Table 5 shows the evaluation criteria of the estimated outflow hydrographs by the LMM-L and VEP-NLMM-L. Similarly to Case 1, the VEP-NLMM-L still produces more accurate routing outflows in terms of all criteria than the LMM-L and the accuracy increases with the number of inflow dividing levels L. All models can estimate the peak outflow on the correct time interval (DPOT = 0). The VEP-NLMM-L with L = 4 also gives the closest peak outflow to the observed value.
Evaluation criteria of estimated outflows for River Wyre flood, October 1982
| Model . | DPO . | DPOT . | η . | MARE . | MAE . | |
|---|---|---|---|---|---|---|
| LMM-L (O'Donnell 1985) | 1.10 | 0 | 98.39 | 0.101 | 3.231 | |
| VEP-NLMM-L (this study) | L = 1 | 1.88 | 0 | 99.81 | 0.030 | 0.950 |
| L = 2 | 0.29 | 0 | 99.90 | 0.024 | 0.665 | |
| L = 3 | 0.17 | 0 | 99.91 | 0.025 | 0.689 | |
| L = 4 | 0.07 | 0 | 99.93 | 0.023 | 0.662 | |
| L = 5 | 0.16 | 0 | 99.94 | 0.021 | 0.624 | |
| Model . | DPO . | DPOT . | η . | MARE . | MAE . | |
|---|---|---|---|---|---|---|
| LMM-L (O'Donnell 1985) | 1.10 | 0 | 98.39 | 0.101 | 3.231 | |
| VEP-NLMM-L (this study) | L = 1 | 1.88 | 0 | 99.81 | 0.030 | 0.950 |
| L = 2 | 0.29 | 0 | 99.90 | 0.024 | 0.665 | |
| L = 3 | 0.17 | 0 | 99.91 | 0.025 | 0.689 | |
| L = 4 | 0.07 | 0 | 99.93 | 0.023 | 0.662 | |
| L = 5 | 0.16 | 0 | 99.94 | 0.021 | 0.624 | |