Objective identification and tracking of multicentre cyclones in the ERA-Interim reanalysis dataset

We present a novel cyclone identification and tracking method that explicitly recognizes multicentre cyclones (MCCs), defined as a cyclonic system with two or three sea-level pressure minima within its outermost contour. The method allows for the recognition of cyclone merger and splitting events in a natural way, and provides a consistent measure of the cyclone extent. Using the ERA-Interim reanalysis dataset, we compute a climatology using this method and show that MCCs occur in about 32 of all cyclone tracks and are much more prevalent in more intense storms. We also show that the method permits reconnection of tracks that would have been spuriously split using a conventional method. We present spatial maps of cyclone mergers, splitting, genesis and lysis using the method and also compute statistics of precipitation falling within cyclones, showing that it is strongly concentrated in the most intense cyclones. Copyright (c) 2011 Royal Meteorological Society