Floods, politics and engineers

There's always been the very strong whiff of a Right wing witch hunt against the dam operators following the Brisbane flood. It would seem this happened with unusual haste because:

a. there is Labor government in power;
b. a theory instantly developed that, being a Labor government, it had fretted too much about releases from Brisbane's water supply because it believed climate change warnings that longer droughts are coming;
c. people like to blame someone, if it is at all possible, when natural disasters happen.

Why the government should be penalised at all, even if there was any evidence (I don't think there is) that it withheld recommended water releases out of concern for future droughts is a complete puzzle, given this report from October 2010, barely 3 months before the flood, when water was being released due to spring rains:

Seqwater said all south-east dams were receiving heavy inflows from surrounding catchments after heavy rainfalls across the region.

The water body’s decision to open the flood gates at the officially 100 per cent full Wivenhoe came under fire from the state opposition.

Opposition spokesman Jeff Seeney told parliament the dam was not completely full.

“Is not this release of water from Wivenhoe Dam, when it is holding only 40 per cent its available storage capacity, a clear indication that the government has learnt nothing from the water crisis,” Mr Seeney said.

But Natural Resources Minister Stephen Robertson said the extra capacity was needed to prevent a repeat of the 1974 floods.

“What Mr Seeney on behalf of the LNP suggests is that Wivenhoe Dam should not be used for flood mitigation purposes,” Mr Robertson said.

“As a result, that puts into jeopardy the very safety of people in Brisbane and surrounding areas.”

And on 20 December, only a few weeks before the flood, Deputy Opposition Leader Springborg repeated in detail the same view (that the dams should store more water, not less) on Brisbane radio.

The decision was made (post flood) to reduce dam levels seeing there is still a La Nina weather pattern hanging around, but I can't say that I have noticed evidence that anyone influential was being outspoken about reducing dam levels in anticipation for summer 2010/11.

In any event, the flood enquiry has taken a curious turn, after having finished its report and sent it to the printers, by virtue of journalist Hedley Thomas (who was running around promoting an "independent" expert or two who were pointing the finger at dam management almost before the flood subsided.)

It turns out that the inquiry seems to have missed that in emails and notes circulating at the time of the flood, the dam engineers were not talking about the same response levels as their later formal report to the Commission indicated.

The dam engineers have therefore been recalled and grilled over how they wrote their report: the accusations of fraud and cover up have flown fast and thick from the Counsel Assisting the inquiry: I have had the feeling that their new found aggression is partly due to the fact that a journalist has shown up the well paid lawyers, as much of this evidence was already before the inquiry, just its importance seems to have been missed.

So, the basic problem is that although the manual gives escalating classifications of response, each involving discretionary faster releases of water, on the weekend before the flood, the engineers were being rather careless to record what level their actual response was at. It even seems a bit unclear whether they recognised the level they were at. So (if I understand it correctly) when writing the report, after an incredibly tiring and stressful period, they looked at how much water they had started releasing, combined with other inflows coming into the river, and petty much retrospectively nominated that they had moved to level W3 by the Saturday morning.

This isn't an ideal way to demonstrate that you were working in accordance with a manual. In fact, in evidence on the last day, one engineer seems to have acknowledged (unwisely, if you ask me) that not knowing what level your response was at would constitute a "breach of the manual". This raises a good philosophical question: if you do the same things a manual would have required just based on your own judgement, you may not have been "following the manual," but have you actually "breached" it?

As far as I could tell from some of the figures, the move out of the lower W1 response did, for much of the weekend, involve not a whole lot more water than the maximum W1 release. (I could be wrong on that, though, as the chief engineer said it was clear that they skipped level W2 and went straight to W3.)

But - given that other independent engineers have already said they think the dam operators mitigated the flood as best they could - getting too hooked up on demonstrating compliance with the manual should surely not triumph over the practical outcome of how they operated it.

As I understand it, each response classification is triggered by the dam reaching certain levels, and it seems the dam engineers certainly recognised the significance of the threshold levels being reached. In other words, they did increase water flow as levels grew, and they did make decisions as to how fast to release water based on how fast the dam level was responding. The manual even at level W3 allows them to consider problems caused downstream by flooding the highest bridges near Fernvale, and they also took that into account in deciding rates of release.

The thing is, as far as I can tell, the manual does not say (at least at the first 3 levels - W4 is doing whatever must be done to save the dam) "once level X is reached, reduce dam levels by Y metres as soon as conceivably possible." And you wouldn't want it to. If there was blue sky forecast for the next week, you wouldn't want to be flooding Brisbane for no real reason.

So: surely you are always going to have to rely on judgements of the dam engineers as to rate of release based on a variety of factors that it is probably difficult to define precisely for all circumstances.

The importance is the outcome, and those who say they should have released a lot more water starting on the Saturday are, of course, doing this with the advantage of hindsight about what was soon to become record inflows into the dam. There must be hundreds of ways to model how releases could have been done differently - but high early releases would have caused earlier flooding of the lowest areas, and how do you recognize at the time the release rates which will turn out to strike the "ideal" balance?

I therefore await the inquiry's findings in this regard with some interest. Non compliance with the manual is said to have significant legal implications, as it would allow class actions. I wonder, however, whether a finding of non compliance might allow legal cases which nonetheless fail due to inability to prove negligence, or flood levels that would have been significantly lower. Surely insurance companies won't get paid much, or at all, if some modelled difference amounts to less than (say) 30 cm? And hydrology seems a rather imprecise science anyway. Lots of Brisbane flooded on land the Council did not expect would flood in a repeat of the 1974 flood - and this one peaked lower.

So the engineers have my sympathy, as do the politicians; the lawyers and the journalists - not much at all. In fact, I suspect Mr Thomas may only be giving false hope to a bunch of witch hunters.

UPDATE: as Hedley Thomas and the Courier Mail are hell bent on criticising the engineers (and just about everyone else associated with the enquiry,) you have to read another media outlet to get the same point I was making. From the ABC:

If the commission finds the engineers breached the operating manual, then it opens the Government to a class action, which law firm Maurice Blackburn estimates could exceed $1 billion.

Insiders question that figure, but regardless of the record keeping both Mr McDonald and an independent hydrologist have found the four engineers released the appropriate amounts of water and that has not yet been challenged.

If it stays that way, it means the only damage from these allegations are to the reputations of the four men.