Joint Centre for Mesoscale Meteorology (JCMM)

Interaction of mesoscale tropopause depressions with a warm-core cyclone:  observational and model analyses, and sensitivity to analysis errors
K A Browning, A J Thorpe, A Montani, D Parsons, M Griffiths, P Panagi and E M Dicks

This paper focuses on the coupling between the warm core of an ex-tropical cyclone and 2 pre-existing mesoscale tropopause depressions (TDs), ie downward intrusions of stratospheric air.  The TDs approached the cyclone from widely separated sources after becoming cut off from different upper-level troughs upstream of it.

The first part of the study reported here combines Meteosat imagery with products from the limited-area version of the operational UK Met Office Unified Model to reveal the 3-dimensional structure and evolution of the mesoscale features.  Each TD was a potential vorticity (PV) anomaly characterised by dry-intrusion air descending beneath an upper level jet streak.  Each TD generated its own cloud head and each is believed to have contributed to the deepening of the surface cyclone.

Later parts of the paper identify errors in model forecasts and attribute them to analysis errors in the position of one of the TDs.  We have used two methods to locate the analysis errors.  Both are capable of being used in real time.  The first is the identification via satellite imagery of an error in the Met Office model's water vapour analysis, and its translation into thermal and kinematic information via PV inversion.  The second method, using singular vectors calculated for the ECMWF model, involves the identification of sensitive regions where any analysis error would be expected to grow rapidly.  In the case studied here the regions highlighted by these two methods were essentially co-located.  The present case is unusual in that the most sensitive region was in the upper troposphere and apparently associated with a tropopause depression.  Forecast reruns made with the Met Office and ECMWF models after modifying the upper-level PV in the analysis showed some limited improvements.

Things to do now



Page navigation