Vadis

 


Model name

VADIS

 

Full model name

Pollutant dispersion in the atmosphere under variable wind conditions

 

Intended field of application

Simulation of flow and air pollutants dispersion (inert gases, particle matter and heavy gases) near obstacles (buildings and trees) in urban areas. Simulation of the vegetative canopy effects on wind flow and dispersion.

 

Model type and dimension

VADIS is a Computational Fluid Dynamic (CFD) model Reynolds-averaged Navier-Stokes (RANS) type with a Lagrangian module for pollutants dispersion.

 

Model description summary

VADIS functioning is based on two modules, FLOW and DISPER. In the first module a k-e turbulence model calculates the 3D wind field, the turbulent viscosity, the pressure, the TKE (Turbulent Kinetic Energy) and the temperature fields affected by a set of obstacles defined in a Cartesian grid. The solution of the Navier-Stokes equations is made through the SIMPLE solver. The DISPER module uses the data provided by the previous module, namely the wind field, and estimates the 3D concentration field, based on the Lagrangian approach. This methodology assumes that the pollutant spatial and temporal dispersion is conveniently represented by a large number of numerical particles released in the flow.

 

Model limitations/approximations

- No topography - No humidity - No roughness - No chemistry - No radiation - No clouds - No nesting - Uses cartesian grid

 

Resolution

Temporal resolution

Depends on available meteorological data and domain cell sizes. Typical time step: 0.1 to 5 seconds Typical simulated time period: minutes to several hours

 

Horizontal resolution

Typically 0.5 to 5 m, but highly dependent on machine performance.

 

Vertical resolution

Typically 0.5 to 5 m, but highly dependent on machine performance.

 

Schemes

Advection & Convection

The model does not use advection.

 

Turbulence

k-e turbulence scheme. This scheme corresponds to a one-and-a-half order closure that retains the prognostic equations for the zero-order statistics such as mean wind, temperature, humidity and the variances of the referred variables. The TKE equation is used in place of the velocity variance equations. A highly-parameterized prognostic equation for the dissipation rate is included in addition to the equation for TKE.

 

Deposition

The model does not use deposition.

 

Chemistry

The model does not use chemistry.

 

Solution technique

Wind field: SIMPLE solver, tri-diagonal matrix algorithm;

Dispersion field: Lagrangian approach using the Langevin equation. This methodology assumes that the spatial and temporal dispersion of the mass of pollutant emitted is conveniently represented by a large number of numerical particles randomly released in the flow. In each time step, each particle displacement is calculated by the sum of a deterministic component obtained from the velocity field, the stochastic component related with the local turbulence translated by the Langevin stochastic theory and the influence of the fluctuation forces, represented by the Langevin equation.

 

Input

Availability and Validation of Input data

Information not available. For more details, please, refer directly to the contact person.

 

Emissions

Multiple point, line, area, volume sources with different positions, dimensions and emission rates; sources 3D extreme coordinates.

 

Meteorology

Wind velocity and direction at a reference height, wind velocity profile, air temperature, turbulence at the entrance of the domain.

 

Topography

The model is applicable for flat terrain.

 

Initial conditions

The model does not use Initial conditions.

 

Boundary conditions

Wind velocity and direction and background pollutant concentrations at the entrance of the domain.

 

Data assimilation options

The model does not use Data assimilation options.

 

Other input requirements

Volumetry: Buildings and trees 3D extreme coordinates; angle between obstacle and the grid; obstacle temperature.

 

Output quantities

3D fields of wind, pressure, temperature, turbulence and viscosity; Numerical particles 3D position and concentration fields.

 

User interface availability

VADIS model interface for a friendly user access is available only for research activities.

 

Previous applications

1.Application type

Urban

Application description

1) Project title

SUTRA Project

2) Relevant references

BORREGO, C.; TCHEPEL, O.; COSTA, A.M.; AMORIM, J.H. and MIRANDA, A.I. – Emission and dispersion modelling of Lisbon air quality at local scale. Atmospheric Environment: Elsevier, Vol. 37 (2003), p. 5197-5205.

