AIR-EIA was funded under INFO2000
by the European Commssion (PUB-1221).

AIR-EIA:

air pollution and environmental impact assessment:
the multimedia information source

On-line Lecture:       Air Quality Modeling
        in Environmental Impact Assessment

 

LECTURE OBJECTIVES

The aim of this lecture is to introduce to the fundamentals of modelling air quality impacts of a typical project that requires a mandatory Environmental Impact Assessment (EIA): an incinerator for domestic waste.

The dispersion of emissions in the atmosphere is a complex spatio-dynamic process, where both the driving forces including control or decision variables as well as the impacts are distributed in space, and thus amenable to GIS analysis.

This introductory lecture is based on a simple Gaussian model commonly used for regulatory purposes, and associated GIS functions that together with the model results can help to forecast impacts for the assessment process.

In this lecture, a series of guided exercises will demonstrate how to use a basic Gaussian air quality model in a concrete problem setting:

Environmental Impact Assessment for a Waste Incinerator.

This will be done by using the simulation model to

  • generate estimates of a spatially distributed impact for each possible source location (the project alternative),

  • derive simple scalar indicators from these spatially explicit data, and

  • use them to rank the alternative locations for the assessment results, i.e., a preferred project.





What you should learn to do

After completing this lecture you should be able to
  1. Identify a decision problem where a pollution dispersion model or any similar spatially distributed model could provide useful information in the context of an environmental impact assessment;

  2. Understand the basic of the steady-state Gaussian air quality model;

  3. Define the logic and the GIS operations that are used to transform the raw data into useful information for spatial decisions;

  4. Use the spatial information coupled with a pollution dispersion model to feed the decision making process with relevant information and impact assessment evaluation.





Table of Contents

1.

Introduction

2.

Air pollution problems

3.

From emissions to immissions

4.

Emissions and pollutants

5.

Air quality assessment

6.

Simulation of air pollution

6.1

Dispersion modelling

6.2

The basic Gaussian model equations

6.3

Impact assessment

7.

The guided exercises:
Impact assessment of siting decisions

8.

Summary and The Final Test ...





1. INTRODUCTION

As we have seen in the previous lectures, a waste incinerator is a necessary, but generally unwanted facility; its location is subject to the so-called NIMBY syndrome (Not In My Back-Yard).

To find an acceptable, if not optimal location, we want to evaluate and compare several alternative sites as possible locations, and ultimately select one of the as the best, or least damaging, or generally accepted, candidate. To do this we use a multi-criteria analysis where criteria relating to economics, politics and environment are combined in an integrated assessment.

We have already seen how the nuisance due to the transport traffic generated by the incinerator could be traded off with the investment & operation cost of the facility. Environment is also impacted in another even more direct way by an incinerator: this type of facility will concentrate in a single source a very high output of air pollutants. Therefore the choice of a proper location of an incineration plant should take into account the impact on ambient air quality and the number of residents that will be exposed to higher levels of pollution concentration.

Using a Gaussian air pollution dispersion model, we will simulate the impacts of the same source, representing the incinerator, for each of the five possible locations and then compare the impacts to rank the alternatives. Impacts will be measured in terms of the ambient concentration of air pollutants caused by the source. To simplify the problem, a single pollutant, namely NOx, will be used.

The lecture is organised as follows:

  • we first recall the air pollution situation in a city like Geneva;

  • then we recall the fundamental principles used in the building of a Gaussian air dispersion model;

  • finally we go through an analysis of the impact, through the use of a interactive models permitting us to run a variety of dispersion models under different assumptions.





2. Air pollution problems

Most economic activities, involving the use and conversion of energy, and transportation prominently among them, are accompanied by emissions of air pollutants, thus degradating the environment, and in particular the urban environment.

Urban air pollution, in turn, is the source of a range of problems, including health risks with inhalation of gases and particles, accelerated corrosion and deterioration of materials, damage to historical monuments and buildings, and damage to vegetation in and near the city. Estimates have between 9.5 and 18 Million people spend a considerable part of their working day in roadside settings where air quality standards are most likely to be exceeded.

