Large-eddy Model Research Articles

APPROXIMATION OF THE LARGER EDDIES IN FLUID MOTIONS II: A MODEL FOR SPACE-FILTERED FLOW

We present two modifications of continuum models used in large eddy simulation. The first modification is a closure approximation which better attenuates small eddies. The second modification is a change in the boundary conditions for the large eddy model from strict adherence to slip with resistance. For model consistency, the resistance coefficient is a function of the averaging radius so that the model's boundary conditions reduce to a no-slip as the averaging radius decreases to zero.

Numerical Simulation of the Lock-In Effect of a Fixed-Fixed Tube in Cross Flow

This paper deals with the development of a numerical calculation code that is able to simulate the three-dimensional flow through a heat exchanger tube bundle and therefore allows a coupled calculation of fluid-structure interaction between the flow and the tube bundle. The incompressible flow field is calculated by a Navier-Stokes solver using a first-order power law scheme, a SIMPLEC algorithm to calculate the pressure and velocity correction fields, and a line-by-line Gauss-Seidl tridiagonal algorithm to solve the linearised system of equations. The transient parts of the Navier-Stokes equations are discretised by a second-order forward finite differencing scheme. The turbulence is examined with the aid of a large-eddy turbulence model. The transient fluid forces acting on the tubes are calculated by integration of all local flow pressure values on the surfaces of the tubes. As an example a single fixed-fixed cylinder in a flow channel is considered using the structural calculation part already developed as well as the new flow field and flow forces subroutines. The time series of the tube's motion and the fluid forces acting on the tube are analysed by Fourier's transformation. The lock-in effect occurring when the vortex shedding frequency approaches the first natural frequency of the tube can be excellently demonstrated by varying the inflow velocity over a wide range of Reynolds numbers.

First radar detection of a misocyclone in a UK tornadic storm

fully developed to an underlying urban area. Laboratory experiments would enable more controlled measurements that would complement such an observational programme. Many urban areas are too small for the boundary layer to adjust fully, so that it is important to investigate the transition from rural terrain to the urban area. Here large-eddy modelling is the crucial tool, perhaps starting by investigating how the boundary layer responds to simple changes in surface properties. New theoretical concepts are required if the urban areas are to be parametrized effectively both in the largeeddy modelling studies and NWP. Here the expertise amongst the UK community developed in studying boundary-layer flow over hills and through vegetation canopies might well provide clues. Methods for representing the average effects of vegetation on the boundary layer as 'canopies' are now reasonably well developed, and a similar approach to representing buildings within urban areas might prove fruitful. Results from these theoretical studies could suggest further large-eddy model simulations and thence further observational studies. These studies of mixing and transport in urban areas will provide the basis for studying the complex chemical transformations that control pollutants from urban sources. Clearly such an integrated programme of work requires careful co-ordination and planning. UWERN is the ideal vehicle to perform this task.

Sensitivity to the representation of precipitating ice in CRM simulations of deep convection

A Cloud resolving model (CRM) is a useful tool for providing a proxy for observed data, against which parameterizations of convection in global and regional models can be compared. The parameterization of microphysics in CRMs has been shown to be crucial for the simulation of the evolution of heat and moisture profiles during deep convection. The aim of the study described herein is to validate the microphysics scheme in the U.K. Met. Office's Large Eddy Model as a prelude to future work where simulations of deep convection will be compared to a single column version of the Met. Office's Unified model. Three observed cases of convection are simulated each with different precipitation production mechanisms, as deduced by the multiparameter radar at Chilbolton. Four simulations are carried out of each case, using different representations of precipitation. The aim is to determine whether a flexible microphysics scheme is capable of modelling the evolution of the precipitation in each of the cases without recourse to tuning coefficients for case-specific conditions. The simulations are validated against radar observations by comparing spatial distribution of radar reflectivity and the type of precipitation at the melting layer. A number of systematic errors occur in simulations using a `single-moment' microphysics scheme (where graupel and snow mass concentrations are each represented with one variable and the number concentrations are prescribed). Whereas a `double-moment' microphysics scheme (that predicts both the mass and number concentrations of snow and graupel) produced simulations consistent with radar observations for all three cases. The sensitivity of important quantities relating to the parameterization of convection in GCMs to the microphysical parameterization has been examined. The total precipitation at the ground and the amount of cloud ice are compared between simulations that differ only in their representation of precipitation. The total precipitation is found to vary by up to 40% and the total cloud ice by up to 200% between simulations of the same case.