BORREGO, C.; MIRANDA, A.I.; COSTA, A.M.; TCHEPEL, O.; AMORIM, J.H. and MARTINS, H. - Air Quality Modelling in European Cities: a local scale perspective. In Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes, 8th, Sofia, Bulgaria, 14-17 October 2002, - Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes Proceedings, eds. Ekaterina Batchvarova and Dimiter Syrakov, pp.244-248.

Borrego, C.; Tchepel, O.; Salmim, L; Amorim, J.H.; Costa, A.M. and Janko, J., 2004. Integrated modelling of road traffic emissions: application to Lisbon air quality management. Cybernetics and Systems: An International journal. Taylor & Francis, Vol. 35, Numbers 5-6, p. 535-548.

BORREGO, C.; MIRANDA, A.I.; COSTA, A.M. and AMORIM, J.H. - VADIS Street Canyon Model: Methodology description. Departamento de Ambiente e Ordenamento, Universidade de Aveiro: Setembro 2002, AMB-QA-09/02. Deliverable D04.3 of SUTRA Project (EVK4-CT-1999-00013).

BORREGO, C.; MIRANDA, A.I.; COSTA, A.M. and AMORIM, J.H. - VADIS Street Canyon Model: Operational Prototype. Departamento de Ambiente e Ordenamento, Universidade de Aveiro: Setembro 2002, AMB-QA-10/02. Deliverable D04.4 of SUTRA Project (EVK4-CT-1999-00013).

3) Project’s short description

The primary objective of SUTRA was to develop a consistent and comprehensive approach and planning methodology for the analysis of urban transportation problems, that helps to design strategies for sustainable cities. This included an integration of socio-economic, environmental and technological concepts including the development, integration, and demonstration of tools and methodologies to improve forecasting, assessment and policy level decision support. VADIS model was one of the numerical tools used in the above mentioned project. The model was applied to the Lisbon city centre in order to evaluate the air pollution associated to road traffic. The study domain covers an area of 450 m x 450 m and is characterised by strictly perpendicular streets including several one-way roads and a pedestrian zone. During the simulation period, wind direction was mainly from Northwest, with velocities varying between 1 m s-1 and 6 m s-1. Background concentrations entering the model domain were based on CO average concentrations measured at an urban monitoring station that is not directly influenced by the emission sources. Hourly simulation were conducted with VADIS to obtain CO concentration levels for a typical summer day, which was chosen using a statistical meteorological approach. Several traffic management scenarios were developed and evaluated using the TREM (Transport Emission model for Line Sources) and VADIS results.

2.Application type

Urban

Application description

1)Project title

Work developed under the 27th ITM Conference (www.dao.ua.pt/itm).

2) Relevant references

BORREGO, C.; TCHEPEL, O.; COSTA, A.M.; MARTINS, H.; FEREEIRA, J. and MIRANDA, A.I. - Traffic-related particulate air pollution exposure in urban areas. Atmospheric Environment: Elsevier, Vol. 40, pp. 7205-7214.

3) Project’s short description

VADIS was applied to the Lisbon city centre for the estimation of flow and PM10 concentrations due to traffic emissions. The study domain has an area of 1000 m2, which was divided in cells of 25x25 m2 and is characterized by a set of 29 buildings, with an average height of 12 m, and 8 main roads. Hourly simulations were conducted for three days (February 28th, 29th and March 1st) with boundary meteorological conditions and PM10 background concentrations given by the mesoscale meteorological and dispersion model MEMO. Hourly PM10 traffic emissions were estimated by TREM (Transport Emission Model for Line Sources) using local information on traffic counting data and fleet composition. Several statistic factors were applied to compare modelling results with air quality data measured on the study domain, showing an underestimation of the simulated concentrations. In fact, the application of both air quality models, MEMO and VADIS, shows an underestimation of the modelling results when compared with the air quality observations. These results can be explained by possible missing PM10 sources, that might have led to an incorrect estimation of PM10 emissions. Recent studies show that a Sahara sand storm episode occurred during the simulation period affecting the south of Portugal with direct consequences in the PM10 concentration values.

