An analysis of simulated and observed global mean near-surface air temperature anomalies from 1979 to 1999: trends and attribution of causes

The 1979–1999 response of the climate system to variations in solar spectral irradiance is estimated by comparing the global averaged surface temperature anomalies simulated by a 2-D energy balance climate model to observed temperature anomalies. We perform a multiple regression of southern oscillation index and the individual model responses to solar irradiance variations, stratospheric and tropospheric aerosol loading, stratospheric ozone trends, and greenhouse gases onto each of five near-surface temperature anomaly data sets. We estimate the observed difference in global mean near-surface air temperature attributable to the solar irradiance difference between solar maximum and solar minimum to be between 0.06 and 0.11 K, and that 1.1–3.8% of the total variance in monthly mean near-surface air temperature data is attributable to variations in solar spectral irradiance. For the five temperature data sets used in our analysis, the trends in raw monthly mean temperature anomaly data have a large range, spanning a factor of 3 from 0.06 to 0.17 K/decade. However, our analysis suggests that trends in monthly temperature anomalies attributable to the combination of greenhouse gas, stratospheric ozone, and tropospheric sulfate aerosol variations are much more consistent among data sets, ranging from 0.16 to 0.24 K/decade. Our model results suggest that roughly half of the warming from greenhouse gases is cancelled by the cooling from changes in stratospheric ozone. Tropospheric sulfate aerosol loading in the present day atmosphere contributes significantly to the net radiative forcing of the present day climate system. However, because the change in magnitude and latitudinal distribution of tropospheric sulfate aerosol has been small over the past 20 years, the change in the direct radiative forcing attributable to changes in aerosol loading over this time is also small.