Radical Precursors and Related SpeciesRadical Precursors and Related Speciesfrom Traffic as Observed and Modeledfrom Traffic as Observed and Modeled

at an Urban Highway Junctionat an Urban Highway Junction

Bernhard RappenglückBernhard Rappenglück11,,Graciela LubertinoGraciela Lubertino22, Sergio Alvarez, Sergio Alvarez11, Julia Golovko, Julia Golovko11,,

Beata CzaderBeata Czader11, Luis Ackermann, Luis Ackermann11

1) Department of Earth and Atmospheric Sciences, University of Houston1) Department of Earth and Atmospheric Sciences, University of Houston2) Houston-Galveston Area Council, Houston2) Houston-Galveston Area Council, Houston

• The Houston region is in non-attainment for: 1997 1-hr, and 2008 8-hr ozone standard

• Study focus on ozone precursors: HONO, HCHO, CO, NO/NO2/NOx

• Observational data is compared to modeling data (MOBILE6 and MOVES) at an urban highway junction in Houston

BackgroundBackground

NO

O3

RCH3

CO

RCHO

OH*

O(1D)

+ H2O

O2

O

CO2

HO2*

H*

NO2

O

+O2

H2O

RCH*

RCH2O2 *

+O2

+O2RCH2O*

H2O

RCO*

+O2

RC(O)O*

`

RC(O)O2 *

+O2

PAN’sPANPPN (anthropogenic)MPAN (biogenic)….

Termination byNOx

HNO3

Termination byROx + ROx

+HO2*+HO2*

H2O2

RCH2OOH

NOx-----VOC Photolysis ofO3, RCHO, HONO, H2O2, ClNO2

+ alkene ozonolysis

Daytime Gas-Phase Air ChemistryDaytime Gas-Phase Air Chemistry

Ozone (O3)

NO

O3

RCH3

CO

RCHO

OH*

O(1D)

+ H2O

O2

O

CO2

HO2*

H*

NO2

O

+O2

H2O

RCH*

RCH2O2 *

+O2

+O2RCH2O*

H2O

RCO*

+O2

RC(O)O*

`

RC(O)O2 *

+O2

PAN’sPANPPN (anthropogenic)MPAN (biogenic)….

Termination byNOx

HNO3

Termination byROx + ROx

+HO2*+HO2*

H2O2

RCH2OOH

NOx-----VOC Photolysis ofO3, RCHO, HONO, H2O2, ClNO2

+ alkene ozonolysis

Daytime Gas-Phase Air ChemistryDaytime Gas-Phase Air Chemistry

Ozone (O3)

NO

O3

RCH3

CO

RCHO

OH*

O(1D)

+ H2O

O2

O

CO2

HO2*

H*

NO2

O

+O2

H2O

RCH*

RCH2O2 *

+O2

+O2RCH2O*

H2O

RCO*

+O2

RC(O)O*

`

RC(O)O2 *

+O2

PAN’sPANPPN (anthropogenic)MPAN (biogenic)….

Termination byNOx

HNO3

Termination byROx + ROx

+HO2*+HO2*

H2O2

RCH2OOH

Photolysis ofO3, RCHO, HONO, H2O2, ClNO2

+ alkene ozonolysis

6

5

4

3

2

1

0

ppbV

/hr

04:005/19/2009

08:00 12:00 16:00 20:00 00:005/20/2009

04:00 08:00 12:00 16:00 20:00 00:005/21/2009

CST

HOX_O3 HOX_HONO HOX_HCHO HOX_H2O2

Moody Tower3x3 cells, PBL

Major HOx production mechanisms: morning: HONO; mid morning: HCHO; throughout the day: O3

Daytime Gas-Phase Air ChemistryDaytime Gas-Phase Air Chemistry

Houston, TX Houston, TX

minor

mostly nighttime

?

radical precursors

Ozone (O3)

6

5

4

3

2

1

0

pp

bV

/hr

04:006/5/2009

08:00 12:00 16:00 20:00 00:006/6/2009

04:00 08:00 12:00 16:00 20:00 00:006/7/2009

CST

Moody Tower3x3 cells, PBL HOX_O3

HOX_HONOHOX_HCHOHOX_H2O2

6

5

4

3

2

1

0

pp

bV

/hr

04:006/5/2009

08:00 12:00 16:00 20:00 00:006/6/2009

04:00 08:00 12:00 16:00 20:00 00:006/7/2009

CST

Moody Tower3x3 cells, PBL HOX_O3

HOX_HONOHOX_HCHOHOX_H2O2

HOX_O3HOX_HONOHOX_HCHOHOX_H2O2

Czader et al.,JGR, 2013

NOx-----VOC

Roadside MeasurementsRoadside Measurements

Houston, TX

Roadside MeasurementsRoadside Measurements

Houston, TX

Houston, Galleria

Houston, TX

Houston, Galleria

TxDOT, 2009

(per day)