3.Application type

Urban

Application description

1) Project title

Work developed under the NATO Advanced Research Workshop on Air, Water and Soil Quality Modelling for Risk and Impact Assessment.

2) Relevant references

COSTA, A.M., MIRANDA, A.I., BORREGO, C. – Dispersion modelling of atmospheric contaminants resulting from terrorist attacks and accidental releases in urban areas. NATO Advanced Research Workshop on Air, Water and Soil Quality Modelling for Risk and Impact Assessment, Tabakhmela, Georgia, 16-20 September, 2005 – Book of Abstracts, pp. 19.

3) Project’s short description

An exposure module to chemical agents was developed and integrated in the CFD model VADIS, in order to estimate the cumulative exposure and the number of persons exposed above specific limit values of Sulphur Mustard HD agent. The numerical system was applied to a selected case study in the Lisbon urban area, in order to determine the effects on the population, as a result of a terrorist attack scenario with chemical agent Sulphur Mustard HD. The meteorological conditions of a typical weekday in Lisbon (1st of March 2000), with hourly values of wind velocity and direction, were considered in the simulation. A terrorist attack scenario was simulated through the release of 100 kg of Sulphur Mustard Agent HD, during 10s, in Entrecampos Square . The air flow around this specific urban area, for the tested meteorological conditions and the evolution of the HD agent plume dispersion were determined. The new development in VADIS model allowed the determination of the number of persons that can be affected if a chemical agent is released in that specific urban area. Output concentration files of HD agent were generated with a time interval of 10s after the release.

 

Documentation status

PhD thesis written in Portuguese with VADIS description:

Martins, JM, 1998, Dispersão de poluentes na atmosfera em condições de vento fraco, PhD thesis, Dep. Ambiente e Ordenamento, Universidade de Aveiro.

Master dissertation written in Portuguese with model application:

COSTA, A.M. – Avaliação da Qualidade do Ar ao Nível Local: contributo para o desenvolvimento urbano sustentável. Dissertação apresentada à Universidade de Aveiro para obtenção do grau de Mestre em Poluição Atmosférica, Departamento de Ambiente e Ordenamento, Universidade de Aveiro, Aveiro, Portugal. Fevereiro 2003.

 

SUTRA European Project (EVK4-CT-1999-00013) Deliverables:

D04.3 – VADIS Street Canyon Model: Methodology Description

D04.4 – VADIS Operational Street Canyon Model (operational prototype).

 

 

Validation and evaluation

A model validation exercise was conducted in the scope of COST Action 732 (http://www.mi.uni-hamburg.de/Home.484.0.html). Franke, J.; Hellsten, A.; Schlünzen, H.; Carissimo, B.; Baklanov, A.; Barmpas, P.; Bartzis, J.; Batchvarova, E.; Baumann-Stanzer, K.; Berkowicz, R.; Borrego, C.; Britter, R.; Brzozowski, K.; Burzynski, J.; Costa, A.M., et al., 2007. Best practice guideline for the CFD simulation of flows in the urban environment. COST Action 732: Quality assurance and improvement of microscale meteorological models. COST Office Brussels, ISBN 3-00-018312-4, 51 pp. Britter, R.; Schatzmann, M.; Baklanov, A.; Barmpas, P.; Bartzis, J.; Batchvarova, E.; Baumann-Stanzer, K.; Berkowicz, R.; Borrego, C.; Brzozowski, K.; Burzynski, J.; Costa, A.M.; Carissimo, B.; Dimitrova, R.; Franke, J., et al., 2007. Background and justification document to support the model evaluation guidance and protocol. COST Action 732: Quality assurance and improvement of microscale meteorological models. COST Office Brussels, ISBN 3-00-018312-4, 85 pp. Britter, R.; Schatzmann, M.; Baklanov, A.; Barmpas, P.; Bartzis, J.; Batchvarova, E.; Baumann-Stanzer, K.; Berkowicz, R.; Borrego, C.; Brzozowski, K.; Burzynski, J.; Costa, A.M.; Carissimo, B.; Dimitrova, R.; Franke, J., et al., 2007. Model Evaluation Guidance and Protocol Document. COST Action 732: Quality assurance and improvement of microscale meteorological models. COST Office Brussels , ISBN 3-00-018312-4, 27 pp.

