Fingerprints of anthropogenic and natural variability in global-mean surface temperature
This paper presents an analysis designed to detect greenhouse warming by distinguishing between temperature rises induced by increasing atmospheric concentrations of greenhouse gases and those induced by background variability that are present without changes in atmospheric composition. The strategy is based on the surface temperature field. At each observation time, the projection of the anomalous temperature field on the presumed anthropogenic fingerprint is removed in order to obtain a temperature deviation field; i.e., the temperature anomalies in the phase space orthogonal to the anthropogenic fingerprint, which are presumed to be entirely natural. The time series of the expansion coefficients of the fingerprint a(t) is then regressed on this temperature deviation field to identify the axis in the orthogonal phase space along which the variations are most strongly correlated, and an index n(t) of the temporal variations along that axis is generated. The index a(t) is then regressed upon n(t) and the resulting least squares fit is regarded as the component of a(t) that can be ascribed to natural causes. The analysis was performed for monthly global surface temperature anomaly fields for the period 1900-95. Results indicate that two well defined patterns of natural variability contribute to variations in global mean temperature: the synthetic cold ocean-warm land (COWL) pattern and the El Nino-Southern Oscillation (ENSO). In domains that include surface air temperature over Eurasia and North America, the COWL pattern tends to be dominant. The ENSO signature emerges as the pattern most strongly linearly correlated with global sea surface temperature and with tropospheric layer-averaged temperatures. 24 refs., 3 figs.
- OSTI ID:
- 535505
- Report Number(s):
- CONF-970207--
- Country of Publication:
- United States
- Language:
- English
Similar Records
Interannual variation of global atmospheric angular momentum
Detecting greenhouse-gas-induced climate change with an optimal fingerprint method