Table 5 Multi-scale prediction accuracy statistics for microplastic transport modeling.

From: Spatiotemporal graph neural networks for analyzing the influence mechanisms of river hydrodynamics on microplastic transport processes

Spatiotemporal scale

Sample size

Mean absolute error (mg/L)

Root mean square error (mg/L)

Correlation coefficient

Prediction confidence interval (mg/L)

Short-term (1–6 h)

2847

0.0156

0.0234

0.923

± 0.0298

Medium-term (1–7 d)

1425

0.0287

0.0412

0.847

± 0.0523

Long-term (1–4 w)

368

0.0445

0.0634

0.724

± 0.0789

Seasonal (3–12 m)

96

0.0623

0.0891

0.658

± 0.1127

Back to article page