Bayesian inference for wind field retrieval


In many problems in spatial statistics it is necessary to infer a global problem solution by combining local models. A principled approach to this problem is to develop a global probabilistic model for the relationships between local variables and to use this as the prior in a Bayesian inference procedure. We show how a Gaussian process with hyper-parameters estimated from Numerical Weather Prediction Models yields meteorologically convincing wind fields. We use neural networks to make local estimates of wind vector probabilities. The resulting inference problem cannot be solved analytically, but Markov Chain Monte Carlo methods allow us to retrieve accurate wind fields.

Divisions: ?? 50811700Jl ??
College of Engineering & Physical Sciences > Systems analytics research institute (SARI)
Uncontrolled Keywords: Bayesian inference,surface winds,spatial priors,Gaussian processes
ISBN: NCRG/98/023
Last Modified: 27 Jun 2024 11:55
Date Deposited: 30 Jul 2009 07:57
PURE Output Type: Technical report
Published Date: 1998-10-22
Authors: Cornford, Dan (ORCID Profile 0000-0001-8787-6758)
Nabney, Ian T. (ORCID Profile 0000-0003-1513-993X)
Williams, Christopher K. I.


Export / Share Citation


Additional statistics for this record