Bob Tisdale asks the wrong people the wrong question @wattsupwiththat

Preamble

I started writing this after Bob Tisdale's article at WUWT about climate models (archived here).  Since then Bob has announced he is quitting full time hypothesising about leprechauns warming the oceans and has opted for more gainful employment.  He tells us that he's not able to earn a living from rejecting climate science. He warns that we haven't seen the last of his magical musings, it's just going to be on a part-time basis from here on in.

You'll note that Anthony Watts has implied at WUWT that none of "big oil" and "big coal" and Donor's Trust think much of Bob's "oceans warm by magic" and other equally silly hypotheses. (Archived here.)  They are more strategic with their political investments and don't waste money on mickey mouse denier bloggers preaching only to the denier chorus.  Their focus would be on trying to con normal people and influencing people of influence, not two-bit denier bloggers. (Click here for Robert Brulle's paper in Climatic Change, with 120 page supplement here. And here for the related article in Nature News.)

Greenhouse effect denier Bob Tisdale has come up with seven questions that he wants policy makers to put to climate scientists (archived here).  I don't know if he wants them to ask any or all climate scientists or only those who work with complex earth system models.  He doesn't say.  His article has the title:

Questions Policymakers Should Be Asking Climate Scientists Who Receive Government Funding

I'll argue that Bob is asking the wrong questions of the wrong people.  (I'd also argue that Perennially Puzzled Bob Tisdale is not in a position to be asking questions of anyone until he makes the effort to learn some basic science himself.)

What Bob does is complain that climate models aren't any good.  By that I think he means that climate models are not perfectly aligned with observations. I say to Bob, show me a model, any model, of anything, that is.

This article is long so I've put in a page break.  If' you're on the home page, click here to read on.

First Bob complains about science that is "settled"

But first he makes a very strange statement.  Read this and wonder:

The recent 5th Assessment Report from the Intergovernmental Panel on Climate Change (IPCC) proclaims that global surface temperatures are projected to increase through the year 2100, that sea levels will continue to rise, that in some regions rainfall might increase and in others it will decrease, etc. But those were the same basic messages with the 4th Assessment Report in 2007, and the 3rd Assessment Report in 2001, and the 2nd Assessment Report in 1995. So we’ve received little benefit for all of those tax dollars spent over the past few decades.

Bob seems to think the "basic messages" should have changed.  I don't know what he thinks the latest report should have contained.  Was he wanting to read that global surface temperatures will decrease through the year 2020 - like Pierre Gosselin predicts:

Or maybe he's waiting to be told that seas will stop rising, ice will stop melting, some regions rainfall will increase but in others it will decrease etc.

Does Perennially Puzzled Bob Tisdale really and truly expect that the world (or science) would have changed so dramatically in seven years.  Has he himself not read any scientific papers since 2007 that he is so bamboozled that while scientists fill in more knowledge gaps day by day, the basic science is settled?

Science deniers just hate admitting that much of the science is settled.  They want to deny that fact.

Bob's denialism is showing

Bob likes to use words that he knows will please his science deniers.  He writes (I've put Bob's subliminal dog-whistling denier-speak in bold italics):

Those predictions of the future are based on simulations of climate using numerical computer programs known as climate models. Past and projected factors that are alleged to impact climate on Earth (known as forcings) serve as inputs to the models. Then the models, based on more assumptions made by programmers, crunch a lot of numbers and regurgitate outputs that are representations of what the future might hold in store, with the monumental supposition that the models properly simulate climate.

Which only goes to show that he's either deliberately deceiving WUWT readers or he hasn't read the IPCC report -  which is the same thing, because he's making believe he has read it.

The IPCC report devotes an entire chapter to discussing the merits and deficiencies of climate models. It's Chapter 9: Evaluation of Climate Models.  Bob already has a lot wrong, and he's barely started his long-winded article.  Let's see now.

