1 National Environmental Ingenieurbro Lohmeyer Motivation - - PDF document

1

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 1 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

UPDATE OF THE ROMBERG-APPROACH AND SIMPLIFIED NO / NO2 CONVERSION MODEL UNDER CONSIDERATION OF DIRECT NO2- EMISSIONS

• I. Düring1, W. Bächlin2, M. Ketzel3, A. Baum4, S. Wurzler5

1Ingenieurbüro Lohmeyer GmbH & Co. KG, Mohrenstraße 14, D- 01445 Radebeul 2Ingenieurbüro Lohmeyer GmbH & Co. KG, An der Roßweid 3, D-76229 Karlsruhe 3Department of Atmospheric Environment, National Environmental Research

Institute, Aarhus University, Frederiksborgvej 399, DK-4000 Roskilde, Denmark

4Bundesanstalt für Straßenwesen, Brüderstraße 53, D-51427 Bergisch Gladbach 5Landesamt für Natur, Umwelt und Verbraucherschutz NRW, Wallneyer Straße 6,

D- 45133 Essen

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 2 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Outline

• 1. Motivation
• 2. The Romberg approach
• 3. Updating the Romberg approach
• 4. OSPM Chemistry Model
• 5. Simplified chemistry model based on annual average values
• 6. Conclusions

2

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 3 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Motivation

Jagtvej NO2 (µg/m

3)

10 20 30 40 50 60 70 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 urban background NO2 (measured) urban background O3 (measured) measured Jagtvej modelled with COPERT4 emissions Jagtvej NOX (µg/m

3)

20 40 60 80 100 120 140 160 180 200 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 urban background (measured) measured Jagtvej modelled with COPERT4 emissions

NO2-limit value: 40µg/m³ (annual mean) Air quality measurements at street stations show often reduction in NOx levels, but no significant reduction in NO2 – NOx = NO + NO2 [ppb] – NO + O3 => NO2 +O2

… … more chemistry … …

limit value

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 4 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Motivation

According to the EU directive the EPA in co-operation with local authorities has to prepare action plans to assure compliance by 2010

Concentration and Emission Frankfurt-Leipziger Straße

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 year Concentration-annual mean [µg/m³] 200 400 600 800 1000 1200 1400 1600 1800 2000 Emission [mg/(mh)] NO2-concentration (annual mean) NO-concentration (annual mean) NO2-emission NOx-emission

3

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 5 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

NOx- chemistry

Formation of NO2 from NO is a complex photochemical process. Mainly depending on the total amount of available NOx and

• zone.

NO2 concentration is also depending on the primarily emitted NO2. Besides NO-emission, ozone concentrations as well as primary NO2 emission are changing from year to year.

(Platt U., 2008 )

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 6 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Updating the Romberg approach

] [ ] [ ] [ ] [

2 x x x

NO C B NO NO A NO ⋅ + + ⋅ =

Statistical concentration parameters parameter of function Romberg et al. (1996) A B C

annual average 103 130 0.005 98-percentile 111 119 0.039

Bächlin et al. (2008) A B C

annual average 29 35 0.217 98-percentile 40 20 0.170

• 19. highest hourly value

43 10 0.151

The new constants for the given period of 2004 to 2006 (German measurement sites) along with the earlier parameters from 1996:

4

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 7 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Comparison old / new

20 40 60 80 100 120 140 50 100 150 200 250 300 350 400 450 NOx [µg/m3] NO2 [µg/m3] measurements Romberg et al. (1996) Bächlin et al. (2008)

88 µg/m³ 81 µg/m³

limit value

The approach by Romberg et al. (1996) results in too low NO2-values for high NOx-values. The annual average value of 40 µg NO2/m3 would be reached by an annual mean NOx-concentration of approx. 88 µg/m3 with the new approach.

NO2 = f (NOx p, O3)

1996 2008

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 8

Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

OSPM Chemistry Model

More realistic conversions can be obtained using a chemistry model. A simplified relation scheme is often used to describe the chemical conversion in the equilibrium state of NO2, NO, and ozone. The considered reactions are: NO + O3 => NO2 + O2 (k) NO2 + hv => NO + O3 (J) k (ppb-1 s-1) is the reaction coefficient J (s-1) is the photolytic frequency of NO2,

5

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 9 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

OSPM Chemistry Model

Assuming that the equilibrium is quickly reached, the differential equations have an analytical solution for the concentration:

[ ] [ ] [ ] [ ]

( )

( )

τ + = k / NO NO NO 4

• B
• B

5 . NO

n 2 O 2 x 2 2

With the variables :

[ ] [ ] [ ]B

2 V 2 n 2

NO NO NO + =

[ ] [ ] [ ]B

n O

O NO NO

3 2 2

+ =

[ ] [ ]

      + + + = τ 1 J k 1 NO NO B

O 2 x

The two terms [NO]V and [NO2]V are the amount being produced by traffic emissions as an increase of concentration above the background

• concentration. [NOx]V is calculated from the difference in NOx between

the traffic station and the background station as

[ ] [ ] [ ] ( )

B x x V 2

NO

• NO

p NO =

[ ] [ ] [ ] ( )

B V

NO

• NO

NO =

with p = NO2/NOx being the fraction of NO2 in the direct traffic emissions and [NOx] the concentration, which is estimated at the traffic station by measurements or dispersion model.

