Earth System Processes - Global Meeting (June 24-28, 2001)

Paper No. 0
Presentation Time: 3:50 PM

EVIDENCE FOR SHORT-TERM FORCING OF LAND-SURFACE AIR TEMPERATURE BY ANTHROPOGENIC CO2 EMISSIONS


HUGHES, Gary B.1, GIEGENGACK, Robert2, JOHNSON, Arthur H.2 and KRITIKOS, Haralambos N.3, (1)Raytheon Infrared Operations, 75 Coromar Drive, B01-23, Goleta, CA 93117, (2)Department of Earth and Environmental Science, Univ of Pennsylvania, Room 251 Hayden Hall, 240 S. 33rd Street, Philadelphia, PA 19104, (3)Department of Electrical Engineering, Univ of Pennsylvania, Philadelphia, PA 19104, gbhughes@west.raytheon.com

We compare records of atmospheric CO2 concentration (1), land surface-air temperature (2), and anthropogenic CO2 emissions (3) during the period 1968-1993. Time-series data were de-trended by linear regression. Seasonal and ENSO-related interannual variations of temperature and CO2 were removed by spectral filtering (4). The de-trended, filtered residuals of temperature and CO2 vary synchronously; maximum correlation of 40% between the two residual signals occurs when the curves are shifted 122 days in the direction of a CO2 lead, temperature lag relationship. A 'greenhouse' cause-effect link is one plausible explanation for the observed correlation. The magnitude of this hypothesized effect is indicated by the correlation to be 5.0 °C for a CO2 doubling, which is marginally consistent with model predictions (5). A similar analysis also demonstrates that changes in anthropogenic CO2 emissions track together with atmospheric CO2 variations; maximum correlation of 45% between the two signals occurs when the curves are shifted 338 days in the direction of an emissions lead, atmospheric CO2 lag relationship. The magnitude of the effect indicated by the correlation is 1.8 Pg C/ppmV CO2, compared to mass balance estimates of 1.0 Pg C/ppmV CO2 (6). By the naïve approach of transitive and direct comparison, anthropogenic CO2 emissions statistically account for 20% of observed land-surface air temperature residual variability within the period of time described by these data. This result does not rely on atmospheric model predictions.

(1) Conway, T.J., and Tans, P.P., NOAA Cooperative Flask Sampling Network, 1967-1993. (2) Easterling, D.R., Karl, S.A., et al., US Global Historical Climatological Network. (3) Marland, G., Boden, T., and Andres, B., Carbon Dioxide Information Analysis Center. (4) Hughes, G.B., Giegengack, R., and Kritikos, H.N., 1999, International Journal of Climatology, v. 19, pp. 813-819. (5) Boer, G.J., Arpe, K., et al., 1992, Journal of Geophysical Research, v. 97, p. 12,771. (6) Keeling, R.F., Piper S.C., and Heimann, M., 1996, Nature, v. 381, n. 6579, pp. 218-221.