Wednesday 26 July 2006

Climate change - open letter to MPs

The NZ Climate Science Coalition have sent an Open Letter to MPs today "urging them to support the Coalition's request to the Prime Minister for a Royal Commission on the Science and Economics of Climate Change." The full text appears here:
The New Zealand Climate Science Coalition has taken the unusual step of delivering (Tuesday 25 July) to every MP in Parliament an Open Letter to Members of the House of Representatives, urging them to support the Coalition's request to the Prime Minister for a Royal Commission on the Science and Economics of Climate Change.

The reasons justifying a Commission are set out in the Open Letter and its explanatory summary, and also in the attached media release.
In anticipation of untrue and entirely unjustified attacks on our members by the likes of Greenpeace claiming that we are simply tools of the fossil fuel industry, etc, we attach also for your background information, details of the qualifications and motivations of the signatories to the letter.

Recent developments make it clear that the science of climate change is far from settled. Our Coalition of qualified people believes that a deep, thorough and independent investigation by a Royal Commission is a necessary prerequisite for setting new national climate and related energy policies. Given the strength of the vested interests involved, we assert that such a Commission is the best way in which to separate fact from fiction and emotion.

We hope that the press will report, analyse and give editorial support to our call for a Royal Commission into climate change.

Terry Dunleavy
Hon Secretary
New Zealand Climate Science Coalition.
Accompanying the letter is a fourteen-page report backing up their call, which includes a summary of the nine reasons that require it:
Nine Reasons why the New Zealand Climate Science Coalition believes that a Royal Commission into the science and economics of climate change is justified:
  1. There is no scientific consensus about man-made global warming (page 2);
  2. The extent to which the globe is warming is questionable (page 3);
  3. New Zealand climate data shows little evidence of warming and no evidence that humans are affecting the temperature (page 4);
  4. Projections of changes in the climate are based on unproven computer models (page 6);
  5. The IPCC uses circumstantial argument, vagueness and ambiguity to hide the fact that direct evidence to support man-made global warming is lacking (page 7);
  6. Carbon dioxide is essential to life on earth and an increase in its atmospheric concentration may be beneficial (page 8);
  7. There is a poor relationship between carbon dioxide concentration in the atmosphere and global temperatures (page 9);
  8. The Kyoto Protocol will not prevent global warming, and is already projected to cost the nation over $1 billion more than was originally estimated (page 10);
  9. Advice to the government has thus far been one-sided, with little provision for contestability or open debate (page 11).
If you agree that a Royal Commission is justified, you might like to write to your MP yourself to support this call.

LINKS: MP's contact details - Parliamentary website
Climate Science Coalition website
Open Letter to members of NZ Parliament - NZ Climate Science Coalition [14-page PDF]

RELATED: Environment, Global Warming, Politics-NZ, Science

11 comments:

Anonymous said...

Fairly disingenuous list. For example (relevant to the list):

1. If by consensus they mean 'There is no complete agreement' then, yes, that's true. If my consensus they mean 'a majority of climate scientists think...' then there is consensus that global warming is aided by human actions.

3. If global warming is occurring then it doesn't mean we should see uniform changes all over the globe; indeed, with current weather patterns it is still possible for global temperatures to rise but local temperatures in some localities fall. Thus New Zealand's climate record is only useful in comparison to the world and not a good indicator of policy (if we are concered for the global environment (which arguably we should be as climate change elsewhere will eventually affect us)).

5. This is true of both sides of the debate.

7. Whilst it can be argued that there is a poor direct relationship between CO2 levels and climate change it is enough to suggest that the correlation is causal (this is similar to the 'smoking cause lung cancer' issue that Tobacco companies are fond of).

8. Whilst it may not prevent global warming research seems to show that it will mitigate its effects, which is better than inaction at this stage (and inaction will be more costly long term).

9. This sounds like the ID slogan 'Teach the controversy.' Debates in the Sciences are rarely 'even-handed' in that where a clear consensus (most, not all) is held that is usually taken to the norm for related action. We don't think that Flat Earthers should be allowed to oppose NASA's funding just because their scientific view opposes that of other cosmologists, do we?

Libertyscott said...

Typical gutlessness when someone anonymous posts. To equate climate change skeptics to flat earthers is just ludicrous.

As one who has advised the government and been alongside those dealing with these issues, you get looked at askance when you raise doubts about whether certain measures to reduce emissions should not proceed - because the net economic impact is negative, and the impact on CO2 emissions is very marginal indeed (the literal piss in the ocean).