NO2 = f (NOx , p, O3 B, NOx B, NO2 B k, J, τ)

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 10 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

OSPM Chemistry Model

Used in OSPM for hourly time series. The parameter is calculated from meteorological values (e.g. wind speed and turbulence) as well as the road geometry (height of the street canyon) and represents the typical residence time the pollution is trapped inside the street canyon and is available for chemical reactions (see Berkowicz et al., 1997). Typical values for are in the range of 80s to 150s.

τ

τ

y = 1.01x - 0.30 R2 = 0.89 20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160 180 200 NO2-1h-mean measured [µg/m³] NO2-1h-mean modelled [µg/m³]

Comparison of hourly mean NO2-concentrations, calculated with the OSPM chemistry model from observed NOx data, and values from the measurement site Corneliusstraße in Duesseldorf for the year 2006.

NO2 modelled (based on measured NOx !)

6

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 11 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Simplified chemistry model based on annual average values Strictly speaking, the above equations of the chemistry model can only be used in time series calculations, because the parameters J and k are dependent on meteorological parameters. However, based on several research projects, these equations can also be applied for annual mean concentrations using the following parameters: J = 0.0045 s-1 k = 0.00039 m³ (ppb s)-1 = 100 s (street canyons) or 40 s (free dispersion)

τ

The input data is the same as for the Romberg approach: NOx- annual mean at traffic station (observed value

• r estimated by dispersion model)

NOx- annual mean at background station NO2- annual mean at background station as well as two additional values: Ozone- annual mean at background station p = share of primary NO2 emissions for all NOx emissions.

NO2 = f (NOx , p, O3 B, NOx B, NO2 B k, J, τ)

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 12 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Cottbus-Bahnhofstraße

10 20 30 40 50 60 70 80

C_Bhf_1997 C_Bhf_1998 C_Bhf_1999 C_Bhf_2000 C_Bhf_2001 C_Bhf_2002 C_Bhf_2003 C_Bhf_2004 C_Bhf_2005 C_Bhf_2006 C_Bhf_2007

NO2-GB

NO2-GB (measured) NO2-GB (Romberg '96) NO2-VB simplified chemistry model NO2-GB (Bächlin '08)

7

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 13 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Düsseldorf- Corneliusstraße

D-Corneliusstraße

10 20 30 40 50 60 70 80

DDC-1997 DDC-1998 DDC-1999 DDC-2000 DDC-2001 DDC-2002 DDC-2003 DDC-2004 DDC-2005 DDC-2006

NO2-GB

NO2-GB (measured) NO2-GB (Romberg '96) NO2-VB simplified chemistry model NO2-GB (Bächlin '08)

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 14 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Comparision (Data 1997 - 2007; 30 German measurement sites)

Romberg (1996) Bächlin (2008) Simplified chemistry model

8

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 15 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Conclusions

The approach by Romberg et al. (1996) is often conservative for existing data up to the year 2003; however, it cannot reproduce the long term trends. The approach by Bächlin and Bösinger (2008) reproduces the existing measured data better than the approach by Romberg, but it cannot reproduce the long term tendencies very well, either. The simplified chemistry model can best reproduce the tendencies and the absolute values. The main reason for the good fit for trends in the NO2/NOx ratios is the explicit incorporation of the share of primary NO2 emissions. Better NO2-prognosis possible, especially if primary NO2- emission changes (e.g. change of the vehicle fleet mix due to environmental zone, future fleet trend etc.)

• 1. – 4. June 2010

HARMO13 in Paris- NO / NO2 CONVERSION MODEL 16 Ingenieurbüro Lohmeyer GmbH & Co. KG Karlsruhe und Dresden National Environmental Research Institute, Aarhus University

Acknowledgements

Bundesanstalt für Straßenwesen, Bergisch Gladbach Landesamt für Natur, Umwelt und Verbraucherschutz NRW, Essen Ministerium für Ländliche Entwicklung, Umwelt und Verbraucherschutz des Landes Brandenburg, Potsdam