Validation of an Uncertainty Propagation Method for Moving‐Boat Acoustic Doppler Current Profiler Discharge Measurements

Abstract

Introduction Moving-Boat ADCP UncertaintyAcoustic Doppler Current Profilers (ADCPs) are now among the most-used instruments for measuring discharge in rivers throughout the world (Boldt & Oberg, 2015;. The technology and general guidance for making ADCP discharge measurements are presented in various manuals and guides, for example, those established by the U.S. Geological Survey (USGS) (Mueller et al., 2013) or the WMO (2010). The ADCP is mounted on a boat or on a small float that transects a river cross-section. The field procedure can be a stationary deployment (section by section) or a moving-boat deployment, wherein each river crossing provides an individual discharge measurement called a transect. This study deals with moving-boat ADCP discharge measurements. The ADCP uses the transit-time of sound waves and the Doppler shift to measure water velocity and depth. Due to physical limitations of the instrument, velocities are measured throughout a limited portion of the cross-section (Mueller et al., 2013). The measured area is decomposed vertically into cells (or bins) that are distributed horizontally into n ensembles (see Figure 1). The discharge Q i,j through a cell is computed from the vector product of the boat velocity 𝐴𝐴 ⃖⃖ ⃗ 𝑣𝑣𝑖𝑖 and the water velocity 𝐴𝐴 ⃖⃖⃖⃖⃖ ⃗ 𝑤𝑤𝑖𝑖𝑖𝑖𝑖 relative to the ADCP system as:where 𝐴𝐴 ⃖ ⃗ 𝑘𝑘 is the unit vertical vector, dz i,j is the cell height, and dt i is the time interval between ensembles.In the unmeasured areas (see Figure 1), discharge has to be estimated (Mueller, 2016). The missing or invalid cells and ensembles are interpolated based on contiguous valid data. The discharge is extrapolated near the riverbed (bottom discharge), near the water surface (top discharge), and near the banks (right and left discharges). Thus, the total discharge Q k in each transect k is a sum of partial discharges: the measured discharge (the sum of individual cell discharges Q i,j ), the interpolated discharges used to fill missing cells and ensembles, respectively

Uncertainties in measuring and estimating water‐budget components: Current state of the science

“…The greatest contributions to uncertainty are often at channel edges, where velocity estimates are unavailable and extrapolation is required. ADCP-derived streamflow discharges agree within 5% error when compared with values determined with mechanical current meters and stage-discharge relations (Morlock, 1995) and agree within 5%-6% error when compared with dam outfall discharges under good conditions (Despax et al, 2017;Le Coz et al, 2008;Le Coz et al, 2014;Moore et al, 2017). The uncertainty of computed stream discharges collected with ADCPs is determined by the number of transects and the length of observation time (Oberg & Mueller, 2007).…”

Datasets of high-resolution water level and discharge from the Saigon-Dong Nai estuary system impacted by a developing megacity, Ho Chi Minh City - Vietnam

“…The OURSIN method also accounts for systematic versus random errors in the computation of the uncertainty of discharge measurements averaged over repeated ADCP transects. Then, the decomposition of the different error sources allows determining their influence on the overall uncertainty [4] . In addition, the error in ADCP measurements was evaluated at 10% using a minimum error value of 100 m 3 /s given the adverse conditions encountered during the campaigns [ 5 , 6 ].…”

Quantifying and Reducing the Operator Effect in LSPIV Discharge Measurements

“…The discharge reference come from ADCP measurements for videos 1, 2, 3, 5, 6, and 7 and from a dilution measurement for video 8. The 95% uncertainty of the reference U ref is computed from the measurement data using the Oursin method (Despax et al., 2023) for ADCP measurement and the Suny method (Hauet et al., 2020) for the dilution method. For video 5, the wind acting on the surface impacts the vertical velocity profile and thus the top layer extrapolation of the ADCP discharge measurement.…”

“…All these limitations have to be studied to ensure the reliability of the measurement, necessary to the operational deployment of the technique. For instance, the operational deployment of the ADCP followed several steps with first the validation of the method against precise references in various laboratory and field conditions (Oberg & Mueller, 2007), then the quantification of the measurement uncertainties through specific computation frameworks (Despax et al., 2023; González‐Castro & Muste, 2007; Moore et al., 2017) supplemented by repeated‐measures experiments (a.k.a. intercomparisons) (Despax et al., 2019; Le Coz et al., 2016).…”