Model intercomparison

1. Wind flow simulations were performed by VADIS and compared with CHENSI model under ATREUS network.

1) Project title

Advanced Tools for Rational Energy Use towards Sustainability with emphasis on microclimatic issues in urban applications (ATREUS) network (http://aix.meng.auth.gr/atreus/).

2) Relevant references

S. Vardoulakis et al. Intercomparison of CFD models within ATREUS: Single building configuration, presented to the ERCOFTAC Meeting on Urban Scale CFD, Nottingham (UK), 9-10 September 2004.

K. Richards et al. A wind tunnel investigation of thermal effects within the vicinity of a single block building with leeward wall heating, accepted to the ERCOFTAC Meeting on Urban Scale CFD, Nottingham (UK), 9-10 September 2004.

S. Vardoulakis, R. Dimitrova, K. Richards, D. Hamlyn, G. Camilleri, M. Weeks, J-F. Sini, R. Britter, C. Borrego, M. Schatzmann, N. Moussiopoulos, Numerical model inter-comparison for a single block building within ATREUS, accepted for oral presentation at the 10th International Conference on Harmonisation within Atmospheric Dispersion, 17-20 October, 2005, Crete, Greece.

3) Project’s short description

The project research objectives include:

• The study of the urban energy budget taking into account the local and microclimatic conditions,

• Use of the knowledge gathered by latest studies on wind flow modifications by urban structures, their geometry and dimensions,

• Development of city maps to allow the determination of optimum arrangements of groups of buildings to optimise the exchange processes for an area of the city or for the city as a whole,

• Study of the flow and turbulence characteristics within a street canyon with special emphasis in the boundary layers of building walls and roofs,

• Investigation of the thermal effects on flow modification within street canyons with special regard to low wind speed conditions around buildings,

• Evaluation of the wind field around buildings,

• Determination of the exploitable RES potential on the urban areas, and

• Determination of heating and cooling loads of the buildings, and their impact on the urban microclimate.

4) more details on Model performance

Both the micro-scale numerical models used (VADIS and CHENSI) were extensively validated against the experimental data for both the isothermal (cold cube) and thermal cases. Both codes made representative predictions of the mean velocity field for the cold cube case but tended to over predict mean turbulent kinetic energy in regions of impingement, a common problem when using RANS codes with variants of the standard k-e turbulence model. In both cases improvements in overall predictions were observed when non-uniform inflow boundary conditions, the same as that recorded in the wind tunnel were applied. With respect to the thermal cases CHENSI generally performed better at predicting the mean temperature field and resulting modifications in the velocity distribution within fair agreement with the experimental data at the model centre-plane. Predictions were further improved through applying a new thermal wall condition, obtained from the experimental data, based on the heat flux at the heated face of the cube. In general VADIS over-predicted the buoyancy force and mean temperature field in the wake of the model. Difficulties with VADIS in applying the thermal conditions to more complex domain meant only CHENSI was further used to make predictions of wind and temperature fields within the more complex ‘ Lisbon ’ geometry. The VADIS code encountered difficulties with these simulations primarily due to the recent addition of wall functions into the code. VADIS is an in-house code developed at GEMAC/UA and is always under development. This compounded with limited time meant that this problem with VADIS could not be fully resolved within ATREUS. However work will continue in this area.

