Results - Response of Flood Events to Land Use and Climate Change - page 151

Fig. 6.19 Calibration based on Manning's coefficient values

measured close to the Abattoir Bridge. Because of a lack of data the calibration
was only based on one point measurement data. More observation points within
reach could have allowed for flood routing, leading to better calibration of the
model. Taking into account the length of the river and lack of observed data, the
model was considered to have been successfully calibrated. Manual calibration of
the 1D model was employed to constrain friction parameters within a feasible
range. Since precise water level predictions are crucial for dyke breach modeling,
calibration against distributed observations was preferred over the model, fit to
discharge series at the measurement gage. Figure 6.20 represents the Q-Q plot of
regression standardized residual for different friction variables.
Statistically, the best result in terms of R, R-square, and standard deviation error
was belonged to friction values of 0.05 (Table 6.11 ).
The 2008 event was selected because the observed hourly water height was
available. In this research, although there was limited observed data, the results