There are many different forms of climate models designed for different purposes, ranging from fairly simplistic energy balance models (eg for crude estimates of climate sensitivity) to very complex models that have models within models to explore dynamics of the climate/earth system in the past and simulate the future response to forcings. Chapter 9 discusses four types of climate models in some detail.  These are:

9.1.2.1 Atmosphere-Ocean General Circulation Models (AOGCMs) 

AOGCMs were the “standard” climate models assessed in the AR4. Their primary function is to understand the dynamics of the physical components of the climate system (atmosphere, ocean, land, and sea-ice), and for making projections based on future greenhouse gas and aerosol forcing. These models continue to be extensively used, and in particular are run (sometimes at higher resolution) for seasonal to decadal climate prediction applications in which biogeochemical feedbacks are not critical.

9.1.2.2 Earth System Models (ESMs)

ESMs are the current state-of-the-art models, and they expand upon AOGCMs to include representation of various biogeochemical cycles such as those involved in the carbon cycle, the sulphur cycle, or ozone (Flato, 2011). These models provide the most comprehensive tools available for simulating past and future response of the climate system to external forcing, in which biogeochemical feedbacks play an important role.

9.1.2.3 Earth-System Models of Intermediate Complexity (EMICs)

EMICs attempt to include relevant components of the Earth system, but often in an idealized manner or at lower resolution than the models described above. These models are applied to certain scientific questions such as understanding climate feedbacks on millennial time scales or exploring sensitivities in which long model integrations or large ensembles are required (Claussen et al., 2002; Petoukhov et al., 2005). This  class of models often includes Earth system components not yet included in all ESMs (e.g., ice sheets). As computing power increases, this model class has continued to advance in terms of resolution and complexity.

9.1.2.4 Regional Climate Models (RCMs)

Regional climate models (RCMs) are limited-area models with representations of climate processes comparable to those in the atmospheric and land-surface components of AOGCMs, though typically run without interactive ocean and sea ice. RCMs are often used to dynamically ‘downscale’ global model simulations for some particular geographical region to provide more detailed information (Laprise, 2008; Rummukainen, 2010). By contrast, empirical and statistical downscaling methods constitute a range of techniques to provide similar regional or local detail.

All models are wrong...

By now I hope you're seeing what I am seeing - the enormous contrast between Bob Tisdale's "all the models are wrong so let's throw them out" approach and the IPCC's approach - what's good, what's not so good and how models are being improved.  It's chalk and cheese.  As climate modelers often say, quoting George E. P. Box,

"Essentially, all models are wrong, but some are useful."

Bob doesn't allude to the fact that there are different models for different purposes.  He "alleges" that there is a "monumental supposition that the models properly simulate climate".  By contrast, the IPCC devotes an entire chapter to discussing the strengths and weaknesses of different types of models as well as different individual models.

The IPCC report explains that models used in CMIPs (Coupled Model Intercomparison Projects) are looked at closely, although understandably there has been much more analyses to date of the earlier CMIP3 models than the more recent CMIP5 models.

What is parametization?

Bob Tisdale refers to "assumptions".  It's not clear whether he is referring to known physics or not.  He might be referring to parametizations.  My understanding is that can be thought of as models within models. Scientists model separately processes that are too complex or operate at too fine a scale to be directly incorporated into the larger scale earth system models.

I doubt there would be anyone who would reject the notion that the earth is a complex place.  To help make sense of it and figure out how a change in one thing can affect another, it helps to think of earth as a system with a whole lot of different parts that affect and are affected by changes in other parts of the system.

Now if you were to build a model that calculated every single thing that happened in the earth system then you'd effectively be building another earth.  Think about all the things that happen every second.  Even the fact that you are reading this means you've breathed in an oxygen rich mixture and exhaled a CO2 rich mixture, which has made a minute if temporary change to the composition of the air. (Plants do the reverse, they take in CO2 and produce oxygen - during sunlight hours.)

How much difference have you as an individual made to the climate by breathing or eating? None that matters.  There are bigger changes happening that have swamped the effect of your breath or digestive system.  So don't worry. You're not about to absorbed into any model à la The Matrix.

Plants in aggregate can make a difference, however.  Some advanced climate models do simulate biological and geochemical processes.  Think about what happens in a cloud, or in a field or in the ocean.  There are small scale changes that when added together have a noticeable effect and these can add up and affect other things in such a way that they even help shape the climate.

