Monday, 2 April 2007

NIWA should be shut down - Auer

Nonsensical comments over Northland's flooding show that the Government's National Institute of Water and Atmosphere (NIWA) should be shut down, says Augie Auer, chair of the science panel of NZ's Climate Science Coalition. "So simplistic, it's silly" is how Dr Auer describes the statement by NIWA climate scientist Dr Jim Salinger over the flooding.

"As an explanation of the cause and consequences of last week's Northland rains, Dr Salinger's statement ... is as unscientific as it is incorrect. "
At a time when MetService and NIWA are at such loggerheads that a mediator has had to be appointed, it's time for the Government and the Opposition parties to disband NIWA to stop its nonsense statements and return its functions to the agency where they rightly belong, MetService. At least MetService deals in the real world of observation of actual weather events, and forecasts based on those events, as well as its links with World Meteorological Organisation, as New Zealand’s official representative on that body. This is the real world of weather, as opposed to the computerised ‘statistical models’ on which Dr Salinger says NIWA relies” said Dr Auer.
RELATED: Global Warming, NZ Politics


  1. PC, let's kick this one around for a bit then, shall we?

    Augie Auer calls for Niwa to be 'shut down' for releasing 'simplistic' statements of science to the press. I don't intend to go into the accuracy of Salinger' simple soundbite, apart from to make the following points:
    - Soundbites by scientists to the press are often simple of necessity.
    - The comment was not made by way of attribution for the current weather event, it was made generally about the expected frequency of weather events looking forward.

    Turning briefly to NIWA, most people are aware that it is a body which focuses on quite a bit more than simply climate science. Aquaculture, biotechnology, oceanic and coastal habitat research, etcetera.

    Now Augie feels that this level of 'dumbing down' is bad enough to actually 'shut down NIWA' and return climate issues back to his ancestral home at Met. We know, as Augie points out, that Met doesn't actually do any long or medium range forecasting. So the assumption is that Met resumes taking up long or medium range forecasting; however the tacit suggestion is that long range forecasting is not necessary for climate study - that observation is the key:

    "At least MetService deals in the real world of observation of actual weather events, and forecasts based on those events [...] as opposed to the computerised ‘statistical models’..."

    So what Augie is not actually driving at is the punishment of a governmental scientific body for dumbing down science, but in fact the replacement of computer modelling with observational data - essentially, the forsaking of long-range forecasting as used internationally.

    Whilst one can certainly posit that Augie is dealing in unreasonable hyperbole (shutting down a large, diverse and internationally respected scientific organisation for an isolated comment, dealing with only one of many issues) the scarier implication is his proposed shackling of climate study.

    The bottom line I guess, without wishing to enter the fraught area of equivalence, is that a retired meteorologist is calling for the disbandment of a highly successful, respected, and credible organisation, with the only suggestion for restitution of it's many functions being that climate study should be taken up by the government body for which he was previously the head. And that study should be severely curtailed.

    In a word - ridiculous.


  2. Auguie Auer said...
    This is the real world of weather, as opposed to the computerized ‘statistical models’ on which Dr Salinger says NIWA relies” said Dr Auer.

    I disagree here with Prof. Auer. There is nothing wrong with computer modeling at all, because most gadgets and new emerging technologies have been developed using computer modeling. There are millions of these examples. The design of DSP chipset (digital signal processor) which is found in most modern telecommunication technologies of today (mobile phones, computers, PDAs, fibre-optic communications systems, TV, stereo systems, medical technologies - lasers, MRI, etc, etc) are by way of mathematical & statistical modeling. DSP chip must be simulated, by sending a mathematical signal to the model and test how well the chip responses to this particular mathematical signal's frequency. Can the mathematical chipset model where it only exists on computer screen and not a real physical proto-type is able to de-noise the hypothetical mathematical noise signal or not? If not, then change the architecture of the chip and try again. This process is iterative until the best model is discovered. The final model is then transferred to the factory floor for production. The majority of the finished products do perform with in the margin of error of the computer model. Car manufactures built mathematical cars on screen (virtual cars) and try to crash test those virtual ones and apply "what-if-scenarios" using Newton mechanics, which then give them hints of where the weaknesses in some structures are located. Change those parameters then run again. Again it is an iterative process. The performance of a real car built according to the simulation specification would not be far away from the performance of the mathematical car on screen. NASA does a lot of this stuff. They built mathematical rocket and then test the potential thousands of variables affecting its operations.

    All those mathematical simulations avoided the costly building real proto-type and then damaging them during a real test. This is completely unaffordable. Without mathematical & statistical simulations in today, technology would be very slow to come out to fruition from research lab into commercial environment.

    The difference between climate modeling & the engineering ones above are that climate is more complex to model than model a DSP chip. So, one would always cautious about claims from models in complex systems such as climate. In comparison to say a claim made by Intel Corporation that their researchers and academic partners are currently developing a one square centimeter size Photonic Chip that will replace the 2-meter high phone exchange cabinet in the near future, I would take their claim a bit more serious, since it is more likely that you're going to see this in a commercial environment.

    NIWA doesn't need to be shut down. It needs to be transformed, perhaps a panel of scientific experts & from our Universities in disciplines as Mathematics, Statistics, Physics, Engineering (particularly Mechanical & Electronics), Computer Science, Geology, Geography, should spearhead a government program to do exactly this.

    The problem is not the mathematics itself, but it is the wrong model formulations & insufficient model assumptions that went to building those computer models. A government lead scientific expert panel perhaps might advise NIWA on things that they don't have expertise on. This will make NIWA much better in their line of work, plus they should pull away from being alarmists.

    I am pro-mathematical modeling and anti-global-warming-alarmists. You can’t condemn mathematical modeling because of AGW. They are 2 different things and I am sure that Augie is wrong here.