INIBAP — Annual Report 2003

Future traffic scenarios inKaliningrad have beenevaluated with the

configured ARIA Regionalmodeling system in order toevaluate various emissionabatement strategies to reduceair pollution from traffic. Theaverage target year for theimplementation of the differenttraffic scenarios.was set to the year 2015. Six different scenarios wereelaborated by grouping thespectrum of policy guidelinesemerged during meetings withECAT and the Stakeholder’sreference group (SRG), with thetechnical support of ARIA andARIANET experts. Scenario 1referred to the situation“business as usual” year 2015,while the other scenarios couldgenerally be identified as eithermeasures related to changes ininfrastructures and trafficconditions (Scenario 2 to 4) orto improvements in vehicle fleetand fuels (Scenario 5 and 6). The expected increasing trendin traffic volumes will in thefuture generate a significantload on the present road

network. However, consideringpollutant emissions, this effectwould to some extent becounterbalanced by theintroduction of newer vehiclesequipped with more efficienttechnologies for pollutantreduction. The measures related to theincrease of publictransportation (Scenario 4) haverevealed the best results, withclear improvements of airquality over a substantial partof the city. Renewal of vehiclefleet and fuels (Scenario 5and 6) are also expected to leadto improvements of air qualityover the whole urban area. Theimpact observed was generallymore significant in the centralparts of the city than in thesuburbs. Even though infrastructuralupgrades of the network(renewal and construction ofbridges, completion of theKaliningrad ring road), certainlywill rationalize the distributionof traffic flow locally, they’renot expected to significantlychange the overall load on thenetwork.

In order to know the actual airpollution situation in variousparts of the city, fieldexperiments with passivesamplers (diffusion tubes) werecarried out at 50 different points.Measurments of NO2 and BTEX(benzene, toluene, ethylbenzeneand xylene) were made duringtwo periods of 15 days chosen inorder to represent the typicalconditions during both warmand cold episodes.

ECAT-Kaliningrad, ARIA andASCAL (French laboratory,

Forbach) selected jointly the measurement points inorder to account for different areas of interest (neartraffic, urban, periurban, rural, schools and hospitalsvicinity …). ECAT staff were trained to use, handle andtransport the passive tubes by ASCAL specialists.

The passive tubes were then prepared by ASCALspecialists and installed by ECAT at the selected points.At the end of the campaign the samples were analyzedby ASCAL and the results were compared with theEuropean and Russian legislation as well as withmodeling results.

Model results were found consistent with the controlfield experiments with hot spots observed along themain roads and in the city center. The FARM model yieldsresults that are generally close to measurements whereasthe model output from ARIA Impact corresponds toapproximately 40-50% of the measured concentrationsof NO2. The difference is mainly related to thecontribution of industrial sources not included in thestudy with ARIA Impact. To make a more detailedcomparison of model results with measurement data amore complete inventory of the emission sources inKaliningrad and neighboring regions would have to becarried out, integrating harbor activities, all the mainpoint sources and background concentrations of NO2

(long distance transport of pollutants). This work washowever not included in the scope of the project.

Passive samplers

Field experiments

Evaluation of scenarios

Comparison of total traffic emissions over the road network for scenarios 1 to 6The selected scenario were :1. Business as usual2. Construction and renovation of bridges and bridge passages 3. Development of city road network and optimization of traffic flows 4. Improvement of public transport5. Renewal and improvement of the Kaliningrad vehicle fleet 6. Improvement of fuel quality

Difference in NO2 concentration (µg/m3) for year 2015 “business as usual”with respect to year 2006

ECAT-KaliningradMunicipal Institution Environmental CentreGorkogo 25 PO Box: 236029Kaliningrad Russia Tel.: +7 401 2 957711 Fax: +7 401 2 965380 E-mail: ecatk@mail.ru

