Spring 2011, Purdue University Calumet,
Hammond, IN 46323, USA

Using CFD to Study Air Quality in Urban Microenvironments

Varun Khare
Purdue University Calumet
Hammond, IN, USA

|

Abstract
The project is concentrated on the study of the plume height coming out of buildings, such as restaurants and cooling towers around the office buildings, in an urban microenvironment, along with the placement of air intakes and exhausts on buildings which can significantly affect the overall indoor air quality. Earlier studies on the effects of building air intakes have been limited to relatively simple situations, unable to treat the complex envelope of most buildings and building groups. Computational Fluid Dynamics (CFD) is a tool that assists in modeling the airflow and dispersion of pollutants among complex urban geometries on the scale of a section of a building’s exterior up to several city blocks. This tool allows more accurate predictions of impacts over a range of meteorological scenarios and alternative building designs and placements relative to roadways and other pollutant sources. The steps in a CFD application are presented including geometry and mesh creation, simulation of meteorological conditions, handling of pollutant sources, and post-processing visualization.

INTRODUCTION
Design and placement of a building’s outside air intake is a very important building design concern, especially in high rise buildings that are placed so close together in an urban environment. According to the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE), “Most buildings using natural ventilation in the United States are high-rise residential buildings that often have no form of outdoor air intake other than operable windows. This results in buildings with inadequate ventilation, because occupants often leave the windows closed in order to run the air conditioning, keep out noise, etc.”[1] According to the Environmental Protection Agency (EPA), the national consensus standard for outside air ventilation is ASHRAE Standard 62.1-200, Ventilation for Acceptable Indoor Air Quality. (Details of the ASHRAE Standard can be found online via www.ashrae.org and its published Addenda.) This standard is often incorporated into state and local building codes, and specifies the amounts of outside air that must be provided by natural or mechanical ventilation systems to various areas of the school, including classrooms, gymnasiums, kitchens and other special use areas [2].
Localized air quality will benefit from an intake system design that provides protection of outside air intakes from direct line of sight impacts. This design strategy will reduce re-entrainment of building exhausts. Proper intake placement can also reduce the need for filtration at the intake with regard to odor, dust and other compounds. The amount of protection provided by an intake location is difficult to quantify because it is dependent on several factors including building geometry (dimensions, corners, rooftop obstructions, setback from roof edge), exhaust type, exhaust location, and intake size. Both numerical and physical modeling methods are useful in assessing these factors, such as Computational Fluid Dynamics (CFD). CFD is a branch of fluid mechanics that uses numerical methods and algorithms to solve and analyze problems that involve fluid flows. Computers are used to perform the calculations required to simulate the interaction of liquids and gases with surfaces defined by boundary conditions. Even with high-speed supercomputers, only approximate solutions can be achieved in many cases. Using numerical and physical modeling, basic principles of outside air intake placement can be demonstrated with practical recommendations that can be applied to the design of building intake systems [3]. Nomenclature
Keywords include:
Indoor Air Quality (IAQ)
Carbon monoxide (CO)
Volatile Organic Compounds (VOC)
Carbon dioxide (CO2)
Parts per million (ppm)
Computational Fluid Dynamics (CFD)
Cubic feet per minute (cfm)

