Editing Openai/6938382f-f9e4-8003-a158-6d812c1bce84 (section)

===== Main claims (in my own words) =====

The paper argues that:
* GCM global-mean temperature projections are basically linear extrapolations of greenhouse-gas (GHG) forcing. This is represented by an emulation equation (Eq. 1) relating projected ΔT to fractional change in “water-vapor-enhanced” GHG forcing. Propagation_of_Error_and_the_Re…
* Cloud simulations in CMIP5 models have a large systematic error in global total cloud fraction (TCF), about ±12.1% RMS. This is mapped to an error of ±4 W/m² in long-wave cloud forcing (LWCF). Propagation_of_Error_and_the_Re…
* That ±4 W/m² per year of “thermal flux uncertainty” is ~114× larger than the annual increase in GHG forcing (~0.035 W/m²/yr). Propagation_of_Error_and_the_Re…
* Treating GCM projections as linear in forcing and then linearly propagating that ±4 W/m² error every year leads to an uncertainty of about ±15°C by the end of a century-scale projection. Propagation_of_Error_and_the_Re…
* Therefore, no anthropogenic warming signal can be detected in climate observations, and all past and present GCM air-temperature projections are claimed to be physically unreliable. Propagation_of_Error_and_the_Re…