Improvement of direct contact condensation model of RELAP5/MOD3.1 for passive high-pressure injection system

A simple set of the transition criterion of the condensation regimes and the heat transfer coefficients on the direct contact condensation in the core makeup tank was developed, and implemented in RELAP5/MOD3.1. The condensatin regimes were divided into two ones: supply limit and condensation limit. In modeling the transition criterion between two regimes, a large-eddy model developed by Theofanous was used. The modified code better predicted the experiments on the core makeup tank using small scale test facility than the original code did.

Large-Eddy Simulation of Contrails

Numerical simulations of contrails have been performed to investigate the role of various external parameters and physical processes in the life cycle of contrails. The general idea underlying the model is that of a large-eddy model. The model explicitly represents the large-scale three-dimensional motions (10-m grid resolution), while small-scale turbulence is parameterized; it contains a detailed microphysical model and it takes into account infrared radiative cooling in cloudy conditions and the vertical shear of the ambient flow. The model is applied to conditions typical for those under which contrails could be observed, that is, in an atmosphere which is supersaturated with respect to ice and at a temperature of 220 K. The simulations begin in the late dispersion phase (i.e., about 103 s after exhaust) and trace the evolution of the contrails for a half-hour period. Coherent structures can be identified within these clouds with vertical velocity fluctuations of the order of 0.1 m s−1 that are generated mainly by buoyancy due to latent heat release. In addition, the sensitivity runs undertaken as a test of the model to a change in significant physical processes or external parameters indicate that the contrail evolution is controled primarily by humidity, temperature, and static stability of the ambient air and secondarily by the baroclinicity of the atmosphere. Moreover, it turns out that the initial ice particle concentration and radiative processes are of minor importance in the evolution of contrails, at least during the 30-min simulation period.

Evaluation of parametrization schemes for the convective boundary layer using large-eddy simulation results

A large-eddy model has been used to simulate the dry atmospheric boundary layer across the range of stabilities between free convection and neutral shear flow. Results for the similarity coefficients, Ai, Bi and Ci are presented. The results are also used to evaluate the performance of a local mixing length model, and of two modified exchange coefficient models which attempt to incorporate some non-local effects.

Dependence of a Plume Heat Budget upon Lateral Advection

The structure and heat budget of a plume created by a linear heat source is investigated, demonstrating their dependence upon the larger-scale cross-source flow. A large-eddy model simulates individual thermal eddies produced by a source 200 m wide having an average surface temperature flux of 0.20 K m s−1, parameters appropriate to an Arctic lead—that is, to a break in the polar ice field. When the large-scale cross-flow is relatively weak, maximum turbulent development occurs over the source itself and the mean updraft is relatively strong. Thermally induced lateral inflow then concentrates the vertical heat transport over the source, and vertical advection accounts for over half the total heat transfer. Horizontal turbulent heat transport is important throughout the mean updraft. Turbulent vertical heat flux dominates the heat budget only in the near-surface region. With increasing externally forced cross-flow, mean and turbulent vertical velocities are weakened. Transport mechanisms are significantly altered when the cross-flow is sufficiently strong to move the region of maximum turbulence beyond the source. Turbulent vertical heat flux then becomes important through the entire plume depth, and its heat transfer becomes larger than that due to mean vertical transport. Maximum turbulent development occurs horizontally where the parcel transit time from the source centerline is approximately half the Brunt–Väisälä period of the surrounding environment. When this occurs over the source, its position coincides with that of the mean updraft. A criterion for strong updraft development, and consequent large vertical advective transport, is for the transit time of a parcel over the source to be less than the Brunt–Väisälä period.