Indoor air quality (IAQ) is defined as the air quality within and around buildings and structures, especially as it relates to the health and comfort of building occupants.IAQ can be affected by microbial contaminants (mold, bacteria), gases (including carbon monoxide (CO), radon (Rn), volatile organic compounds (VOC)), particulate matter (PM), or any mass or energy stressor that can induce adverse health conditions. Indoor air is becoming an increasingly more concerning health hazard than outdoor air. The primary methods for improving indoor air quality in most buildings are: using ventilation systems, diluting contaminants, filtration devices, and source control.
According to a study conducted by the United States EPA in 1995, they found that urban Americans spend about 90% of their time indoors. Many office buildings have significant indoor air pollution sources. These sources include furnishings, occupant activities, housekeeping practices, pesticide applications, and microbial contamination. A factor greatly influencing the effect of these sources and the overall quality of indoor air in offices is the ventilation system design, operations and maintenance. People generally have less control over the indoor environment in their offices than they do in their homes. As a result, there are large numbers of reported health problems associated with office buildings [4].
For the past several years, there have been many debates among indoor air quality specialists about the proper definition of indoor air quality and specifically what constitutes "acceptable" levels of indoor air quality. A way to quantitatively ensure the health of indoor air is by the frequency of effective turnover of interior air by replacement with outside air. In the UK, for example, classrooms are required to have 2.5 outdoor air changes per hour. In halls, gym, dining, and physiotherapy spaces, the ventilation should be sufficient to limit carbon dioxide to 1,500 ppm.
In the USA, and according to ASHRAE Standards, ventilation in classrooms is based on the amount of outdoor air per occupant plus the amount of outdoor air per unit of floor area, not air changes per hour. The ASHRAE standard 62-1989 recommends fresh air intake of 20 cfm per person. Since carbon dioxide indoors comes from occupants and outdoor air, the adequacy of ventilation per occupant is indicated by the concentration indoors minus the concentration outdoors. The value of 615 ppm above the outdoor concentration indicates approximately 15 cubic feet per minute of outdoor air per adult occupant doing sedentary office work where outdoor air contains 385 ppm, the current global average atmospheric CO2 concentration. In classrooms, the requirements in the ASHRAE standard 62.1, Ventilation for acceptable IAQ, would typically result in about 3 air changes per hour, depending on the occupant density. Of course the occupants are not the only source of pollutants; therefore outdoor air ventilation may need to be higher when unusual or strong sources of pollution exist indoors. When outdoor air is polluted, then bringing in more outdoor air can actually worsen the overall quality of the indoor air and exacerbate some occupant symptoms related to outdoor air pollution. Figure one illustrates an example of a building’s position of the outdoor air intake.

Figure 1: Example of a building’s outdoor air intake
(Reference: www.high-performance-hvac.com)
The use of air filters can trap some of the air pollutants. According to The Department of Energy's Energy Efficiency and Renewable Energy section, "Air filtration should have a Minimum Efficiency Reporting Value (MERV) of 13 as determined by ASHRAE 52.2-1999." Air filters are used to reduce the amount of dust that reaches the wet coils. Dust can serve as food to grow molds on the wet coils and ducts and can reduce the efficiency of the coils.
Recently according to ASHRAE’s website, they felt the need to strengthen the existing outdoor air intake requirements to ensure adequate ventilation and their corresponding IAQ benefits that are available to building occupants. The IAQ procedure, which allows for the calculation of the amount of outdoor air necessary to maintain the levels of indoor air contaminants below recommended levels, has been made more robust by increasing requirements for using the “similar building” design approach and clarifying other requirements. Per ASHRAE officials, “The standard now contains, in informative Appendix B, a table of volatile organic compounds that designers might want to consider as possible contaminants of concern. To encourage designers to consider ‘additivity’ when applying the IAQ Procedure, some guidance from the American Conference of Governmental Industrial Hygienists has been included.”
Commercial buildings, and sometimes residential, are often kept under slightly-positive air pressure relative to the outdoors to reduce infiltration. Limiting infiltration helps with moisture management and humidity control. Dilution of indoor pollutants with outdoor air is effective to the extent that outdoor air is free of harmful pollutants. Ozone (O3) in outdoor air occurs indoors at reduced concentrations because ozone is highly reactive with many chemicals found indoors. The products of the reactions between ozone and many common indoor pollutants include organic compounds that may be more odorous, irritating, or toxic than those from which they are formed. These products of ozone chemistry include formaldehyde, higher molecular weight aldehydes, acidic aerosols, and fine and ultrafine particles, among others. The higher the outdoor ventilation rate, the higher the indoor ozone concentration and the more likely the reactions will occur, but even at low levels, the reactions will take place. This suggests that ozone should be removed from ventilation air, especially in areas where outdoor ozone levels are frequently high. Recent research has shown that mortality and morbidity increase in the general population during periods of higher outdoor ozone and that the threshold for this effect is around 20 parts per billion (ppb).
These statistics are why the positioning of a building’s outdoor air intake is so crucial. Air handling units (AHU’s) and exhaust stacks on buildings should be carefully analyzed to ensure that building inhabitants are not exposed to irritants, carcinogens, and odors that originate from outdoor sources or the building’s own exhausts. Intake locations should be designed to minimize the probability of exhaust from nearby contaminant sources entering the building’s air supply system. There are three outdoor air intake locations on a building: ground level, upper level (raised), or rooftop. There are different design issues associated with ground level, upper level and roof level intakes, which are dependent on the types of exhaust sources present at a particular site.
Ground level intakes are beneficial when the majority of exhaust sources are at roof level. However, in cases where there are nearby contaminant sources at ground level, ground level intake placement with respect to air quality becomes an important issue. Some sources that should be considered in placing ground level intakes include:
-loading docks
-bus stops
-emergency generators
-lawnmowers
-automobile traffic
-designated smoking areas