 

2. Comparisons were made against FLUENT model

1) Project title

SUTRA Project

2) Relevant references

BORREGO, C.; TCHEPEL, O.; COSTA, A.M.; AMORIM, J.H. and MIRANDA, A.I. – Emission and dispersion modelling of Lisbon air quality at local scale. Atmospheric Environment: Elsevier, Vol. 37 (2003), p. 5197-5205.

3) Project’s short description

The primary objective of SUTRA was to develop a consistent and comprehensive approach and planning methodology for the analysis of urban transportation problems, that helps to design strategies for sustainable cities. This included an integration of socio-economic, environmental and technological concepts including the development, integration, and demonstration of tools and methodologies to improve forecasting, assessment and policy level decision support. VADIS model was one of the numerical tools used in the above mentioned project. The model was applied to the Lisbon city centre in order to evaluate the air pollution associated to road traffic. The study domain covers an area of 450 m x 450 m and is characterised by strictly perpendicular streets including several one-way roads and a pedestrian zone. During the simulation period, wind direction was mainly from Northwest, with velocities varying between 1 m s-1 and 6 m s-1. Background concentrations entering the model domain were based on CO average concentrations measured at an urban monitoring station that is not directly influenced by the emission sources. Hourly simulation were conducted with VADIS to obtain CO concentration levels for a typical summer day, which was chosen using a statistical meteorological approach. Several traffic management scenarios were developed and evaluated using the TREM (Transport Emission model for Line Sources) and VADIS results.

4) more details on Model performance

Hourly CO concentrations obtained with VADIS for the Lisbon city centre were compared with concentration values measured by an air quality station located in the study domain and with FLUENT results. There is a good agreement between predicted and measured values between 0 a.m. and 3 p.m. for the simulated day. Moreover, during the time period when model estimations and measurements show some discrepancy, the results obtained with FLUENT and VADIS are very similar, indicating that the trend of both models to underestimate the CO concentrations could be justified by a possible inaccuracy associated with initialisation data. Due to the low CO background concentration (77 µg m-3), the modelling results are not affected by the background values. A quantitative analysis to determine modelling uncertainties has been applied to the estimated CO concentration values. The analysis is based on the definition of maximum deviation of the measured and calculated levels during the considered period. To be compared with modelling quality objectives established by the Directive (200/69/EC), 8-hour average CO values were evaluated. Based on this approach, the average uncertainty of the model prediction for this study corresponds to 15 %, achieving 52 % as a maximum for the 8-hour average period from 1 to 8 p.m.. This value slightly exceeds the 50 % acceptability limit defined by the Directive.

 

Portability and computer requirements

Portability

Workstation 2.33 GHz 4GB RAM

 

CPU time

Extremely dependent on number of computational cells and grid resolution. Typical CPU run time of ~14h for a 30 million cells simulation on a Cluster with 2 processors 2.33 HHz and 4GB of RAM.

 

Storage

Typically around 25 Mb per simulation.

 

Availability

The model is not a public domain programme. Information on the conditions for obtaining VADIS can be provided by Prof. Carlos Borrego.

 

Other references

MARTINS J.M. - Study of the atmospheric dispersion of aerosols generated during a fire, Workshop on European Research on Industrial Fires, EUR15340EN, Apeldoorn (1993).

BORREGO C., MARTINS J.M. - Simulation of the atmospheric dispersion of gases and aerosols, Final report of the EC Project MISTRAL, Department of Environment and Planning, University of Aveiro (1994).

MARTINS J.M., BORREGO C. - Simulation of the atmospheric dispersion of gases and aerosols, 2nd Workshop on European Research on Industrial Fires, EUR15967EN, Cadarache (1994).