If you were to build a machine, you'd probably build the casing in one factory and  the small scale computer components in another factory.  Then you fit them together so that the computer can fit into and affect (and be affected by) the large scale components.

That's how I understand parametization.  One system is built to model large scale dynamics of the ocean-atmosphere-surface and other systems are designed to model, for example, the microphysics in clouds and ocean chemistry.  The small scale components are "parametised".  Thus the large scale models are "tuned" with these small scale parametisations to get a more realistic picture of the earth system.  (If I'm using the terms wrongly, maybe a reader will help explain.)

I'll add this explanation of parameterizations in the context of climate models, from an excellent ArsTechnica article on climate models:

Climate models are, at heart, giant bundles of equations—mathematical representations of everything we’ve learned about the climate system. Equations for the physics of absorbing energy from the Sun’s radiation. Equations for atmospheric and oceanic circulation. Equations for chemical cycles. Equations for the growth of vegetation. Some of these equations are simple physical laws, but some are empirical approximations of processes that occur at a scale too small to be simulated directly.

Cloud droplets, for example, might be a couple hundredths of a millimeter in diameter, while the smallest grid cells that are considered in a model may be more like a couple hundred kilometers across. Instead of trying to model individual droplets, scientists instead approximate their bulk behavior within each grid cell. These approximations are called “parameterizations.”

Connect all those equations together and the model operates like a virtual, rudimentary Earth. So long as the models behave realistically, they allow scientists to test hypotheses as well as make predictions testable by new observations.

What is meant by "tuning"?

Deniers will often claim that models are "tuned" to give the impression that climate scientists are simply making up stuff.  Deniers are wrong, of course.  Here is how the procedure of "tuning" is described in Chapter 9 of the IPCC report (WG1):

Individual model components (e.g., the atmosphere, the ocean, etc.) are typically first evaluated in isolation as part of the model development process. For instance, the atmospheric component can be evaluated by prescribing sea surface temperature (Gates et al., 1999) or the ocean and land components by prescribing atmospheric conditions (Barnier et al., 2006; Griffies et al., 2009). Subsequently, the various components are assembled into a comprehensive model, which then undergoes a systematic evaluation. At this stage, a small subset of model parameters remain to be adjusted so that the model adheres to large-scale observational constraints (often global averages). This final parameter adjustment procedure is usually referred to as “model tuning”. Model tuning aims to match observed climate system behaviour and so is connected to judgments as to what constitutes a skilful representation of the Earth’s climate. For instance, maintaining the global-mean top-of-the-atmosphere energy balance in a simulation of pre-industrial climate is essential to prevent the climate system from drifting to an unrealistic state.

The purpose of "tuning" is to ensure that the model adheres to large scale observational constraints such as observed global averages. The example provided is that models need to maintain the observed top of atmosphere energy balance. As for how the procedure is implemented in different models, this is how the IPCC describes it:

...it is clear that tuning involves tradeoffs; this keeps the number of constraints that can be used small and usually focuses on global-mean measures related to budgets of energy, mass, and momentum. It has been shown for at least one model that the tuning process does not necessarily lead to a single, unique set of parameters for a given model, but that different combinations of parameters can yield equally plausible models (Mauritsen et al., 2013). Hence the need for model tuning may increase model uncertainty. There have been recent efforts to develop systematic parameter optimization methods, but due to model complexity they cannot yet be applied to fully coupled climate models (Neelin et al., 2010).

How do we know that models aren't only correct because of tuning?  Well, here is how Gavin Schmidt explains it to ArsTechnica:

There’s obviously more to a climate model than a graph of global average temperature. Some parameterizations—those stand-ins for processes that occur at scales finer than a grid cell—are tuned to match observations. After all, they are attempts to describe a process in terms of its large-scale results. But successful parameterizations aren’t used as a gauge of how well the model is reproducing reality. “Obviously, since these factors are tuned for, they don't count as a model success. However, the model evaluations span a much wider and deeper set of observations, and when you do historical or paleoclimate simulations, none of the data you are interested in has been tuned for,” Schmidt told Ars.

Back to Bob's questions

 Bob's first question relates to global surface temperature:

1. After decades of climate modeling efforts, why do the current generation of climate models simulate global surface temperatures more poorly than the prior generation?