Only Treasury provides robust debate about this in the public sector, the arguments that climate change measures should only be taken if they deliver net overall benefits are not ones that environmentalists argue - since they tend to oppose economic benefit/cost analysis.

Anonymous said...

“Anonymous said...”
[#1. If by consensus they mean 'There is no complete agreement' then, yes, that's true. If my consensus they mean 'a majority of climate scientists think...' then there is consensus that global warming is aided by human actions.]

Consensus does not mean reality. Mathematical models are the reality in the eyes of researchers. So, this is where blindly accepting the consensus of the majority has to be questioned. Most scientists now recognize climate dynamics as truly exhibiting "Complex Systems" behaviour, which originated from "Theory of Complexity". Such systems include but not limited to Economic systems, Biological systems, Human colonies (society), and so forth.

“Complex Systems”
http://en.wikipedia.org/wiki/Complex_system

Complex systems is typically modelled by smaller sub-systems that interacts with each other thus forming complex behaviour which is stochastic (indeterminate but predictable). This means that if you do a multiple run of the same input test data to the same model of an "Evolving Complex Dynamical Systems", you will always have outcome that varies from the current test to the next ones. So, from here, you can pretty much figure out why 2 different researchers would tend to disagree on each other’s models.

“Anonymous said...”
[# 3. If global warming is occurring then it doesn't mean we should see uniform changes all over the globe; indeed, with current weather patterns it is still possible for global temperatures to rise but local temperatures in some localities fall. Thus New Zealand's climate record is only useful in comparison to the world and not a good indicator of policy (if we are concerned for the global environment (which arguably we should be as climate change elsewhere will eventually affect us)).]

Of course it does not mean we should see uniform changes all over the globe if global warming is taking place. What has been established, as fact is that we are seeing an uptrend in temperature? What has not been established for certainty, if this is caused by human actions? I emphasized on the quote “What has not been established for certainty ”, to mean that I have seen scientific publications which have appeared within the last 2 or more years recently, questioning the consensus. Green Peace call this refutation from those publications as researchers funded by big corporate whom are anti-global warming. This is a typical comment from Green Peace (GP) spokesperson, Vanessa Atkinson that I think she is more of a tree hugger, rather than a GP mouthpiece.

The main report that GP and environmental lobbyists campaigned for has been the ‘Hockey-Stick Model’, published by Man et al. I have a summary of this report and I can see some weak data analysis techniques used. As a person who specializes in mathematical computing and data analysis, I can pick out some flaws in their choice of algorithms (mathematical techniques). They used a multivariate statistical technique called PCA (Principal Component Analysis), which is a linear method that de-correlates independent variables. If you don’t understand PCA, in my criticism of the ‘Hockey-Stick Model’, don’t bother, you are already using it everyday via your computer. PCA is solved using a linear algebra method called ‘eigen-decomposition’ which is exactly what Google search engine algorithm is. Google uses an algorithm called ‘PageRank’ invented by their founders (Larry Page & Sergey Brin), which is solved using ‘eigen-decomposition’. So, every time you use Google search, you are using a form of PCA.

“Principal Component Analysis”
http://en.wikipedia.org/wiki/Principal_components_analysis

First, PCA works well in data that is linear (all variables are assumed to be independent), but the world which includes climate and weather data are non-linear, therefore the Hockey-Stick team should have used one of the thousands of non-linear models available. There is a version of PCA that is non-linear which is called KPCA (Kernel Principal Component Analysis) and I wondered why Man et al, did not use KPCA for their fraudulent ‘Hockey-Stick’. I assumed that they have no clue at all to KPCA or even aware that there is such a model. Now, this is my main issue with the consensus of scientists. How many scientists those have agreed with the general consensus that global warming is caused by human actions rather than a natural cycle, who are truly capable of understanding complex mathematical modelling? I believe that only a handful of them out of the so called ‘few hundreds’. The reason, I brought this up, because climate modelling is mainly mathematical modelling. This domain of mathematical computing is also known as “scientific computing”. The reason, it is called scientific computing is because it is the same fundamental mathematics (numerical methods) that is used in climate modelling is also used in particle physics, molecular biology dynamics simulations, computational economics, satellite image analysis, and so forth. Not only, there is a huge amount of computer processing power needed for this type of modelling, where such simulations can run for hours or even days using super-computers. There are a lot of those doing scientific computing who are dedicated to weather simulation as can be found on the internet from different institutions that I have but I doubt that the majority of consensus scientists do even understand differential calculus. In my opinion, I think that most of them just jump the bandwagon and agreed for the sake of agreeing. I will check this out over time as I just have come to know some names. Scientists names these days can be done with a Google mouse-click, where it will give you either a publication title from a recognized journal in weather review or cross-disciplinary journal as “Physics Review Letters” or “International Journal of Fluid Dynamics”, etc, where I have seen an abstract of a paper in that journal for modelling related to ‘atmospheric heat convection flow’. You can obviously tell if the publication uses simple models or complex models, which landed it to high-speed number-crunching simulations. If a consensus author publishes simple model, then it can be concluded that such author does not comprehend complex mathematical modelling and therefore he/she is one of those that just jumped the bandwagon, because his colleagues are in the consensus camp.

