14 December 2009: Answers to many “chicken and egg” type climate questions could have a significant impact on our understanding of both the climate system and manmade global warming. One key question is if the Global Circulation ("climate") Models (GCMs) used by the UN's International Panel on Climate Change (IPCC) are correctly programmed to reflect scientific fact, or if the mechanism is merely assumed to be correct. In an invited talk today (16 December 2009) at the American Geophysical Union’s fall meeting in San Francisco, Dr. Roy Spencer from The University of Alabama in Huntsville will discuss that challenge of answering questions about cause and effect (also known as forcing and feedback) in the climate. Spencer’s interest is in using satellite data and a simple climate model to test the simulated feedback processes contained in climate models that are used to forecast global warming.
At present, modelers viewing satellite data see warm years and fewer clouds, then assume that warmth caused the clouds to dissipate. This would be a positive feedback in the climate system, which would theoretically lead to strong global warming. GCMs are programmed to reflect this assumption. Spencer argues that scientists can't measure feedbacks the way that people have traditionally tried to do it. “My question to them was, ‘How do you know it wasn’t fewer clouds that caused the warm years, rather than the other way around?’ It turns out they didn’t know. They couldn’t answer that question.”
“Feedbacks will determine whether the manmade portion of global warming ends up being catastrophic or barely measurable,” Spencer said recently.
One problem is the simplicity of the climate models. Because cloud systems are so complex and so poorly understood, all of the climate models used by the IPCC use greatly simplified cloud parameters to represent clouds. But the calculations that set those parameters are based on assumed cause-and-effect relationships. Those assumptions might be working in the wrong direction, Spencer said. “What we have found is that cloud cover variations causing temperature changes dominate the satellite record, and give the illusion of positive feedback.”
Using satellite observations interpreted with a simple model, Spencer’s data support negative feedback (or cooling) better than they support positive feedback. “This critical component in global warming theory – cloud feedback – is impossible to measure directly in the real climate system,” Spencer said. “We haven’t figured out a good way to separate cause and effect, so we can’t measure cloud feedback directly. And if we don’t know what the feedbacks are, we are just guessing at how much impact humans will have on climate change.
“I’m trying to spread the word: Let’s go back to basics and look at what we can and cannot do with measurements of the real climate system to validate both climate models and their predictions.”
A former NASA scientist, Spencer is a principal research scientist in University of Alabama in Huntsville’s Earth System Science Center.
For more information:
Dr. Roy Spencer, (256) 961-7960,
Phillip Gentry, (256) 961-7618,
We are pleased to provide a mirror post to Dr Roy W Spencer's invited presentation on estimating cloud feedbacks from satellite measurements available here in PDF form. These are copies of Dr Spencer's draft slides for the presentation. The presentation title is: On the Diagnosis of Radiative Feedback in the Presence of Unknown Radiative Forcing ... or ... Observational Evidence that Cloud Feedbacks could be Negative, Not Positive
This is a highly technical presentation, and appears to build on a paper ("Potential Biases in Feedback Diagnosis from Observational Data: A Simple Model Demonstration," in PDF format) published by Spencer & Braswell in the Journal of Climate in early 2008.
The evidence from Spencer's research suggests that climate system feedbacks appear to be negative (causing cooling), while most if not all GCMs now in use rely on positive ("warming") feedbacks for their projections of future temperatures and other climate conditions.