Road transport contributes, on average, more than half of the nitrogen oxides emissions, particulate matter, and about 35 % of VOC emissions. And, in contrast to stationary industrial and domestic sources which are decreasing, traffic generated emissions show a continuous increase in most areas.

The example of Geneva

Geneva is a city of about 200,000 inhabitants located at the end of Lake Leman, in a low-lying area, surrounded on three sides by mountain ranges. In winter, under windless high pressure anticyclones, a thermal inversion layer appears which traps the pollutants emitted by the city.

The most important sources of air pollution in Geneva are related to transport (71% of NOx emissions), heating systems (15% of NOx emissions) and industry (5% of NOx emissions).





3. From emissions to immissions

An emission is the compound or pollutant as it enters the atmosphere from an emission source, e.g., flue gas as it comes out of an exhaust or chimney. Often emissions are also called primary pollutants. Once emitted the pollutants are dispersed in the atmosphere, more or less quickly depending on the weather conditions. Numerous chemical reactions take place during this dispersion. This produces more stable compounds that can be detected and analysed in measuring stations. These are the "immissions", the pollutants that we breathe or which are deposited on the ground. In epidemiological studies one relates the immissions with health risks for the exposed population.





Background reading:

Fedra, K. (in print)
Model-based Decision Support for Integrated Urban Air Quality Management. In: Air Quality Management (Advances in Air Pollution Series), WIT Press.
On-line Paper (HTML)

Fedra, K., Greppin, H., Haurie, A., Hussy, C., Dao, Hy, and Kanala, R. (1996)
GENIE: An Integrated Environmental Information and Decision Support System for Geneva. Part I: Air Quality. Arch.Sci.Geneve, Vol. 49, Fasc.3, pp 247-263.
On-line Abstract




4. Emissions and pollutants

Many different substances and compounds are considered air pollutants. They are generally of anthropogenic origin, and result, in their majority, from combustion processes of fossil fuels. There are also natural pollutants such as gases and dust from volcanos, forest fires, or dust entrained by storms. While these may be of considerable importance on a global scale, it is the anthropogenic pollutants we are most concerned with in the urban environmental context.

From the perspective of spatial analysis, we can discriminate between:

  • fixed or mobile sources;

  • point sources,

  • line sources,

  • diffuse or area sources.

A few categories of pollutants have special significance:

CO2 and CFCs

These are substances with a global effect by contributing to the greenhouse effect and/or stratospheric ozone depletion.

Switzerland is active in promoting a reduction of GHGs (greenhouse gases). The cantons should be associated to this global goal.

CO2 results from any combustion process, but is also part of the natural photosynthetic and respiratory cycles. All vegetation, and forest in particular, fix and store CO2 as part of the photosynthesis and production of organic material, and the role of the oceans in this system may be a determining one.

When emanating from combustion processes, CO2 sources may be well defined point sources such as the stacks of power plants, or more diffuse, such as all the chimneys in a city, but also forest fires or the burning of organic residues.

Sulphur dioxide (SO2)

As one of the main atmospheric acidifiers, sulphur dioxide is perhaps the worst culprit in the gradual destruction of buildings and other objects of historical and cultural heritage exposed to ambient air.

However, in most industrialised countries considerable progress has been made in the reduction of sulphur emission through the use of low-sulfur fuels, and desulfurisation technologies cleaning effluent gas from industrial stacks.

Nitrogen dioxide (NO2)

Nitrogen dioxide is one of two nitrogen oxides emitted by car traffic and the combustion of various fuels such as natural gas, coal, and oil. It causes respiratory illnesses and damage to lung tissue and contributes to acid rain and smog. It also corrodes stone buildings, statues, and monuments including ancient sites of cultural heritage.

