
Comparison of model and observed regional temperature changes during the past 40 years
Author(s) -
Russell Gary L.,
Miller James R.,
Rind David,
Ruedy Reto A.,
Schmidt Gavin A.,
Sheth Sukeshi
Publication year - 2000
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2000jd900156
Subject(s) - greenhouse gas , northern hemisphere , environmental science , climatology , atmospheric sciences , climate model , forcing (mathematics) , southern hemisphere , troposphere , atmosphere (unit) , climate change , meteorology , geology , geography , oceanography
Results are presented for six simulations of the Goddard Institute for Space Studies (GISS) global atmosphere‐ocean model for the years 1950–2099. There are two control simulations with constant 1950 atmospheric composition from different initial states, two greenhouse gas (GHG) experiments with observed greenhouse gases up to 1990 and compounded 0.5% CO 2 annual increases thereafter, and two greenhouse gas plus sulfate (GHG+SO 4 ) experiments with the same varying greenhouse gases plus varying tropospheric sulfate aerosols. Surface air temperature trends in the two GHG experiments (with the control simulations climate drift subtracted out) are compared between themselves and with changes in the observed temperature record between 1960 and 1998. All comparisons show high positive spatial correlation in the Northern Hemisphere except in summer when the greenhouse signal is weakest. The GHG+SO 4 experiments show weaker correlations. In the Southern Hemisphere, correlations between any experiments or observations are either weak or negative, in part because the model's interannual variability of southern sea ice cover is unrealistic. The model results imply that temperature changes due to forcing by increased greenhouse gases have risen above the level of regional interannual temperature variability in the Northern Hemisphere over the past 40 years. This period is thus an important test of the reliability of coupled climate models.