The non-Gaussian statistics of the velocity field in low-resolution large-eddy simulations of homogeneous turbulence

A low-resolution (643) large-eddy model of forced homogeneous turbulence is numerically simulated using Kraichnan's eddy viscosity. The introduction of a reliable statistical estimate of the ζp exponents allows one to perform a detailed statistical analysis of the velocity field and shows that the probability distribution functions, the structure functions and the power-law exponents ζp agree with previous numerical and experimental results obtained at much higher effective resolution. This result shows how a simple modelling of the energy transfer produces self-similar dynamics extending to the small scales and obtains the right statistical properties of the velocity field.

Direct and large-eddy simulations of turbulence in fluids

As a result of the increasing power of supercomputers numerical simulation of turbulent flows has become feasible. In the present paper we give a short review of the requirements for such simulations. First we discuss so-called direct numerical simulation (DNS), where the equations of motion for a turbulent flow are solved in all detail. This application is illustrated with two examples, viz. the transition from laminar to turbulence in a differential heated cavity and a fully developed turbulent pipe flow. The second application of turbulence simulation is large-eddy modelling where only the large scales of turbulence are numerically resolved and the small scales are parameterized by a turbulence model. As example for this case we discuss the atmospheric boundary layer and in particular the dispersion of pollutants by atmospheric turbulence. We close our contribution with a discussion of the computer requirements necessary for turbulence simulation together with an outlook in the future in which we consider the pro's and contra's of massively parallel systems.

Turbulent transport from an arctic lead: A large-eddy simulation

The upward transfer of heat from ocean to atmosphere is examined for an Arctic “lead”, a break in the Arctic ice which allows contact between the cold atmosphere and the relatively warm ocean. We employ a large-eddy model to compute explicitly the three-dimensional turbulent response of the atmosphere to a lead of 200 m width. The surface heat flux creates a turbulent “plume” of individual quasi-random eddies, not a continuous updraft, which penetrate into the stable atmosphere and transport heat upward.

A large-eddy simulation of a line source in a convective atmospheric boundary layer—II. Dynamics of a buoyant line source

We consider a large-eddy simulation of a buoyant line source in a convective boundary layer. The motion of the line source as a function of time is influenced by two contributions: internal buoyancy (plume rise) and ambient turbulence. The advantage of large-eddy modelling with respect to laboratory and atmospheric experiments is that the simulations allow us to distinguish between these two contributions. For the part due to internal buoyancy we formulate an integral model for plume rise. A new feature in this model is the fact that plume looping, which is characteristic for plume dispersion in a convective boundary layer, promotes entrainment and therefore reduces plume rise. The contribution by ambient turbulence to plume motion is modelled in terms of the standard deviation of vertical velocity fluctuations. The results of these models are consistent with our large-eddy simulations.

The Behavior of Passive and Buoyant Plumes in a Convective Boundary Layer, as Simulated with a Large-Eddy Model

Abstract With a large-eddy model the large-scale flow structure of the convective boundary layer is simulated in a box of (5 × 5 × 2) km. The calculation is run till the turbulence has reached a quasi-steady state. At that time we introduce a line-source of contaminants and the calculation is continued with an additional equation for the concentration. We consider both passive and buoyant sources. The latter are simulated by increasing the temperature of the line-source with respect to the ambient temperature. We present data for two dimensionless release heights: zs/h = 0.15 and zs/h = 0.48. For a passive source our results agree well with the results of Willis and Deardorff and the results of the CONDORS experiments. With respect to a buoyant source we found that the influence of buoyancy on the plume parameters can be described in terms of the dimensionless buoyancy parameter F*; this conclusion is based on runs for F* = 0.01 and F* = 0.02. The simulations for buoyant plumes are compared with the labor...