Particulates

Particulates are tiny particles so small that they remain suspended in the air for long periods. Particulates are released by the burning of wood, diesel and other fuels, by industrial plants, and by agriculture: ploughing, which raises dust, and the burning-off of fields, which produces smoke. In cities, road traffic - especially vehicles with diesel engines such as trucks and buses - is one of the main sources. Particulates are also formed by the chemical reaction of gases like SO2 and NO2 with moisture in the air. These are called secondary particulates and are as least as important as primary particulates.

Some classes of particulates are especially detrimental. Particles less then 10 millionth of a meter in diameter are called PM-10s. They have been linked to severe and sometimes fatal respiratory and cardiac ailments, higher rates of infant mortality, and cancer.

Fossil fuel combustion also gives off soot, sometimes known as black smoke. Soot consists of extremely fine carbon-containing particles that, in addition to being a health hazard, are responsible for the blackening of buildings and other outdoor structures in many cities.

Metals

Various metals can have adverse effects on health. Cadmium, mercury, arsenic, nickel and lead are emitted from industrial processes, energy production and transport. Mercury's most toxic derivative, methylmercury, enters the food chain and is toxic to the nervous system. Cadmium, arsenic, nickel and lead are considered carcinogenic. Lead also causes digestive problems and damage to the nervous systems, especially in children.

Benzene

Benzene is emitted mainly by motor vehicles and the chemical industry. Benzene depresses the central nervous system and is a well-known carcinogen.

Ozone

One well-known form of pollution is summer smog, which is mostly ozone (a form of oxygen with three atoms bound together in a molecule instead of two). Ozone is not emitted as such: it is a secondary pollutant that is formed from Nitrogen oxides and organic compounds through photochemical reactions.

The stratospheric ozone layer which exists at an altitude about 20 km above the earth's surface, is a vital protection against ultraviolet radiation. Yet the gas is severely detrimental to human health when inhaled and should not exist in any significant quantities close to the earth's surface.

Other pollutants, mostly nitrogen oxides and organic vapors, undergo chemical reactions in the air that produce ozone. These reactions increase in warm temperatures and strong sunlight, which is why many cities experience ozone problems in the summer.

Ozone causes breathing problems, reduced lung function, asthma, eye irritation, nasal congestion, and reduced resistance to colds and other infections. Ozone can be especially dangerous for the elderly and the young. It can also damage plants and trees and cause deterioration of rubber and fabrics.

Toxics

A typical example is dioxine, one of the toxic substances that can be found in the flue gas from waste incinerators if they are not operated correctly, i.e., at too low combustion temperatures.

Many other substances are polluting the atmosphere, but usually in very small amounts that may, at best, be of local significance.

The latest European Directive (Framework Directive on Air Quality Assessment and Management (96/62/EC) lists 13 substances of concern:

  1. Sulphur dioxide (SO2)

  2. Nitrogen dioxide (NO2)

  3. Fine particulate matter such as soot (including mw 10)

  4. Suspended particulate matter

  5. Lead

  6. Ozone (O3)

  7. Benzene

  8. Carbon monoxide (CO)

  9. Poly-aromatic hydrocarbons (PAH)

  10. Cadmium

  11. Arsenic

  12. Nickel

  13. Mercury

See, for example: European Commission, The ambient air quality framework directive - Clean air for Europe's cities. Luxembourg: Office for Official Publications of the European Communities, 1998 - 18 pp., ISBN 92-828-1599-4

For the moment we restrict our attention to NOx and in particular the immissions of NO2.





Background reading:

For a summary introduction to the European Unions' regulations concerning air quality, browse this on-line SLIDE SHOW.



5. Air quality assessment

The assessment of air quality is usually based on monitoring data. Regular, continuous measurement are mandatory in most countries for ambient air quality, but also of emissions from major sources such as power plants above a certain size threshold, e.g., 50 MW.

Monitoring, however, is not a viable approach if we want to evaluate the impacts of a source that is not yet in operation, like the incinerator for which we wish to find a feasible location: this requires a forecasting method like simulation modelling.

WHO, EU regulations

Swiss federal law

The Swiss federal law on the protection of the environment has been enacted in 1985 and revised in 1995. The ordinance on the protection of air (OPAIR) sets limits for the immissions of pollutants. We give below the OPAIR norms concerning NO2.

