Table 4 Model predictions and uncertainty metrics across different population density classifications.

From: Global high-resolution ultrafine particle number concentrations through data fusion with machine learning

Classification

Population limit (km−2)

Mean PNC (cm−3)

Mean 95% CI

Mean Percentage Error

Rural

250

2,606

1,852 ± 56

91 ± 3

Suburban

800

6,360

2,165 ± 72

35 ± 3

Urban

1,900

14,992

3,715 ± 182

29 ± 2

  1. Areas are classified as rural (<250 people/km2), suburban (250–800 people/km2), or urban (>800 people/km2). For each class, the table shows mean PNC values, 95% coverage intervals, and percentage errors (presented as mean  ± standard error).
Back to article page