2024 – Is Conventional Hydrological Calibration Methodology Fully Reliable for Estimating Extreme Events? Application to Six Catchments in VIC

Monte Azmi, Kosta Tsourdalakis, Nam Nguyen

Determining the values of delay and nonlinearity parameters of hydrological models is a challenging task due to the lack of a clear methodology and benchmark. In routing-based hydrological models, the default value of the nonlinearity parameter is mostly used where existing empirical equations are applied to get a plausible range for the delay parameter. In Australia, for extreme hydrological modeling purposes, models are routinely calibrated with a focus on 1% Annual Exceedance Probability (AEP) peaks; and then
calibrated models (i.e. calibrated routing parameters) are employed for extreme events up to and including probable maximum precipitation flood (PMPF). . In this research, a comparative assessment was undertaken based on two different calibration approaches (namely conventional and alternative). The default value of nonlinearity was adopted for the conventional approach, while for the alternative approach, the calibrated value was determined to replicate at-site FFA. For the alternative approach, a new criterion was introduced to give a plausible range for changes in the nonlinearity parameter. In terms of validation and verification, in addition to observed events, 2-dimensional hydraulic models were applied. The conventional and alternative approaches were then tested on six catchments in Australia VIC with a range of areas from 30km2 to just under 1000km2. Results showed that for three catchments, where nonlinearity values were considerably different from a default value (m=0.8), significant differences in extreme hydrological outcomes were identified (up to 50% differences for PMPF). Considering the outcomes, the assumption of a fixed nonlinearity parameter for the full range of events is questionable. This parameter is closely related to the hydraulic behavior of catchments, therefore with changes from frequent to extreme events (since the hydraulic behavior of catchments changes), the update of the nonlinearity parameter was reasonable. It was noted that depending on the project types (e.g., urban, infrastructure, dams), the impact and consequences can potentially be significant.