• Flow Discharge Measurements Using Small Unmanned Aerial Systems

      Duan, Jennifer G.; Cadogan, Ammon F.; Engel, Frank L.; Lansey, Kevin E.; Ferré, P.A. T. (The University of Arizona., 2021)
      The accurate measurement of river discharge during flooding events has long been a challenging and dangerous task in the Southwestern United States where flows are often flashy and have a high sediment concentration. Small unmanned aerial systems (sUAS) can be deployed to access unsafe field sites and capture flow measurements remotely. Using the footage collected with the sUASs along with Large Scale Image Velocimetry techniques, a detailed surface velocity distribution can be obtained. Therefore, the use of sUASs for obtaining flow discharge measurements has become increasingly common in recent years. However, the measured velocity distribution at water surface must be transformed into the depth averaged velocity distribution in order to obtain the overall river discharge. It is common practice to multiply the surface velocity by a surface velocity correction coefficient (also referred to as the velocity index, α, in this thesis) to obtain the depth averaged velocity. As one of the central components in a discharge measurement is the depth averaged velocity, the quality of discharge calculation is directly correlated with the accuracy of the velocity index. Until now, most approaches for selecting the velocity index in field applications are qualitative and empirical based upon channel characteristics (such as depth, bed type, etc.). This research aims to improve the estimation of the velocity index by incorporating turbulence properties (e.g., energy dissipation rate) measured at the surface, channel roughness, and channel geometric properties. Surface velocity measurements were collected from sUASs from eight field sites (including concrete, earthen, and natural channels) near established United States Geological Survey (USGS) streamgaging stations. The velocity index was calculated for each of the sites by estimating the turbulence dissipation rate at the surface from the instantaneous velocity distributions. To determine the accuracy of the method, discharge estimates were compared against the flow measurements collected by USGS personnel. Errors in discharge obtained from calculating the velocity index ranged from 4% to 91%. While the turbulence dissipation rate method of estimating the velocity index introduces many sources of error into the velocity index estimates, it still has potential for calculating accurate discharge values when seeding conditions are ideal and discharge is high. While limited in scope, this research shows that under ideal seeding and flow conditions, discharge can be calculated from surface flow velocities using the velocity index estimates from the turbulence dissipation rate method.