Yearly average 30 µg/m3
95% of hourly average over a year 100 µg/m3
Daily average, not to be exceeded more than once a year 80 µg/m3





6. Air Quality Modelling

The factors that affect the transport, dilution, and dispersion of air pollutants can be grouped into:

  • emission or source characteristics

  • the nature of the pollutant material

  • meteorological characteristics

  • the effects of terrain and anthropogenic structures.

Source characteristics

Most industrial pollution is discharged vertically from a stack or duct into the open air. As the contaminated gas stream is emitted, the plume (body of polluted air) expands and Plume means the body of polluted air, Wind, that is horizontal air movement will bend the plume in the downwind direction. AT some distance from the source, the plume will level off. While the plume is rising, bending, and starting to move with the wind in the downwind direction, the flue gas is being mixed and diluted by the ambient air. As the gas is being diluted by increasing volumes of air, the contaminant will eventually reach the ground.

The initial rise of the plume is due to the upward inertia of the gas stream exiting the stack, and by its buoyancy. The vertical inertia is related to the exit velocity and mass of the gas. The buoyancy is related to the density relative to the surrounding air, primarily determined by temperature. Increasing exit velocity, and increasing exit temperature will increase the plume rise.

The plume rise, together with the physical stack height, is called the effective stack height

For a given set of stack and discharge conditions, the ground level concentration is proportional to the mass flux, i.e., the amount emitted per unit time. Increasing emission rates will therefore lead to a proportional increase in ambient concentrations.

Downwind distance

The greater the distance from the discharge point, the greater the volume of air available for dilution. However, since the plume starts above the ground and needs some time to reach the ground (by bending and spreading), there is no concentration observable in the immediate vicinity of the stack, then we can observe an increase for some distance as the plume approaches the ground. After this, the ground-level concentration will decrease with increasing distance from the emission source.

Wind speed and direction

The wind direction will determine the direction in which the plume will move across local terrain. Wind speed affects the plume rise (fast wind will bend the plume faster), and will increase the rate of dilution.

Thus, the effects of wind speed work in two opposite directions:

  • increasing wind speed will decrease plume rise, thus increase ground level concentrations;

  • increasing wind speed will increase mixing, thus decreasing ground level concentrations.

Depending on the specific conditions, one or the other of these phenomena will prevail. These effects also determine the distance from the source where the maximum concentration will occur.

Atmospheric stability

The tendency of the atmosphere to resist or enhance vertical motion and thus turbulence is termed stability. Stability is related to both the change of temperature with height (the lapse rate) and wind speed.

A neutral atmosphere neither enhances nor inhibits mechanical turbulence. An unstable atmosphere enhances turbulence, whereas a stable atmosphere inhibits mechanical turbulence.

The turbulence of the atmosphere is by far the most important parameter affecting dilution of a pollutant. The more unstable the atmosphere, the greater the dilution.

Stability classes are defined for different meteorological situations, characterised by wind speed and solar radiation (during the day) and cloud cover during the night. The so called Pasquill-Turner stability classes (based on D. Bruce Turners Workbook of Atmospheric Dispersion Estimates include six stability classes:
1 A very unstable
2 B unstable
3 C slightly unstable
4 D neutral
5 E stable
6 F very stable

Mixing height

Inversions result from the vertical temperature profile of the air: while temperature normally decreases with altitude (on average, at a rate of one degree Centigrade per 100 meters), an inversions describes an increase of temperature with increasing height. This results in a stable temperature profile (the colder air is stable under the upper warmer layer), restricting vertical mixing, and exacerbating pollution episodes through restricted mixing volumes.





Exploring the behaviour of a single point source

To explore the behaviour of a single source (industrial point source with an elevated stack), try this Java applet that demonstrates the interdependency of meteorological and stack parameters and shows estimated concentrations downwind from a stack in 1D or 2D graphs, respectively.

