We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Modelling versus measurement in Queensland

Computer modellers have long been accused of inventing extreme weather events at some deliberately vague point in the future. But here is an Australian story about how another kind of modelling invented extreme weather conditions at an exact moment in the recent past, that other forms of actual measurement didn’t register.

The claim by SEQWater in its official report that a “one-in-2000-year” rainfall event occurred over the Wivenhoe Dam at a critical stage on January 11 has been widely reported in the media and cited by senior public servants to justify the near loss of control of the dam at the time.

But no such rainfall event was measured by any rainfall gauges. Instead, the claim was manufactured by SEQWater after it modelled the rapid rise of levels in the dam, repositioned rainfall data to an area immediately upstream of the dam, and then doubled it.

After extrapolating in this unusual way to achieve an extreme number, the SEQWater report states: “Rainfall of this intensity and duration over the Wivenhoe Dam lake area at such a critical stage of a flood event was unprecedented.

“The resulting run-off could not be contained without transition to (an operating strategy that led to the operator opening the dam’s gate for huge releases).”

But:

Senior independent engineer Michael O’Brien, who has spent the past nine weeks analysing the performance of the dam and SEQWater, said that while the rainfall was heavy, he did not believe it was extreme and he doubted it was ever close to the range claimed by the operator.

This is no mere academic spat. SEQW’s allegedly flawed decision making contributed hugely to the serious flooding that recently hit Queensland.

Mr O’Brien, who has mounted a strong case that the devastating floods in and near Brisbane would have been almost completely avoided with better management of the dam, said the one-in-2000-year event was an “invention” that could not be taken seriously.

The modelling-trumps-measurement vibe to all this is the reason that climate skeptics like Anthony Watts are already onto this.

Delingpole hasn’t yet had a gloat about it all, but doubtless he will, because this just begs to be amplified into a big story, of the sort that the world’s Old School Media will either run with, or make further climate-prats of themselves by ignoring.

Tweet about this on TwitterShare on FacebookShare on LinkedInShare on TumblrShare on RedditShare on Google+Share on VK

4 comments to Modelling versus measurement in Queensland

  • Gene

    This looks like a huge story but I’m disappointed that the journalism about it isn’t better. What I’ve read does not make the story very clear, and really I saw no discussion of why SEQWater did anything that it did.

  • It’s pretty simple. The I’ve seen a report that the government told SEQ water to run the dam fuller than it should have been for flood control purposes in order to use cheaper dam water than expensive desalinated water for South East Queensland’s water supply.

  • Dyspeptic Curmudgeon

    From comments at various Oz sites, it appears that everyone was so caught up in the ‘global-warming-will-cause-endless-drought’ mind-set, that the dam was not opened at an early stage (60-70% capacity) so the water “would not be wasted”. By the time the dam was opened ( at 100+% of usual capacity), the waterflow was horrendous and damage downstream was inevitable.

    One should always remember Murphy’s Fifth Law: Mother Gaia is a bitch.

  • Paul Marks

    Errrr – the fact remains.

    The fact that the company LIED about the level of rainfall – in order to cover up its own incompetance.

    If the “mainstream” media ignore this LIE then there is no point in the msm continuing to exist.