Bridenstine, Climate Scientists Are Not Noble, Stop Paying Them

All climate models are wrong. Can't predict tomorrow much less to 2100. The latest paper out on models admits the models run hot. The first crack in the dogma and some of the defenders are mighty pissed.

 

You might be interested in this: http://web.mit.edu/rpindyck/www/Papers/Climate-Change-Policy-What-Do-the-Models-Tell-Us.pdf

"When it comes to the damage function, however, we know almost nothing, so developers of IAMs can do little more than make up functional forms and corresponding parameter values. And that is pretty much what they have done."

"Most IAMs (including the three that were used by the Interagency Working Group to estimate the SCC) relate the temperature increase T to GDP through a “loss function” L(T ), with L(0) = 1 and L 0 (T ) < 0."

The major problem here is that the loss function is completely unknowable today. The temperature increase is seen in the median value. The median value can change from seeing increasing maximum temperatures, from seeing more days of moderate temperatures (i.e. a longer growing season), or from higher nighttime and winter temperatures. Two of these three would conceivably be *positive* impacts (i.e. the "loss function could be negative) because of more food (longer growing seasons) and less energy spent during colder temperatures.

The climate models will have to start providing *much* more information than just "median" values in order for informed judgments to be made.

The US, at least the central and eastern portions, have recently been found to be global warming "holes". While the median temperatures are going up the maximum temperatures are actually going down. The State Climatologist for Iowa has published a graph showing the number of record max temps in Iowa has gone down for the past five years. A preliminary look at my data shows we had one day over 100degF in Kansas this year (I haven't imported it into my database yet so I can't do any searches on it). The number of cooling-days in Kansas and Nebraska has been going down for the past five years.

Since the median still appears to be going up an objective observer would have to conclude one of the positive aspects of global warming is happening here in the central and eastern US. Rather than global warming being a disaster it could turn out to be a boon to food production!

 

I like the positive side of warming, and as we know past warming has benefited man, not hurt him. Until now, warming was good, cooling bad.

 

At least get the spelling right!

you said: "You started by claiming that climate models shouldn't be believed when predicting climate change over the next few decades and you used the accuracy of hurricane track predictions more than a day or two out as evidence."

I said: "The models do *not* match with satellite and weather balloon data. But that does *NOT* mean they don't use GCM models!"

If the hurricane models, which use GCM, can't predict the movement of a hurricane more than a day or two ahead then how can climate models using the same GCM's accurately predict the climate decades from now?

The "cone" of the hurricane models is based on the "uncertainties" associated with the models ability to predict based on the information we have and the knowledge of the physics of the sea, land, and atmosphere. How often do you see a "cone of uncertainty" associated with a climate model?

And I pointed out to you how that reference was garbage! It says "Climate modeling is also fundamentally different from weather forecasting. Weather concerns an initial value problem: Given today's situation, what will tomorrow bring? Weather is chaotic; imperceptible differences in the initial state of the atmosphere lead to radically different conditions in a week or so. Climate is instead a boundary value problem — a statistical description of the mean state and variability of a system, "

That is garbage! Do you know *anything* about partial differential equations representing any kind of physical situation? You have to establish BOTH initial conditions *and* boundary conditions - FOR BOTH CLIMATE AND WEATHER MODELS!

Do you need me to provide you a basic tutorial on describing physical situations with partial differential equations?

Take a rod at temperature T0 isolated from the environment (wrapped in insulation perhaps. Attach a cooling device at one end to maintain the temperature there at T1 and a heating device at the other end to maintain the temperature at T2. The Initial Condition, IC, is equal to T0. The Boundary Conditions, BC, are equal to BC1 = T1 for one end of the bar and BC2 = T2 for the other end of the bar. '

The partial differential equation for this is ut = (alpha**2) * (uxx)

ut is the partial first derivative of temperature with respect to time (rate of change of temperature with time)
uxx is the partial second derivative of temperature with respect to x (a measure of the concavity of the curve over x)

This is a *very* simple representation of how the elements of a GCM are built up using vertical columns of the atmosphere. The Navier Stokes equations are much more involved and there are interactions between the columns and along the columns that must be accounted for.

I'm not gong to get into actually solving the PDE. Suffice it to say that the shape of the curve for ut along the rod will be a sinusoid.

EXACTLY THAT ONE. The one that says weather forecasting is based on initial conditions and climate forecasting is based on boundary conditions. That *is* pure mathematical GARBAGE!

And I understand why there is a cone of uncertainty associated with a hurricane track! If there is that much error in solving the Navier Stokes equations and the parameterizations used for processes that can't be described well then how do you think climate models can do any better when run for decades?

I *am* making sense. You keep on trying to defend an untenable position!

