Response of the interglacial climate to different astronomical forcing

Interglacials of the last one million years are a few thousands to tens of thousands of years long. Transient climate simulation is probably the best way to understand their climate behavior. In the mean time, before transient simulation is available, snapshot simulation at particularly well suited dates helps to understand how the interglacial climate behaves. Two dates are evident: the peak of the interglacial which generally corresponds to a time when northern hemisphere fall occurs at perihelion, and the date when northern hemisphere summer occurs at perihelion just preceding the interglacial peak. The simulation at the date when northern hemisphere summer at perihelion has the advantages: 1. to maximize the forcing signal; 2. to help comparing more easily the interglacials between them together as the only astronomical difference arises from obliquity and eccentricity; and 3. to take into account the fact the climate system response to the astronomical forcing with a lag of about 3-5 thousands of years. The simulation at the peak of the interglacials gives a first instantaneously view of the climate in equilibrium with the forcing whatever it is at that particular date. The purpose of this paper is to compare both simulations at each interglacial and the interglacials between each other. The differences in the simulated results between these two dates are quite large at regional and seasonal scales. For example, compared with the simulation of northern hemisphere summer at perihelion, the annual climate at the peak of the interglacials is much cooler in northern mid-high latitudes and warmer in tropical-subtropical regions. These climatic differences between the two dates show the importance of the chronology when proxy records and simulation results are compared. Transient simulations are expected to smooth out this difficulty.