The scale of this chart makes it useless as a means of estimating how accurate the models are. May I point out to you that all models 'predict' past temperature with near perfection!, because they are fit to past data and have so many parameters that they can be made to fit any data. The fitting to past data is where their parameter values come from!
I have a model that 'predicts' with 100% certainty the past winners of the World Series. I sell it to you.
The following comment is
from: http://environ.andrew.cmu.edu/m3/s2/subsect/predict.htm
It greatly understates the problems with current models and their unreliability.
Unfortunately, there are still major areas of uncertainty that render the climate-modeling process highly complicated and far from totally accurate. Eric J. Barron, chair of the USGCRP Forum on Global Change Modeling, writes:
I have a model that 'predicts' with 100% certainty the past winners of the World Series. I sell it to you.
The following comment is
from: http://environ.andrew.cmu.edu/m3/s2/subsect/predict.htm
It greatly understates the problems with current models and their unreliability.
Unfortunately, there are still major areas of uncertainty that render the climate-modeling process highly complicated and far from totally accurate. Eric J. Barron, chair of the USGCRP Forum on Global Change Modeling, writes:
"Predictions of future climate are imperfect because they are limited by significant uncertainties that stem from: (1) the natural variability of climate; (2) our inability to predict accurately future greenhouse-gas and aerosol emissions; (3) the potential for unpredicted or unrecognized factors, such as volcanic eruptions or new or unknown human influences, to perturb atmospheric conditions; and (4) our as-yet incomplete understanding of the total climate system."
(from "Climate Models: How Reliable are Their Predictions?" Consequences, Vol. 1 No. 3, 1995.)
Last edited: