The present situation of CWE(Computational Wind Engineering) and the papers presented to the CWE 2000 Symposium are reviewed from the following viewpoints; 1) topics treated, 2) utilization of commercial code (software), 3) incompleteness of CWE, 4) remaining research subjects, 5) prediction accuracy, 6) new fields of CWE application, etc. Firstly, new tendencies within CWE applications are indicated. Next, the over-attention being given to the application field and the lack of attention to fundamental problems, including prediction error analysis, are pointed out. Lastly, the future trends of CFD (Computational Fluid Dynamics) applications to wind engineering design are discussed.

Accidental gaseous losses from industrial processes can pose considerable health and environmental risks but assessing their health, safety and environmental impact is problematic. Improved understanding and simulation of the dispersion of emissions in the vicinity of storage tanks is required. The present study aims to assess the capability of the turbulence closures and meshing alternatives in a commercially available CFD code for predicting dispersion in the vicinity of cubes and circular cylindrical storage tanks. The performance of the $k-{\varepsilon}$ and Reynolds Stress turbulence models and meshing alternatives for these cases are compared to experimental data. The CFD simulations are very good qualitatively and, in many cases, quantitatively. A mesh with prismatic elements is more accurate than a tetrahedral mesh. Overall the Reynolds stress model performs slightly better than the $k-{\varepsilon}$ model.

Both a Finite Volume and a Discrete Vortex technique to solve the unsteady Navier-Stokes equations have been employed to study the air flow around long-span bridge decks. The implementation and calibration of both methods is described alongside a quasi-3D extension added to the DVM solver. Applications to the wind engineering of bridge decks include flow simulations at different angles of attack, calculation of aerodynamic derivatives and fluid-structure interaction analyses. These are being presented and their specific features described. If a numerical method shall be employed in a practical design environment, it is judged not only by its accuracy but also by factors like versatility, computational cost and ease of use. Conclusions are drawn from the analyses to address the question of whether computer simulations can be practical design tools for the wind engineering of bridge decks.

A mesoscale meteorological model is applied to simulate turbulent airflow and eddy shedding over the Isle of Arran, SW Scotland, UK. Under conditions of NW flow, the mountain ridge of Kintyre, located upwind of Arran, induces gravity waves that also affect the airflow over the island. The possibility to nest domains allows description of the airflow over Arran with a very high resolution grid, while also including the effects of the surrounding mainland of Scotland, in particular of the mountain ridge of Kintyre. Initialised with a stably stratified NW flow, the mesoscale model simulates quasi-stationary gravity waves over the island induced by Kintyre. Embedded in the larger scale wave trains there is continuous development of small-scale transient eddies, created at the Arran hill tops, that move downstream through the stationary wave field. Although the transient eddies are more frequently simulated on the northern island where the terrain is more pronounced, they are also produced over Tighvein, a hill of 458 m on the southern island where measurements of surface pressure and 2 m meteorological variables have been recorded at intermittent intervals between 1996 and 2000. Comparison between early observations and simulations so far show qualitatively good agreement. Overall the computations demonstrate that turbulent flow can be modelled with a horizontal resolution of 70 m, and describe turbulent eddy structure on wavelength of only a few hundred metres.

This paper presents some of the results of a project whose aim has been to produce a full simulation model which would determine the efficacy of pesticides for use by both farmers and the bio-chemical industry. The work presented here describes how crop architecture can be mathematically modelled and how the mechanics of pesticide droplet capture can be simulated so that if a wind assisted droplet-trajectory model is assumed then droplet deposition patterns on crop surfaces can be predicted. This achievement, when combined with biological response models, will then enable the efficacy of pesticide use to be predicted.

Numerical simulations based on the ALE finite element method are carried out to examine the aerodynamics of an oscillating circular cylinder when the separated shear flows around the cylinder are stimulated by periodic jet excitation with a shear layer instability frequency. The excitation is applied to the flows from two points on the cylinder surface. The numerical results showed that the excitation with a shear layer instability frequency can reduce the negative damping and thereby stabilize the aerodynamics of the oscillating cylinder. The change of the lift phase seems important in stabilizing the cylinder aerodynamics. The change of lift phase is caused by the merger of the vortices induced by the periodic excitation with a shear layer instability frequency, and the vortex merging comes from the high growth rate, the rapid increase of wave number and decrease of phase velocity for the periodic excitation in the separated shear flows.

Predictions of the pedestrian level wind speeds for the downtown area of Auckland that have been obtained by wind tunnel and computational fluid dynamic (CFD) modelling are presented. The wind tunnel method involves the observation of erosion patterns as the wind speed is progressively increased. The computational solutions are mean flow calculations, which were obtained by using the finite volume code PHOENICS and the $k-{\varepsilon}$ turbulence model. The results for a variety of wind directions are compared, and it is observed that while the patterns are similar there are noticeable differences. A possible explanation for these differences arises because the tunnel prediction technique is sensitivity to gust wind speeds while the CFD method predicts mean wind speeds. It is shown that in many cases the computational model indicates high mean wind speeds near the corner of a building while the erosion patterns are consistent with eddies being shed from the edge of the building and swept downstream.

Results of measurements of surface pressure and of velocity field made on a full-scale 6 m cube in natural wind are reported. Comparisons are made with results from boundary-layer wind-tunnel studies reported in the literature. Two flow angles are reported; flow normal to a face of the cube (the $0^{\circ}$ case) and flow at $45^{\circ}$. In most comparisons, the spread of wind-tunnel results of pressure measurements spans the full-scale measurements. The exception to this is for the $0^{\circ}$ case where the roof and side-wall pressures at full-scale are more negative, and as a result of this the leeward wall pressures are also lower. The cause of this difference is postulated to be a Reynolds Number scale effect that affects flow reattachment. Measurements of velocity in the vicinity of the cube have been used to define the mean reattachment point on the roof centre line for the $0^{\circ}$ case, and the ground level reattachment point behind the cube for both $0^{\circ}$ and $45^{\circ}$ flow. Comparisons are reported with another full-scale experiment and also with wind-tunnel experiments that indicate a possible dependency on turbulence levels in the approach flow.

Computation solutions for the flow around a cube, which were generated as part of the Computational Wind Engineering 2000 Conference Competition, are compared with full-scale measurements. The three solutions shown all use the RANS approach to predict mean flow fields. The major differences appear to be related to the use of the standard $k-{\varepsilon}$, the MMK $k-{\varepsilon}$ and the RNG $k-{\varepsilon}$ turbulence models. The inlet conditions chosen by the three modellers illustrate one of the dilemmas faced in computational wind engineering. While all modeller matched the inlet velocity profile to the full-scale profile, only one of the modellers chose to match the full-scale turbulence data. This approach led to a boundary layer that was not in equilibrium. The approach taken by the other modeller was to specify lower inlet turbulent kinetic energy level, which are more consistent with the turbulence models chosen and lead to a homogeneous boundary layer. For the $0^{\circ}$ case, wind normal to one face of the cube, it is shown that the RNG solution is closest to the full-scale data. This result appears to be associated with the RNG solution showing the correct flow separation and reattachment on the roof. The other solutions show either excessive separation (MMK) or no separation at all (K-E). For the $45^{\circ}$ case the three solutions are fairly similar. None of them correctly predicting the high suctions along the windward edges of the roof. In general the velocity components are more accurately predicted than the pressures. However in all cases the turbulence levels are poorly matched, with all of the solutions failing to match the high turbulence levels measured around the edges of separated flows. Although all of the computational solutions have deficiencies, the variability of results is shown to be similar to that which has been obtained with a similar comparative wind tunnel study. This suggests that the computational solutions are only slightly less reliable than the wind tunnel.

Previous studies have shown that Computational Wind Engineering (CWE) is still in its infancy and has a long way to go to become truly useful to the design practitioner. The present work focuses on more recent studies to identify progress on outstanding issues and improvements in the numerical simulation of wind effects on buildings. The paper reviews wind loading and environmental effects; it finds that, in spite of some interesting and visually impressive results produced with CWE, the numerical wind tunnel is still virtual rather than real and many more parallel studies - numerical and experimental - will be required to increase the level of confidence in the computational results.

When predicting unsteady flow and pressure fields around a structure in a turbulent boundary layer by Large Eddy Simulation (LES), velocity fluctuations of turbulence (inflow turbulence), which reproduce statistical characteristics of the turbulent boundary layer, must be given at the inflow boundary. However, research has just started on development of a method for generating inflow turbulence that satisfies the prescribed turbulence statistics, and many issues still remain to be resolved. In our previous study, we proposed a method for generating inflow turbulence and confirmed its applicability by LES of an isotropic turbulence. In this study, the generation method was applied to a turbulent boundary layer developed over a flat plate, and the reproducibility of turbulence statistics predicted by LES computation was examined. Statistical characteristics of a turbulent boundary layer developed over a flat plate were investigated by a wind tunnel test for modeling the cross-spectral density matrix for use as targets of inflow turbulence generation for LES computation. Furthermore, we investigated how the degree of correspondence of the cross-spectral density matrix of the generated inflow turbulence with the target cross-spectral density matrix estimated by the wind tunnel test influenced the LES results for the turbulent boundary layer. The results of this study confirmed that the reproduction of cross-spectra of the normal components of the inflow turbulence generation is very important in reproducing power spectra, spatial correlation and turbulence statistics of wind velocity in LES.

Recently, the prediction of wind environment around a building using Computational Fluid Dynamics (CFD) technique comes to be carried out at the practical design stage. However, there have been very few studies which examined the accuracy of CFD prediction of flow around a high-rise building including the velocity distribution at pedestrian level. The working group for CFD prediction of wind environment around building, which consists of researchers from several universities and private companies, was organized in the Architectural Institute of Japan (AIJ) considering such a background. At the first stage of the project, the working group planned to carry out the cross comparison of CFD results of flow around a high rise building by various numerical methods, in order to clarify the major factors which affect prediction accuracy. This paper presents the results of this comparison.

Two-dimensional formulations for wind forces on elongated bodies, such as bridge decks, are reviewed and links with expressions found in two-dimensional airfoil theory are pointed out. The present research focus on indicial lift responses and admittance functions which are commonly used to improve buffeting analysis of bluff bodies. A computational fluid dynamic (CFD) analysis is used to derive these aerodynamic functions for various sections. The numerical procedure is presented and results are discussed which demonstrate that the particular shapes of these functions are strongly dependent on the evolution of the separated flows around the sections at the early stages.

In the recent years flow around bridges are investigated using computer modeling. Selvam (1998), Selvam and Bosch (1999), Frandsen and McRobie (1999) used finite element procedures. Larsen and Walther (1997) used discrete vorticity procedure. The aeroelastic instability is a major criterion to be checked for long span bridges. If the wind speed experienced by a bridge is greater than the critical wind speed for flutter, then the bridge fails due to aeroelastic instability. Larsen and Walther (1997) computed the critical velocity for flutter using discrete vortex method similar to wind tunnel procedures. In this work, the critical velocity for flutter will be calculated directly (free oscillation procedure) similar to the approaches reported by Selvam et al. (1998). It is expected that the computational time required to compute the critical velocity using this approach may be much shorter than the traditional approach. The computed critical flutter velocity of 69 m/s is in reasonable comparison with wind tunnel measurement. The no flutter and flutter conditions are illustrated using the bridge response in time.

Research being undertaken at the University of Auckland has enabled Vortec Energy to improve the performance of the Vortec 7 Diffuser Augmented Wind Turbine. Computational Fluid Dynamic (CFD) modelling of the Vortec 7 was used to ascertain the effectiveness of geometric modifications to the Vortec 7. The CFD work was then developed to look at new geometries, and refinement of these led to greater power augmentation for a given diffuser exit area ratio. Both full scale analysis of the Vortec 7 and a wind tunnel investigation of the development design have been used for comparison with the CFD model.

