Estimating the Uncertainty of Streamgauging Techniques Using In Situ Collaborative Interlaboratory Experiments

Self Cite

Stage‐fall‐discharge (SFD) rating curves are traditionally used to compute streamflow records at sites where the energy slope of the flow is variable due to variable backwater effects. We introduce a model with hydraulically interpretable parameters for estimating SFD rating curves and their uncertainties. Conventional power functions for channel and section controls are used. The transition to a backwater‐affected channel control is computed based on a continuity condition, solved either analytically or numerically. The practical use of the method is demonstrated with two real twin‐gauge stations, the Rhône River at Valence, France, and the Guthusbekken stream at station 0003⋅0033, Norway. Those stations are typical of a channel control and a section control, respectively, when backwater‐unaffected conditions apply. The performance of the method is investigated through sensitivity analysis to prior information on controls and to observations (i.e., available gaugings) for the station of Valence. These analyses suggest that precisely identifying SFD rating curves requires adapted gauging strategy and/or informative priors. The Madeira River, one of the largest tributaries of the Amazon, provides a challenging case typical of large, flat, tropical river networks where bed roughness can also be variable in addition to slope. In this case, the difference in staff gauge reference levels must be estimated as another uncertain parameter of the SFD model. The proposed Bayesian method is a valuable alternative solution to the graphical and empirical techniques still proposed in hydrometry guidance and standards.

“…We further assume that stage errors are negligible compared to discharge errors:

true{\begin{array}{l} left & {\overset{h}{true}}_{1, i} = h_{1, i} \\ left & {\overset{h}{true}}_{2, i} = h_{2, i} \\ left & {\overset{Q}{true}}_{i} = Q_{i} + ϵ_{Q_{i}} with ϵ_{Q_{i}} \overset{indep.}{\sim} scriptN false(0, u_{Q_{i}} false) \end{array}

where the standard deviations

u_{Q_{i}}

are assumed to be known and the discharge errors

ϵ_{Q} = (ϵ_{Q, 1}, \dots, ϵ_{Q, N})

Section: Methodsmentioning

confidence: 99%

Bayesian analysis of stage‐fall‐discharge rating curves and their uncertainties

Mansanarez¹,

Coz²,

Renard³

et al. 2016

Self Cite

“…The interaction of factors is the systematic error related to the ( ij ) combination of factor levels. The linear combination of the explanatory variables (equation ) is a classic model (Kutner et al, ) which allows for uncertainty budgets in compliance with the Guide to the expression of Uncertainty in Measurement, according to ISO 21748 (ISO, ; Le Coz et al, ).

Q_{i j m} = μ + δ + α_{i} + β_{j} + γ_{i j} + ϵ_{i j m}

where μ is the true discharge such as

\overset{Q_{• • •}}{true} = μ + δ

with

\overset{Q_{• • •}}{true}

the grand mean of all single‐transect discharge measurements.…”

Section: Experiments and Methodsmentioning

confidence: 99%

“…Table ). For instance, during the same Génissiat 2010 experiments, the uncertainty of discharges measured close to the dam was substantially larger than that of discharges measured more downstream, which could be explained by a pulsating, nonfully established flow directly downstream of the dam turbine outlets (Le Coz et al, ). Although the previous experiments propose a range of uncertainty estimates, they do not allow the decomposition and the quantification of uncertainty sources due to operator, instrument, and cross‐section effects separately.…”

Section: Introductionmentioning

confidence: 99%

“…As a complementary approach to uncertainty propagation methods, repeated measures experiments, also known as interlaboratory comparisons, provide a useful tool to investigate the performance of a measurement technique and for empirically estimating the uncertainty of stream gauging techniques in given measurement conditions (Despax et al, ; Le Coz et al, ; WMO, ). It is based on repeated measurements of the same variable (the discharge) by several participants, or laboratories, using the same measurement procedure.…”

Section: Introductionmentioning

confidence: 99%

“…Since 2007, ADCP comparison experiments (or ADCP regattas) have been conducted around the world, for instance, in Germany, England, Croatia, Canada, the United States, New Zealand, Australia, South Korea, and France (Le Coz et al, ). Only a few of them, however, have been processed as repeated measures experiments to compute the uncertainty (cf.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Decomposition of Uncertainty Sources in Acoustic Doppler Current Profiler Streamflow Measurements Using Repeated Measures Experiments

Despax

Coz

Hauet

et al. 2019

Repeated measures experiments can be conducted to empirically estimate the uncertainty of a stream gauging method, such as the widespread moving‐boat acoustic Doppler current profilers (ADCPs) approach. Previous ADCP repeated measures experiments, also known as interlaboratory comparisons, provided a credible range of uncertainty estimates reflecting the quality of the site conditions. However, the method, which is a one‐way analysis of variance, only addresses the uncertainty of one lumped factor that combines several distinct factors: instrument, operator, procedure, and cross‐section effects. To decompose the uncertainty of ADCP streamflow measurements due to cross‐section selection and team effects, a large repeated measures experiment has been conducted in the Taurion River (France). The experiment design was crossed and balanced, with two sets of 24 teams circulated over two sets of 12 cross sections. A constant flow rate was released from a dam, located immediately upstream of the experimental site. Prior to the statistical analysis, a data quality review was performed using the U.S. Geological Survey QRev software to clean the data set from avoidable errors and to homogenize the discharge computations. A two‐way analysis of variance was applied to quantify the cross‐section effect, the team effect, and their interaction, which was found to dominate the pure cross‐section effect. It was then possible to predict the average uncertainty of multiple‐transect ADCP discharge measurements, depending on the number of teams, cross sections, and repeated transects included in the discharge average. The method opens interesting avenues for documenting difficult‐to‐estimate uncertainty sources of stream gauging techniques in other measuring conditions, especially the most adverse ones.

Streamflow Observations From Cameras: Large‐Scale Particle Image Velocimetry or Particle Tracking Velocimetry?

Tauro

Piscopia

Grimaldi

2017

Image‐based methodologies, such as large scale particle image velocimetry (LSPIV) and particle tracking velocimetry (PTV), have increased our ability to noninvasively conduct streamflow measurements by affording spatially distributed observations at high temporal resolution. However, progress in optical methodologies has not been paralleled by the implementation of image‐based approaches in environmental monitoring practice. We attribute this fact to the sensitivity of LSPIV, by far the most frequently adopted algorithm, to visibility conditions and to the occurrence of visible surface features. In this work, we test both LSPIV and PTV on a data set of 12 videos captured in a natural stream wherein artificial floaters are homogeneously and continuously deployed. Further, we apply both algorithms to a video of a high flow event on the Tiber River, Rome, Italy. In our application, we propose a modified PTV approach that only takes into account realistic trajectories. Based on our findings, LSPIV largely underestimates surface velocities with respect to PTV in both favorable (12 videos in a natural stream) and adverse (high flow event in the Tiber River) conditions. On the other hand, PTV is in closer agreement than LSPIV with benchmark velocities in both experimental settings. In addition, the accuracy of PTV estimations can be directly related to the transit of physical objects in the field of view, thus providing tangible data for uncertainty evaluation.