Skip to Main Content
The ADCP-IMU integration algorithm corrected 836 ensembles (4.8% of the total number of ensembles) potentially affected by compass errors. Table 1 and Figure 5 show the statistical and spatial distribution of the detected errors. The differences in obtained with and without compass correction (all other processing steps held constant) were significant in statistical terms , but subtle in physical terms for all cross sections and horizontal planes analysed (Table 2).
Table 1

Errors in ADCP compass heading , Total Station based positioning , boat velocity estimation and water velocity error introduced by kriging

 MeanMedianStandard deviationSample size
Compass correction 
 (deg) 2.59 1.68 3.47 836 
Spatial data referencing 
 (m) 0.021 0.016 0.018 13,543 
 (ms–1–0.001 –0.001 0.075 
 (ms–10.047 0.028 0.058 
Kriging cross validation 
 (ms–1–0.001 –0.001 0.075 1,000 
 (ms–10.057 0.045 0.049 
 (ms–1–0.002 0.002 0.080 
 (ms–10.058 0.044 0.055 
 (ms–10.000 0.000 0.022 
 (ms–10.016 0.013 0.015 
 MeanMedianStandard deviationSample size
Compass correction 
 (deg) 2.59 1.68 3.47 836 
Spatial data referencing 
 (m) 0.021 0.016 0.018 13,543 
 (ms–1–0.001 –0.001 0.075 
 (ms–10.047 0.028 0.058 
Kriging cross validation 
 (ms–1–0.001 –0.001 0.075 1,000 
 (ms–10.057 0.045 0.049 
 (ms–1–0.002 0.002 0.080 
 (ms–10.058 0.044 0.055 
 (ms–10.000 0.000 0.022 
 (ms–10.016 0.013 0.015 
Table 2

Effects of the suggested ADCP data correction techniques on the water velocity magnitude and the area-weighted vorticity ; p-values <0.05 show statistically significant effects of the respective techniques on

Section/plane (ms–1)
ΓABSATOT (s–1)
MinMaxMeanStandard deviationSample sizep-value
All corrections applied 
 Cross 0.012 0.917 0.130 0.116 1,133 – 0.061 
0.019 0.487 0.285 0.107 526 – 0.052 
0.006 0.543 0.203 0.159 433 – 0.034 
 Horizontal at depth (m) 0.35 0.003 0.938 0.217 0.149 8,104 – 0.066 
1.10 0.003 0.598 0.195 0.129 6,030 – 0.073 
No compass correction (all other corrections applied) 
 Cross 0.012 0.870 0.130 0.114 1,133 0.000 0.061 
0.020 0.488 0.287 0.107 526 0.000 0.051 
0.006 0.543 0.200 0.160 433 0.002 0.034 
 Horizontal at depth (m) 0.35 0.002 0.904 0.217 0.150 8,104 0.000 0.066 
1.10 0.003 0.611 0.195 0.129 6,030 0.000 0.073 
No BT replacement (all other corrections applied) 
 Cross 0.013 0.799 0.131 0.111 1,133 0.428 0.061 
0.018 0.523 0.291 0.113 526 0.007 0.068 
0.004 0.499 0.198 0.156 433 0.000 0.032 
 Horizontal depth (m) 0.35 0.002 0.881 0.211 0.144 8,104 0.000 0.066 
1.10 0.007 0.592 0.193 0.126 6,030 0.243 0.073 
Conventional 3D velocity estimation instead of Vermeulen et al. (2014) (all other corrections applied) 
 Cross 0.003 0.607 0.119 0.093 1,133 0.000 0.057 
0.065 0.574 0.289 0.112 526 0.050 0.057 
0.008 0.444 0.203 0.156 433 0.808 0.027 
 Horizontal at depth (m) 0.35 0.003 0.864 0.219 0.146 8,104 0.000 0.061 
1.10 0.003 0.825 0.200 0.135 6,030 0.000 0.067 
No corrections applied 
 Cross 0.009 0.716 0.122 0.097 1,133 0.001 0.061 
0.027 0.584 0.301 0.109 526 0.000 0.079 
0.010 0.466 0.204 0.159 433 0.001 0.029 
 Horizontal at depth (m) 0.35 0.003 0.868 0.219 0.148 8,104 0.019 0.067 
1.10 0.002 0.735 0.199 0.137 6,030 0.000 0.072 
Section/plane (ms–1)
ΓABSATOT (s–1)
MinMaxMeanStandard deviationSample sizep-value
All corrections applied 
 Cross 0.012 0.917 0.130 0.116 1,133 – 0.061 
0.019 0.487 0.285 0.107 526 – 0.052 
0.006 0.543 0.203 0.159 433 – 0.034 
 Horizontal at depth (m) 0.35 0.003 0.938 0.217 0.149 8,104 – 0.066 
1.10 0.003 0.598 0.195 0.129 6,030 – 0.073 
No compass correction (all other corrections applied) 
 Cross 0.012 0.870 0.130 0.114 1,133 0.000 0.061 
0.020 0.488 0.287 0.107 526 0.000 0.051 
0.006 0.543 0.200 0.160 433 0.002 0.034 
 Horizontal at depth (m) 0.35 0.002 0.904 0.217 0.150 8,104 0.000 0.066 
1.10 0.003 0.611 0.195 0.129 6,030 0.000 0.073 
No BT replacement (all other corrections applied) 
 Cross 0.013 0.799 0.131 0.111 1,133 0.428 0.061 
0.018 0.523 0.291 0.113 526 0.007 0.068 
0.004 0.499 0.198 0.156 433 0.000 0.032 
 Horizontal depth (m) 0.35 0.002 0.881 0.211 0.144 8,104 0.000 0.066 
1.10 0.007 0.592 0.193 0.126 6,030 0.243 0.073 
Conventional 3D velocity estimation instead of Vermeulen et al. (2014) (all other corrections applied) 
 Cross 0.003 0.607 0.119 0.093 1,133 0.000 0.057 
0.065 0.574 0.289 0.112 526 0.050 0.057 
0.008 0.444 0.203 0.156 433 0.808 0.027 
 Horizontal at depth (m) 0.35 0.003 0.864 0.219 0.146 8,104 0.000 0.061 
1.10 0.003 0.825 0.200 0.135 6,030 0.000 0.067 
No corrections applied 
 Cross 0.009 0.716 0.122 0.097 1,133 0.001 0.061 
0.027 0.584 0.301 0.109 526 0.000 0.079 
0.010 0.466 0.204 0.159 433 0.001 0.029 
 Horizontal at depth (m) 0.35 0.003 0.868 0.219 0.148 8,104 0.019 0.067 
1.10 0.002 0.735 0.199 0.137 6,030 0.000 0.072 
Figure 5

(a) Spatial distribution and magnitude of the detected absolute ADCP compass error ; the dash-point line denotes the extent of the study area. (b) Statistical distribution of (n = 836).

Figure 5

(a) Spatial distribution and magnitude of the detected absolute ADCP compass error ; the dash-point line denotes the extent of the study area. (b) Statistical distribution of (n = 836).

Close Modal

or Create an Account

Close Modal
Close Modal