Roadside MeasurementsRoadside Measurements

Houston, GalleriaHighway junction I-59 South/610TxDOT pumping station

Roadside MeasurementsRoadside MeasurementsMeasurements (10 min)

Houston, GalleriaHighway junction I-59 South/610TxDOT pumping station

Parameter Method Instrument

NO Chemiluminescence TE 42i TL

NO2 Chemiluminescence/ photolytic conversion

TE 42i TL / BLC

CO Gas Filter correlation TE 48i TLE

CO2 Differential, non-dispersive infrared absoprtion

LI-7000

HCHO Hantzsch/ fluorescence AL4021

HONO Long Path Absorption Photometer

LOPAP 03

Speciated PANs GC/ECD Mod. Metcon GC/ECD

Met. Parameters various Vaisala WXT510

Global radiation Pyranometer Vaisala QMS101

0

200

400

600

800

1000

1200

1400

1600

1800

0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00

CST

PA

N [

pp

tv*2

], H

ON

O [

pp

tv],

CO

[p

pb

v],

NO

x [p

pb

v*10

], G

lob

al R

adia

tio

n [

W m

-2]

0

2

4

6

8

HC

HO

[p

pb

v]

Global Radiation

PAN

HONO

CO

NOx

HCHO

I

III

II

Average Diurnal Variation (Weekdays)Average Diurnal Variation (Weekdays)

Morning rush hour

Photochemical processes

Elevated levels throughout the day due to ongoing traffic

0

200

400

600

800

1000

1200

1400

1600

1800

0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00

CST

PA

N [

pp

tv*2

], H

ON

O [

pp

tv],

CO

[p

pb

v],

NO

x [p

pb

v*10

], G

lob

al R

adia

tio

n [

W m

-2]

0

2

4

6

8

HC

HO

[p

pb

v]

Global Radiation

PAN

HONO

CO

NOx

HCHO

I

III

II

Average Diurnal Variation (Weekdays)Average Diurnal Variation (Weekdays)

Morning rush hour

Observational data screened for:

(i) weekdays(ii) Rush hour time 4-8 am CST(iii) Global radiation < 10 W m-2 (iv) PAN < 50 pptv(v) No precipitation(vi) RH > 80%

Roadside MeasurementsRoadside Measurements

Strong relationship of CO vs NOx (r2=0.91). Slope of 6.01 ppbv CO / 1 ppbv NOx is in agreement with Parrish et al. [2009] for rush hour times in selected cities.

Emission ModelingEmission Modeling

Emission ModelingEmission Modeling

MOBILE6: hourly Harris county emission factors for on-road

• Observed meteorology at the Galleria site for the model day: September 28, 2009• 2009 local registration distribution• 2009 local diesel fractions• 2009 local VMT per hour• Local inspection and maintenance program• Anti-tampering program• Reformulated gasoline

Emission ModelingEmission Modeling

MOVES:

> MOVES2010a used to calculate EF for on-road and off

network for NOx, CO, VOC, HCHO, CO2 (atm), NO, NO2.> MOVES2010b used to calculate HONO.

• Avgspeeddistribution• Dayvmtfraction • Fuelformulation • Fuelengfraction• Fuelsupply • Hourvmtfraction• Hpmsvtypeyear• Imcoverage • Monthvmtfraction• Roadtypedistribution• Sourcetypeagedistribution• Sourcetypeyear• Zonemonthhour

Using the Texas TransportationInstitute suite of programs:

• EFs adjusted for Tx Low Emission Diesel and the motorcycle rule

• Emissions calculated multiplying the hourly adjusted emission factors (according to speed) by the hourly VMT per link, using the 2009 hourly VMT mix.

• Output link-level emissions by vehicle type

Emission ModelingEmission Modeling

Time of day- Facility Type

Diesel percentage

Gasoline percentage

AM_Arterial 7% 93%

AM_Freeway 5% 95%

MD_Arterial 11% 89%

MD_Freeway 9% 91%

PM_Arterial 6% 94%

PM_Freeway 5% 95%

OV_Arterial 7% 93%

OV_Freeway 7% 93%

(i) Meteorology of 09/28/2009(ii) 2009 registration distr.(iii) 2009 diesel fractions(iv) 2009 VMT mix(v) 2009 VMT per hour(vi) 2009 trip length distrib.

