, seismic surveys and water needed to produce energy from gas) in areas prospective for unconventional gas resources.  Areas of concern for each activity, stressor, process and endpoint in the causal network are identified using tables and maps. For more detailed description of this image contact bioregionalassessments@awe.gov.au.

Images in this figure are representative only as many are too detailed for interpretation. They are available from the causal network for the Beetaloo GBA region in GBA Explorer

Element: GBA-BEE-3-527

Last updated: