Tom Fuller’s Interview with John Christy

Here is an excellent interview with a well-respected atmospheric scientist. Please note the comments with regard to the accuracy of global climate models (GCMs). In the comments section, one questioner asks why we don’t receive short term (seasonal) forecasts from the GCMs. I can answer that one: Because they are horribly inaccurate. That is the dirty secret of the climate models. They can’t make accurate forecasts for 90 days but we are supposed to believe their forecasts for 90 years.

2 thoughts on “Tom Fuller’s Interview with John Christy

  1. Mike – I agree with your skepticism of the climate models and their accuracy, but I agree for a different reason. Short-term weather prediction models are fundamentally different than long-term climate models, so you can't directly compare them (i.e. a 90 day forecast by a climate model). However, the amount we don't know about climate dynamics and feedbacks is staggering, so it's likely that the climate models are way off simply because we have more knowledge to acquire. However…Christy's point about being ready for historical weather/climate extremes is a good one. We're not adequately prepared to deal with floods/heat/hurricanes/etc that have happened before and will happen again, so putting all our efforts to preventing a potential scenario 100 years in the future may not be the best use of resources…

  2. Joel,

    I see three items here, so allow me to comment on each.

    The meteorological "guts" of the climate models are the same as the day-to-day forecast models except at lower resolution and more parameterization. If you go back and look at Hansen's 1988 testimony, he said his model indicated that "lower Manhattan would be under 3-4' of water by 2000" — a testable forecast that busted. The IPCC climate model forecasts have consistently overforecast the amount of warming. If they cannot get relatively short term forecasts (ten months or ten years) right, I cannot think of a scientific reason why their 90-year forecasts (2100) should be treated as if they had perfect accuracy.

    You are correct that what we don't know is staggering! This includes things that aren't significantly important in 24-48 hour forecasts (arctic ice albedo, ocean temperatures, etc.) and non-atmospheric influences like volcanoes, sunspots, etc. This is a second reason to take the climate models with a huge grain of salt.

    That said, you and I are in complete agreement about historical weather and climate extremes. There is much more to be gained from mitigating things that are right in front of us ( http://meteorologicalmusings.blogspot.com/2010/07/where-is-big-environment.html ) than trying to "fix" the climate 90 years from now by any rational cost/benefit analysis.

    Thanks for the comment, Joel.