Home About HyMeX
Motivations
Science questions
Observation strategy
Modelling strategy
Target areas
Key documents
Organisation
International coordination
Working groups
Task teams
National contributions
Endorsements
Resources
Database
Data policy
Publications
Education and summer schools
Drifting balloons (BAMED)
SOP web page
Google maps data visualisation
Workshops Projects
ASICS-MED
MOBICLIMEX
MUSIC
IODA-MED
REMEMBER
FLOODSCALE
EXAEDRE
Offers Links Contacts
Science & Task teams
Science teams
Task teams
Implementation plan
Coordination
International Scientific Steering Committee (ISSC)
Executive Committee for Implementation and Science Coordination (EC-ISC)
Executive Committee - France (EC-Fr)
HyMeX France
HyMeX Italy
HyMeX Spain
Archive
by Vich, M.-D.-M., Romero, R. and Homar, V.
Abstract:
In Part I of this work, an ensemble prediction system (EPS) based on different combinations of model physical parameterizations was compared against another ensemble based on perturbing initial and boundary conditions through the Potential Vorticity (PV) field. This comparison was done for western Mediterranean cyclonic situations associated with high-impact weather phenomena such as heavy rain and showed a better performance of the PV-perturbed ensemble over the more traditional multiphysics approach. The current study extends the comparison to another ensemble based on perturbing initial and boundary conditions through the PV field but guided by the MM5 adjoint derived sensitivity zones (PV-adjoint) instead of by the three-dimensional PV features showing intense values and gradients as was done in Part I (PV-gradient). The PV-adjoint and PV-gradient EPSs perturb specific areas of the cyclonic development using a PV error climatology that typifies PV errors in the initial and boundary conditions to provide the appropriate error range. The non-hydrostatic MM5 mesoscale model nested in the ECMWF forecast fields is used to provide all predictions. For the studied cases, 19 cyclonic events associated with heavy rain, the verification results show that both PV-perturbed are skillful, the PV-gradient being the best. Therefore, for our testbed, the extra computational cost of running the MM5 adjoint model does not provide a significant ensemble skill improvement.
Reference:
Vich, M.-D.-M., Romero, R. and Homar, V., 2011: Ensemble prediction of Mediterranean high-impact events using potential vorticity perturbations. Part II: Adjoint-derived sensitivity zonesAtmospheric Research, 102, 311-319.
Bibtex Entry:
@Article{Vich2011b,
  Title                    = {Ensemble prediction of Mediterranean high-impact events using potential vorticity perturbations. Part II: Adjoint-derived sensitivity zones},
  Author                   = {Vich, M.-D.-M. and Romero, R. and Homar, V.},
  Journal                  = {Atmospheric Research},
  Year                     = {2011},

  Month                    = {November},
  Number                   = {3},
  Pages                    = {311-319},
  Volume                   = {102},

  Abstract                 = {In Part I of this work, an ensemble prediction system (EPS) based on different combinations of model physical parameterizations was compared against another ensemble based on perturbing initial and boundary conditions through the Potential Vorticity (PV) field. This comparison was done for western Mediterranean cyclonic situations associated with high-impact weather phenomena such as heavy rain and showed a better performance of the PV-perturbed ensemble over the more traditional multiphysics approach. The current study extends the comparison to another ensemble based on perturbing initial and boundary conditions through the PV field but guided by the MM5 adjoint derived sensitivity zones (PV-adjoint) instead of by the three-dimensional PV features showing intense values and gradients as was done in Part I (PV-gradient).

The PV-adjoint and PV-gradient EPSs perturb specific areas of the cyclonic development using a PV error climatology that typifies PV errors in the initial and boundary conditions to provide the appropriate error range. The non-hydrostatic MM5 mesoscale model nested in the ECMWF forecast fields is used to provide all predictions.

For the studied cases, 19 cyclonic events associated with heavy rain, the verification results show that both PV-perturbed are skillful, the PV-gradient being the best. Therefore, for our testbed, the extra computational cost of running the MM5 adjoint model does not provide a significant ensemble skill improvement.},
  Copublication            = {3: 3 Es},
  Doi                      = {10.1016/j.atmosres.2011.07.016},
  Owner                    = {hymexw},
  Timestamp                = {2016.01.08},
  Url                      = {http://www.sciencedirect.com/science/article/pii/S016980951100247X}
}