• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.103-103
• /
• 2015

유류유출사고와 같이 하천 수표면의 흐름에 따라 이동 확산하는 부유성 오염물질의 혼합해석을 위해 많은 연구자들은 입자추적모형을 사용한 혼합모의를 수행해왔다. 입자추적모형에서 오염물질의 혼합은 평균 유속 분포에 의한 결정론적인 이동과 난류유동에 의한 무작위적인 혼합으로 나타내며, 난류혼합에 의한 수평확산은 난류확산계수로 조절한다. 따라서 표면흐름에 의한 난류확산계수의 산정을 위해 많은 연구자들은 부유성 입자를 이용한 실내실험을 수행하여 수평확산계수를 산정했고(Engelund, 1969; Cederwall, 1971), 최근에는 GPS의 발전으로 인해 해양영역에서 GPS를 장착한 표면부자를 활용한 확산실험을 통해 수평확산계수를 산정한 바 있다(Kjellson and $D{\ddot{o}}{\ddot{o}}s$, 2012; Alpers 등, 2013). 하천수질오염사고의 약 43.5%가 유류유출에 의한 것이며(환경부, 2013), 이에 따라 표면흐름에 의한 오염물질 혼합해석이 필요하나, 하천에서 수평확산계수 산정을 위한 현장실험연구는 부족한 상황이다. 따라서 본 연구에서는 낙동강 본류에서 GPS부자를 이용한 입자추적실험을 수행하여 표면흐름에 의한 확산계수를 산정했다. GPS부자를 이용한 입자추적실험은 낙동강의 강정고령보 하류와 구미보 하류의 각각 세 지점에서 수행되었다. GPS부자는 바람에 의한 교란을 최소화하기 위해 지름 10 cm의 구형으로 제작하였으며 시범테스트를 통해 입자의 주 궤적 변화가 크지 않은 지점에 GPS부자를 투입했다. GPS부자는 오염물질의 사고유출을 가정하여 한 지점에 투입했고 GPS부자 사이의 간섭을 최소화하기 위해 25 ~ 35개의 GPS부자를 이용했다. 표면흐름에 의해 이동하는 부자의 위치는 GPS에 시계열로 저장됐고 ADCP를 이용하여 실험당시의 수리량을 측정했다. 입자위치의 시계열자료로부터 GPS부자의 확산범위의 시간변화를 계산했고 단순 모멘트법을 이용하여 종, 횡 방향 확산계수를 계산했다. 그 결과, 종 방향 확산계수는 $0.003{\sim}0.041m^2/s$로 계산되었고 횡 방향 확산 계수는 $0.001{\sim}0.012m^2/s$로 계산되어, 흐름방향의 유속성분에 의한 확산이 지배적인 것으로 나타났다. 지류 합류부에서는 이송이 지배적인 혼합이 발생되었고(Pe>1) Pe의 증가에 따라 수평확산계수가 감소되었다. 25~35개 GPS부자 궤적의 앙상블 평균으로부터 계산한 Integral time scale은 모멘트법으로부터 계산한 종, 횡 방향 확산계수와 비례하는 것으로 나타나, Taylor(1921)의 이론과 일치했다. 또한 실험수로에서 수행된 기존연구결과와 비교한 결과, 하폭 대 수심비, 마찰항의 증가에 따라 수평확산계수가 증가하는 경향을 나타내었다.

• Water for future
• /
• v.29 no.4
• /
• pp.187-198
• /
• 1996

A numerical model for unsteady dispersion of horizontal line source in turbulent shear flow is developed. A fractional step finite difference method is used which splits the unsteady two-dimensional advective diffusion equation into the longitudinal advection and the vertical diffusion equations, and solves them alternately for half time intervals by the Holly-Preissmann scheme and the Crank-Nicholson scheme, respectively. The developed numerical model is verified using a semi-analytic solution for steady dispersion in turbulent shear flow. Dispersion of an instantaneous plane source in turbulent shear flow is analyzed using the model. The degree of mixing at the same dimensionless time is almost the same regardless of the friction factor, and the travel distance required to reach a certain degree of mixing is inversely proportional to the square root of the friction factor.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.2
• /
• pp.191-199
• /
• 1993

Diffusion of a steady horizontal line source in a turbulent shear flow is simulated by numerically solving a steady two-dimensional advective diffusion equation. The computational result is compared with the analytic solution for uniform velocity and diffusivity distributions over the depth. The analytic solution for constant velocity and diffusivity overestimates the degree of vertical mixing. The normalized equation indicates that friction factor is the only physical parameter that governs the vertical diffusion process. Sensitivities of the diffusion process to the friction factor and initial source position are analyzed. The rate of vertical mixing varies approximately as the square root of the friction factor. The optimal source position, which gives the most rapid mixing, lies above the mid-depth and moves toward the water surface as the friction factor increases.

