Amplitude-Phase Characteristics of the Annual Cycle of Surface Air Temperature in the Northern Hemisphere

Abstract

The amplitude phase characteristics (APC) or surface air temperature (SAT) annual cycle (AC) in the Northern Hemisphere are analyzed. From meteorological observations for the 20th century and meteorological reanalyses for its second half, it is found that over land negative correlation of SAT AC amplitude with annual mean SAT dominates. Nevertheless, some exceptions exist. The positive correlation between these two variables is found over the two desert regions: in northern Africa and in Central America. Areas of positive correlations are also found for the northern Pacific and for the tropical Indian and Pacific Oceans. Southward of the characteristic annual mean snow ice boundary (SIB) position, the shape of the SAT AC becomes more sinusoidal under climate warming. In contrast, northward of it, this shape becomes less sinusoidal. The latter is also found for the above-mentioned two desert regions. In the Far East (southward of about 50°N), the SAT AC shifts as a whole: here its spring and autumn phases occur earlier if the annual mean SAT increases. From energy balance climate considerations, those trends for SAT AC APC in the middle and high latitudes are associated with the influence of the albedo SAT feedback due to the SIB movement. In the Far East the trends are attributed to the interannual cloudiness variability, and in the desert regions, to the influence of a further desertification and/or scattering aerosol loading into the atmosphere. In the north Pacific, the exhibited trends could only be explained as a result of the influence of the greenhouse-gases loading on atmospheric opacity. The trends for SAT AC APC related to the SIB movement are simulated reasonably well by the climate model of intermediate complexity (IAP RAS CM) in the experiment with greenhouse gases atmospheric loading. In contrast, the tendencies resulting from the cloudiness variability are not reproduced by this model. The model also partly simulates the tendencies related to the desertification processes.