Classification of Dansgaard-Oeschger climatic cycles by the application of similitude signal processing

In this work we have performed a detailed analysis, using signal processing tools, to study time series of data (temperature proxy) extracted from the GRIP ice-core records and we relate it with the evolution of atmospheric CO2 within the last glacial period. Our method is based in considering the warm periods known as Dansgaard-Oeschger (D/O) events and posterior returns to the cold stage as a climatic cycle. After the warming phase, D/O events relax to the initial cold state in three different ways, what gives rise to three classes of cycles. Also, the Younger/Dryas-Bolling/Allerod (Y/D-B/A) cycle corresponds to one of the classes obtained. We have found that all cycles start with identical warming phases which seem completely unrelated to variations in CO2 concentration. We discuss on the consequences for global climatology of such steady pattern of cycles. (c) 2007 Elsevier B.V. All rights reserved.