“Anonymous said...”
[#7. Whilst it can be argued that there is a poor direct relationship between CO2 levels and climate change it is enough to suggest that the correlation is causal (this is similar to the 'smoking cause lung cancer' issue that Tobacco companies are fond of).]

WRONG. As I have described in the previous paragraphs, that weather and climate phenomena exhibit traits of a complex systems and therefore should be treated with ‘complex non-linear dynamical systems’ model. This means that high-school statistics method as ‘correlation’ is also out. I don’t know how familiar are you with modelling of dynamical systems, that is a MIMO (multi-input & multi-output) systems. In short, you can input many variables to the model, and then observe the many output variables, responses. So, correlation of one to one (one input to one output) must be thrown out, as it does not apply to complex dynamical systems as weather. I have quoted and algorithm called KF (Kalman Filter) in a previous comment on a different thread, which is a MIMO method. KF can be formulated as a MIMO via a technique called ‘State-Space’, which was originated in the field of robotics & control systems design. Just look at the link shown below, then scroll down to the sub-heading “Feedback with setpoint input” , where you can see a block-diagram on the right hand side of the page for a SISO model (Single Input – Single Output). In modelling MIMO systems as weather data using a Kalman Filter or a different ‘State-Space’ algorithm, then your model can be built to be complex as a circuit board with many components inter-connecting to other components where their interaction is via non-linear signal flow. That is why simple one-to-one ‘correlation’ does not apply here and must be thrown out completely as irrelevant. Weather is a many-to-many variable relationship.

“State space (controls)”
http://en.wikipedia.org/wiki/State_space_%28controls%29


“Anonymous said...”
[#8. Whilst it may not prevent global warming research seems to show that it will mitigate its effects, which is better than inaction at this stage (and inaction will be more costly long term).]

It is important to spend money wisely on something we know that is truly happening and not spend any because some Green Peace tree hugger lunatics have drummed up fraudulent model, such as ‘Hockey-Stick’ and presented it as fact.

“Anonymous said...”
[#9. This sounds like the ID slogan 'Teach the controversy.' Debates in the Sciences are rarely 'even-handed' in that where a clear consensus (most, not all) is held that is usually taken to the norm for related action. We don't think that Flat Earthers should be allowed to oppose NASA's funding just because their scientific view opposes that of other cosmologists, do we?]

Irrelevant. Intelligent Design (ID) is not science, period. ID has no physical hypothesis that can be tested at all. Climate & Weather modelling obeys the laws of Physics (they are indeed a manifestation of interaction of physical entities) and therefore can be tested according to those known laws. However the proper model has to be developed and outputs have to be properly examined using advanced methods and don’t prematurely jump to conclusions.

Berend de Boer said...

David Parker: Climate Change Minister David Parker rejected the proposal for a Royal Commission, saying that "by far the majority" of climate scientists in the world agreed there was no longer any doubt the climate was changing due to human activity.

Consensus science, the greatest innovation in science since the Novum Organum.

Anonymous said...

Anon: 'Whilst it can be argued that there is a poor direct relationship between CO2 levels and climate change it is enough to suggest that the correlation is causal (this is similar to the 'smoking cause lung cancer' issue that Tobacco companies are fond of).'

Weak argument and even weaker example.

Smoking, however less than salubrious, does not 'cause' lung cancer per se - if it did, every smoker would eventually get it and they don't.

Secondly, that analogy has never explained the fact that 80% of Japanese men smoke and yet have only 20% the lung cancer rates of the west.

The dichotomy is much better explained by the high prevalence of fish oil in the Japanese diet.

But, like global warming, why let the facts get in the way of a good (sensationalist) story, eh - particularly one where loads of public money is up for grabs ...

Anonymous said...