The main concern associated with the placement of upper level intakes, is the presence of proposed upper level exhausts. However, ground level exhaust sources and sources on nearby surrounding buildings also play a part in the optimal placement of upper level air intakes. As with ground level sources, roof level intakes on the sidewall of buildings should not be placed on the side of a building that faces ground level sources or surrounding buildings with upper level exhausts. The benefit of upper level air intakes is that they provide protection from direct line of sight impacts from rooftop exhausts.
For rooftop air intakes, penthouses and other rooftop features should be used as protection from the rooftop exhaust sources. High re-entrainment impacts can occur at an unprotected rooftop intake location. As a general rule of thumb, the face area of a rooftop obstruction should be larger than the area of the exhaust plume to protect the intake from exhaust re-entrainment.
The area of the exhaust plume at the rooftop obstruction can be estimated using the following relationship:

A=4σy+φ4σz+φ
A = area of the exhaust plume (m2); σy= horizontal standard deviation from plume centerline (m) σz= vertical standard deviation from plume centerline (m) φ= stack exit diameter (m)
Various studies of airflow around buildings generally involve the positioning of cooling towers, exhausts and air-handling units. These projects have ranged from assisting architects with the placement of large cooling towers and air-handling units to addressing odor complaints due to exhaust sources near to air intakes.
Until recently, analytical methods were most often used to model the airflow in these types of projects. The ASHRAE static model is commonly used to determine roof recirculation and turbulence zones and downwind building recirculation zones (Fackrell, 1984). A picture of the descriptive ”zones” which can be calculated by this method is shown in Figure 2 below.

Figure 2: Building wake boundary, recirculation zones, turbulence zones, and streamlines calculated by static analytical methods. (Reference: Wilson, 1979)

Consultants have also utilized theatrical smoke releases in existing situations to detail the flow of sources to intakes. Wind tunnels have also been popular tools in the industry to model airflow and pollutant dispersion in scale models of urban environments, especially for proposed buildings.
Each of these traditional methods has advantages and disadvantages. The disadvantages of these models require a better method of analysis. Simple mathematical methods cannot account for complex urban spatial relationships or meteorological conditions. Theatrical smoke release can achieve this, but can require an extended effort to capture more than one or two conditions and cannot yield quantitative results without intricate instrumentation. Theatrical smoke is also limited to examination of nearby existing conditions. Wind tunnels offer more control, but are quite costly. Meteorological conditions are difficult to imitate in the wind tunnel environment because arrangements of upstream obstacles must be placed strategically to simulate incoming turbulence.
A new tool has become more widely available that overcomes many of the disadvantages of traditional methods Computational Fluid Dynamics (CFD). Computational fluid dynamics (CFD) is a branch of fluid mechanics that uses numerical methods and algorithms to solve and analyze problems that involve fluid flows. Computers are used to perform the calculations required to simulate the interaction of liquids and gases with surfaces defined by boundary conditions. With high-speed supercomputers, better solutions can be achieved [5].

Figure 3: Sample CFD generation

Now that CFD has become more available through computer software programs and CFD websites, it has been applied to aider range of problems, most recently urban microenvironments. With CFD the urban spatial geometry can be represented in computer simulation with reasonable accuracy, including more than just the building surfaces. The topography, vegetation, and mechanical features such as building stacks and air intakes, are easily detailed. CFD also offers the ability to experiment with solutions by making it easy to alter site geometry and features or stack and inlet parameters. A host of meteorological conditions including wind directions and magnitudes, temperatures, and atmospheric stabilities can be simulated in a CFD project in a reasonable time.
On the local scale, in urban microenvironments around buildings and sets of buildings, there are six pollutant sources of the highest concern: ozone (O3), carbon monoxide (CO), lead (Pb), sulfur dioxide (SO2), nitrogen oxides (NOx), and particulate matter (PM).
The majority of CFD projects are performed to address HVAC positioning, health or odor complaints, or exposure analyses will focus on one, several, or all of these.
Types of building exhaust include: * Lab hood exhaust * Cooling tower exhaust * Automobile exhaust * Odor sources (from kitchen stacks, garbage storage, industrial exhausts, etc.) * Allergens such as grass clippings or cigarette smoke * General building exhaust

