River Monitoring with Ground-based Videography - Fluvial Remote Sensing for Science and Management - page 363

001; St-Charles River, R 2

of the unfiltered PIV measurements were respectively
−

p 001).

However, the slopes of the ordinary least square regres-
sions were less than one (River Wharfe, slope = 0 . 43;
St-Charles River, slope

< 0 . = 0 . 80, p < 0 .

4 . 29 cm/s and ± 8 . 44 cm/s on the River Wharfe and

− 96 cm/s on the St-Charles River.

On the St-Charles River, the large underestimation of
surface flow velocities by the PIV method is thought to
be related to the bed effect due to the clear and shallow
water. This bed effect was less important on the River

30 . 92 cm/s and ± 9 .

58), indicating a consis-
tent underestimation in the LSPIV measurements
(Figure 16.5a). Using the current meter measurements
as the true velocities, the overall accuracy and precision

= 0 .

(a)

(b)

Figure 16.4 The spatial probability density function (PDF) of instantaneous velocity vectors in the x (streamwise) and y (lateral)
directions: a) prior to filtering; and b) after filtering.