Comparison of initial perturbation methods for the mesoscale ensemble prediction system of the Meteorological Research Institute for the WWRP Beijing 2008 Olympics Research and Development Project (B08RDP)

Different initial perturbation methods for the mesoscale ensemble prediction were compared by the Meteorological Research Institute (MRI) as a part of the intercomparison of mesoscale ensemble prediction systems (EPSs) of the WorldWeather Research Programme (WWRP) Beijing 2008 Olympics Research and Development Project (B08RDP). Five initial perturbation methods for mesoscale ensemble prediction were developed for B08RDP and compared at MRI: (1) a downscaling method of the Japan Meteorological Agency (JMA)’s operational one-week EPS (WEP), (2) a targeted global model singular vector (GSV) method, (3) a mesoscale model singular vector (MSV) method based on the adjoint model of the JMA non-hydrostatic model (NHM), (4) a mesoscale breeding growing mode (MBD) method based on the NHM forecast and (5) a local ensemble transform (LET) method based on the local ensemble transform Kalman filter (LETKF) using NHM. These perturbation methods were applied to the preliminary experiments of the B08RDP Tier-1 mesoscale ensemble prediction with a horizontal resolution of 15 km. To make the comparison easier, the same horizontal resolution (40 km) was employed for the three mesoscale model-based initial perturbation methods (MSV, MBD and LET). The GSV method completely outperformed the WEP method, confirming the advantage of targeting in mesoscale EPS. The GSV method generally performed well with regard to root mean square errors of the ensemble mean, large growth rates of ensemble spreads throughout the 36-h forecast period, and high detection rates and high Brier skill scores (BSSs) for weak rains. On the other hand, the mesoscale model-based initial perturbation methods showed good detection rates and BSSs for intense rains. The MSV method showed a rapid growth in the ensemble spread of precipitation up to a forecast time of 6 h, which suggests suitability of the mesoscale SV for short-range EPSs, but the initial large growth of the perturbation did not last long. The performance of the MBD method was good for ensemble prediction of intense rain with a relatively small computing cost. The LET method showed similar characteristics to the MBD method, but the spread and growth rate were slightly smaller and the relative operating characteristic area skill score and BSS did not surpass those of MBD. These characteristic features of the five methods were confirmed by checking the evolution of the total energy norms and their growth rates. Characteristics of the initial perturbations obtained by four methods (GSV, MSV, MBD and LET) were examined for the case of a synoptic low-pressure system passing over eastern China. With GSV and MSV, the regions of large spread were near the low-pressure system, but with MSV, the distribution was more concentrated on the mesoscale disturbance. On the other hand, large-spread areas were observed southwest of the disturbance in MBD and LET. The horizontal pattern of LET perturbation was similar to that of MBD, but the amplitude of the LET perturbation reflected the observation density.

[1]  Takemasa Miyoshi,et al.  Applying a Four-dimensional Local Ensemble Transform Kalman Filter (4D-LETKF) to the JMA Nonhydrostatic Model (NHM) , 2006 .

[2]  Clifford F. Mass,et al.  Aspects of Effective Mesoscale, Short-Range Ensemble Forecasting , 2005 .

[3]  T. Miyoshi,et al.  Data assimilation experiments of precipitable water vapour using the LETKF system: intense rainfall event over Japan 28 July 2008 , 2011 .

[4]  Donald W. Burgess,et al.  The Sydney 2000 World Weather Research Programme Forecast Demonstration Project: Overview and Current Status , 2003 .

[5]  Kazuo Saito,et al.  Mesoscale Data Assimilation Experiment in the WWRP B08RDP , 2010 .

[6]  Isztar Zawadzki,et al.  Analysis of scale dependence of quantitative precipitation forecast verification: A case‐study over the Mackenzie river basin , 2006 .

[7]  D. Nychka.  Data Assimilation” , 2006 .

[8]  Ryan D. Torn,et al.  Boundary Conditions for Limited-Area Ensemble Kalman Filters , 2006 .

[9]  D. Stensrud,et al.  Reliable Probabilistic Quantitative Precipitation Forecasts from a Short-Range Ensemble Forecasting System , 2007 .

[10]  H. Niino,et al.  An Improved Mellor–Yamada Level-3 Model with Condensation Physics: Its Design and Verification , 2004 .

[11]  Istvan Szunyogh,et al.  Efficient data assimilation for spatiotemporal chaos: A local ensemble transform Kalman filter , 2005, physics/0511236.

[12]  F. Molteni,et al.  The ECMWF Ensemble Prediction System: Methodology and validation , 1996 .

[13]  Munehiko Yamaguchi,et al.  Typhoon Ensemble Prediction System Developed at the Japan Meteorological Agency , 2008 .

[14]  Istvan Szunyogh,et al.  A Local Ensemble Kalman Filter for Atmospheric Data Assimilation , 2002 .

[15]  Neill E. Bowler,et al.  Comparison of error breeding, singular vectors, random perturbations and ensemble Kalman filter perturbation strategies on a simple model , 2006 .

