Abstract
Suspended sediment concentrations are typically estimated using either the advection-diffusion equation or sediment rating curves in a deterministic manner. This study attempts to develop a stochastic approach for quantifying the probabilistic characteristics of sediment concentrations that can account for the uncertainty associated with flow randomness. Turbulence is a primary cause of particle diffusion in a flow. Impacts from such flow randomness on particle diffusion can be observed from two aspects: (1) the degree of spreading of a particle cloud, and (2) variability in concentration curves observed in different releases of particles under the same turbulence intensity. While the former diffusion has been extensively studied, the latter has not been fully investigated, in spite of its significance in terms of identifying the uncertainty associated with estimating concentrations. Herein, the effect of probabilistic characteristics attributed to turbulence on sediment concentrations is evaluated through multiple realizations of a Lagrangian-based stochastic differential equation for particle trajectory. Both the resuspension and deposition of particles are considered in the transport processes. Sediment concentrations can then be estimated from the spatial distribution of particles. As a result of the ensemble standard deviation, it is found that estimating higher-concentration regions is subject to a higher uncertainty. The coefficient of variation representing the extent of variability relative to their mean in lower-concentration regions is found to be more variable than that in higher-concentration regions. It is observed that when the ensemble standard deviation of the concentration is normalized by the square root of the particle number, the magnitude of the variability of concentration curves tends to approach asymptotically to one single curve.
| Original language | English |
|---|---|
| Article number | 04015031 |
| Journal | Journal of Hydraulic Engineering |
| Volume | 141 |
| Issue number | 12 |
| DOIs | |
| Publication status | Published - 2015 Dec 1 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Civil and Structural Engineering
- Water Science and Technology
- Mechanical Engineering
Cite this
}
Quantifying the uncertainty associated with estimating sediment concentrations in open channel flows using the stochastic particle tracking method. / Oh, Jungsun; Tsai, Christina W.; Choi, Sung-Uk.
In: Journal of Hydraulic Engineering, Vol. 141, No. 12, 04015031, 01.12.2015.Research output: Contribution to journal › Article
TY - JOUR
T1 - Quantifying the uncertainty associated with estimating sediment concentrations in open channel flows using the stochastic particle tracking method
AU - Oh, Jungsun
AU - Tsai, Christina W.
AU - Choi, Sung-Uk
PY - 2015/12/1
Y1 - 2015/12/1
N2 - Suspended sediment concentrations are typically estimated using either the advection-diffusion equation or sediment rating curves in a deterministic manner. This study attempts to develop a stochastic approach for quantifying the probabilistic characteristics of sediment concentrations that can account for the uncertainty associated with flow randomness. Turbulence is a primary cause of particle diffusion in a flow. Impacts from such flow randomness on particle diffusion can be observed from two aspects: (1) the degree of spreading of a particle cloud, and (2) variability in concentration curves observed in different releases of particles under the same turbulence intensity. While the former diffusion has been extensively studied, the latter has not been fully investigated, in spite of its significance in terms of identifying the uncertainty associated with estimating concentrations. Herein, the effect of probabilistic characteristics attributed to turbulence on sediment concentrations is evaluated through multiple realizations of a Lagrangian-based stochastic differential equation for particle trajectory. Both the resuspension and deposition of particles are considered in the transport processes. Sediment concentrations can then be estimated from the spatial distribution of particles. As a result of the ensemble standard deviation, it is found that estimating higher-concentration regions is subject to a higher uncertainty. The coefficient of variation representing the extent of variability relative to their mean in lower-concentration regions is found to be more variable than that in higher-concentration regions. It is observed that when the ensemble standard deviation of the concentration is normalized by the square root of the particle number, the magnitude of the variability of concentration curves tends to approach asymptotically to one single curve.
AB - Suspended sediment concentrations are typically estimated using either the advection-diffusion equation or sediment rating curves in a deterministic manner. This study attempts to develop a stochastic approach for quantifying the probabilistic characteristics of sediment concentrations that can account for the uncertainty associated with flow randomness. Turbulence is a primary cause of particle diffusion in a flow. Impacts from such flow randomness on particle diffusion can be observed from two aspects: (1) the degree of spreading of a particle cloud, and (2) variability in concentration curves observed in different releases of particles under the same turbulence intensity. While the former diffusion has been extensively studied, the latter has not been fully investigated, in spite of its significance in terms of identifying the uncertainty associated with estimating concentrations. Herein, the effect of probabilistic characteristics attributed to turbulence on sediment concentrations is evaluated through multiple realizations of a Lagrangian-based stochastic differential equation for particle trajectory. Both the resuspension and deposition of particles are considered in the transport processes. Sediment concentrations can then be estimated from the spatial distribution of particles. As a result of the ensemble standard deviation, it is found that estimating higher-concentration regions is subject to a higher uncertainty. The coefficient of variation representing the extent of variability relative to their mean in lower-concentration regions is found to be more variable than that in higher-concentration regions. It is observed that when the ensemble standard deviation of the concentration is normalized by the square root of the particle number, the magnitude of the variability of concentration curves tends to approach asymptotically to one single curve.
UR - http://www.scopus.com/inward/record.url?scp=84947782700&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84947782700&partnerID=8YFLogxK
U2 - 10.1061/(ASCE)HY.1943-7900.0001045
DO - 10.1061/(ASCE)HY.1943-7900.0001045
M3 - Article
AN - SCOPUS:84947782700
VL - 141
JO - Journal of Hydraulic Engineering
JF - Journal of Hydraulic Engineering
SN - 0733-9429
IS - 12
M1 - 04015031
ER -