Statistical Compression of Wind Speed Data

• Activity Number: 157 - Compressing Climate Model Data: Lowering Storage Burden While Preserving Information
• Type: Topic Contributed
• Date/Time: Monday, July 31, 2017 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Felipe Tagle* and Stefano Castruccio and Marc G. Genton and Paola Crippa
• Companies: Newcastle University and Newcastle University and KAUST and Newcastle University
• Keywords: skew-t distribution ; wind energy ; stochastic wind generator ; internal climate variability

Abstract

A recent study of the wind fields over Saudi Arabia demonstrated that an important component of the interannual variability of available wind power is the internal variability of the climate system. Typically, the quantification of internal climate variability relies on multi-member ensembles such as the Large Ensemble (LENS) project developed at NCAR comprising more than 30 fully-coupled multi-decadal simulations. In this talk, we propose a spatio-temporal model of daily wind speed, based on a multivariate skew-t distribution, that requires only a few of these simulations for training and can accurately reproduce the internal variability present in the LENS.