Little regulation beyond the common law of nuisance exists for these pollutants or the conditions that make them a problem. Vague guidelines for exhaust stack heights, filtering, and safety are incorporated in different standards documents but no one standard can be applied for the world of circumstances that can occur at the local scale. Also, these types of exhausts affect different people indifferent ways. For example, an allergen source near an air intake may have a dramatic impact on sensitive individuals while non-allergic persons will not be affected. An odor from kitchen stack may be quite pleasing to one passing on the street, but may be highly objectionable to the nearby office worker exposed to the exhaust for hours on end. [6]

Lab Hood Exhaust Stacks
Exhaust from lab hood stacks can potentially be harmful depending on the agents used in the lab. Various chemicals with toxic or acidic properties may be released from these stacks, and in some cases, biological hazards may be emitted. This type of exhaust may be the most dangerous and it is very important to keep this exhaust away from air intakes or any place where the public may encounter it.
The volumes of these stacks are often quite small (ranging from 1,000 – 10,000 cfm).They are often small in diameter, ranging from 6 inches to 2 feet and are often placed in-groups or in lines along building roofs. The relatively low volume of flow means that the effective stack heights (H) of these exhausts may be quite low during stronger winds. Thus, the heights of these stacks often need to be quite high to avoid entrainment into recirculation zones which may bring the exhaust to the surface or concentrate it in regions near intakes, etc. Calculation for exhaust height (H):

H = hphysical + Δh where hphysical = physical stack height; Δh = the plume rise

CFD can assist with the analysis and determining the placement and height of these stacks.

Figure 4: Example of proper placement of Exhaust Hood
(Reference: Strobic Air Corporation)

Figure 4 above details the Tri-Stack laboratory fume hood exhaust systems at the University of Notre Dame’s South Bend, Indiana Galvin Laboratory Complex that prevent re-entrainment of laboratory workstation fume hood exhaust into the building’s ventilation air intakes.

Odorous Sources
Odorous sources are highly variable in their make-up. The most common of these are stacks similar to lab hood exhausts that emit kitchen odors, lab odors, locker room odors, or industrial odors. Airflow volumes from these stacks can be quite large (>30,000 acfm) such as those from larger kitchens or industrial sources. Other sources of odor may be from street level garbage bins or diesel exhaust from idling vehicles. The impact of each odor source is highly dependent on the amount of dilution of the exhaust that is needed before the odor is not offensive. However, this is a highly subjective area, considering that odor sensitivity varies considerably throughout the population. Standard descriptions of odors are based on the number of dilutions of the original source odor that are needed to render the sample gas odorless (that is, below the thresh hold of detection) to a majority of a panel of 7–13 average individuals. Thus, an odor’s strength may be cited as “1000 dilutions per threshold” (d/t). Typical odor strengths for an industrial source might be 2,000 to 5,000 d/t, while a kitchen exhaust might be 5,000 to 10,000 d/t. Because this represents the odor strength evaluated by 50% of the population it may be necessary to go to more than twice the dilutions of the measured detection threshold to avoid notice by all but a very few in the general population.
The best way to handle these types of odors in CFD modeling is by examining the dilution of the exhaust at locations throughout the CFD domain from the concentration at the initial source. Exhaust dilutions can then be compared to the d/t measurement in odor studies to estimate impacts. For visualization purposes, different outlines of the plume can be illustrated at various dilution levels of the exhaust to assess areas of impact [7].

Motor Vehicle Exhaust
Emissions from motor vehicles are one of the main sources of pollutants in the atmosphere, contributing to emissions of nitrogen oxides, carbon monoxide, benzene, and many other chemicals. Therefore, it is ideal to place building air intakes away from busy intersections or roads where exhaust may buildup in adverse traffic or wind conditions. Street canyons can concentrate motor vehicle exhaust under certain wind conditions or funnel high levels of exhaust to sensitive regions.CFD can simulate street canyons and other recirculation zones to assist in the placement of the air intakes of a building on the street.