ARIA Technologies SA 8/10, rue de la Ferme

92100 Boulogne-BillancourtFrance

Tel.: +33 (0)1 46 08 68 60Fax: +33 (0)1 41 41 93 17

E-mail: info@aria.fr Web: www.aria.fr

ARIANET s.r.l.Via Gilino, 9

20128 MilanoItaly

Tel.: +39-02/27007255Fax: +39-02/25708084

E-mail: info@aria-net.it Web: www.aria-net.it

ASCAL SA Parc d’Activités Forbach Ouest

57600 ForbachTél. 03 87 83 48 00 Fax. 03 87 83 48 35

E-mail: contact@ascal.fr

Clean Air for Kaliningrad

PlaquetteKALININGRADenglishV7:Maquette RIO Renault V7 6/02/09 21:59 Page 2

The urbanair pollutioninduced bytraffic in

Kaliningrad is raising increasing concern.KALAIR is a LIFE-Third Countries Project,co-funded by the European Community,designed to aid the city in the evaluationof abatement strategies for reducing airpollution caused by traffic. The projectwas implemented at the KaliningradMunicipal Environmental Centre,ECAT, in collaboration withARIA Technologies, during theyears 2007-2008.

Clean Airfor Kaliningrad

T he main objective of theproject was to jointlyconstruct a modeling tool

to be used by ECAT in order tostudy the mechanisms or airpollution induced by traffic inthe Kaliningrad environment.The modeling tool has beenvalidated through on-siteexperiments, involvingcomparisons with available airpollution datasets fromcontinuous monitoring stations,as well as with specific fieldexperiments using passivesamplers (diffusion tubes). The validated modeling systemcan be used by ECAT to studythe effect of urbanmanagement decisions in orderto reduce air pollution levelsobserved in the City, such aschanges in fuel composition,improvements in the car fleetquality, increases in public

transportation offers, changesin the city traffic scheme andtraffic permissions, etc. An operational modelingsystem, including a trafficmodel, a meteorologicalmodel, an emission model andair pollution dispersion modelswere delivered to ECAT. Thesetools allow the simulation ofvarious emission abatementstrategies as well as theirconsequences on air pollutionlevels. By comparing the modelresults with the costs of eachreduction strategy the optimalmanagement decisions canthereafter be obtained. The outcome expected in long-term is a reduction of airpollution levels induced bytraffic at the "hot spots" of thecity of Kaliningrad, through theapplication of optimal emissionreduction strategies.

on critical substances and onthe correct representation ofparticles. Dispersion modelinghas been performed through astate of the art modelingsystem, and the outputconcentration fields have beencompared with two categoriesof data: existing air pollutiondata at federal monitoringstations, and data from specificfield experiments using anetwork of passive samplers(diffusion tubes) for benzeneand NO2.

Traffic countingwas carried outby ECAT ondifferent roadsin Kaliningradduring threedifferent days

and at five different time periods ofthe day. Based on this informationthe Kalingrad vehicle fleetcomposition could be defined. Thetraffic counting was divided intodifferent vehicle categories (cars,trucks, trams, trolleybuses, buses andminibuses). The observations weregathered in a GIS databasecontaining the Kaliningrad roadnetwork.

CARUSO

Traffic simulations were realized withthe CARUSO (CAR Usage SystemOptimization) traffic model. Theobjective was to configure anddeliver a flexible PC-based trafficmodelling system providing resultsthat could further be used forcalculating emissions from traffic inthe city. CARUSO is a static modelbased on one of the majoralgorithms for determining trafficflow on a transport network and toestimate Origin/Destination (O/D)matrixes.

The model was populated withKaliningrad data, tests of thesimulation quality was carried out

and ECAT personnel were trained onthe theory and practice of using themodel. By using real Kaliningradtraffic data, ECAT technicians had thepossibility to start working directlyon a concrete case and to apply theirown experience on the local trafficsituation.

The original GIS road network forKaliningrad was composed of morethan 900 streets. According to ahierarchical classification, a subset of410 streets was selected andintegrated in the CARUSOsimulation. The traffic density wasbased on the average vehicle flowtaken from traffic counting at threedifferent hours of the day: 08:00,12:00 and 17:30.

The results from the trafficsimulation show that the main tripsoccur within the city of Kaliningradand that only a smaller amount isdue to the exchange with thesurrounding regions.

Traffic data

Assessing the traffic

Results of CARUSO simulation

TREFIC

TREFIC (“TRaffic Emission FactorImproved Calculation”) has beenconceived to answer to the manyspecific requests which arise whencalculating the atmospheric pollutantemissions from on-road vehicles. Themodel calculates the emissions fromtraffic for ~120 different pollutants.It integrates data related to:

• Road Network (including trafficflow, percentage of heavy vehicles,vehicle speed, road type…)

• Fleet distribution (dividing vehiclesinto more than 100 sub-categories)

• Emission factors (Based onCOPERT IV methodology)

• Fuel characteristics (basedon annual fuel sales inKaliningrad)

The calibration of theTREFIC model wasperformed for a situationcorresponding to year2006 during peak hourin Kaliningrad. Thespecification of Russianfuels and the Kaliningradvehicle fleet was based ondata provided by theFederal Inspection on TrafficSafety and the FederalStatistical Service in theKaliningrad Oblast(Kaliningrad Stat).