Berend de Boer said...
[David Parker: says that "by far the majority" of climate scientists in the world agreed there was no longer any doubt the climate was changing due to human activity.]

I think that the minister (David Parker) is a sucker and naive person. He gets conned easily by Green Peace & Green Party supporters into believing this bullshit global warming consensus which they say is final.

Anonymous said...

Falafulu Fisi said...
[...climate modelling is mainly mathematical modelling. This domain of mathematical computing is also known as “scientific computing”. The reason, it is called scientific computing is because it is the same fundamental mathematics (numerical methods) that is used in climate modelling is also used in particle physics, molecular biology dynamics simulations, computational economics, satellite image analysis, and so forth. Not only, there is a huge amount of computer processing power needed for this type of modelling, where such simulations can run for hours or even days using super-computers.]


Here is what the scientific computing group at NIWA are doing with super-computers for weather simulations.

"NIWA High Performance Computing Facility"
http://www.niwascience.co.nz/rc/hpcf/


They use Fortran-90 computer language in the development of their simulation software. Knowing that developing numerical codes are complex and time-consuming, I suspect that NIWA simulation scientists do use industry standards, publicly available open source parallel fortran-90 library available from Sandia Corporations (US government Dept) with packages as Trillinos and LAPACK (linear algebra package). LAPACK is embedded in all major computer chip that is available today. Intel, Sun MicroSystems and others have modified the standard LAPACK for processing speed. I doubt that the NIWA people will spend huge amount of money to develop their own linear algebra package from ground up, instead of using a stable (or almost bug-free) library which is already available in the public domain.

However some scientific computing developers still use parallel Fortran-90 for speed of execution there are a lot of researchers moving in to modern language as Java. Because of Java's multi-thread (multi-task) built-in to the language, which any long running simulations (hours or days) can temporarily stop regularly during its course to write the simulation results to disc. In this way if something went wrong with the computer during the long running simulation, there is already data saved. So, if the simulation is restarted, then the simulation is continued on the point with where it stopped rather than starting from the beginning. Fortran-90 is not multi-thread at all. A particularly favourite free-downloadable high performance numerical Java library that I use in my development is called COLT developed at CERN (European Organization for Nuclear Research), which they use in particle physics simulation.

"COLT"
http://dsd.lbl.gov/~hoschek/colt/index.html


COLT is popular worldwide amongst numerical computing Java developers & researchers. I have noted from the COLT mailing list, which I am subscribed to, in that JPL (Jet Propulsion Laboratory) of NASA does use COLT library for their rocket flight aero-dynamics design simulations. Those researchers do frequently ask questions to the list. Their simulations involve testing the stability of structures before any proto-type is built, thus saving money. They develop codes for ‘finite element analysis’ (FEA) technique for their analysis. Civil engineers do frequently use FEA for analysis of large structures as in bridge design. Auckland University department of Bio-informatics (marriage of computer science, biology & mathematics) do use Java in their software development for DNA sequencing & gene discovery simulations. They have been renting the spare time of Peter Jackson's WETA workshop super-computer to run their simulations, which usually take hours. Perhaps, internal university politics that forces a crown organization (Bioinformatic Department) to go outside looking for resources. I know that the Auckland University has only one supercomputer housed at the School of Engineering but its always fully booked in advance by other departments such as, the Electrical Engineering (large scale electromagnetic simulations), Mechanical Engineering (intensive computational fluid dynamics simulations) & Engineering Science.

“Bio-Informatics Institute”
http://www.bioinformatics.org.nz/


From NIWA’s description in high performance computing it says that they do Monte Carlo simulations, for estimating models of fish population. This is a popular algorithm for general data analysis. The Bioinformatics researchers at Auckland are using it to detect the anomaly of defects or abnormal gene sequence. Knowing anomaly gene sequence can help the diagnosis of certain diseases. I myself do develop Monte Carlo codes for financial modelling of Option Pricing (European Put & Call). Typical number of trial run using my PC is 15 millions max in 17 minutes, and if go above this max, my machine crashes, because of not enough memory. The higher the number of trial runs, the more precise the price of the calculated option. Perhaps I do need to use a super-computer such as the one at NIWA to run my codes and stretch the trial run even upto 100 millions or more. Typical Monte Carlo trial runs that are frequently done in institution as OCCF (Oxford Centre for Computational Finance) can go up to half a billion or a billion. In this range of trial runs, it will take number of hours or days to run. Well, they do indeed use a super-computer to crunch this sort of numeric intensive data analysis.

