River cross-section measurement using unreviewed aerial vehicle with an improved bathymetry instrument

Accurate river cross-section (RCS) data is critical for hydrological research and applications; however, in most rivers, hydrographic stations are extremely limited and RCS data are scarce. A new method for mobile mea-surements of RCS is herein proposed based on an unreviewed aerial vehicle (UAV) platform, i.e., using UAV aerial survey to obtain above-water cross-sections and a self-made UAV bathymetric instrument to attain un-derwater cross-sections. Underwater cross-sectional measurement is the focus of this paper, namely UAV ba-thymetry (UAVB). The effectiveness of this method has been verified in rivers with different widths, water depths, turbidity, and flow velocities. The results show that the UAVB cross-section is consistent with the un-derwater cross-section measured in the field via conventional surveying. The surface velocity is the main factor affecting the accuracy of UAVB, and the corresponding measurement accuracy is-4%,-8%, and-18% of the actual water depth when the surface velocity is less than 1 m/s, 1-2 m/s, and greater than 2 m/s, respectively. From the combination of hydraulic parameters obtained by UAV aerial survey and the calculated discharge results of three types of RCSs (measured cross-section, UAVB cross-section, and generalized cross-section), our results indicate that discharge calculated from the measured cross-section has the highest accuracy with a mean relative error (MRE) of-5%, followed by the UAVB cross-section with MRE of-15%, and the generalized cross-section has the worst accuracy with MRE-35%.

Automated summary

This study proposes a new method for measuring river cross-section data using unmanned aerial vehicles (UAVs). The effectiveness of this method has been verified in different rivers, showing good results. The accuracy of the measurements is affected by the surface velocity, with measurement errors of -4%, -8%, and -18% of the actual water depth for surface velocities less than 1 m/s, 1-2 m/s, and greater than 2 m/s, respectively. The results also indicate that discharge calculated from the measured cross-section has the highest accuracy, followed by the UAVB cross-section, and the generalized cross-section has the worst accuracy.