Time of day- Facility Type

Diesel percentage

Gasoline percentage

AM_Arterial 7% 93%

AM_Freeway 5% 95%

MD_Arterial 11% 89%

MD_Freeway 9% 91%

PM_Arterial 6% 94%

PM_Freeway 5% 95%

OV_Arterial 7% 93%

OV_Freeway 7% 93%

(i) Meteorology of 09/28/2009(ii) 2009 registration distr.(iii) 2009 diesel fractions(iv) 2009 VMT mix(v) 2009 VMT per hour(vi) 2009 trip length distrib.

Local 2009 diesel/gasoline split on the arterial and freeway facility types

Something to consider:Something to consider:Total NOTotal NOxx emissions from Light Duty Gasoline (LDG) vehicles are about the same as emissions from Light Duty Gasoline (LDG) vehicles are about the same as

from High Duty Diesel (HDD), although LDG vehicles are about 80% of the total from High Duty Diesel (HDD), although LDG vehicles are about 80% of the total registration, while HDD are about 7%....registration, while HDD are about 7%....

6 am – 9 am CST6 am – 9 am CST

9 am – 3 pm CST9 am – 3 pm CST

3 pm – 7 pm CST3 pm – 7 pm CST

7 pm – 6 am CST7 pm – 6 am CST

Diurnal variation of VMT for the Diurnal variation of VMT for the Galleria site study area September 28, 2009Galleria site study area September 28, 2009

ResultsResults

Model results:

For the morning rush hour, MOBILE6 overestimates the CO/NOx ratio by almost a factor of 2, while MOVES is 30% higher

Observations:

3.67±0.09 kg CO / kg NOx

Obs

ResultsResults

Model results:

MOBILE6 largely underestimates Form/CO ratio

MOVES calculates a very high ratio for very early morning due to heavy duty diesel off-road emissions (idling and starting trucks)

Observations:

3.14±0.14 g HCHO / kg CO

Obs

ResultsResults

Model results:

The differences in CO/NOx and HCHO/CO ratios are largely due to higher NOx and HCHO in MOVES (30% and 57% more than in MOBILE6), while CO emissions are about the same for both models.

ResultsResults

Model results:

MOVES shows a constant HONO/NOx ratio based on a tunnel study done more than 15 years ago.

The observed HONO/NOx ratio is twice the modeled.

Observations:

0.017±0.0009 kg HONO / kg NOx

Obs

ResultsResults

Model results:

Due to the underestimation of HONO/NOx, MOVES also underestimates the HONO/CO ratio. Only at very early morning hours MOVES results are closer to the observations.

Observations:

0.0046±0.0002 kg HONO / kg CO

Obs

ResultsResults

Model results:

Earlier studies showed a NO2/NOx ratio of 5%. MOVES shows a ratio of 9.3% for the rush hour time. The observation ratio is about twice.

Observations:

0.16±0.01 kg NO2 / kg NOx

Obs

ResultsResults

Model results:

MOVES calculates 3 times higher CO/CO2 than observed.

Observations:

0.0033±0.0002 kg CO / kg CO2

Obs

ResultsResults

Model results:

MOVES calculates 3 times higher CO/CO2 than observed.

It seems that MOVES overestimates the CO/CO2 from light duty gasoline vehicles

Observations:

0.0033±0.0002 kg CO / kg CO2

CO/ NOx: observations: 6.01±0.15 ppbv CO / 1 ppbv NOx (r2 = 0.91) in agreement with other studies.

MOBILE6 and MOVES, overestimate the corresponding observed emission ratio. MOVES gets closer, but 30% above the observed value.

HCHO/CO: observations: 3.14±0.14 g HCHO / kg CO. While MOBILE6 largely underestimates this ratio, MOVES calculates higher ratios, but is lower than the observed ratio. MOVES shows high HCHO/CO ratios during the early morning hours due to heavy duty diesel off-road emissions (potential reasons are idling and starting trucks).

Differences of the modeled CO/NOx and HCHO/CO ratios: largely due to higher NOx emissions in MOVES (30% increased from MOBILE6) and

higher HCHO emissions in MOVES (57% increased from MOBILE6); CO emissions about the same in both models.

ConclusionsConclusions

HONO/NOx:

observations: 0.017±0.0009 kg HONO / kg NOx which is twice as high as in MOVES.

NO2/NOx:

observations: 0.16±0.01 kg NO2 / kg NOx, which is ~70% higher than in MOVES.

CO/CO2: observations: 0.0033±0.0002 kg CO / kg CO2

MOVES overestimates by a factor of 3 compared with the observations.

Findings indicate: overestimation of CO for light duty gasoline vehicles underestimation of HONO and NO2 for heavy duty diesel vehicles in MOVES.

ConclusionsConclusions

Acknowledgements: - Houston Advanced Research Center (HARC)- TxDOT, Houston and Austin - TCEQ, Austin