Error message

  • Unable to create CTools CSS cache directory. Check the permissions on your files directory.
  • Unable to create CTools CSS cache directory. Check the permissions on your files directory.
  • The file could not be created.

Telling climate sensitivity by the Sun

As (almost) everyone knows, if CO2 continues on its present pathway the resulting rise in radiative forcing (RF) will dominate climate change during this century.  There is however considerable uncertainty as to what temperature that pathway is likely to cause, due to large non-CO2-related fluctuations during the 20th century in surface temperature (ST, represented here by HadCRUT4) that make it hard to estimate climate sensitivity (CS) based on 20th century observations, and that for all we know may be even larger this century than last. Furthermore neither equilibrium climate sensitivity (ECS) nor transient climate response, the two IPCC-sanctioned notions of CS, describe any climate scenario observed in recent centuries.

We greatly improve precision of existing analyses of 20th century ST by filtering out the higher frequencies to leave just RF and total solar irradiance (TSI) and applying multiple regression to fit both to ST with a residual whose standard deviation is 5 mK.  For comparison fitting RF or TSI separately yields respective residuals of 15 mK and 35 mK.  Inspection of the fit makes it clear that fitting RF alone yields a considerably more pessimistic forecast of climate in this century based on the last than when TSI is included.

We define prevailing climate sensitivity (PCS) as for TCR but with the 1% rise in total CO2 replaced by a 2% rise in the excess over 280 ppm, which agrees to within 5 ppm with the past 250 years of CO2 data and also with RCP8.5 to 2100.  The above fit gives separate coefficients of RF and TSI, permitting estimates of PCS both with and without feedbacks.  This is the first empirically based estimate of no-feedback CS of any kind applicable to modern climate, all previous estimates having been based on numerical climate models.

Vaughan Pratt

Date & time

10–11am 9 May 2017


Room: A105 Seminar Room


Professor Emeritus Vaughan Pratt


Updated:  10 August 2021/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing