Table 3.
Summary statistics of measured BOD and mean error statistics of BOD model predictions (concentrations and specific loads) in the calibration and validation datasets. Model parameters are in Table 1. MAE = mean absolute error. RMSE = root mean Square Error.
Calibration |
Validation |
||||
---|---|---|---|---|---|
Concentration mg O2/L |
Load t/km2 |
Concentration mg O2/L |
Load t/km2 |
||
Observations | Sample size | 2156 | 2157 | 1131 | 1134 |
5th percentile | 0.73 | 0.21 | 0.66 | 0.22 | |
Mean | 2.29 | 0.81 | 1.75 | 0.62 | |
95th percentile | 5.65 | 1.92 | 3.62 | 1.21 | |
Max | 41.18 | 36.14 | 32.8 | 8.34 | |
Prediction errors | MAE | 1.72 | 0.52 | 1.21 | 0.37 |
RMSE | 5.47 | 1.20 | 3.04 | 0.79 | |
Percent bias % | 19.6 | 0.6 | 24.2 | 8.9 |