Abstract

The ERA40 and NCEP/NCAR data over 1958-1998 were used to estimate the sensitivity of ampli- tude-phase characteristics (APCs) of the annual cycle (AC) of the surface air temperature (SAT) T s . The results were compared with outputs of the ECHAM4/OPYC3, HadCM3, and INM RAS general circulation models and the IAP RAS climate model of intermediate complexity, which were run with variations in greenhouse gases and sulfate aerosol specified over 1860-2100. The analysis was performed in terms of the linear regres- sion coefficients b of SAT AC APCs on the local annual mean temperature and in terms of the sensitivity char- acteristic D = br 2 , which takes into account not only the linear regression coefficient but also its statistical sig- nificance (via the correlation coefficient r ). The reanalysis data were used to reveal the features of the tendencies of change in the SAT AC APCs in various regions, including areas near the snow-ice boundary, storm-track ocean regions, large desert areas, and the tropical Pacific. These results agree with earlier observations. The model computations are in fairly good agreement with the reanalysis data in regions of statistically significant variations in SAT AC APCs. The differences between individual models and the reanalysis data can be explained, in particular, in terms of the features of the sea-ice schemes used in the models. Over the land in the middle and high latitudes of the Northern Hemisphere, the absolute values of D for the fall phase time and the interval of exceeding exhibit a positive intermodel correlation with the absolute value of D for the annual-har- monic amplitude. Over the ocean, the models reproducing larger (in modulus) sensitivity parameters of the SAT annual-harmonic amplitude are generally characterized by larger (in modulus) negative sensitivity values of the semiannual-harmonic amplitude T s , 2 , especially at latitudes characteristic of the sea-ice boundary. In contrast to the averaged fields of AC APCs and their interannual standard deviations, the sensitivity parameters of the SAT AC APCs on a regional scale vary noticeably for various types of anthropogenic forcing.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call