“Oxford Centre for Computational Finance”
http://www.occf.ox.ac.uk/

The yacht designer for Team New Zealand (TNZ) for the America’s cup in 2003 used Monte Carlo in some of their simulations. The design engineer was interviewed on TV and he described how they usually leave the simulations to run overnight in every night, where they retrieved the results in the morning for analysis usually by visualization softwares. He described of how to find a best geometry of the keel by knowing exactly which shape will give less drag, which will slow, down the yacht. So , this team were doing a fluid dynamic analysis. The drag is worst when voids (vortex) of fluid (swirling air pockets) formed between the interface of the keel and the seawater and hanging around a bit longer (in the ranges of 1 to 4 seconds). Voids are created and destroyed continuously as the floating object (keel) is moving. Average lifetime of voids depend on the their sizes, densities, swirling velocities which varies, shape of floating object and many other variables. The TNZ designers configured the software to continually morph into different shapes, where hypothetical random generated voids of different sizes, densities, and swirling velocities are bombarded against the morphed geometry. This randomly generations of voids and tracking their random walk paths in the simulation is a “Monte Carlo” process. The result is then retrieved and analysed to find out the shape (geometry) that produces the least amount of drag.

Monte Carlo is heavily used in Weather Simulation for convectional directional flow simulations and many others. Scientific computing can vary across disciplines, but the majority of the fundamental mathematics is the same, whether it is economics, bio-dynamics or rocket design, or whatever.

I hope that numerical computing team at NIWA will tell their boss(es) to convey to the government that global warming debate is not final yet. There needs to be more analysis and modelling done before policy implementations is rushed thru parliament. The team at NIWA and any scientists or software developer who understand mathematical computer modelling can stand up and challenge those consensus scientists as I believe that majority of them don’t understand complex modelling, they just want to be part of a camp that happens to be the majority, because they think its cool to be in the right side of the public opinion.

Anonymous said...

Falafulu Fisi said...
[Green Peace tree hugger lunatics have drummed up fraudulent model, such as ‘Hockey-Stick’ and presented it as fact.]


Now here is the link for the controversial paper from Prof. Michael Mann on 'Hockey-Stick' global temperature computer model. It is free to download the PDF format. The paper was originally published in Nature science journal.

"Global-scale temperature patterns and climate forcing over the past centuries."
http://www.geo.umass.edu/faculty/bradley/mann1998.pdf

As I have mentioned in my criticism of this model that if any of the many non-linear models available from mathematical literatures such as K-PCA (kernel principal component analysis) been used instead of straight linear PCA that Mann et al adopted, then I believe that the 'Hockey-Stick' shape might look different or completely disappeared. I have just emailed Prof. Mann requesting his data for that paper so I can run a non-linear analysis on it. Prof. von Storch of Plank Institutes have found that Mann et al analysis methods were flawed. Also other researchers have confirmed this. I basically want to see it for myself if I can run some analysis using Mann's original dataset, provided that he will be kind enough to send me the data. Others who challenged Mann's flawed 'Hockey-Stick' have found that Mann and the rest of his team have refused in the past to give their data to other researchers. Some have concluded that Mann et al , feared that other researchers might reproduced their claim made in their publication and found out the holes and flaws in their methods & theory.

Here is Prof. Mann's home page:
http://holocene.meteo.psu.edu/Mann/

Anonymous said...

Just for interest:

MM5 is an advanced, and free, model used by the metservice - you can find more information about it here. It'll run on a fairly grunty Linux machine if have one lying around.

Also, if anyone saw the NZ Herald article by Geoff Austin (prof of geophysics, Auckland Uni) you can read it here. He is currently working on a variation of the MM5 model for predicting weather patterns on Mars. Very interesting stuff.

Anonymous said...

Polemic said...
[Geoff Austin - prof of geophysics, Auckland Uni]


You are right here. I read Prof. Austin's article the other day on the NZ Herald. Here is someone who has been there and involved in research during similar claim by alarmists made in the 1970s about the end of the world by the turn of the century (year 2000). Now its 2006, and we are still enjoying watching the All Blacks beating the Wallabies on TV. Its good that scientist like him speaks out so as to counter-balance the Green Peace nutters and their team of consensus scientists.

Hey Polemic, was Prof. Austin became Physics HOD at some stage ? I had long gone when he was HOD in the mid to late 1990s or around that time.

Cheers,

Falufili Fisi

Anonymous said...

If he was, he is not anymore, although I think he went away for a while. I'm actually taking one of his courses this year (trying to finish my long neglected degree) and it's been an eye opener.