A two dimensional discrete vortex method (DIVEX) has been developed at the Department of Aerospace Engineering, University of Glasgow, to predict unsteady and incompressible flow fields around closed bodies. The basis of the method is the discretisation of the vorticity field, rather than the velocity field, into a series of vortex particles that are free to move in the flow field that the particles collectively induce. This paper gives a brief description of the numerical implementation of DIVEX and presents the results of calculations on a recent suspension bridge deck section. The results from both the static and flutter analysis of the main deck in isolation are in good agreement with experimental data. A brief study of the effect of flow control vanes on the aeroelastic stability of the bridge is also presented and the results confirm previous analytical and experimental studies. The aeroelastic study is carried out firstly using aerodynamic derivatives extracted from the DIVEX simulations. These results are then assessed further by presenting results from full time-dependent aeroelastic solutions for the original deck and one of the vane cases. In general, the results show good qualitative and quantitative agreement with results from experimental data and demonstrate that DIVEX is a useful design tool in the field of wind engineering.

An efficient large eddy simulation algorithm is used to compute surface pressure distributions on an eleven story (target) building on the NIST campus. Local meteorology, neighboring buildings, topography and large vegetation (trees) all play an important part in determining the flows and therefore the pressures experienced by the target. The wind profile imposed at the upstream surface of the computational domain follows a power law with an exponent representing a suburban terrain. This profile accounts for the flow retardation due to friction from the surface of the earth, but does not include fluctuations that would naturally occur in this flow. The effect of neighboring buildings on the time dependent surface pressures experienced by the target is examined. Comparison of the pressure fluctuations on the single target building alone with those on the target building in situ show that, owing to vortices shed by the upstream buildings, fluctuations are larger when such buildings are present. Even when buildings are lateral to or behind the target, the pressure disturbances generate significantly different flows around this building. A simple grid-free mathematical model of a tree is presented in which the trunk and the branches are each represented by a collection of spherical particles strung together like beads on a string. The drag from the tree, determined as the sum of the drags of the component particles, produces an oscillatory, spreading wake of slower fluid, suggesting that the behavior of trees as wind breakers can be modeled usefully.

An important goal of computational wind engineering is to impact the design process with simulations of flow around buildings and bridges. One challenging aspect of this goal is to solve the Navier-Stokes (NS) equations accurately. For the unsteady computations, an adaptive finite element technique may reduce the computer time and storage. The preliminary application of a p-version as well as an h-version adaptive technique to computational wind engineering has been reported in previous paper. The details on the implementation of p-adaptive technique will be discussed in this paper. In this technique, two posteriori error estimations, which are based on the velocity and vorticity, are first presented. Then, the polynomial order of the interpolation function is increased continuously element by element until the estimated error is less than the accepted. The second through sixth orders of hierarchical functions are used as the interpolation polynomials. Unequal order interpolations are used for velocity and pressure. Using the flow around a circular cylinder with Reynolds number of 1000 the two error estimators are compared. The result show that the estimated error based on the velocity is lower than that based on the vorticity.

A numerical investigation on the turbulent flows over a three-dimensional steep hill is presented. The numerical model developed for the present work is based on the finite volume method and the SIMPLE algorithm with a non-staggered grid system. Standard $k-{\varepsilon}$ model and Shih's non-linear model are tested for the validation of the prediction accuracy in the 3D separated flow. Comparisons of the mean velocity and turbulence profiles between the numerical predictions and the measurements show good agreement. The Shih's non-linear model is found to predict mean flow and turbulence better than the Standard $k-{\varepsilon}$. Flow patterns have also been examined to explain the difference in the cavity zone between 2D and 3D hills.

The along-wind response of a surface-mounted elastic fence under the action of wind was investigated numerically. In the computations, two sets of equations, one for the simulation of the unsteady turbulent flow and the other for the calculation of the dynamic motion of the fence, were solved alternatively. The resulting time-series tip response of the fence as well as the flow fields were analyzed to examine the dynamic behaviors of the two. Results show that the flow is unsteady and is dominated by two frequencies: one relates to the shear layer vortices and the other one is subject to vortex shedding. The resulting unsteady wind load causes the fence to vibrate. The tip deflection of the fence is periodic and is symmetric to an equilibrium position, corresponding to the average load. Although the along-wind aerodynamic effect is not significant, the fluctuating quantities of the tip deflection, velocity and acceleration are enhanced as the fundamental frequency of the fence is near the vortex or shedding frequency of the flow due to the occurrence of resonance. In addition, when the fence is relatively soft, higher mode response can be excited, leading to significant increases of the variations of the tip velocity and acceleration.

