Statistical Correction of a Deterministic Numerical Weather Prediction Model

The forecasting skill of meteorologists is determined largely by their ability to interpret output from deterministic numerical weather prediction (NWP) models in the light of local conditions. Biases in the deterministic model may arise for many reasons, including the inability to account for physical processes at a scale smaller than the grid used in the numerical solution of the model equations. A statistical method for correction and interpretation of NWP model output is the model output statistics technique (MOS), where regression analysis is used to relate observations and NWP model predictions. We describe a Bayesian hierarchical approach to MOS that was motivated by the need to develop sensible statistical corrections for recently opened stations with a short historical data record. The strength of the Bayesian hierarchical formulation is its ability to combine information from stations with short and long observational records in a sensible way. Markov chain Monte Carlo methods are used for computation, and our approach is illustrated by using daily maximum temperature data from a network of 29 stations in the Sydney area.

[1]  J. Holton.  An introduction to dynamic meteorology , 2004 .

[2]  H. Glahn,et al.  The Use of Model Output Statistics (MOS) in Objective Weather Forecasting , 1972 .

[3]  R. Tibshirani,et al.  Generalized additive models for medical research , 1986, Statistical methods in medical research.

[4]  Andrew Harvey,et al.  Forecasting, Structural Time Series Models and the Kalman Filter , 1990 .

[5]  R. Kohn,et al.  On Gibbs sampling for state space models , 1994 .

[6]  S. Frühwirth-Schnatter.  Data Augmentation and Dynamic Linear Models , 1994 .

[7]  N. Shephard,et al.  The simulation smoother for time series models , 1995 .

[8]  Robert L. Vislocky,et al.  Generalized Additive Models versus Linear Regression in Generating Probabilistic MOS Forecasts of Aviation Weather Parameters , 1995 .

[9]  L. Wasserman,et al.  A Reference Bayesian Test for Nested Hypotheses and its Relationship to the Schwarz Criterion , 1995 .

[10]  A. O'Hagan,et al.  Fractional Bayes factors for model comparison , 1995 .

[11]  J. Berger,et al.  The Intrinsic Bayes Factor for Model Selection and Prediction , 1996 .

[12]  Peter Green,et al.  Markov chain Monte Carlo in Practice , 1996 .

[13]  M. Clyde,et al.  Prediction via Orthogonalized Model Mixing , 1996 .

[14]  Thomas S. Shively,et al.  Variable Selection and Function Estimation in Additive Nonparametric Regression Using a Data-Based Prior , 1999 .

[15]  David Ruppert,et al.  Variable Selection and Function Estimation in Additive Nonparametric Regression Using a Data-Based Prior: Comment , 1999 .

[16]  Robert Kohn,et al.  Nonparametric regression using linear combinations of basis functions , 2001, Stat. Comput..