When it comes to how the actual climate system might respond to extra greenhouse gases, we’re out of luck in terms of “proof” because the climate’s complexities are innumerable and poorly understood.
Climate science is a murky science. When dealing with temperature variations and trends, we do not have an instrument that tells us how much change is due to humans and how much to Mother Nature. Measuring the temperature change over long time periods is difficult enough, but we do not have a thermometer that says why these changes occur.
We cannot appeal to direct evidence for the cause of change, so we argue.
The real climate system is so massively complex we do not have the ability to test global-size theories in a laboratory. Without this ability, we tend to travel all sorts of other avenues to confirm what are essentially our unprovable views about climate. These avenues tend to comfort our souls because we crave certainty over ambiguity.
It is a fundamental characteristic of the scientific method and, therefore, of the confidence we have in our theories, that when we finally understand a system, we are able to predict its behavior.
All 102 model runs overshot the actual temperature change on average by a factor of three. Not only does this tell us we don’t have a good grasp on the way climate varies, but the fact that all simulations overcooked the atmosphere means there is probably a warm bias built into the basic theory — the same theory we’ve been told is “settled science.”
To me, being off by a factor of three doesn’t qualify as “settled.”
Others might look to certain climate anomalies and convince themselves that humans are the cause. I often hear claims that extreme weather is getting worse. Now, here we do have direct evidence to check. Whether it’s tornadoes (no change over the past 60 years), hurricanes (no changes over the past 120 years) or droughts and heat waves (not as bad as they were during the past 1,000 years), the evidence doesn’t support those claims. So, we argue.
Without direct evidence and with poor model predictability, what other avenues are available to us? This is where things get messy because we are humans, and humans tend to select those avenues that confirm their biases. (It seems to me that the less direct evidence there is for a position, the more passion is applied and the more certainty is claimed.)