Figures 5: Example of vehicle exhaust emissions
(Reference: Environment Waikato Regional Council)
Simulating motor vehicle exhaust in the CFD environment is difficult. Tailpipes from vehicles are placed at different heights depending on the type of vehicle and diesel engines have different kinds and amounts of emissions from gasoline engines. Also, the vehicle movement creates its own airflow that disperses the pollution itself. This poses a challenge in CFD modeling. Perhaps the best way to account for this is by using a volume source in the CFD domain, with emissions calculated from the EPA vehicle emissions models, such as MOBILE 6. According to the EPA, MOBILE6 is an emission factor model for predicting gram per mile emissions of Hydrocarbons (HC), Carbon Monoxide (CO), Nitrogen Oxides (NOx), Carbon Dioxide (CO2), Particulate Matter (PM), and toxics from cars, trucks, and motorcycles under various conditions [8].
If only the plume path needs to be analyzed, an array of plume streamlines released from the traffic exhaust volume region may be adequate for completing CFD.

Cooling Tower Exhaust
One would think of cooling tower exhaust as a harmless aerosol, consisting of water mist. However, the possibility of bacterial growth in the cooling tower water stream makes it a possible source of contamination. The infection of a cooling tower by bacteria makes it a biohazard us source of emissions.

Figure 6: Tennessee Valley Authority (TVA) power plant counterflow cooling tower with water vapor emission plumes
(Reference: http://www.tva.gov/sites/caledonia.htm#)

Steps taken to reduce the chance of bacterial infection may make the tower a chemically hazardous source depending on the antibacterial agents used in the stream. Ammonia, bromine, and chlorine are common chemicals used to treat cooling tower water which make the exhaust potentially hazardous to nearby receptors. Mist drift from cooling towers has been implicated as a source of the infectious proteobacteria legionella pneumophila. The conditions in cooling towers can be ideal for the growth of legionella, which is present in low concentrations in most water supply systems. The conditions which promote growth of the bacteria are * Water temperatures between 95 and 115 degrees F * Sediment and food sources in the water which support the growth of algae, protozoa, etc. * The presence of l-cysteine-HCl and iron salt

Legionella belongs to an unusual group of bacteria with special properties that can defeat the respiratory disease response system, a group which includes tuberculosis and salmonella. Legionella is widely distributed and occurs in five different varieties. Infections commonly appear with only two of the forms, one occurring relatively infrequently but manifesting as a mild respiratory disease in approximately 95% of those exposed and the other a more troublesome form that only matures in 2 to 5% of those exposed. The milder form causes flu-like symptoms that pass in less than a week. The other results in severe symptoms, often requires hospitalization and is fatal in about 10%of the cases.
One of the best known cases of disease outbreak (an American Legion convention in Philadelphia and the Oakland County Health Department in Pontiac, Michigan) the building air intakes were close to the infected cooling tower. However even with significant separation it is judicious to determine the likelihood of cross-contamination in order chose relative locations that will minimize the opportunity for infection. Because a cooling tower uses sprays of water to cool the working liquid, the exhaust air from a cooling tower contains fine droplets of water, called mist, which can drift with the exhaust air away from the cooling tower in a plume. If the cooling tower water has developed a legionella growth, the mist will contain the bacteria. The mist will evaporate quickly in warm dry conditions, but may remain as droplets for quite a distance in humid conditions. Even when dried, legionella can retain its infective capability [9].

CFD Project Study
In the project an office building is located adjacent to a chemical analysis lab. The lab has a high volume fume exhaust stack located on the roof of the building which emits sulfur dioxide at a rate of 2.5 m/s. The office building has an air intake off of a mechanical room on the third floor and on a mechanical penthouse on the building roof. A CFD air quality analysis is completed by analyzing the change in concentration levels of sulfur dioxide due to changes in wind velocities and changes in stack heights at the air intake on the third floor, in order to improve the indoor air quality for the people located inside the office building.