Wind tunnel pressure measurements and numerical simulations based on the Reynolds Stress Model (RSM) are compared with full and model scale data in the flow area of impingement, separation and wake for $60^{\circ}$ and $90^{\circ}$ wind azimuth angles. The phase averaged fluctuating pressures simulated by the RSM model are combined with modelling of the small scale, random pressure field to produce the total, instantaneous pressures. Time averaged, rsm and peak pressure coefficients are consequently calculated. This numerical approach predicts slightly better the pressure field on the roof of the TTU (Texas Tech University) building when compared to the wind tunnel experimental results. However, it shows a deviation from both experimental data sets in the impingement and wake regions. The limitations of the RSM model in resolving the intermittent flow field associated with the corner vortex formation are discussed. Also, correlations between the largest roof suctions and the corner vortex "switching phenomena" are observed. It is inferred that the intermittency and short duration of this vortex switching might be related to both the wind tunnel and numerical simulation under-prediction of the peak roof suctions for oblique wind directions.

Mean surface pressures and overall wind loads on hemispherical domes immersed in a boundary layer were obtained by numerical simulation. The effects of alternative turbulence models, Reynolds Number and surface roughness were examined and compared with earlier studies. Surface pressures on dual hemispherical domes were also calculated for three wind orientations ($0^{\circ}$, $45^{\circ}$, and $90^{\circ}$) to evaluate flow field interactions. Calculated values were compared to wind-tunnel measurements made in equivalent flow conditions.

This paper describes a CFD investigation into the flow over the cab of a bluff-fronted lorry. Several different simulations were undertaken, using the commercial codes: CFX, Fluent and PowerFLOW. Using the $k-{\varepsilon}$ turbulence model, the flow over the cab was symmetric, however, using more accurate turbulence models such as the RNG $k-{\varepsilon}$ model or the Reynolds Stress Model, the flow was asymmetric. The paper discusses whether this phenomenon is a real effect or whether it is a solver artefact and the study is supported by experimental evidence. The findings are preliminary, but suggest that it has a physical origin and that it may be aspect ratio-dependent.

The application of Large Eddy Simulation (LES) in a curvilinear coordinate system to the flow around a square cylinder is presented. In order to obtain sufficient resolution near the side of the cylinder, we use an O-type grid. Even with a curvilinear coordinate system, it is difficult to avoid the numerical oscillation arising in high-Reynolds-number flows past a bluff body, without using an extremely fine grid used. An upwind scheme has the effect of removing the numerical oscillations, but, it is accompanied by numerical dissipation that is a kind of an additional sub-grid scale effect. Firstly, we investigate the effect of numerical dissipation on the computational results in a case where turbulent dissipation is removed in order to clarify the differences between the effect of numerical dissipation. Next, the applicability and the limitations of the present method, which combine the dynamic SGS model with acceptable numerical dissipation, are discussed.

Numerical flow computations around an aeroelastic 3D square cylinder immersed in the turbulent boundary layer are shown. Present computational code can be characterized by three numerical aspects which are 1) the method of artificial compressibility is adopted for the incompressible flow computations, 2) the domain decomposition technique is used to get better grid point distributions, and 3) to achieve the conservation law both in time and space when the flow is computed a with moving and transformed grid, the time derivatives of metrics are evaluated using the time-and-space volume. To provide time-dependant inflow boundary conditions satisfying prescribed time-averaged velocity profiles, a convenient way for generating inflow turbulence is proposed. The square cylinder is modeled as a 4-lumped-mass system and it vibrates with two-degree of freedom of heaving motion. Those blocks which surround the cylinder are deformed according to the cylinder's motion. Vigorous oscillations occur as the vortex shedding frequency approaches cylinder's natural frequencies.