User selected parameter are:

  • Emission rate (g/s)

  • Wind speed (m/s)

  • Air temperature (degree C)

  • Stability Class (Pasquill)

  • Stack height (m)

  • Stack diameter (m)

  • Exit temperature (degree C)

  • Distance from stack (display parameter)

  • 1D/2D switch (display parameter)

The applet shows ground-level concentration versus downwind distance from the stack, as well as the plume rise versus the physical stack height. Two scenarios (current and last) can be compared in the 1D case.

Use the blue button to start the model applet - make sure your browser is Java enabled !






6.1 Dispersion modelling

A dispersion model is a mathematical description of the meteorological transport and dispersion processes, using source and meteorological parameters, for a specific period in time. The model calculations result in estimates of pollutant concentration for specific locations and times.

The NOx emission rate of a waste incinerator is a (technical) data. Among the criteria used by the decision makers to evaluate the different options there is the increase of population exposure to high concentration of NO2. Therefore, in order to transform the data concerning the NOx emissions into information, useful for the decision process, one needs to use a model that simulates the impact of locating the NOx source in a specific place on the average increase of NO2 concentrations in different locations of the urban community. This is typically performed by a model of air pollution dispersion which simulates the spatial dynamics of the pollution dispersion process.

One calls "plume" the picture representing the dispersion of a pollutant in given region. We can use a model to compute a short term or a long term average plume. The short term representation depends on the particular weather conditions, in particular the wind direction and speed, the temperature, the humidity, etc... The long term average is influenced by the distribution over a long period, typically a year, of different weather conditions. The wind rose is a graphical representation of the frequencies of the different wind directions.

To see the average air pollution plume resulting from NOx emissions by a waste incinerator of capacity 120'000 tones/year, you can run the ISC3 long-term model: press the blue Long-Term button to start the modeling tool for the incinerator as the emission source.

The default parameter describe a 30 meter stack with 60 cm diameter, exit temperature of 80 degrees Centigrade and an exit velocity of 1 meter per second. The emission rate is set to 100 g/s (the real emission of Cheneviers is closer to 14 g/s) and the impact threshold is set to 10 µg per cubic meter.

The plume is being computed through the use of a Gaussian dispersion model. The plume is averaged over all the weather conditions observed over a period of one year.

Air pollution dispersion models are representations of the complex spatial dynamics of particulate emissions under different wind orientation and stability classes. The topic 1.5 is dedicated to a more detailed presentation of the different classes of air pollution models. We can see on the map of Figure 1 the Geneva population density layer, as represented in the GIS of AIDAIR. The intersection of the immission and population layers will permit an assessment of the increased population exposure to NO2 concentrations, for the different site selections.



6.2 The Gaussian steady-state dispersion model

The most commonly used model for regulatory purposes is the so-called Gaussian steady-state model. It provides a steady-state solution to the transport and diffusion equations (transport plus diffusion = dispersion). Steady-state implies that the basic assumption is a constant emission and constant meteorological conditions:

The basic Gaussian diffusion equations assumes:

  • that atmospheric stability and all other meteorological parameters are uniform and constant throughout the layer into which the pollutants is discharged, and in particular that wind speed and direction are uniform and constant in the domain;

  • that turbulent diffusion is a random activity and therefor the dilution of the pollutant can be described in both horizontal and vertical directions by the Gaussian or normal distribution;

  • that the pollutant is released at a height above the ground that is given by the physical stack height and the rise of the plume due to its momentum and buoyancy (together forming the effective stack height);

  • that the degree of dilution is inversely proportional to the wind speed;

  • that pollutant material reaching the ground level is reflected back into the atmosphere;

  • that the pollutant is conservative, i.e., not undergoing any chemical reactions, transformation or decay.

The model equations

The spatial dynamics of pollution dispersion is described by the following type of equation in a Gaussian model:

where

  • C(x, y, z) : pollutant concentration at point ( x, y, z );

  • U : wind speed (in the x "downwind" direction, m/s)

  • σ represents the standard deviation of the concentration in the x and y direction, i.e., in the wind direction and cross-wind, in meters;

  • Q is the emission strength (g/s)

  • Heff is the effective stack height, see below.