Answer: Is Bob's question based in fact?  I don't know. I have prepared a quick comparison from KNMI climate explorer but warn that it's not to be relied upon.  There are far too many choices to make so it's not a full comparison.  More importantly, the chart below doesn't include uncertainty for model runs or for observations. As always, click the chart to enlarge it.

Data Sources: NASA and KNMI Climate Explorer 

While the CMIP5 multimodel mean is higher than that of CMIP3 in parts of the first half of last century, I couldn't say whether it is more or less accurate. In more recent years, the CMIP5 multimodel mean seems to get closer to observations such as the volcanic dips. But they are both pretty good, aren't they.  (The observations themselves are likely to be more accurate in more recent years.)

Anyway I just did that little exercise to see if Bob's question was a strawman or not.  Bob bases his question on a paper by Kyle Swanson.  He quotes from the paper:

As a result, the current generation (CMIP5) model ensemble mean performs worse at capturing the observed latitudinal structure of warming than the earlier generation (CMIP3) model ensemble. This is despite a marked reduction in the inter-ensemble spread going from CMIP3 to CMIP5, which by itself indicates higher confidence in the consensus solution. In other words, CMIP5 simulations viewed in aggregate appear to provide a more precise, but less accurate picture of actual climate warming compared to CMIP3.

So it's talking about "latitudinal structure" of warming rather than the composite picture.  Bob's question didn't bring out that distinction.  Anyway, I don't know about that.  The paper is a bit odd really.  Kyle Swanson's hypothesis is that there is selection bias in favour of models that more closely reflect the recent warming in the Arctic at the expense of a wider spread of diversity among models. But then he says:

It is beyond this (and probably any) study to explain in detail why this apparent convergence towards some common solution is emerging among these CMIP5 project climate change simulations.

Kyle Swanson doesn't appear to provide any evidence for his hypothesis.  He is arguing for diversity but has pointed out that models used in both CMIP3 and CMIP5 closely simulate observations at least until the last few years.

Anyway, I'll not be distracted by that paper.  It doesn't seem to have distracted anybody else (except Bob Tisdale) as it's not been cited. In any case, the IPCC report goes into quite some detail in Chapter 9 in regard to evaluating climate models.

Why aren't models a perfect simulation of Earth in all respects?

The rest of Bob's questions can be boiled down to "why can't climate models simulate absolutely everything perfectly".  Here's his mish-mash of a wish list.  Bob is calling for perfection in modelling and he wants it now.  His wish list includes perfection in modelling:

• sea ice gains and losses
• the dynamic response in the Northern Hemisphere following a volcanic eruption
• detailed regional patterns of drought and precipitation
• seasonal sea surface temperature variations and other expression of the Atlantic Multidecadal Oscillation
• ENSO
• "halt" in global warming.

I gather Bob's not unhappy with the overall aggregate, apart from the fact that since around 2005 climate models appear to be on the high side when it comes to global surface temperature.  Then again, climate models used for projecting global surface temperature don't claim great precision year to year. They are best used for looking ahead on a multi-decadal time horizon not month to month or year to year.

Scientists of course are working hard to improve climate models.  But not everything that happens in the real world is important.  Some things are more important than others and that's what scientists first determine and then try to emulate.  As an example, here is how Tony Del Genio explains it to ArsTechnica:

“The real world is more complicated than any model of it,” Del Genio told Ars. “Given the limited computing and human resources, we have to prioritize. We try to anticipate which processes that are missing from the model might be most important to include in the next-generation version (not everything that happens in the atmosphere is important to climate).”

I very much doubt that Bob Tisdale would have a clue about what is important and what is not.  After all, to him the oceans warm by magic so he doesn't need any models to explain things.

If anyone has scrutinized the working of climate models, it's climate scientists

This article is a bit of a mish mash in its own right.  In addition to Chapter 9 of the IPCC AR5 WG1 report, I've referred a few times to the excellent article on climate models at ArsTechnica.  I'll quote one more item from it:

Skepticism is certainly not an unreasonable response when first exposed to the concept of a climate model. But skepticism means examining the evidence before making up one’s mind. If anyone has scrutinized the workings of climate models, it’s climate scientists—and they are confident that, just as in other fields, their models are useful scientific tools.

