HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Boer, E. P. J. Articles by Stein, A. Search for Related Content PubMed PubMed Citation Articles by Boer, E. P. J. Articles by Stein, A. Agricola Articles by Boer, E. P. J. Articles by Stein, A. Related Collections Spatial Distribution Statistics Air Pollution

Optimization of a Monitoring Network for Sulfur Dioxide

^a Wageningen University & Research Centre, Mathematical and Statistical Methods Group, Dreyenlaan 4, 6703 HA,Wageningen, the Netherlands
^b RIVM, National Institute of Public Health and the Environment, P.O. Box 1, 3720 BA, Bilthoven, the Netherlands

View larger version (31K): [in a new window]   Fig. 1. Monitoring network of 74 stations in the Netherlands in 1993 coded with the annual average SO2 concentration (µg m-3).

View larger version (30K): [in a new window]   Fig. 2. Estimation locations with corresponding weights based on population density at each location in number of inhabitants per km2 (Criterion II, left). Estimation locations with weights based on residuals of locally weighted regression (Criterion III, right).

View larger version (16K): [in a new window]   Fig. 3. Plots of cross validation values (N = 74) for a range of smoothing parameters for the tricube and the McLain weight function, both isotropic as anisotropic (top). The weight functions with the optimized smoothing parameters (bottom).

View larger version (42K): [in a new window]   Fig. 4. Maps of interpolated SO2 values (N = 74) of the annual average concentration of SO2 by locally weighted regression, hf = 70. Tricube weight function (top) and McLain weight function (bottom).

View larger version (13K): [in a new window]   Fig. 5. Values of (n) of a monitoring network where sequential monitoring stations are dropped from the original network of 74.

View larger version (32K): [in a new window]   Fig. 6. Monitoring networks (black squares) obtained from the network of 74 monitoring stations by the drop–add algorithm . Tricube weight function, p = 1, anisotropic, and hf = 120 (km), for the three design criteria (A–C) (see the Specification of Different Design Criteria section). Same for McLain weight function, hf = 136 (km) (D–F).