A panoramic view of the Brisbane River flooding between Goodna and Gailes at the flood peak 12 January 2011.

Australia has just emerged from some of the most widespread flooding this country has seen since European settlement.

These events are certainly tragic and have impacted many of our fellow Australians. The cleanup costs are being measured in billions of dollars. However, no sooner had the flood peak passed than we began to hear an outcry from those insisting existing flood management infrastructure and planning were somehow to blame.

Opinion: By Tony Weber, Manager, Water Quality-Associate BMT WBM

In my view the real debate should be around some very different questions. Should we simply learn to live with such floods? Can we deal with the uncertainties of predicting floods? What might be the cost impacts of changing how we live with floods in the future?

We will always have floods - it is as simple as that - and we will inevitably face significantly worse events than this one.  After all, most of us live in floodplains, and for very good reasons. As our forefathers found, the flatter areas of land on floodplains were easier to build ports on to facilitate trade, and cheaper to build on too, leading to rapid development in those areas. Not surprisingly then, the majority of our major trading infrastructure is centred on these floodplains: consider the Port of Brisbane, The Australia Trade Coast, Botany Bay, The Port of Melbourne and other examples. Quite logically located at the end of river systems such as the Cooks and Georges in Sydney, the Yarra in Melbourne and the Brisbane River in Brisbane, the development of these ports encouraged major settlement growth on adjacent land. As a result, nearly all of our major settlements have been centred on rivers. This allows us to know with confidence that we will continue to experience flooding in our economic and social epicentres in this country.

The Brisbane to Ipswich Railway line severed by floodwaters 12 January 2011.So the question becomes, if we face a high risk of future flood events, how best should we manage that risk? On the one hand, the science involved in predicting the likelihood of future events has improved significantly: we have some of the world’s best hydrologic scientists in this country and have been at the forefront of modelling science, with eWater CRC and predecessor, the CRC for Catchment Hydrology, playing a leading role.

One the other hand, even as our science and models have improved, uncertainty has remained our biggest challenge. These recent flood events provide an excellent example. Pundits were quick to call the events the “worst ever”, or to classify them “bigger than a 1 in 100 year flood” and so on. In truth, we simply don’t have a long enough data record to make such calls. Some countries, like South Korea, have been collecting rainfall records for more than 500 years, but in this country we only have data going back between 100 to 150 years. When we only may have experienced one or two extreme events in a century, how can we possible determine what has been “the worst ever” or a “1 in a 100 year” flood? If we could travel back in time to collect data for the previous century we might learn of even larger events. Indeed, if one looks at the history of flooding in Brisbane, there was evidence to suggest that in the 1800s, Brisbane experienced at least 10 floods of a magnitude equal to or greater than that experienced this time (see http://www.bom.gov.au/hydro/flood/qld/fld_history/brisbane_history.shtml). If this was the case then, what was it like 100 years before that? 

If we can’t escape uncertainty in our attempts to manage flood risk, it seems to me that we must closely examine all that is involved in living with this risk, recognising that flooding – just like drought and bushfires – will always be with us.

A good starting point would be a re-appraisal of urban planning guidelines that allow for building on floodplains and on flood-affected land. As higher-quality land parcels become scarcer, pressure to use what was previously considered marginal land inevitably grows. Many of us will also continue to want to live near waterways, and so prime riverfront properties will continue to attract high premiums despite the flooding risk. It is past time, surely, to assess whether existing planning controls take adequate account not only of future risk, but also the degree of uncertainty in our understanding of that risk. Just as overdue is an assessment of whether we have appropriately costed the economic impact of that risk and its corresponding uncertainty. 

A local caravan park completely innundated 12 January 2011.To gauge the magnitude of such costs, one simply has to look at the review of dam safety conducted by ANCOLD, the Australian National Committee on Large Dams several years ago. Resultant efforts to reduce the risk of dam failure by accounting for much larger events (from 1:10,000 year probability to 1:100,000 year) have prompted widespread investment in works to improve the structural integrity of dam walls, including the widely-discussed “fuse plug” weirs on Wivenhoe Dam in Brisbane. Across the nation such infrastructure improvements have cost in the order of several billion dollars. Imagine then the cost impost of upgrading other infrastructure such as road, rail, power and telecommunications infrastructure to improve its flood immunity in the interests of reducing potential future flood impacts.  Do we have sufficient data to truly understand whether this would be cost beneficial, given our uncertainties in understanding the potential magnitude of future floods? 

The inquiry into the Brisbane flooding represents a golden opportunity to consider such questions.

Such deliberations would also represent a useful progression from the somewhat shallow debate which has absorbed some sections of the media. Take, for example, those media reports which alleged that the operators of Wivenhoe Dam exacerbated the flooding. It is fascinating that none of those reports considered the cost-benefit equations of the actions they claim should have been taken.

Assume, for instance, that, as some pundits have called for, 25 per cent of Wivenhoe Dam’s water supply volume had been released prior to the flood event in an attempt to attempt to mitigate the flood peak. At the most conservative estimate of the value of the water - $1 per kilolitre – the water lost would have cost around $300 million. It would take a brave dam manager indeed to release that much revenue downstream on the chance (with its associated uncertainty) of increased rainfall falling within the dam catchment in the following week. And how could any dam manager, given our existing state of knowledge, be certain that such a move would have reduced downstream flood damage by a similar amount?

This is what flood risk management is all about: understanding the cost-benefit of reducing risk by improving infrastructure. These are the real questions for an inquiry and for the community more broadly: how much are we willing to pay, both economically and socially, to live with risk?

 Tony Weber is a visiting fellow at the Integrated Catchment Assessment and Management Unit at the Fenner School at ANU and is the National Practice Leader for Water Quality at consultants BMT WBM

The height of the flood peak in Goodna, approximately 2.5m less than the 1975 flood, but devastating nonetheless