Optimal estimation of the surface temperature response to natural and anthropogenic climate forcings over the past century

Space-time optimal filtering is used to estimate the response of the Earth's surface temperature to both natural and anthropogenic climate forcings over the past century. A linear two-dimensional energy balance model with an ocean mixed-layer is developed to generate the hypothesized space-time patterns of response to the climate forcings. Optimal filters are constructed using these space-time patterns along with estimates of the background climate noise from 1000-year control runs from three general circulation models (GCMs). The filters are then applied to 100 years of observed surface temperature data. The search is limited to the "solar cycle" frequency band (periods from 7.49 to 16.67 years) to reduce possible model errors from EL Nino-Southern Oscillation (ENSO) events and low-frequency oscillations from the deeper ocean. A method is introduced to isolate the response of each climate forcing from the other three. The surface temperature response to volcanic aerosols, greenhouse gases, and tropospheric aerosols is detected with a high level of confidence. The response to solar cycle forcing is detected with a moderate level of confidence.;The same four climate forcings are used, along with an estimate of the surface temperature response to ENSO events, in an investigation of the forced climate variability over the past 100 years. The sources of variability in the quasi-decadal frequency band are examined. It is found that a majority of the variance of the surface temperature can be explained by the model response plus the ENSO cycle. This supports the use of the energy balance model for the generation of the climate signals. In addition, these results provide information about the natural (unforced) level of climate variability in the surface temperature record.