Roadshow 1.3 Causes and causal pathways

Causes and causal pathways – assessing direct and indirect impacts using causal networks 

Click here to open transcript

Dr Luk Peeters: Hi. I'll be presenting on Causes and Causal Pathways - assessing direct and indirect impacts using causal networks. So, where this fits in, is into the final stage of the GBA program, where we're doing the risk analysis and evaluation of all the potential impacts on water and the environment, from unconventional gas resource development. As part of that, we have been looking at what development might look like, what water use might look like. But we're also looking at, can we integrate impacts, avoidance, mitigation or offset measures? How can management and monitoring occur? Throughout the direct exercise, we've been identifying what the knowledge gaps are and what limitations from the assessment.

One of the crucial pieces of information summarising the baseline information is this kind of conceptual diagram. It's a cartoon representation, pretty much all the knowledge that we gathered and digested during throughout stage two is capturing the geology, the landscape, the ecology, but also what kind of activities are proposed in the landscape. This is an invaluable tool in communication and really quickly summarising a lot of information and to highlight the complexity of the system. However, turning that into a systematic risk assessment is not a trivial exercise. So that's why we developed the causal network approach.

So starting from this conceptual diagram, we started out systematically identifying all of the cause-and-effect relationships that are linking unconventional gas resource development with the endpoints, the aspects of water in environments that we care for and that we want to know the impact on. Now, this conceptual diagram is depicting the Cooper Basin, is for the Cooper region. What this is showing is what is a causal network that we came up with, that we developed for the Beetaloo GBA region. So I'm not going to walk you through all of this. But it's just to highlight that on the left hand side, we've got the resource development that leads to activities, things like drilling, transport of materials, which in turn then leads to stressors on the environment, can be things like groundwater extraction, vehicle movement, those things then affect processes in the landscape, air pollution, soil contamination, which in turn affect the endpoints in blue. As you can see, these boxes are connected by quite a lot of links.

It's just highlighting the complexity of the system that we're studying and how many things are related to each other. In the session after this, we'll walk you through the GBA Explorer, which is the web interface that we developed that allows you to explore the content. I'm just going to highlight one of the features is that rather than showing that entire complexity, you can focus on one single node and see what pathways are associated with that. So what I'm highlighting here is the waste and wastewater management activity, shows which stressors it's associated with and ultimately, how that activity can affect what are the logical pathways for it to affect any of the endpoints that we have studied.

What I want to emphasise is that it's not just these boxes that you see on the screen. Behind each of these boxes, these nodes, there is quite extensive documentation. So in this case, the Explorer, if you click on it, you'll see this kind of information sheet popping up, describing exactly what we mean by waste and wastewater management. That also includes maps of where these activities might occur in the landscape. Now, you've seen that all of these links have a different colour. That's where the risk assessment comes in.

So for each of the links in this network, so where the link is an arrow connecting those two boxes, we systematically went through this flow chart of questions where we were looking at is this link, is this possible, does it lead to a material change? Can it be avoided? If not, if it cannot be avoided, can it be mitigated? So systematically going through these, we are addressing the core aspects of any risk assessment. Namely, looking at likelihood, consequence and then avoidance or mitigation. Depending on the answers on this flowchart, you get a code value of between zero and four that we could take further in our assessment. Again, we've spent a lot of time doing that evaluation and finding justification for our evaluations.

We've tried to capture that as part of the documentation as well. Again we'll show that later on in the GBA Explorer. If you go into this, what you'll see is that we provide that information sheet detailing exactly how we arrived at an evaluation. One important aspect of it is that we also did this spatially. So this evaluation can change depending on where you are in the landscape. So in this case, what I'm showing here is the link from resource development to drilling and showing that map in the Beetaloo GBA region. If you want to develop your conventional gas resources, it's unavoidable that you need to drill wells, it's the only way to get the gas out.

However, if the region is not prospective or at least not considered prospective in our assessment, we said, "Well, that's where it's not possible to be drilling because there's no resources." So that's where that distinction comes from, where the area is prospective for gas, it's unavoidable. Where it's not prospective, it's not possible. However, as you can see, there's areas indicated in grey on the map as well. Already the NT Government provides quite a lot of areas where drilling is not allowed in legislation.

So that's what we've captured here. It's mostly around this buffer distance around motorways and reserve blocks. So in those areas, while it is possible to be drilling, we consider it can be avoided because it's not allowed under legislation. So what I want to highlight here is that we get quite a nuanced image or a nuanced evaluation in space of what is possible, where in the landscape now that was for each link individually. Now, what we're actually interested in is how that pans out over that entire network, how does that drilling activity affect things like the Gouldian finch. That's where we start, once we've got that code for each link, we can start combining that systematically. Basically what we'll be doing is along each pathway, we'll take the minimum. The logic behind it is quite straightforward. For instance, if you're in a location where drilling is not possible, then nothing that follows after the drilling can happen. So the minimum of that is it's not possible.

Now, but there might be other pathways possible at a certain location. So when we look at different pathways, we're looking at a maximum. So although drilling might not be possible, it might be possible that you're building an access road, that you're still going to do vegetation removal for that. So combining all of that together, we end up with what we call these maps of level of concern, where a very low level of concern corresponds to things, pathways that are not possible, while potentially high concern is what corresponds to things that cannot be mitigated. Now, the example I'm showing here is for Gouldian finch, it's pretty uniform. It's basically saying that impact is possible, material cannot be avoided, but there are ways of mitigating that impact on the Gouldian finch from the various activities.

What this tool allows you is to also start digging deeper into that, which pathways exactly are of concern. So what I'm showing here are two different pathways. One is the vegetation removal. As you can see, there is certain areas in the Beetaloo, for instance, reserved blocks, where there's protection in terms of vegetation removal, as I mentioned before. So in those areas, we say that the impact on the Gouldian finch is of low concern because it's already legislated that vegetation removal is not allowed. Therefore, the impact is avoidable. However, invasive carnivores, they might not keep to these kind of administrative orders. So even though there is protection under law in those areas, invasive carnivores spreading to the landscape associated with unconventional gas resource development, might still spread into that region.

Again, this is just to illustrate that there's a lot of information in these causal networks that lets us build a very nuanced image of what is possible and what is often potential risks to the endpoints. So in conclusion, what we developed here with causal network allows you to do a risk assessment. The key features of that is that it is really systematically looking at all the potential impacts. It's considering management and mitigation measurements. Based on this network, it allows you to identify where to focus on for detailed local skill assessments, but also where your monitoring and compliance can focus on. Thank you.

Additonal information 

About the presenter

Dr Luk Peeters 


Luk is a Senior Research Scientist with over 15 years of experience in water resources management, with an emphasis on conceptualisation, numerical modelling and uncertainty analysis, geostatistics and machine learning. He obtained his PhD in geology from the Katholieke Universiteit Leuven (Belgium) and joined CSIRO in 2010. 

Last updated:
15 November 2021