DATE2016-05-27 13:59:21
IDABSTRACT20160527135921-1019
CONTACTleone.cavicchia@cmcc.it
PRESENTATIONORAL-PARALLEL
INVITED0
IDSESSION1
TITLEMEDITERRANEAN EXTREME PRECIPITATION: A MULTI-MODEL ASSESSMENT
AUTHORSLeone Cavicchia (1), Enrico Scoccimarro (1,2), Silvio Gualdi (1,2), Paola Marson (11), Bodo Ahrens (3), Ségolène Berthou (5), Dario Conte (1), Alessandro Dell'aquila (6), Philippe Drobinski (5), Vladimir Djurdjevic (7), Clotilde Dubois (8), Clemente Gallardo (9), Laurent Li (5), Paolo Oddo (4), Antonella Sanna (1), Csaba Torma (10)
AFFILIATIONS
  1. Cmcc Bologna (Italy)
  2. Ingv Bologna (Italy)
  3. Institute For Atmospheric And Environmental Sciences, Goethe University Frankfurt Am Main (Germany)
  4. Centre For Maritime Research And Experi- 14 Mentation Sto Nato (cmre) La Spezia (Italy)
  5. Ipsl/laboratoire De Meteorologie Dynamique Palaiseau (France)
  6. Enea Rome (Italy)
  7. Institute Of Meteorology, University Of Belgrade Belgrade (Serbia And Montenegro)
  8. Météo France Toulouse (France)
  9. Universidad De Castilla-la Mancha Toledo (Spain)
  10. Ictp Trieste (Italy)
  11. Ca' Foscari University Venice (Italy)
ABSTRACTExploiting the added value of the ensemble of high-resolution model simulations provided by the Med-CORDEX coordinated initiative, an assessment of Mediterranean extreme precipitation as represented in different observational, reanalysis and modelling datasets is presented. A spatio-temporal characterisation of the long-term statistics of extreme precipitation is performed, using a number of different diagnostic indices. Employing a novel approach based on the timing of extreme precipitation, a number of physically consistent subregions are defined. Extreme precipitation diagnostics over the Mediterranean domain and physically homogeneous sub-domains are discussed, focussing on the impact of different model configurations (resolution, coupling and physical parameterisations) on the performance in reproducing observed precipitation. It is found that the agreement between the observed and modelled long-term statistics of extreme precipitation is more sensitive to the model convective parameterisation than to resolution or coupling.
PAGE1
STATE1