M

Models, General Circulation Models (GCM)
Posted by Administrator (admin) on 7th February, 2010
M >>

Models are at the epi-center of the Anthropogenic Global Warming (AGW)-hypothesis. It is the results of mathematical models, that makes some scientists to be concerned about future rise in temperatures, not the actually measured temperatures. Models are trained on previous temperatures but are they any good at forecasting?

Below is a graph that shows the IPCC's best estimate of surface temperatures over the 20th century (grey), problematic as they are (see Urban Heat Island effect). The pink graph is the average of 17 IPCC model simulations, trained to reproduce the temperature. As can be seen, the models have successfully been trained to fit in the beginning and in the end, but what is interesting about this graph is, that the models CANNOT REPRODUCE the steep increase between 1910 and 1940, as well as the decrease between 1940 and 1978. WHY??

 

Spencer: HadCRUT3 temperatures and 17 IPCC model average.
 

Actually, it is not supricing the models get it wrong:

How to make a climate model

Cartoon by Josh

 

The models are fed with the scientist's knowledge about the earths energyflow, for instance incoming solar radiation, albedo effect from clouds and ice, greenhouse effect from gasses, etc. The capacities of some of the worlds largest computers are streched to the limit in order to try to make these huge simmulations of energy flow.

After feeding the computers with as much known causality as we have, and as computers can handle, the simulations show a difference to the measured temperature, i.e. there is some unknown forcing. Scientists then conclude, that this part must be caused by carbon and other human emmisions:

 

Step 1:

Step 2:

 Step 3:

 Step 4:

 Step 5:

Step 6:

 Graphics by courtesy of Joanne Nova.

Spencer then plots the difference between models and measured temperatures (grey), i.e. the variations that the models cannot explain. In the same figure is plottet 3 known modes of natural variability: the Pacific Decadal Oscillation (PDO, in blue); the Atlantic Multidecadal Oscillation (AMO, in green); and the negative of the Southern Oscillation Index (SOI, in red):

As can be seen, the three climate indices all bear some level of resemblance to the unexplained temperature variability in the 20th Century.

When Spencer adjusts the IPCC models with a linear combination of the PDO, AMO, and SOI that best matches the models’ “unexplained temperature variability”, he gets a much better fit:

As can be seen, the 2 curves now fit much better until 1970. After this date the modelled temperatures show a 74% greater trend than measured, indicating that the original models have a 74% greater CO2-warming programmed than actually needed, when natural variations of PDO, AMO, and SOI is taken into account.

 So, IPCC models:

  • fail to reproduce the steep increase between 1910 and 1940
  • fail to reproduce the decline between 1940 and 1978
  • fail to reproduce the decline between 2002 and 2010

Further, running the models on the last millenium, they

Temperatures have been rising linearily with oscillations since around 1600.
 

 

Are the models good for forecasting?

So while the experts are pouring money into more advanced and more expensive models, trying to get CO2 to explain as much as possible, and trying to understand the energy balance of earth, maybe we should try to look at a less complicated model, that doesn't claim to explain everything, but apparently is doing a better job in forecasting temperatures than the IPCC models:

Akasofu model.

 

Inside the yellow square is shown the measured temperatures, and outside to the right the predictions of the IPCC models (upper) and of Akasofu (lower). Dr. Akasofu draws a linear trend from the end of LIA until 2000 and continues that until 2100. As temperarures have been increasing linearily since around 1600, Akasofu assumes this trend to continue until 2100. (This might not be true, see "Solar thery, a cold future?". On top of that he superimpose oscillations from natural variabilies. The green arrow and the red dot shows where we are now, below the IPCC predictions.

So, have all the many millions of dollars spend on advanced models, supercomputers and top scientists been wasted? As far as a tool for forecasting, YES! Models failed to reproduce the past and they failed to predict the future. Not a very good tool for deciding the earth's future. It is now clear that the models are too simple, complicated and expensive as they are. This is also recognised by Kevin Trenberth, head of the Climate Analysis Section at the National Center for Atmospheric Research and a lead author of the 2001 and 2007 IPCC Scientific Assessment of Climate Change:

  • Trenberth to Mann: The fact is that we can’t account for the lack of warming at the moment and it is a travesty that we can’t. The CERES data published in the August BAMS 09 supplement on 2008 shows there should be even more warming: but the data are surely wrong. Our observing system is inadequate.
  • Trenberth to Wigley: Hi Tom How come you do not agree with a statement that says we are no where close to knowing where energy is going or whether clouds are changing to make the planet brighter. We are not close to balancing the energy budget. The fact that we can not account for what is happening in the climate system makes any consideration of geoengineering quite hopeless as we will never be able to tell if it is successful or not! It is a travesty! Kevin

As a playing tool for scientists trying to understand the energy balance of earth, models are still valuable, for forecasting they are worthless.

Humorsist Storm Petersen once said:

  • It is difficult to forecast, especially the future

Last changed: 17th February, 2010 at 19:04:01

Back

Comments

None Found

Add Comment