From the above equation one can deduce, in steady state, the concentration in any point ( x, y, z ) in the model domain, from a constant emission rate.

The effective stack height (physical stack height plus plume rise) can be computed, for example, using the formula below:

where:

 

  working on it ....




 

6.3 Impact assessment

The pollution (emission) and its impacts (immissions) caused by a given source depends upon:

  • The source characteristics like emission rate, stack height, and stack parameters like diameter, emission temperature and the momentum of the flue gas (its exit velocity);

    Since these parameters can be assumed to be the same for each location (they are spatially invariant), they will not enter the analysis. They will be kept constant in a reference scenario.

  • The weather characteristics: depending on the distance of the different sites from each other, we can assume that they are subject to the same meteorological conditions (if they are close to each other), or that for each, a different micro-climate applies (e.g., a different wind rose);

    Therefor, the meteorological conditions may or may not be spatially invariant, depending on location.

  • The location of the source relative to the target areas. While the solution of the basic Gaussian equation is again spatially invariant (i.e., the results for the same emission and meteorological parameters will always be the same irrespective of the absolute location of the source), the impacts depends on the characteristics of the target area such as landuse and population density: If the area near a source, or downwind of it is heavily populated (an urban center) the impacts will be considered greater than if that area is under, say, agricultural landuse.

Thus, to measure impacts, we can define an impact function that expresses the relationship between concentration and target area. For example, we could simply count the number of people near the source, e.g., in a given radius around it.





Question:   which standard GIS operation would you use to do this ?
Try the multiple-choice test !



Alternatively, the impact function F that we want to minimise, can be much more complicated, for example:

where we sum, over N spatial elements (e.g., grid cells) the (positive) difference between the ambient concentration minus an no-effects threshold or standard, raised to some power to express the non-linear nature of exposure, and multiply this exposure index with a special weight derived from the land use in each parcel or grid cell W.

The resulting value will describe the impact of a given level of pollution distributed over a given area or domain.

The five locations are (we assume they have been found in a pre-feasibility study as the only possible candidates in the region):

  • Cheneviers

  • Bois de Bay

  • Z.I. Meyza

  • Velodrome

  • les Rupières

The basic difference of these locations that will determine their relative impacts is their position relative to the sensitive areas, namely the populated zones of the City, and the orography, that will influence the wind fields. If we would be using the data from several meteorological stations, each possible location would also have a somewhat different wind rose that drives the transportation of pollutants.





QUESTION:   which of the above features are relevant when a Gaussian model is used ?

QUESTION: How can we justify using a single pollutant ?
Try the multiple-choice test !



7. The guided exercises:
    Impact assessment of siting decisions

To make the comparison and ultimate decision between stations easy, we will use a single variable describing impact: the number of people exposed to concentrations above a threshold value such as those defined by the OPAIR ordinance of 1992, introduced above.

This value will be determined by a post-processor of the model, the calculates the number of people from their spatial distribution, and the concentration field.





Question:   which standard GIS operation is used for this analysis ?



This indicator for environmental and public health impact could be computed for an infinite number of possible weather and emission scenarios. To make the comparison practical, we have to choose standard scenarios and apply them to each source location in turn. These scenarios should be based on meaningful conditions:

  • on the one hand, they should be related to the air quality standards and in particular, the averaging periods defined there:

    • one hour

    • one day

    • one year

  • on the other hand, we may wish to define extreme conditions (worst case) to make sure that the location selected also meets our expectations under these conditions, even if they should occur less frequently as the 95% rule of the air quality standard suggests.

In consequence, we will use the model for:

  • a one-hour averaging period, representing worst-case assumptions;

  • the annual average based on long-term weather frequency data.





Question:   for a single point emission source from a high stack, what would constitute a worst-case scenario (in terms of maximum or minimum values of meteorological and emission parameters) ?
Can this situation be realistically simulated with a Gaussian model ? Does the model formulation introduced above allow us to represent this situation ?