 

Finding out that the newest, most accurate ocean temperature devices show that the ocean is several degrees colder than the older, less accurate devices means that future sea level rise has been way overestimated as well! And, of course, NASA revised the newer, more accurate temperatures UP to make them match the older, less accurate temperatures. That way the fact that the sea level rise has been overestimated can be hidden from the public!

 

The fact that the true believers are so invested that they shun good news for a belief in the bad is kinda sick.

 

Boy a little bit of knowledge sure goes a long way to making a fool of yourself.

Here, misinterpret these.

https://www.nap.edu/read/13430/chapter/16


https://earthscience.stackexchange....ces-between-weather-models-and-climate-models

 

Have you looked at the root of that website. It's just some hacker site posing as something legit. Sorry!

 

Over 3 evenings of watching tv documentaries about numerous nuclear tests in the early days of fission experimentation; the industrial revolution and the thousands of chimneys belching out toxic smoke day and night, year after year; oil wells on fire for weeks on end whether accidental or intentional; footage of various wars using chemicals on a grand scale by the feuding sides; and the exponential increase of the exhaust fumes from squillions of motor vehicles and almost as many aircraft, is it any wonder that nature can't cope any more?

 

Sure.

 

From your first reference: "The essence of this “seamless” approach to weather and climate prediction is that they both share common processes and mechanisms, and the interactions across time and space scales are fundamental to the climate system."

Btw, did you happen to look at the date on the first reference? It was written in 2012, Five years ago! Much of what this article talks about have been implemented since the article was written!

Let's go on with the first reference:

"or weather prediction, detailed analyses of the observed state of the atmosphere are required, but uncertainties in this initial state grow rapidly over several days. Other components of the climate system are typically fixed as observed. For climate predictions, the initial state of the atmosphere makes less difference, but the initial states of other climate system components are necessary. For predictions of a season to a year or so, the upper ocean state, sea-ice extent, soil moisture, snow cover, and state of surface vegetation over land can all be important. For the decadal prediction problem, a full-depth global ocean initial state could be essential (Meehl et al., 2009; Shukla, 2009; Smith et al., 2007; Trenberth, 200. Initial conditions for the global ocean could conceivably be provided by existing ocean data assimilation exercises. However, hindcast predictions for the 20th century, which are desirable to test models, are severely hampered by poor salinity reconstructions prior to the early 2000s when Argo floats began to provide much better depictions of temperature and salinity in the upper 2,000 m of the near-global ocean. Challenging research tasks are to develop optimal methods for initializing climate model predictions with the current observational network and identifying an optimal set of ocean observations to use for initializing climate predictions (Hurrell et al., 2009)."

In both INITIAL CONDITIONS *are* required. For long term climate predictions the initial conditions of the atmosphere are less important than in weather conditions BUT THEY MUST STILL BE PROVIDED! The climate model requires other initial conditions for other processes to be set which become more important over the long term. For short term weather predictions these processes are just assumed to be fixed BUT THEY STILL MUST BE CONSIDERED!

LIKE I SAID WEATHER MODELS AND CLIMATE MODELS USE THE SAME EQUATIONS THAT ARE SOLVED IN THE SAME MANNER! GCM's are the same for both!

Now, let's look at your second reference:

"At their core lie the same set of primitive equations, but from here there are many differences.

A weather model only (skillfully) predicts about 10 days into the future, while a climate model integrates forward in time for hundreds of years. The main difference here is that in a weather model, we care about the when and where of a storm or front. In a climate model, you get weather, but you don't really care too much where or exactly when the weather is as you are looking for a long term means" (bolding mine, upside)

"Weather models vary from global models to very localized regional models, which can in some cases be very idealized. Climate models tend to be global. This doesn't change the physics involved, but can influence the specific forms of the equations. A global model will solve in spherical coordinates and many use spectral methods. Regional weather models will use Cartesian coordinates and may make other assumptions that simplify the physics for the specific purpose the model (e.g. a storm scale idealized weather model may neglect Coriolis)."

So, once again, we see EXACTLY what I said. You don't use different equations (or algorithms as you like to call them), you use different forms of the same equation. You use different step intervals. You use different coordinate systems. But none of these impact the actual physics which determine the equations. For a short time step like in a weather model you may assume the land/ocean temperature is fixed instead of varying based on some function but the temperature is *still* included in the model. It *has* to be in order to calculate the energy level feeding the weather! If TO is the temperature of the ocean then perhaps TO(t) = C + D(sin(t)) + E(t**2) can be used to describe the ocean temp over time. Then for T0(0) you get C. And that is what the weather prediction may use for the model, it is *still* using the equation but it is just solved for one time point since the ocean temp doesn't vary much over a day or so. For a climate model with time steps in days, weeks, or years, the equation will provide different temperatures for each step. E.g. for 200 days later you would get [C + D(-.34) + E(200**2). It's still the same equation and it is still solved the same way in both!