BORREGO C., MARTINS J.M. - Simulation of the atmospheric dispersion of gases and aerosols - Comparison between experiments and numerical models results, MISTRAL 2 - progress report, Department of Environment and Planning, University of Aveiro (1995).

MARTINS J.M., BORREGO C., HODIN A. AND MEJEAN P. – Comparison of real scale, wind tunnel and numerical model results of atmospheric dispersion in the vicinity of a building in 3rd Workshop on European Research on Industrial Fires Proceedings, Roskilde (1996).

BORREGO C., MARTINS J.M. - Simulation of the atmospheric dispersion of gases and aerosols - A simple numerical simulation for low wind speed dispersion, MISTRAL 2 - progress report, Department of Environment and Planning, University of Aveiro (1996).

MARTINS J.M. - Dispersão de poluentes em condições de vento fraco (Air pollutant dispersion in calm wind conditions), PhD Thesis, University of Aveiro (1998).

MARTINS J.M. AND BORREGO C. – Describing the dispersion of pollutants near buildings under low wind speed conditions: real scale and numerical results in Envirosoft 98 – Development and application of computer techniques to environmental studies, WIT, Las Vegas (1998).

BORREGO C., MARTINS J.M., PINTO C., AND CARVALHO A. – Modelling flow and dispersion around buildings under varying wind direction in Abstracts of 3rd Saturn Workshop, pp. 5, Aveiro (1999).

BORREGO, C.; MIRANDA, A.I.; COSTA, A.M.; TCHEPEL, O.; AMORIM, J.H. and MARTINS, H. - Air Quality Modelling in European Cities: a local scale perspective. In Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes, 8th, Sofia, Bulgaria, 14-17 October 2002, - Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes Proceedings, eds. Ekaterina Batchvarova and Dimiter Syrakov, pp.244-248.

Borrego, C.; Tchepel, O.; Costa, A.M.; Amorim, J.H. and Miranda, A.I., 2003. Emission and dispersion modelling of Lisbon air quality at local scale. Atmospheric Environment: Elsevier, Vol. 37, p. 5197-5205.

BORREGO, C.; MIRANDA, A.I.; TCHEPEL, O.; COSTA, A.M.; AMORIM, J. and MAGALHÃES, S. – Development of an integrated air quality management system for urban areas. In EUROTRAC-2 Symposium 2002, Garmisch-Partenkirchen , Germany , 11-15 March, 2002: – Proceedings EUROTRAC-2 Symposium 2002: Eds. P. Midgley, M. Reuther, Margraf-Verlag, Weikersheim 2002.

Borrego, C.; Tchepel, O.; Salmim, L; Amorim, J.H.; Costa, A.M. and Janko, J., 2004. Integrated modelling of road traffic emissions: application to Lisbon air quality management. Cybernetics and Systems: An International journal. Taylor & Francis, Vol. 35, Numbers 5-6, p. 535-548.

Borrego, C.; Tchepel, O.; Costa, A.M.; Martins, H.; Ferreira, J. and Miranda, A.I., 2006. Traffic-related particulate air pollution exposure in urban areas. Atmospheric Environment: Elsevier, Vol. 40, pp. 7205-7214.

COSTA, A.M., MIRANDA, A.I., BORREGO, C. – Dispersion modelling of atmospheric contaminants resulting from terrorist attacks and accidental releases in urban areas. NATO Advanced Research Workshop on Air, Water and Soil Quality Modelling for Risk and Impact Assessment, Tabakhmela, Georgia, 16-20 September, 2005 – Book of Abstracts, pp. 19.

 

Vadis application

última atualização a 15-04-2014
Este sítio web utiliza cookies sem recolher informação pessoal que permita a identificação dos utilizadores. Ao navegar neste sítio está a consentir a sua utilização.saber mais
Para que esta página funcione corretamente deve ativar a execução de Javascript. Se tal não for possível, algumas funcionalidades poderão estar limitadas.