Bob Tisdale isn't examining the evidence.  What he is doing is looking to find evidence that supports his contention that "climate models are wrong" and ignoring evidence that "climate models are useful".  He is not following any scientific method.  Bob is following the disinformation method of the fake sceptic.

The really egregious aspect of Bob Tisdale complaining that climate models don't simulate the earth system perfectly in all details is that Bob himself is a climate science denier who doesn't even accept the physics of the greenhouse effect.  He's not in any position to criticise climate scientists.  

A Question for Policy Makers

As for Bob's call for policy makers to question climate scientists let me turn Bob's question on its its head:

Questions Policymakers Climate Scientists Should Be Asking Climate Scientists Policy Makers Who Receive Government Funding

It's we who need to question policy makers who receive government funding by way of salary, perks and benefits.  Policy makers have asked questions of climate scientists.  Climate scientists have been providing the answers for decades.

Yet almost half a century after taxpayer-funded policy makers were first directly warned by scientists about the consequences of burning fossil fuels, some policy makers are still denying the science.  We need to be holding policy makers to account for not taking enough decisive action to reduce CO2 emissions and shift to clean energy production.

I'm thinking about what Carl Sagan said more than seventeen years ago.

We've arranged this society based on science and technology in which nobody understands anything about science and technology.  And this combustible mixture of ignorance and power sooner or later is gonna blow up in our faces.  

Who is running the science and technology in a democracy if the people don't know anything about it?

From the WUWT comments

M Courtney, son of richardscourtney, is probably worried about his coal investments and says:

January 2, 2014 at 6:08 am
Section 3 also raises questions over the cost of SO4 scrubbers on coal-fired power plants as well as futuristic geoengineering proposals.
Perhaps the costs associated with coal power could be lowered?
Cheaper energy would be good for all.

Robin Hewitt is a little bit sceptical of the "climate science hoax" conspiracy theory and says:

January 2, 2014 at 6:41 am
Do these squandered billions include genuine researchers who have merely put “climate change” in their grant request so they can get funding?
Do they include all those wonderful Argo floats and satellites that Willis loves playing with?
Surely all the money cannot be quietly trousered by scallywags, some of it must advance the cause of science.

John W. Garrett is wrong when he says:

January 2, 2014 at 6:48 am
This piece should be required reading for the incumbent President of the United States, all 435 members of the House of Representatives, every member of the Senate, every state Governor, every state legislator and every member of the 4th Estate (i.e., the media).

Regretfully, I fear that a substantial proportion of that group is functionally illiterate.

If he'd said that "a substantial proportion of that group is scientifically illiterate" I'd have agreed with him on that score.

Doug Proctor goes for a variation of Godwin's Law and says:

January 2, 2014 at 7:52 am
The IPCC “scientific” position is that multiple weak indicators is as good as one strong indicator. It is a principle that sends many an innocent to prison or, as Stalin would have preferred, to the firing squad.

Alcheson is caught up in billions and trillions and wants to spend billions and trillions on fusion energy research - and says:

January 2, 2014 at 10:15 am
Seems you won’t be getting any comments on this excellent article by Bob from the CAGW peeps. Nothing like trying to defend the indefensible and looking like a fool. Maybe Mosher wants to chime in on why all of the billions have been well spent and why we should keep wasting hundreds of billions or even trillions generating this crap? Billions spent every year on flawed computer models that get worse with every new generation based on comparison to real world data, When it comes to research, it would be far better spent on FUSION energy research, the real solution to our energy needs of the future.

Steve Richards is probably right on the money when says:

January 2, 2014 at 2:45 pm
I believe this is far to technical to be absorbed by policy makers.
It needs an executive summary, bullet style, to access the intellectual level and time pressure on the worlds policy makers

John West also unwittingly echoes Carl Sagan and says (excerpt):

January 2, 2014 at 3:30 pm
With all due respect, Bob, these questions are not suitable for policymakers. They’re too complicated and technical for the average policymaker to fully appreciate the significance and nuances of the question or understand the answer well enough to be able to discern truth from BS.