Sun, sea and pollution
By Steve GartnerMay 30th, 2013
In an effort to better understand the impacts of human activity on Australia’s coastal environments, scientists are using powerful computer models to investigate and predict the water quality in coastal bays, gulfs, harbours and river estuaries.
30 May 2013
Glen Paul: G’day, and welcome to CSIROpod. I’m Glen Paul. The ever increasing human population, escalation of land use and development of coastal regions are the primary causes of decline in coastal water quality and ecosystem health. Nutrient sources in stormwater runoff, agricultural inputs, and effluent discharge, result in complex interactions with the marine food web, which is then stirred up by tides, coastal currents, and waves.
While Scientists do have an excellent understanding about many of the basic processes of these interactions, it’s basically impossible to observe them all. This is why computer modelling has been developed, to allow a far greater understanding of environmental change brought on by human activities. Known as Marine Biogeochemical Modelling, these models act as a prediction tool that can be used to investigate the behaviour of coastal systems.
Joining me on the line is CSIRO’s Dr Emlyn Jones. Emlyn, why is it so difficult to understand exactly how humans are impacting coastal water quality? After all, we’ve had plenty of practice at polluting the place up over the years.
Dr Jones: Well I suppose the largest issue is that we really don’t understand in a lot of situations what’s actually driving exactly the decrease in water quality in specific regions. It’s very difficult to tease apart the interaction between the effluent coming out of say sewage treatment plants versus the diffuse loads coming in from agriculture. And then when you put on top of that the effects of aquaculture in these coastal regions, you’ve got one user group suggesting that it’s the others that are causing it. And what we’re really trying to do is pin down the exact causes for declines in coastal water quality, and then come up with prediction tools to allow managers to make better use of limited funding to increase and improve the water quality of the coastal regions.
Glen Paul: So what do you see then as the main marine water quality issues facing the waters around Australia?
Dr Jones: Well at the moment there’s a lot of focus being put on sediment loads and nutrient inputs in the Great Barrier Reef region. There’s a number of stresses on the Reef at the moment, and you’ve got influences from industry, you’ve got influences from agriculture, and then you’ve got natural influences such as tropical cyclones that come through and potentially cause quite large localised damage to the Reef. But at the same time you’ve then got this effect of global climate change superimposed on the top of it.
So when we try and sort of make decisions within this very complex domain you really need to take into account all of these influences, and at the same time we need to be putting estimates of uncertainty on these predictions so that we don’t actually start making decisions using overly confident models.
Glen Paul: OK. Now just on that, where do you get the data used in the modelling, and where’s the limit to what can be considered acceptable or trustworthy data?
Dr Jones: So much of the data that comes in, in the coastal region, is really from a number of sources. Many of the State Governments run monitoring programs, whereby they’d go out say weekly to monthly timescales to monitor a certain number of set points. We then take observations from the Integrated Marine Observing System, they’ve sort of nicknamed IMOS, and that’s a federally funded marine initiative that then looks at not so much the coastal, but more so the shelf scale dynamics. We then also use a number of products that come in via satellite remote sensing. And then all of this is augmented by us deploying Slocum Gliders in the coastal zone, which again are a fairly new source of data, but are starting to become very important in understanding a lot of the drivers.
So there’s a real mix of different sources of the observations, and the challenge in many ways is to try and bring them together. But really the crux of this issue is it’s very difficult to completely observe the marine environment, and ultimately if we start trying to base all of our decisions on observations alone we end up getting quite a skewed picture of what’s going on. Two reasons for that being the sparsity of the observations, but in many cases what we try and do is make predictions sort of outside of the envelope that observations exist for. For example, we might be interested in trying to understand the outcome of a different management strategy, and there’s no way we can actually observe an alternative management strategy, so we need to model that. And the other trick is we can’t observe the future, so we really need to rely on the models to predict the future.
So when it comes down to it we need to come up with tools to merge the models and data to really get an understanding of where the errors lie, and how good our predictions or alternative management scenarios are likely to be.
Glen Paul: With the complexity of what you’re describing there, what will be required then to make it a commonplace resource tool for environmental managers?
Dr Jones: Ideally what we want to be aiming for is to come up with a coastal information system that is built on a number of levels. First up, the simple physics that drive the circulation of the coastal areas needs to be run, and that’s called the hydrodynamics. And then the next level in the hierarchy is the sediment dynamics, because a lot of biological processes are in many ways linked to sediment dynamics. And then the next level in the hierarchy is the biogeochemical model, so there’s at least two models that then feed into the biogeochemical model.
And what really needs to happen for these models to be taken up and used by managers, we need to be running these models routinely in the coastal zone, they need to be fit for the purpose that the managers want to use them for. Some models are not designed to answer certain questions, so we need to be very careful when defining what the model should be used for. We then need to plug all these models into an operational framework whereby the models required to force these models are all available to it, so there’s also this data and information infrastructure built around it.
So these coastal information systems, while they might seem sort of fairly simple in concept, the mechanics that goes on behind the scenes to join all the dots basically is fairly complex, and it’s projects like eReefs which are really trying to link the models that we’ve got with the data and information sources that we need, and then we have user reference groups which then define exactly what products the managers require, and we then work with the models and make sure that we can come up with a product which is fit for purpose for these guys to actually make useful decisions with.
Glen Paul: OK. Well hopefully you’ll be able to pull that all together and the modelling will go a long way in tidying up the mess. I appreciate your time today, Emlyn.
Dr Jones: Thanks very much, Glen.
Glen Paul: Dr Emlyn Jones. And to find out more about the research, or to follow us on other social media, just visit www.csiro.au.