Correlations between pollutants using Kriging estimates at the base
| CO | PM10 | NO2 | SO2 | O3 | |
|---|---|---|---|---|---|
| Across bases (N = 935 base*year observations) | |||||
| CO | 1 | ||||
| PM10 | 0.5303 | 1 | |||
| NO2 | 0.5106 | 0.3998 | 1 | ||
| SO2 | 0.0821 | 0.0684 | 0.2414 | 1 | |
| O3 | −0.0208 | 0.228 | 0.1447 | 0.3593 | 1 |
| In sample (weighted by population in base N = 159,275) | |||||
| CO | 1 | ||||
| PM10 | 0.5494 | 1 | |||
| NO2 | 0.1982 | 0.089 | 1 | ||
| SO2 | 0.2727 | 0.3377 | 0.4165 | 1 | |
| O3 | −0.0307 | 0.1438 | 0.1438 | 0.2723 | 1 |
These correlations are based on spatial predictions using Kriging and thus might be quite different in reality. But since monitors for different pollutants do not overlap, the real correlations at particular locations are not known.