Science a GoGo's Home Page
Posted By: Canuck How Reliable are those climate models????? - 05/11/08 01:38 AM
You know, those wonderful computer models, which tell us we're doomed, and the earth will turn into Venus in 100 years. They've been around for a while now, haven't any of you wondered how well they've been able to predict the 90's and thus far into the 2000's?

Apparently Demetris Koutsoyiannis (from the Department of Water Resources and Environmental Engineering, National Technical University of Athens) has wondered that. So did some digging and compared the output of a couple GCMs to observed temperature data.

What did he find? Here's the link

Since I love giving out's the conclusions grin
• All examined long records demonstrate large overyear variability (long‐term fluctuations) with no systematic signatures across the different locations/climates.
• GCMs generally reproduce the broad climatic behaviours at different geographical locations and the sequence of wet/dry or warm/cold periods on a mean monthly scale.
• However, model outputs at annual and climatic (30‐year) scales are irrelevant with reality; also, they do not reproduce the natural overyear fluctuation and, generally, underestimate the variance and the Hurst coefficient of the observed series; none of the models proves to be systematically better than the others.
• The huge negative values of coefficients of efficiency at those scales show that model predictions are much poorer that an elementary prediction based on the time average.
• This makes future climate projections not credible.
• The GCM outputs of AR4, as compared to those of TAR, are a regression in terms of the elements of falsifiability they provide, because most of the AR4 scenarios refer only to the future, whereas TAR scenarios also included historical periods.

Here's another quote, which I particularly enjoyed, found on page 18
Climatic models generally fail to reproduce the long‐term changes on temperature and precipitation.

Gee, good thing we're not basing some multi-trillion dollar policy on the output of these things, eh??
Posted By: paul Re: How Reliable are those climate models????? - 05/15/08 10:44 PM
Gee, good thing we're not basing some multi-trillion dollar policy on the output of these things, eh??

Really ... your right I think we should base the trillion dollar policies on the here and now.

while were still here and now.


GISS 2007 Temperature Analysis
The year 2007 tied for second warmest in the period of instrumental data, behind the
record warmth of 2005, in the Goddard Institute for Space Studies (GISS) analysis. 2007 tied
1998, which had leapt a remarkable 0.2°C above the prior record with the help of the “El Nino of
the century”. The unusual warmth in 2007 is noteworthy because it occurs at a time when solar
irradiance is at a minimum and the equatorial Pacific Ocean is in the cool phase of its natural El
Nino – La Nina cycle.
Figure 1 shows 2007 temperature anomalies relative to the 1951-1980 base period mean.
The global mean temperature anomaly, 0.57°C (about 1°F) warmer than the 1951-1980 mean,
continues the strong warming trend of the past thirty years that has been confidently attributed to
the effect of increasing human-made greenhouse gases (GHGs)
( ). The eight warmest years in the
GISS record have all occurred since 1998, and the 14 warmest years in the record have all
occurred since 1990.

IMO, science and knowledge are wonderful things. We need them. However--if you get my drift--I refuse to depend on them, alone.
Posted By: RicS Re: How Reliable are those climate models????? - 05/18/08 04:00 AM
G'day all,

The trouble with climate models is that no one has been able to work out how to create one with any accuracy. They closely equate to weather forcasts. With a great deal of technology you can get weather forcasts for temperate climate about 80% accurate out three days. World average climate models have not yet been able to provide accuracy greater than chance even one year out. Eventually they will. Eventually a great deal of information will be available and will be fed in correctly but as far as I can see that is still years away.

The simplest question in this debate you can ask is "Show me any global climate model that, without correction or tweeking or the like, has been accurate for three years in its predictions, or even two years. If you cannot then the model is not of any use except to tell you perhaps where you have gone wrong."

It is these type of logically sound simple questions that can show up scientific research that often is cloaked in considerable complexity in the language usage or mathamatics for what they are. Guesswork wrapped up to look like research or worse actual reasonable assumptions as to what might happen.

And I still have a real problem with 2007 being a near record year. According to what data? It certainly did not occur at a time that the solar output was at a cyclic minimum nor when there was a cool phase in the Pacific Oscillation. An El Nino was in full effect for part of 2007, petering out in about the end of July and solar activity dropped dramatically in October 2007 but was pretty high for the rest of the year. So it certainly could have been close to a record based on what happened for about eight months of the period and it is the northern springs and summers that have the greater effect on the average for the year. But I still find it hard to work out why 2007 is considered to be such a hot year when the data just doesn't seem to bear this out.


There's a lot to respond to here, and I can't afford the time; but I'd like to add:

It's not so much what all the models "show," it's what the models don't show.

Models don't show any maintenance of stability in climate.
...or they do show "increasing instability," with any business-as-usual forcasts.

Originally Posted By: blog blather
"Self-regulating" Gaia?
Well, sure; but at what cost? Melting ice reserves, shifting oceanic and atmospheric currents, extinctions, and invasions contribute to "regulation" within one climate mode; but eventually, with too much need for "self-correction," the modes will switch....
...and I thought this seemed illustrative of a general direction for a response about models.
© Science a GoGo's Discussion Forums