In a previous blog “CO2 and temperatures: da stelle mer uns janz dumm” I presented a “zero-dollar” model using past global temperatures and CO2 data to estimate the climate sensitivity using the 1850-1945 and 1945-2013 periods, and found the effective climate sensitivity (which should be close to the equilibrium climate sensitivity ECS) to be about 1.34°C. This means that a doubling of atmospheric CO2 mixing ratio with respect to the pre-industrial level would cause a global warming of at most 1.34°C. This is a number much lower than the “consensus” values of the IPCC which suggest a most probable warming of 3°C (1.5 to 4.5 °C range). Many authors disagree with the IPCC estimation. Lewis and Curry for instance find in their recent paper “The implication for climate sensitivity of AR5 forcing and heat uptake estimation” values of 1.33°C for the transient climate response TCR and 1.25 to 2.45 °C (i.e. a central value of 1.85°C) for the equilibrium climate sensitivity ECS. Let me just recall that the lower TCR gives the heating due to a CO2 doubling at the moment where this doubling occurs (assuming this doubling takes 70 years to realize), whereas the ECS gives the far in the future lying definitive warming if all feed-backs and readjustments are finished. Dr. Roy Spencer from the UAH presented in his blog yesterday a new calculation using the 15 years of data from the CERES instruments which are carried by successive NASA satellites. CERES measures out-going and incoming radiation fluxes (in W/m2), and is the best (and practically only) source for these extremely important data. Dr. Spence found in several previous papers that when the globe changes its surface temperature, the atmosphere reacts with a delay of about 4 months with its changes in radiative flux. So he took the available 15 years of CERES data, computed yearly means and plotted these data versus the 4 month time shifted global temperatures of the HadCRUT4 series, with a linear regression of flux (t) = a*T(t-4) + b (t = time in months). This gives the following figure: This linear regression tells us that dF = 2.85*dT (a change of global temperature of 1 °C (or 1 K) corresponds to a forcing of 2.85 W/m2 (and vice-versa). The parameter 2.85 represents the climate feedback lambda. Now the effective climate sensitivity ECS is defined as ECS = F2xCO2/lambda where F2xCO2 is the radiative forcing caused by a doubling of atmospheric CO2, and lambda = feedback factor. Let us accept that F2xCO2 = 5.25*ln(2) = 3.71 W/m2; the number 5.35 is the “consensus” value, which remains subject to discussions, but is more or less accepted by both climate alarmists and realists. So we have: ECS = 3.71/2.85 = 1.3 °C When CO2 mixing ratio reaches 560 ppmV (i.e. the double of the concentration at pre-industrial time, about 1850), we should have a total warming of max. 1.3°. As the globe warmed by about 0.8° since that time, there would be max. 0.5° of coming warming in the pipe-line. Conclusion: All these ECS values derived from observations (and not climate models!) are rather low. Dr. Spencer says that the 1.3°C should be taken as a maximum, and that the real ECS could possibly be much lower (Prof. Lindzen suggested 0.7 – 1°C). Should we worry? No! And should we desperately try to avoid any CO2 emissions? Neither!