[16]  気象庁.  Outline of the operational numerical weather prediction at the Japan Meteorological Agency , 1977 .

[17]  Ross N. Hoffman,et al.  Lagged average forecasting, an alternative to Monte Carlo forecasting , 1983 .

[18]  Tadashi Tsuyuki,et al.  Assimilation of Precipitation Data to the JMA Mesoscale Model with a Four-dimensional Variational Method and its Impact on Precipitation Forecasts , 2005 .

[19]  T. Miyoshi.  The Gaussian Approach to Adaptive Covariance Inflation and Its Implementation with the Local Ensemble Transform Kalman Filter , 2011 .

[20]  Craig H. Bishop,et al.  Adaptive sampling with the ensemble transform Kalman filter , 2001 .

[21]  Neill E. Bowler,et al.  The MOGREPS short‐range ensemble prediction system , 2008 .

[22]  Tiziana Paccagnella,et al.  The COSMO-LEPS mesoscale ensemble system: validation of the methodology and verification , 2005 .

[23]  Xuguang Wang,et al.  A Comparison of Breeding and Ensemble Transform Kalman Filter Ensemble Forecast Schemes , 2003 .

[24]  Eugenia Kalnay,et al.  Ensemble Forecasting at NMC: The Generation of Perturbations , 1993 .

[25]  Ronald M. Errico,et al.  Singular-Vector Perturbation Growth in a Primitive Equation Model with Moist Physics , 1999 .

[26]  A. H. Murphy,et al.  Skill Scores Based on the Mean Square Error and Their Relationships to the Correlation Coefficient , 1988 .

[27]  P. L. Houtekamer,et al.  Ensemble Kalman filtering , 2005 .

[28]  Albert A. M. Holtslag,et al.  Flux Parameterization over Land Surfaces for Atmospheric Models , 1991 .

[29]  E. Kalnay,et al.  Ensemble Forecasting at NCEP and the Breeding Method , 1997 .

[30]  Neill E. Bowler,et al.  Ensemble transform Kalman filter perturbations for a regional ensemble prediction system , 2009 .

[31]  M. Kunii,et al.  Numerical Simulation of Myanmar Cyclone Nargis and the Associated Storm Surge Part II: Ensemble Prediction , 2010 .

[32]  M. Kunii,et al.  Mesoscale Ensemble Experiments on Potential Parameters for Tornado Outbreak , 2009 .

[33]  Guillem Candille,et al.  A Regional Ensemble Prediction System Based on Moist Targeted Singular Vectors and Stochastic Parameter Perturbations , 2008 .

[34]  M. Baldwin,et al.  THE WGNE ASSESSMENT OF SHORT-TERM QUANTITATIVE PRECIPITATION FORECASTS , 2003 .

[35]  Yong Wang,et al.  The Central European limited‐area ensemble forecasting system: ALADIN‐LAEF , 2011 .

[36]  Takemasa Miyoshi,et al.  Localizing the Error Covariance by Physical Distances within a Local Ensemble Transform Kalman Filter (LETKF) , 2007 .

[37]  David J. Stensrud,et al.  Application of Lateral Boundary Condition Perturbations to Help Restore Dispersion in Limited-Area Ensemble Forecasts , 2004 .

[38]  Solving Ax = b using the Lanczos Algorithm with , 2004 .

[39]  Y. Honda,et al.  A pre‐operational variational data assimilation system for a non‐hydrostatic model at the Japan Meteorological Agency: formulation and preliminary results , 2005 .

[40]  Munehiko Yamaguchi,et al.  Using TIGGE Data to Diagnose Initial Perturbations and Their Growth for Tropical Cyclone Ensemble Forecasts , 2010 .

[41]  T. Segawa,et al.  Nonhydrostatic Atmospheric Models and Operational Development at JMA , 2007 .

[42]  I. Jolliffe,et al.  Forecast verification : a practitioner's guide in atmospheric science , 2011 .

[43]  Roberto Buizza,et al.  Tropical singular vectors computed with linearized diabatic physics , 2001 .

[44]  Shigeo Yoden.  Atmospheric Predictability(125th Anniversary Issue of the Meteorological Society of Japan) , 2007 .

[45]  V. Ducrocq,et al.  Mesoscale analysis and impact on simulation of IOP14 of the MAP experiment , 2005 .

[46]  Roberto Buizza,et al.  The Singular-Vector Structure of the Atmospheric Global Circulation , 1995 .

[47]  斉藤 和雄,et al.  WWRP Beijing 2008 Olympics forecast demonstration/research and development project (B08FDP/RDP) , 2010 .

[48]  Yong Wang,et al.  Verification and intercomparison of mesoscale ensemble prediction systems in the Beijing 2008 Olympics Research and Development Project , 2011 .

[49]  X. Deng,et al.  Model Error Representation in an Operational Ensemble Kalman Filter , 2009 .

[50]  Kazuo Saito,et al.  The Operational JMA Nonhydrostatic Mesoscale Model , 2006 .

[51]  Cathy Hohenegger,et al.  Predictability and Error Growth Dynamics in Cloud-Resolving Models , 2007 .