You can continue to try and defend your untenable claim that weather forecasting and climate modeling use different methods and algorithms and whatever but it *is* an untenable position. That may have been the case a decade ago but lots of advancement has been made over the last decade!

 

Most of what you see from old chimney's is particulate matter that soon falls to the ground. You simply can't see CO2! The same applies for the steam engines of the early industrial revolution. Chemicals used in war are *all* heavier than air, otherwise they would be useless!

Cars *do* produce more CO2 than all the power plants on the world combined. So what?

 

Such a lot of work to demonstrate that a core set of algorithm in both weather forecasting and long range climate modelling. which I believe is what I said.

Perhaps you should consider that specific equations are programming objects that do not comprise the totality of either application and indeed each has a slightly different em-PHA-sis on those same syl-ABLEs.

Oh look core code.

Seems you can't get over the fact that two applications can share a core set of code and then apply differing techniques and tools (algorithms if you will) to those outputs to achieve their different objectives and results.

Core code is a fundamental concept in application development theses days, especially for various useful and widely applicable standardized equations.

Depending on O/S you can pick up "objects" and algorithms that deliver all kinds of functionality, most of them either provided for free or in SDKs.

Are you at all familiar with application development?

 

You said:

But you also said:

Spreadsheets and word processors are *totally* different things with totally different application layers. Weather forecasting today and climate modeling today are *NOT* like this. And yet you keep trying to offer references that say they are! Like the one from NASA that said weather forecasting uses initial conditions and climate modeling using boundary conditions! That was *NOT* written by someone that knows how to solve partial differential equations - yet you bought into it!

My guess is that you have no idea of how to solve partial differential equations. It's far more than just coding an equation. You have to make proper judgments on the values of initial conditions and boundary conditions, either by using available data or by deducing what to use by other means.

How does a programming object make a decision as to whether to consider the output of an equation as fixed or variable? Someone has to decide that from outside!

Which was the point *I* made.

Now you are back to trying to say that weather forecasting and cimate models use different equations (it was *YOU* that finally admitted an equation and an algorithm are the same!).

You don't seem to have any idea of what you are talking about. When I was developing VTxxx emulation code for some of the first PC's the same code was used over and over and expanded to handle the differences from a VT100 to a VT220 to a VT320 to etc. When I was developing data collection and analysis and graphing software for my telephone company for one process and for AT&T for another process of course I used the same core code for collecting the data, for graphing it, and for displaying it. All core functions. But the analysis was *NOT* a core function. One was to analyze accounting data off a minicomputer and the other for analyzing traffic data from a central office.

That is *NOT* the case for weather forecasting and climate modeling. They both use GCM models which do the same type of analysis using the very same equations. The only difference is the data given each model!

Which has absolutely NOTHING to do with weather forecasting, climate modeling, and solving the very same partial differential equations!

I've done my share, much of it using machine code on early processors such as the Z80. I used to have *reams* of coding I could reuse. And if I had written two models to use numerical techniques to solve the very same equations then I would have duplicated the code. But the determining factor there is not the code but *equations* that are the same!

 

Look its clear you know NOTHING about application architecture. I on the other hand have some 35 years of experience in the design, development, marketing, distribution and support for DOZENS of business applications on platforms starting back the ol' model T days of CPM and DOS.

OMG. boy are you smrt. gee who knew you knew so much about parameter driven algorithms, let alone incremental iteration.

Can you say SUPERCOMPUTER.





[

once again demonstrating you know nothing about how complex such applications are

VT220? ahahahahahahahahahahahahahah. Oh man I haven't heard since the 70's. Terminal emulation code over 300 baud dial up modems. Yep back in the day it was bleeding edge. I know, I was there.


You are attempting to compare a primitive 8 bit computing environment with state of art supercomputing of today which is like saying you once worked on a model T engine so you can totally figure out how a Formula 1 engine works. that is truly funny.

Z80 oh my the power of that 8 bit processing environment and the power of CPM. You flew with Orville and now you think you know how to fly a space shuttle. Amazing hubris.

 

Doesn't matter. GIGO. The more complex the more prone to error that compounds over iteration.

 

Except it aint garbage going in. Its an entire global array of extremely accurate measurements going in.

I agree complexity does create more opportunity for error

 

None of the measurements are accurate. That is an impossibility. For one, there are so many adjustments to the raw data based on guesses which change the record from one year to the next. For the surface record many areas of the globe have zero coverage. Some records are thermometer, some digital, taken at different times of the day and often have months or even years missing that they have to be filled in with guesses. Even the satellite data has to be adjusted and that record has changed over time.