Exercise No. 1

The purpose of the exercise if to fill a table like the example shown here at the right;

the numbers are representing environmental impact obtained by the model runs (remember, the number of people exposed above a certain threshold value will be used as an indicator of impacts):

Location Short-term impact Long-term impact
Cheneviers    
Bois de Bay    
Z.I. Meyza    
Velodrome    
les Rupières    

To make sure the results are comparable, select a weather scenario and keep it constant for each location ! Since each location may have a different default scenario defined, be careful to edit the scenarios where necessary to make them consistent !

Start with the short-term episode (representing worst case assumptions); you can start the modelling tool with the Short-Term button:

For each run, select a different source location, but keep all other parameters constant: emission rate, stack charactersitsics, and the meteorological scenario.

From the model output, not that impact assessment, expressed as the number of people exposed to concentrations beyond the threshold value (which must be kept constant, of course).

Note this value against the respective site.


This should lead to a table of results similar to the one shown here on the right:

The exercise should result in a single indicator, and therefor also in a clear ranking: in the example above, location No. 5 (les Rupières) has the highest impact, and the Velodrome the lowest, thus suggesting the most (and the least) desirable location by our single criterion.

Location Short-term impact Long-term impact
Cheneviers 4,689  
Bois de Bay 1,633  
Z.I. Meyza 1,901  
Velodrome 1,325  
les Rupières 5,344  

Exercise No.2:

Obviously, the numbers in the table would be very different if you choose different scenario assumptions. There are quite a few parameters represented in each scenario, and all will change the numbers: but will they also affect the ranking (and thus our decision ?

The explore this relationship, test as many parameters as possible to determine whether they change the ranking or not ! The results of this exercise should be another table that lists the parameters, and then notes whether they affect the ranking or not, like in the example shown on the right:

Before you run the model again, think about the likely outcome and mark your assumptions in the Table !

Re-run the short-term episode (worst case):

Parameteraffects ranking
threshold yes/no
threshold yes/no
emission rate yes/no
stack height yes/no
wind direction yes/no
wind speed yes/no
stability class yes/no
inversion height yes/no

QUESTION: Which of the scenario parameters do affect ranking ?

Exercise No. 3

The next step is now to repeat the first test but with the long-term model. Instead of a single episode, this now represents an entire year.

To simulate this, the models uses the relative frequency of different classes of weather (represented by wind speed, wind direction, air temperature, mixing height, and stability class), computed a solution for each class, and than adds the class results weighted by their relative frequencies. This is computationally much more efficient than simulating each of the 8,760 hours of a year one by one !

To summarize the results from the annual average assessment, use the same table as before and complete it now in the third column by adding the exposure values for the long-term model:

  • How do you determine the threshold value to use ?

  • What happens when you use the same threshold as for the short-term model ?

  • Why should the short-term and long-term thresholds be different ?

Location Short-term impact Long-term impact
Cheneviers 4,689  
Bois de Bay 1,633  
Z.I. Meyza 1,901  
Velodrome 1,325  
les Rupières 5,344  

Question:   there may be a good reason to simulate each hour anyways: can you name it ?

Run the long-term model for each location, keeping al other parameters (emission and stack characteristics, meteorological period of one year) constant:

How do the short-term and long-term results compare ?

Why are the concentrations you are getting much lower than in the short-term episode model ?

Do you get the same ranking of the five sites ?

If you get the same ranking, the solution seems obvious: select the location with the minimum impact.

But what to do when the table looks like the example here on the right:

While the least desirable locations remains the same, the one with the minimum impact is different for the short-term and long-term assessment:

This leads to a multi-criteria approach -- and another Lecture !

Location Short-term impact Long-term impact
Cheneviers 4,689 1,872
Bois de Bay 1,633 4,083
Z.I. Meyza 1,901 12,435
Velodrome 1,325 12,206
les Rupières 5,344 14,370





8. Summary and a final test

FINAL TEST: Are you ready to take the final test ?

© Copyright 1995-2002 by:   ESS   Environmental Software and Services GmbH AUSTRIA