Air‐water gas transfer in open channels

Laboratory flume experiments on waterside‐controlled gas transfer, such as reaeration, are described. The experiments are well described by Danckwerts' surface renewal (1951) model, as originally proposed by O'Connor and Dobbins (1958), for an open channel flow. The primary cause of surface renewal in flumes is proposed to be the upwelling of large streamwise vortices and is documented by relating prior velocity measurements of these vortices and observations of surface upwellings to gas transfer measurements. A dimensionless liquid film coefficient (Stanton number) is related to a shear Peclet number and a shear Reynolds number. The shear Reynolds number represents the effectiveness of the large streamwise vortices in penetrating the water surface. The proposed hypothesis is essentially a large eddy model, except that coherent structures in the flow have been identified as the predominant mechanism of surface renewal, rather than random turbulence. The coherent structures of importance to surface renewal are the large streamwise vortices with a size similar to the depth that occurs in an open channel flow. Thus a streamwise vortex model of gas transfer in open channels is developed. The implications of this model on the prediction of gas transfer in open channels is discussed.

Propagation Velocity of Premixed Turbulent Flames

Abstract The problem of propagation of turbulent premixed flame is analyzed using the field equation introduced recently by Kerstein, Ashurst and Williams (1987). The dynamic renormalization group method is applied to this equation and the formula for the turbulent flame velocity is derived in the lowest order in the e-expansion. The formula, which does not include adjustable parameters, agrees well with experimental (Abdel-Gayed et al., 1984) and numerical (Ashurst & Barr 1983) results on flame propagation in high-Reynolds number turbulent media. Ways to design transport and large-eddy (sub-grid) models for simulation of combustion processes, based on the ideas developed in the present paper, are discussed.

A large eddy simulation of buoyant and non-buoyant plume dispersion in the atmospheric boundary layer

The dispersion of contaminants in the convective atmospheric boundary layer is calculated with the aid of a large eddy model. Our initial condition is a model simulation of the quasi-stationary convective boundary layer in a horizontal calculation domain of 5 × 5 km. At t = 0 we introduce in this boundary layer an instantaneous line source of contaminants. Subsequently, we calculate the evolution of the line source by integrating the conservation equation of the contaminant simultaneously with the other governing equations of our large eddy model. We consider both a source of passive material with release heights of z s / h = 0.04 and 0.49 and also a source of buoyant material with a dimensionless buoyancy parameter F ∗ = 0.024 and with release heights of z g/ h = 0.16 and 0.49. The buoyant line source is constructed by increasing, at t = 0, the temperature within the source with respect to the mean ambient temperature. When we interpret our instantaneous line-source results in terms of a continuous point source, we may compare our data with the well-established laboratory results of Willis and Deardorff ( Q. Jl R. Met. Soc. 102, 427–445, 1976a; Atmospheric Environment 15, 109–117,1981; 17, 2933-2447, 1983). The agreement with these experimental data is good for the case of the passive source but not as clear for the buoyant source. In the latter case the effects of plume rise are clearly noticeable in our calculation results but they do not seem to obey the standard plume-rise equations.

The Decay of Convective Turbulence

Abstract Using simulations with a large-eddy model we have studied the decay of convective turbulence in the atmospheric boundary layer when the upward surface sensible heat flux is suddenly stopped. The decay of turbulent kinetic energy and temperature variance scales with the dimensionless time tw*/h. The temperature fluctuations start to decrease almost immediately after the forcing has been removed, whereas the turbulent kinetic energy stays constant for a time t ≈ h/w*. Vertical velocity fluctuations decay faster than horizontal fluctuations. Entrainment persists well into the decay process and may explain departures from similarity. Some evidence suggests a decoupling of large and small scales during the decay.