| Index of Editorials
Climate Science Climate Sensitivity
All Editorials for
CO2 Emissions 
Climate Cycles 
Climate Sensitivity 
Thermal History 
Unsolved Problems 
American Power Act 
Clean and Sustainable 
Nuclear Waste Storage 
Renewable Electricity Standard (RES) 
Surrogate Religion 
Energy Primer for Kids 
Global Climate - International
French Academy 
Anthropogenic Global Warming (AGW) 
Greenhouse Gases 
Ice Cores 
Oceans' Role 
Sun's Role 
Second Hand Smoke 
Arctic Sea Ice 
Atmospheric Temperature Data 
Sea Surface Temperature 
Surface Data 
Statistics Misuse 
Modern Empirical Science
v. Medieval Science 
Climate Research Unit (CRU) 
International Panel on Climate Change (IPCC) 
Nongovernmental International Panel on Climate Change (NIPCC) 
UK Met Office 
World Meteorological Organization (WMO) 
Climate Realism 
Independent Cross Check of Temperature Data 
IPCC Assessment Report 
NOAA State of the Climate 2009 
NRC-NAS Advancing the Science of Climate Change 
West Antarctic Ice Sheet (WAIS) 
Types of Energy
Nuclear Energy 
SEPP Science Editorial #17-2009
(in TWTW Jun 13, 2009)
S. Fred Singer, Chairman and President , Science and Environmental Policy Project (SEPP)
Climate Sensitivity (CS), Negative Feedback (NF), and all that
Jun 13, 2009
Based on empirical evidence, various researchers have concluded that CS is much smaller than the modelderived values quoted by the IPCC. Some of the empirical studies compare observed temperature trends over time with IPCC values [Schwartz, Monckton, etc]; others [Douglass, Singer, NIPCC] compare observed and modeled patterns of temperature trends ("fingerprints")'
CS is conventionally defined as the equilibrium temp rise caused by a doubled forcing of GH gases; it is often taken to be just a doubling of CO2 levels. The "canonical" CS values of the IPCC range from 1.5 to 4.5 C, with a median of 3.0 C. Many model calculations show higher values, depending on assumptions about cloud parameters; for example, Stainforth et al  quote as high as 11.5 C. The empirical values for CS are all well below the IPCC's; some are 0.5 C or even less, corresponding to a trend of Global Mean Sfc Temp (GMST) of only about 0.05 C/decade and a tropical troposphere trend of about 0.1 C/decade. These trends are at or below the limit of detection, because of the interfering effects of aerosol emissions (both natural and anthropogenic), volcanic eruptions, El Niños and other, less dramatic atmosphere-ocean interactions.
The "fingerprint" method can only conclude that anthropogenic effects are not detected [NIPCC], and yields no values for CS - only an upper limit of perhaps 0.3 C, an order of magnitude smaller than the IPCC's median value.
How to account for the huge discrepancy between IPCC and NIPCC? In principle, one can invoke natural forcings, both external (solar) and internal, as well as aerosols that affect the optical properties of the atmosphere. It is tempting, however, to first investigate the possibility of negative feedbacks within the climate system itself, principally the various effects of water in the atmosphere.
Atmospheric water can occur in three different forms: as a gas -- water vapor (WV), as liquid cloud droplets, and as solid ice particles. In principle, one can measure the climate effects of each component, as we shall discuss below.
1. Liquid: The negative feedback effects of water droplets are easiest to visualize [Singer WSJ 1988]. As the oceans warm, increased evaporation can increase cloudiness, increasing optical albedo, and reducing the incidence of solar radiation at the surface - thus reducing any warming caused by increasing GH gases. But measuring such an albedo change is difficult, requiring accuracies of a fraction of a percent and exceptional stability over a number of years. 2. WV: Models all assume a constant relative humidity with altitude; thus WV in the cold upper troposphere (UT) will radiate at a low temperature and contribute little to OLR (outgoing long-wave radiation), with the remainder therefore coming from the warm surface. (Total OLR is fixed and must equal absorbed solar energy.) However, if atmospheric processes manage to achieve a drying of the UT (as GH gases increase) [Ellsaesser, Gray, Lindzen], then WV will radiate at the higher temperature of the boundary layer, contribute the bulk of the OLR, and leave less IR emission from the surface.
Satellite measurements, such as by the AIRS instrument, can resolve the WV bands in the OLR and determine the source temperature. Data would be required versus latitude, and over a number of years. 3. Ice: Convective activity in the tropics can transport moisture to heights near the tropopause where ice crystals would form cirrus clouds, often invisible but having strong absorption properties over the entire effective IR region. A reduction of the area covered by cirrus (iris effect - Lindzen) would permit more escape of IR from the surface and thus produce a cooling -- a negative feedback.
Again, AIRS data could obtain the necessary confirming data by observing long-term trends. NF is not a sure thing; aerosols and/or natural forcings can reduce and even overcome GH warming. At present, one cannot tell which of the possible NF effects is dominant; but the right kind of data could help settle the issue. Establishing the magnitude of NF would independently confirm the low values of CS.
View The Week That Was in which this editorial appeared.
Return to Top of Page