Geostatistical risk mapping with chemical transport model output and ozone station data

Numerical models usually average over a volume of sensibly larger size than the averaging volume of measuring devices. We propose to use a geostatistical change-of-support model to describe how the statistical descriptors of ozone in the area of Paris are altered by averaging. The change-of-support model is used for estimating locally the risk of exceeding a given alert level. The averaging volume which the results refer to is of a size intermediate between the large transport model cells and the comparatively pointwise station measurements. We also describe how the model output can be corrected by kriging with external drift prior to its use in conditioning the local risk estimate. The interpretation of the external drift coefficient is examined. (C) 2004 Elsevier B.V. All rights reserved.