Emission models

From traffic to emissions

Results of TREFIC emission modeling for Kaliningrad

Another key result of theproject was the cross-transferof technology and experiencesbetween European and Russianpartners, obtained throughextended joint work on datacollection, experiments andmodeling. The modeling tools take intoaccount urban effects and sitecharacteristics. Physicalgeographical data (terrain,land-use) has been gathered ina GIS-type environment,together with geographicaldata on the road network,urban landscape and humangeographical data onpopulation and transportationdemand. Traffic data has been collected,including observations ofvehicle flow and statistical dataon traffic demand at varioustimes of typical days. Thesedata have been synthesizedthrough a traffic model,adapting the results fromtraffic demand allocationalgorithms to existingmeasurements on links. Emissions from traffic flowwere computed using the mostrecent techniques applied inEurope for this type ofproblem, with a special focus

DisseminationeventsThe dissemination program ofthe present project wastargeted to the stakeholdersinvolved in the officialdecision making (responsiblefor the definition ofscenarios), all the institutionsand companies dealing withtransport issues as well as thebroad public. The final projectresults were presented at aseminar in the KaliningradCity Hall in October 2008.

A project web-site (www.kalair.ru) has finally beendeveloped including all the project information, detaileddescription of project activities, presentation of theinvolved partners, the main deliverables and characteristicsof the modeling tools applied.

The figures (right) show the modelreproduction of two typicalatmospheric conditions: “unstable”(middle of the day) and “stable”(evening). During unstableconditions, the well-developedturbulence effectively disperses thepollutants emitted near the ground,diluting them over a large volumeof air, and thus decreasing theconcentrations at ground level.The situation is oppositeduring stable conditions,where the emittedpollutants are confinedin a thinner layer,increasing theconcentrations atground level.

Two different dispersion models (ARIA Impactand FARM) have been implemented in order tocompute maps of pollutant concentrationsinduced by traffic emissions. The simultaneoususe of two models allows to choose between aquick Gaussian model able to simulate the long-term impact on air quality over one or severalyears, and a more sophisticated Eulerian modelrequiring longer CPU times but providing betterresults in complex situations.

ARIA Impact

ARIA Impact is a Gaussian model developed byARIA Technologies, designed for evaluating long-term impact of emissions from industrial sites,vehicular traffic and diffuse sources. The modelconforms to the norms of the EnvironmentProtection Agency (EPA), United States.

The principle is to simulate one or several yearsof activity by using representative meteorologicaltime series. The software allows time-dependentmodulations including daily, monthly, and annualvariations of emissions for realistic simulations ofair quality.

Modelling with ARIA Impact over Kaliningradwas performed for the average situationcorresponding to year 2006 in order tocharacterize the long term impact on air quality.ARIA Impact was configured for the Kaliningradsituation (emissions, topography andmeteorological data), tests of the simulationquality was carried out and ECAT personnel weretrained on the theory and practice of using theARIA Impact dispersion model.

ARIA Regional / FARM

The FARM dispersion model was applied for twodifferent meteorological episodes correspondingto “average” and “worst case” conditions. Thefinal goal was to obtain reliable referencesimulations of pollutants dispersion over theKaliningrad area, to be used in the analysis of aset of future emission scenarios related to traffic.

The FARM dispersion model employed is a three-dimensional Eulerian dispersion model includingthe treatment of chemistry in gas and aerosolphase, developed by ARIANET on the basis of theSTEM model, an U.S. academic model with a verylong application record in many parts of theworld. FARM can be applied in different contexts,from scenario evaluation to daily pollutionforecasts, on time scales ranging from national toregional and urban, and it is currently used inItaly to evaluate national-scale policy scenarios,on behalf of Ministry of the Environment.

Dispersion models

From emissions to air pollution

PlaquetteKALININGRADenglishV7:Maquette RIO Renault V7 6/02/09 21:59 Page 5