Numerical Modeling
Computational Domain
The first step in CFD model creation is the establishment of the modeling domain. Careful attention must be paid to the project site to ensure that all possible receptors are included in the domain. Air intakes and open windows are the most critical receptors for all types of exhaust.
After all of the receptors, sources, and buildings have been selected for inclusion in the model, the size of the domain can be established. The edges of the domain must be placed at a distance adequate enough to allow the full development of airflow features around the structures. A rule similar to the “5L” rule is used as a guide in air quality modeling to account for the building wake zone; this has been found to be useful in setting the domain size. Each edge of the domain must be at least Lb+ 4wm the distance from the nearest building edge where wm is the maximum wind speed modeled and Lb is the smaller of building height or projected building width (in consistent units). The domain roof must be tall enough that no signature of turbulence from lower layers affects the roof layers of airflow. A general rule of thumb for the height of the domain is at least three times the height of the tallest building in the domain. However, this can vary depending on thecae. For example, if a tall buoyant plume is being modeled in light winds, it will be necessary to have a height that can account for the high plume rise. If the plume insignificantly buoyant, it may still reach the domain roof. This does not create problems at the surface as long as any re-entrainment of the plume to sensitive receptors is fully accommodated.
The study has two buildings considered in the domain, the office building which is 30 meters in length, 20 meters wide and 30 meters tall, and the chemical lab which is 40 meters in length, 20 meters wide and 7 meters tall. The air intake is located at two locations on the office building one on the roof and other on the third floor of the building. The office building is 30 meters tall, 30 meters long, and 20 meters wide, so the office building Lb is 20. With Lb+ 4wm=52, our corresponding domain sides will be located at 50 meters (rounded to simplify the model creation process) from the office building. The top of the domain will be 90 meters above the surface. The chemical lab is 7 meters tall, 40 meters in length, and 20 meters wide. Then Lb+ 4wm=39 gives us a guideline distance of 40 meters to place the domain walls from the faces of the chemical lab. The computational domain utilized for this study was 180 m in length, 120 m wide and 90 meters tall. The domain mesh with a stack height of 2m was meshed with 80-m tetrahedral cells near the ground and faces of the buildings. These cells were allowed to grow at a rate of 1.16 away from these surfaces to minimize the computational effort associated with excess cells. The meshing scheme created approximately 433,708 cells in the domain.
The domain mesh with a stack height of 4m was meshed with 32-m tetrahedral cells near the ground and faces of the buildings. These cells were allowed to grow at a rate of 1.05 away from these surfaces to minimize the computational effort associated with excess cells. Although hexahedral cells are preferred for computational accuracy and efficiency, the geometry of the domain made it particularly difficult to utilize hexahedral meshing. The meshing scheme created approximately 799,500 cells in the domain.

Geometry

Figure 7: Complete mesh of domain and buildings.

Rooftop air intake
Rooftop air intake

Office Building
Office Building
Chemical lab
Chemical lab
Third floor air intake
Third floor air intake

Figure 8: Closer view of the mesh of the buildings.

Model Details and Boundary Conditions
Species Transport Model was used to study the plum dispersion in the environment at standard temperature of 300K and 1 atmospheric pressure. Fluent was employed in this modeling study. Steady-state computations with the standard k–ε turbulence closure (where k is turbulent kinetic energy and ε is turbulent dissipation rate) and Reynolds stress closure models were utilized.
The k–ε models are known to have weaknesses in complex flows, and incorrectly assume the dispersion coefficients are isotropic. However, it was an appropriate model to begin the computation, as it required less computational effort relative to that of the Reynolds stress model (RSM) or large eddy simulation (LES).
A continuous release of sulfur dioxide was simulated. The wind direction taken was parallel to the X-axis coming in from the right boundary of the domain which was taken as velocity inlet and the left boundary of the domain through which wind was going out was taken as pressure outlet. The front and back of the domain were taken as pressure outlets and the top of the domain was taken as moving wall. The ground and walls of the buildings inside the domain were taken as stationary walls and another velocity inlet was defined inside the stack of the chemical lab through which SO2 was coming out at 2.5 m/s.

Model Cases and Results
To characterize the level of concentration change around the office building, various iterations were run by changing wind velocities and stack heights. Cases were run having SO2 coming out of the 2m diameter stack at a rate of 2.5m/s. There were two stack heights taken (2m and 4m from the roof of the chemical lab) to compare the concentration level of SO2, at different wind velocities of 1m/s,3m/s, and 8m/s blowing in negative X-direction from the right boundary of the domain.

Below are figures of contour plots for the molar concentration of SO2 (kmol/m3).

Figure 9: Stack height 2m wind velocity1m/s.

Figure 10: Stack height 4m wind velocity 1m/s.

Figures 9 and 10 represent the concentration contour plots of SO2 for stack heights of 2m and 4m around the office building at wind velocity of 1m/s. The contour plots show that there is higher concentration of SO2 (red color being the highest and blue color being the lowest) at the air intake on the third floor of the office building for the 2m stack height than the 4m stack height. The plots also show that the concentration of SO2 at the roof of the office building is higher for the 2m stack height than the 4m stack height. The concentration of SO2 at the third floor air intake when the stack height was 2m was 0.000926 kmol/m3 and for a stack height of 4m, the concentration was 0.00069 kmol/m3.

