Predictive skill, predictive capability and predictability in ocean forecasting

We discuss the concepts involved in the evaluation and quantitative verification of ocean forecasts and present two predictive skill experiments to develop and research these concepts, carried out in the North Atlantic and Mediterranean Sea in 2001 and 2002. Ocean forecasting involves complex ocean observing and prediction systems for ocean regions with multi-scale interdisciplinary dynamical processes and strong, intermittent events. Now that ocean forecasting is becoming more common, it is critically important to interpret and evaluate regional forecasts in order to establish their usefulness to the scientific and applied communities. The Assessment of Skill for Coastal Ocean Transients (ASCOT) project is a series of real-time Coastal Predictive Skill Experiment (CPSE) and Rapid Environmental Assessment (REA) experiments and simulations focused on quantitative skill evaluation, carried out by the Harvard Ocean Prediction System group in collaboration with the NATO SACLANT Undersea Research Centre. ASCOT-01 was carried out in Massachusetts Bay and the Gulf of Maine in June 2001. ASCOT-02 took place in May 2002 in the Corsican Channel near the island of Elba in the Mediterranean Sea. Results from the ASCOT exercises highlight the dual use of data for skill evaluation and assimilation, real-time adaptive sampling and skill optimization and present both real-time and a posteriori evaluations of predictive skill and predictive capability.

[1]  A. M. Davies,et al.  Quantitative skill assessment for coastal ocean models , 1995 .

[2]  Richard A. Levine,et al.  Wavelets and Field Forecast Verification , 1997 .

[3]  Roman Krzysztofowicz.  Bayesian Correlation Score: A Utilitarian Measure of Forecast Skill , 1992 .

[4]  Allan R. Robinson,et al.  An interdisciplinary ocean prediction system: Assimilation strategies ana structured data models , 1996 .

[5]  Hernan G. Arango,et al.  Real-time operational forecasting on shipboard of the Iceland-Faeroe frontal variability , 1996 .

[6]  James A. Carton.  How predictable are the geostrophic currents in the recirculation zone of the North Atlantic , 1987 .

[7]  A. H. Murphy,et al.  Skill Scores and Correlation Coefficients in Model Verification , 1989 .

[8]  A. Navarra,et al.  Pattern Analysis of SST-Forced Variability in Ensemble GCM Simulations: Examples over Europe and the Tropical Pacific , 1997 .

[9]  Cambridge Ma.  Advanced Interdisciplinary Data Assimilation: Filtering and Smoothing via Error Subspace Statistical Estimation , 2002 .

[10]  N SwarztrauberPaul,et al.  A standard test set for numerical approximations to the shallow water equations in spherical geometry , 1992 .

[11]  Hernan G. Arango,et al.  Quantitative skill of quasi‐geostrophic forecasts of a baroclinically unstable Iceland‐Faroe Front , 1995 .

[12]  S. Besiktepe.  Coupled physical and biochemical data driven simulations of Black Sea in spring-summer: real-time forecast and data assimilation , 2003 .

[13]  Michael Ghil,et al.  Recent forecast skill for the El Niño/Southern Oscillation , 1998 .

[14]  P. Swarztrauber,et al.  A standard test set for numerical approximations to the shallow water equations in spherical geometry , 1992 .

[15]  P. Stamus,et al.  Application of a Scale-Separation Verification Technique to Regional Forecast Models , 1992 .

[16]  D. Haidvogel,et al.  Numerical models of the coastal ocean , 1998 .

[17]  Scott Glenn,et al.  Experiment evaluates ocean models and data assimiliation in the Gulf Stream , 1994 .

[18]  A. H. Murphy,et al.  General Decompositions of MSE-Based Skill Scores: Measures of Some Basic Aspects of Forecast Quality , 1996 .

[19]  Roberto Buizza,et al.  Impact of Ensemble Size on Ensemble Prediction , 1998 .

[20]  Scott Glenn,et al.  Forecasting Gulf Stream meanders and rings , 1989 .

[21]  Allan R. Robinson,et al.  Real-time regional forecasting , 1996 .

[22]  T. Palmer,et al.  Predictability of a Coupled Model of ENSO Using Singular Vector Analysis. Part II: Optimal Growth and Forecast Skill , 1997 .

[23]  Operational regional prediction , 1997 .

[24]  Pierre F. J. Lermusiaux,et al.  Coupled physical and biogeochemical data-driven simulations of Massachusetts Bay in late summer: real-time and postcruise data assimilation , 2003 .

[25]  Allan R. Robinson,et al.  Physical processes, field estimation and an approach to interdisciplinary ocean modeling , 1996 .

[26]  A. H. Murphy.  Forecast verification: Its Complexity and Dimensionality , 1991 .