• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.14 no.1
• /
• pp.131-141
• /
• 1994

Three characteristics-based split-operator methods were applied to a longitudinal pollutant dispersion problem, and the results were compared with those of several Eulerian schemes. The split-operator methods consisted of generalized upwind, two-point fourth-order and sixth-order Holly-Preissmann schemes, respectively, for the advection calculation, and the Crank-Nicholson scheme for the diffusion calculation. Compared with the Eulerian schemes tested, split-operator methods using the Holly-Preissmann schemes gave much more accurate computational results. Eulerian schemes using centered difference approximations for the advection term resulted in numerical oscillations, and those using backward difference resulted in numerical diffusion, both of which were more severe for smaller value of the longitudinal dispersion coefficient.

• Journal of the Korean earth science society
• /
• v.39 no.4
• /
• pp.317-326
• /
• 2018

Effect of the nonuniform grid on the two-dimensional transport equation was investigated in terms of theoretical analysis and finite difference method (FDM). The nonuniform grid having a typical structure of the numerical weather forecast model was incorporated in the vertical direction, while the uniform grid was used in the zonal direction. The staggered and non-staggered grid were placed in the vertical and zonal direction, respectively. Time stepping was performed with the third-order Runge Kutta scheme. An error analysis of the spatial discretization on the nonuniform grid was carried out, which indicated that the combined effect of the nonuniform grid and advection velocity produced either numerical diffusion or numerical adverse-diffusion. An analytic function is used for the quantitative evaluation of the errors associated with the discretized transport equation. Numerical experiments with the non-uniformity of vertical grid were found to support the analysis.

• Journal of Korea Water Resources Association
• /
• v.45 no.5
• /
• pp.473-480
• /
• 2012

A two-dimensional depth-integrated hydrodynamic and a depth-averaged passive scalar transport models were developed by using a Compact Finite Volume Method (CFVM) which can assure a higher order accuracy. A typical wave current interaction experimental data set was compared with the computed results by the proposed CFVM model, and resonable agreements were observed from the comparisons. One and two dimensional scalar advection tests were conducted, and very close agreements were observed with very little numerical diffusion. Finally, a turbulent mixing simulation was done in an open channel flow, and a reasonable similarity with LES data was observed.

• Journal of Environmental Science International
• /
• v.19 no.3
• /
• pp.269-279
• /
• 2010

The relative contributions of physical and chemical processes to the production of ozone ($O_3$) were evaluated based on an integrated process rate (IPR) analysis using the MM5/CMAQ in a downtown (i.e., Yangsan_U) and suburban area (i.e., Ungsang) on high ozone days during spring and summer in 2006 (28 April and 8 August 2006). The IPR includes a horizontal advection (HADV) and diffusion (HDIF), a vertical advection (ZADV) and diffusion (VDlF), a dry deposition (DDEP), and a chemistry (CHEM). The VDIF in Yangsan_U was found to be the most dominant contributor (29.5% in spring and 32.1% in summer) to high $O_3$ concentrations, followed by the HADV and ZADV. In contrast, the contributions of the HADV (40.3% in spring and 32.3% in summer) in Ungsang were significantly higher than those of VDIF and ZADV. Moreover, $O_3$ production due to the chemical effect in the two areas (especially in Ungsang) during summer was found to be moderately higher than that during spring.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.10 no.1
• /
• pp.37-44
• /
• 1998

In this study a new hybrid method is developed for solving flow-dominated transport problems accurately and effectively. The method takes the forward-tracking particle method for advection. However, differently from the random-walk Lagrangian approach it solves the diffusion process on the fixed Eulerian grids. Therefore, neither any interpolating algorithm nor a large enough number of particles is required. The method was successfully examined for both cases of instantaneous and continuous sources released at a point. Comparison with a surrounding 5-point Hermite polynomial method (Eulerian-Lagrangian method) and the random-walk pure Lagrangian method shows that the present method is superior in result accuracy and time-saving ability.

• Journal of Ocean Engineering and Technology
• /
• v.32 no.6
• /
• pp.474-484
• /
• 2018

In this study, a new numerical modeling system was proposed to predict oil spills, which increasingly occur at sea as a result of abnormal weather conditions such as global warming. The hydrodynamic conditions such as the flow velocity needed to calculate oil dispersion were estimated using a three dimensional hydrodynamic model based on the Navier-Stokes equation, which considered all of the physical variations in the vertical direction. This improved the accuracy compared to those estimated by the conventional shallow water equation. The advection-diffusion model for the spilled oil was combined with the hydrodynamic model to predict the movement and fate of the oil. The effects of absorption, weathering, and wind were also considered in the calculation process. The combined model developed in this study was then applied to various test cases to identify the characteristics of oil dispersion over time. It is expected that the developed model will help to establish initial response and disaster prevention plans in the event of a nearshore oil spill.