Figure 11: Stack height 2m wind velocity 3m/s.

Figure 12: Stack height 4m wind velocity3m/s.

Figures 11 and 12above represent the concentration contour plots of SO2 for the stack heights of 2m and 4m around the office building at a wind velocity of 3m/s. The contour plots show that there is a higher concentration of SO2 at the air intake on the third floor of the office building for the 2m stack height than the 4m stack height. The plots also show that the concentrations of SO2at the roof location, for both of the stack heights, are almost same – same color blue on both plots. The concentration of SO2 found at the third floor air intake when the stack height was 2m was 0.000142 kmol/m3 and for the stack height of 4m the SO2 concentration was 0.000138 kmol/m3. It was determined that the overall concentration of SO2 decreased at both the air intakes, third floor and roof of the office building, when the wind velocity increased from 1m/s to 3m/s.

Figure 13: Stack height 2m wind velocity 8m/s.

Figure 14: Stack height 4m wind velocity 8m/s.

Figures 13 and 14 represent the concentration contour plots of SO2 for stack heights of 2m and 4m around the office building at a wind velocity of 8m/s. The contour plots show that there is a higher concentration of SO2 at the air intake on the third floor of the office building for the 2m stack height than the 4m stack height. The concentration of SO2 at the third floor air intake when the stack height was 2m was 0.00003256 kmol/m3 and for the stack 4m was 0.00002920 kmol/m3. It was determined that the overall concentration of SO2decreased at both the air intakes, the third floor and roof of the office building, when the wind velocity increased from 3m/s to 8m/s.

Below are the Velocity Vector Plots of Molar Concentration of SO2 (kmol/m3). These plots detail the velocity vector contour plots of flow of wind and SO2 around the buildings and the direction of flow of the concentration of SO2 along with the wind. The contour plots for wind velocities 3m/s and 8m/s show large recirculation zones, which can trap a certain amount of SO2 and create a higher concentration of SO2 in between the two buildings at lower ground levels. From the plots it is clear how the direction of the concentration is changed (red being the highest and blue being the lowest) at different wind velocities and stack heights. The plots detail that in all of the cases, as the wind velocity is increasing the concentration level is decreasing. When the stack height is taken into account it is seen that when the stack height level is increased there is less concentration of SO2 around the air intake on the third floor of the office building.

Figure 15: Stack height 2m wind velocity 1m/s.

Figure 16: Stack height 4m wind velocity 1m/s.

Figure 17: Stack height 2m wind velocity 3m/s.

Figure 18: Stack height 4m wind velocity 3m/s.

Figure 19: Stack height 2m wind velocity 8m/s.

Figure 20: Stack height 4m wind velocity 8m/s.

Detailed below are the continuous Pathlines of the Molar Concentrations of SO2 (kmol/m3).

Figure 21: Stack height 2m wind velocity 1m/s.

In Figure 21, when the stack height is 2m the number of SO2 particles entering the air intake on the third floor is higher than when the stack height is 4m with a much higher concentration. The figure also shows that the plume travels to a higher height when the stack is 4m than when the stack is 2m. It is the same method for all of the other wind velocities and stack heights.

Figure 22: Stack height 4m wind velocity 1m/s.

Figure 23: Stack height 2m wind velocity 3m/s.

Figure 24: Stack height 4m wind velocity 3m/s.

Figure 25: Stack height 2m wind velocity 8m/s.

Figure 26: Stack height 4m wind velocity 8m/s.

Detailed below are the SO2 molar concentration XY plots for the different stack heights of 2m and 4m, with a wind velocity of 1m/s.

Figure 27: Plot of SO2 molar concentration versus position for 2m stack height, wind velocity 1m/s.

Figure 28: Plot of SO2 molar concentration versus position for 4m stack height, wind velocity 1m/s.

Some more examples using CFD

Figure 29: Plume centerline streamlines (red streamers)

Originating from vivarium exhaust louvers, flow freely at roof level and do appear to stay at that elevation. However a portion of the dispersing edge of the plume is entrained in the lee recirculation zone and travels down the wall. A measurement plane is placed directly above the surface air intake of the building (the red and yellow streak at ground level on the near side of the building) to measure the concentration of exhaust entering the intake. Exhaust enters the building at a dilution of around 300, a potentially high odorous concentration. (The measurement plane only reports in concentration units, the inverse of dilution. So the 0.0025 on the red end scale is 400 dilutions while the 0.0017 in the yellow is 588 dilutions.)

