Atmospheric pollution forecasting. Applications of horizontal and vertical vector fields

David Pearson (1) , Mireille Batton-Hubert (2, 3)

1 LHC - Laboratoire Hubert Curien [Saint Etienne]
2 SITE-ENSMSE - Centre Sciences, Information et Technologies pour l'Environnement
3 IDEE-ENSMSE - Département Information, Décision et Evaluation Environnementale

In this paper we apply methods from differential geometry, called horizontal and vertical vector fields, to the problem of ozone concentration forecasting. These vector fields allow us to find structures in the data and to define a sort of distance between data points. Our aim is to use this distance to define a confidence measure for the forecast.

Keywords: forecasting; atmospheric pollution; horizontal and vertical vector fields; differential geometry

Domains: Computer Science [cs]; Signal and Image Processing; Engineering Sciences [physics]; Signal and Image processing

David Pearson, Mireille Batton-Hubert. Atmospheric pollution forecasting. Applications of horizontal and vertical vector fields. Journal Européen des Systèmes Automatisés (JESA), 2005, 39 (4), pp.553-569. ⟨10.3166/jesa.39.553-569⟩. ⟨ujm-00394804⟩