Then there are unsolvable issues like the Navier-Stokes equation which creates the need to parametrize things like cloud formation in 100 km grids. Cloud formation is one of the known unknowns in climate change modeling and possibly one of the most important inputs. Then there are the limitations to computing which even super computers have. One of the reasons for parameterization is that limit.

None of the models are accurate or you would only need one.

 

The measurements, as datapoints are extremely accurate. The tools to interpret that data is where all the hard work is taking place.

the modelling challenges you have outlined are consuming the brainpower of literally thousands of mathematicians, physicists, computer scientists, programmers and system designers. they've come a long way but there's much more to learn and understand.

Just because the current state of the art can't produce definitive predictions doesn't mean the predictions they do make are invalid.

 

Actually, according to the 2014 National Climate Assessment, the models have been very accurate - if anything too Conservative.

http://nca2014.globalchange.gov/highlights/overview/overview
It is notable that as these data records have grown longer and climate models have become more comprehensive, earlier predictions have largely been confirmed. The only real surprises have been that some changes, such as sea level rise and Arctic sea ice decline, have outpaced earlier projections.

 

'So what?' Well we are talking about life-threatening environmental pollution and holes in the ozone layer here, that's 'what'! Also, you 'simply can't see' radioactivity, but that doesn't mean it isn't around? I don't really comprehend the net value of your post.

 

So what? You aren't showing any actual knowledge when you claim GCM models used for weather forecasting and for climate modeling are as different as spreadsheets vs word processors!

A supercomputer is just a FASTER computer! It's not "smarter" or able to make judgment calls on what can be considered to be fixed or has to be actually solved over a time period.

Again, you are showing that you are *NOT* that smart concerning computers and applications

Do you have *ANYTHING* of import to offer besides the argumentative fallacy of Poisoning the Well?

When having to use machine language you don't have the benefit of of more advanced languages and optimizing compilers. You have to understand the mathematics *very* well for it is *YOU* implementing directly in registers even things as simple as adding two numbers. *YOU* have to take care of managing register overflows and significant numbers.

Engine thermodynamics can be completely described mathematically. A flat head 8-cyl from 1939 has the same mathematics as a Ferrari v-12 today. Air flows, fuel flows, air/flow mixtures, compression ratios, bore and stroke, energy released from the oxidation of a complex carbon chain, etc all are used to describe the engine. Don't confuse advances in materials used in the engines with the actual physics and thermodynamics of the combustion engine.

You make the very same mistake with engines that you make with weather models and climate models both using Navier Stokes equations and GCM models. An engine with a carburetor works exactly like one with fuel injection as far as the math is concerned thermodynamically. Only the constants used with the equations change because of differences in the air and fuel flows and the resulting air/fuel mixtures. Just like with weather and climate models.

A Z80 can do the exact same things as a quad-core I7 can do. It just takes longer to get it all done. You don't demonstrate enough understanding of the fundamentals to even grasp that most obvious truth!

 

If it isn't garbage going in then why do the outputs of the models continue to deviate from the satellite and weather balloon data?

From the Remote Sensing Systems company:

"Why does this discrepancy exist and what does it mean? One possible explanation is an error in the fundamental physics used by the climate models. In addition to this possibility, there are at least three other plausible explanations for the warming rate differences. There are errors in the forcings used as input to the model simulations (these include forcings due to anthropogenic gases and aerosols, volcanic aerosols, solar input, and changes in ozone), errors in the satellite observations (partially addressed by the use of the uncertainty ensemble), and sequences of internal climate variability in the simulations that are difference from what occurred in the real world. "

 

Even the measurement devices have margins of error. These margins of error continue throughout the models that use the measurements. Yet the climate models *never* make any direct mention of their confidence levels or margins of error for their outputs.

You nailed it with your last sentence! If this were *true* science we would have one model that accurately predicts both the weather and the climate. Instead we have over 50 models which don't agree with each other! And this is while using the same data since there are only about three basic data sets!

 

Land measurements today have a margin of error of about +/- 0.5degC. They are *NOT* extremely accurate. And it gets worse as you go to 3rd world countries around the globe where the the measurement device accuracy may be even less!

go here: wattsupwiththat.com/2011/01/22/the-metrology-of-thermometers/

"My main points are that in climatology many important factors that are accounted for in other areas of science and engineering are completely ignored by many scientists:

1. Human Errors in accuracy and resolution of historical data are ignored
2. Mechanical thermometer resolution is ignored
3. Electronic gauge calibration is ignored
4. Mechanical and Electronic temperature gauge accuracy is ignored
5. Hysteresis in modern data acquisition is ignored
6. Conversion from Degrees F to Degrees C introduces false resolution into data."