Figure 30: Cooling tower impact at a building air intake

The blue and blue-green screen in the top view shows the roof-level exhaust readily entering the air intake. The bottom view shows the raised cooling towers, which reduced the impact of the cooling tower exhaust. Note the change in the scale of the measurement plane between the two views. For the upper view the blue-green portion represents only 125 dilutions while in the lower view even the red portion of the measurement plane represents 10,000 dilutions.

Figure 31: Exhaust plume from kitchen stack showing size of plume.

The inner 2,000 dilution isopleth is represented in yellow and the10,000 dilution outer isopleth is represented in red. The entire office building surface is treated as a measurement plane to determine odor levels at critical points along the building surface such as air intakes and operable windows.
Conclusion
Designing and positioning of a building’s air intake is very critical for healthy indoor air quality. Outdoor air may contain various pollutants from vehicle exhausts, stack exhausts, and other types of exhausts. These pollutants will be re-introduced into the building’s air intake if proper design and installation procedures are not followed. Therefore these pollutants will lead to much illness and infection of humans consuming the polluted air inside the semi-enclosed spaces, such as office buildings.
Computational Fluid Dynamics (CFD) was used by utilizing the k-ε turbulence closure approach for evaluation of the plume movement in urban microenvironments. CFD is a cost-effective tool for analyzing pollutant dispersion around buildings and sets of buildings. CFD is also a useful tool that helps guide the design placement of exhaust stacks and air handling units on buildings, and aids in investigating sources of odor which impact building air intakes. Better initialization schemes for the atmospheric conditions and improved turbulence closure models will improve the CFD results. Our CFD study resulted in SO2 concentration decreasing with increased wind velocities, and the SO2 concentration decreasing with increased stack heights.

References
[1]http://www.ashrae.org/pressroom/detail/17552
[2] http://www.epa.gov/iaq/schooldesign/hvac.html#Codes and Standards
[3]”Air Intake Placement – Recommendations From Years of Modeling Results”, by Aimée L. Smith, M.Eng., and Glenn D. Schuyler, P.Eng. Rowan Williams Davies & Irwin Inc.
[4]http://en.wikipedia.org/wiki/Indoor_air_quality
[5]http://en.wikipedia.org/wiki/Computational_fluid_dynamics
[6] Journal of Applied Meteorology and Climatology Volume 46; Computational Fluid Dynamic Simulations of Plume Dispersion in Urban Oklahoma City, by JULIA E. FLAHERTY*; Laboratory for Atmospheric Research, Department of Civil and Environmental Engineering, Washington State University, Pullman, Washington; DAVID STOCK Department of Mechanical and Materials Engineering, Washington State University, Pullman, Washington; BRIAN LAMB, Laboratory for Atmospheric Research, Department of Civil and Environmental Engineering, Washington State University, Pullman, Washington; (Manuscript received 13 September 2005, in final form 9 May 2006)
[7]www.envirometrics.com/abstracts/CFDUrbanMicro.pdf
[8] Estimating Odor Impact with Computational Fluid Dynamics, by Michael Ruby and J.D. McAlpine, Envirometrics, Inc.
[9] Turner, D. Bruce (1970) Workbook of Atmospheric Dispersion Estimates; AP-26; U.S. Environmental Protection Agency; Research Triangle Park, N Carolina
Thompson, Rhonda (2000) Preliminary Analysis of 5-Minute Maximum Ambient SO2
Concentrations; unpublished report available at www.epa.gov/ttnamti1/files/ambient/so2/techrept.pdf (10/23/03)
[10] Fackrell, J.E., 1984: Parameters characterizing dispersion in the near wake of buildings, Journal of Wind Engineering and Industrial Aerodynamics 16, 97-118.

Annex A
Xy plots

Wind 3m/s stack 2m

Wind 3m/s stack 4m

Wind 8m/s stack 2m

Wind 8m/s stack 4m

Annex B
Concentration results

Stack Height (m) | Wind Velocity (m/s) | SO2 Concentration (kmol/m3) | 2 | 1 | 9.2600E-04 | 4 | 1 | 6.9000E-04 | 2 | 3 | 1.4200E-04 | 4 | 3 | 1.3800E-04 | 2 | 8 | 3.2560E-05 | 4 | 8 | 2.9200E-05 |