• Atmosphere
• /
• v.14 no.4
• /
• pp.3-18
• /
• 2004

Neglecting the vertical transport from the surface, most of the previous studies on the long-range transport of pollutants have only considered the horizontal transport caused by the free atmosphere wind. I used a three dimensional numerical model, MM5 (The fifth generation Penn State Univ./NCAR Mesoscale Model) for the simulation of vertical transport of pollutants and investigated the mechanism of the vertical transport of atmospheric pollutants between planetary boundary layer(PBL) and free atmosphere by fronts. From the three dimensional simulation of MM5, the amount of pollutants transport from PBL to free atmosphere is 48% within 18 hour after the development of front, 55% within 24 hour, and 53% within 30 hour. The ratios of the vertically transported pollutant for different seasons are 62%, 60%, 54%, and 43% for spring, summer, fall, and winter, respectively. The most active areas for the vertical transport are the center of low pressure and the warm sector located east side of cold front, in which the strong upward motion slanted northward occurs. The horizontal advection of pollutants at the upper level is stronger than at the lower level simply because of the stronger wind speed. The simulation results shows the well known plum shape distribution of pollutants. The high concentration area is located in the center and north of the low pressure system, while the second highest concentration area is in the warm sector. It is shown that the most important mechanism for the vertical transport is vertical advection, while the vertical diffusion process plays an important role in the redistribution of pollutants in the PBL.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.2
• /
• pp.495-506
• /
• 2013

This study developed the Time-split Mixing Model (TMM) which can represent the pollutant mixing process on a three-dimensional open channel through constructing the conceptual model based on Taylor's assumption (1954) that the shear flow dispersion is the result of combination of shear advection and diffusion by turbulence. The developed model splits the 2-D mixing process into longitudinal mixing and transverse mixing, and it represents the 2-D advection-dispersion by the repetitive calculation of concentration separation by the vertical non-uniformity of flow velocity and then vertical mixing by turbulent diffusion sequentially. The simulation results indicated that the proposed model explains the effect of concentration overlapping by boundary walls, and the simulated concentration was in good agreement with the analytical solution of the 2-D advection-dispersion equation in Taylor period (Chatwin, 1970). The proposed model could explain the correlation between hydraulic factors and the dispersion coefficient to provide the physical insight about the dispersion behavior. The longitudinal dispersion coefficient calculated by the TMM varied with the mixing time unlike the constant value suggested by Elder (1959), whereas the transverse dispersion coefficient was similar with the coefficient evaluated by experiments of Sayre and Chang (1968), Fischer et al. (1979).

• 한국가시화정보학회:학술대회논문집
• /
• 2005.12a
• /
• pp.99-102
• /
• 2005

This study was conducted for obtaining the optimized data to build the mixer or micro fluid device as measuring the three dimensions flow field in micro mixer. To acquire the rapid diffusion on the region of low Reynolds (Re < 100), the staggered herringbone mixer using chaotic advection was selected in this case. At first, by conducting the numerical analytical virtual experiment using CFD-ACE+, three dimensions flow field in the micro mixer was estimated As this flow field was proven using defocusing particle tracing method, the behavior of micro flow with three dimensional aspects could be analyzed. Numerical analysis and flow pattern in the micro mixer by experimental verification made to be able to analyze the chaotic advection. These can be important sources for building more optimized form. Verifying the information of three dimensional flow structure, these information can be used as the data for developing and improving the $\mu$ -TAS.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.9
• /
• pp.1163-1171
• /
• 1999

Superdiffusive transport motions of passive scalars are numerically considered for various advection velocity fields. Calculated exponents ${\alpha}$ in the superdiffusion-defining relation ${\sigma}^2(t){\sim}t^{\alpha}$ for model flow fields agree to the theoretically predicted values. Simulation results show that the superdiffusion takes place as the tracers' motion become less random, compared to their motion at the pure molecular diffusion. Whether the flow field is random or not, degrees of superdiffusion are directly related to the velocity autocorrelation functions along the tracers Lagrangian trajectories that characterize degrees of randomness of the tracers' motion.