• 한국해양학회지
• /
• v.30 no.1
• /
• pp.13-26
• /
• 1995

In this paper three turbulence models including two-equation model by Blumberg and Mellor (1987), one-equation model with mixing length formula of Blackadar's (1962), and zero-equation model of Prandtl's (1925) were compared in homogeneous, unstratified channel flows. Steady flows which a steep-sided trapezoidal trench with uniform discharge, tidal flow and steady wind-driven flow in finite channels are considered in detail. Steady flows in a trench and tidal flows in a finite channel were reproduced fairly accurately and there was virtually no difference among results of three turbulence models. However, In case of steady wind-driven flow only two-equation model reproduced the important features of experimental data. the other two models underestimated the surface velocity. In tidal and wind-driven flows with negligibly small adjective and diffusive effects, the two-equation model gives rise to parabolic profile of eddy viscosity with maximum at the mid0depth, and the one and zero equation model based on Blackadar formula linear profile with maximum at the surface.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.16 no.2
• /
• pp.92-102
• /
• 2004

In this paper, an analytical solution of calculating the excess temperature field due to a point heat source is presented in the presence of spring-neap modulation of convective alongshore flow. The basic form of the solution is identical to that given by Jung et al. (2003) but the convective term in the exponential kernel function is extended and a spring-neap variation in the horizontal eddy diffusivity is newly introduced. A set of calculations have been performed to examine the sensitivity of the heat build-up to the change in current fields and turbulent dispersion. Results indicate that the excess temperature field is confined within the tidal excursion distance, while the excess temperature field beyond the distance is mainly controlled by the horizontal diffusion. The heat build-up within the distance is considerably affected by the spring-neap variation in the horizontal eddy diffusivity; the relatively high excess temperature more than 1$^{\circ}C$ is extended further when the eddy diffusivity has spring-neap modulation.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2001.05a
• /
• pp.180-185
• /
• 2001

In order to regulate the physical characteristics of an ocean dumping material in the south-eastern East Sea, the diffusion characteristics with the observation, hydraulic experiment and numerical experiment data are investigated. The main results are as follows; (1) Spying CTD observation result of the area of Jung in the East Sea, the ocean dumping area had influenced the Tsushima warm current of high temperature and salinity. Horizontal turbulent diffusivity is 1.913${\times}$10$^{7}$ $\textrm{cm}^2$/sec by drogue tracking. (2) From the experiment of settling, the initial settling velocity of each material is 1.0∼2.7 cm/sec according to the specific gravity and initial concentration. In the pycnocline, particles didn't settle under the pycnocline any more and accumulated. It is signified that calculation of the sedimentation rate of the ocean dumping material including of vertical diffustion must be regard the pycnocline in the ocean area have well-developed pycnocline. (3) Vertical turbulent diffusivity were 2.219${\times}$10$^{-8}$∼8,874${\times}$10$^{-4}$ $\textrm{cm}^2$/sec from the experiment of settling. And, the pycnocline influenced the vertical turbulent diffusivity. (4) From the result of diffusion simulation in the East Sea, the co-concentration line of 0.05 ppm and 0.1 ppm are limited at dumping area after 200 days. The constant concentration line of 0.01 ppm is distributed to the vicinity of Ulleungdo and Tokdo, but isn't distributed to the coastal area of East Sea and southern area of Jung in the East Sea.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.4 no.2
• /
• pp.91-107
• /
• 1992

A numerical formulation is developed to solve the linear three-dimensional hydrodynamic equations which describes wind induced flows in a homogeneous shelf sea. The hydmdynamic equations are at the outset separated into two systems. namely, an equation containing the gradient of sea surface elevation and the mean flow (external mode) and an equation describing the deviation from the mean flow (internal mode). The Galerkin method is then applied to the internal mode equation. The eigenvalues are determined from the eigenvalue problem involving the vertical eddy viscosity subject to a homogeneous boundary condition at the surface and a sheared boundary condition at the sea bed. The model is tested in a one-dimensional channel with uniform depth under a steady, uniform wind. The analytical velocity profile by Cooper and Pearce (1977) using a constant vertical eddy viscosity in channels of infinite and finite length is chosen as a benchmark solution. The model is also tested in a homogeneous, rectangular basin with constant depth under a steady, uniform wind field (the Heaps' Basin of the North Sea scale).