Premium
A model for external drift kriging with uncertain covariates applied to air quality measurements and dispersion model output
Author(s) -
de Kassteele Jan van,
Stein Alfred
Publication year - 2006
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/env.771
Subject(s) - kriging , covariate , interpolation (computer graphics) , bayesian inference , dispersion (optics) , atmospheric dispersion modeling , air quality index , inference , bayesian probability , quality (philosophy) , multivariate interpolation , statistics , computer science , environmental science , econometrics , mathematics , meteorology , geography , air pollution , artificial intelligence , physics , chemistry , organic chemistry , optics , motion (physics) , philosophy , epistemology , bilinear interpolation
We present a method that combines uncertain air quality measurements with uncertain secondary information from an atmospheric dispersion model. The method combines external drift kriging and a measurement error (ME) model, and uses Bayesian techniques for inference. An illustration with simulated data shows what can theoretically be expected. The method is flexible for assigning different error variances to both the primary information and secondary information at each location. Next, we address actual NO 2 data collected at an urban and a rural site in the Netherlands. Uncertainty assessments in terms of exceeding air quality standards are given. The study shows that biased uncertain secondary information can be used successfully in a spatial interpolation study at the national scale. Copyright © 2005 John Wiley & Sons, Ltd.