Download - Transportation System Impacts on Bicyclists' Air Pollution Risks: Considerations for System Design and Use

PSU Friday Transportation Seminar13 February, 2015

Transportation System Impacts on Bicyclists' Air Pollution Risks

Alex BigazziMiguel Figliozzi James PankowWentai Luo

Considerations for System Design and Use

Framework

2

Vehicle Emissions

Air Quality

Traveler ExposureInhalation

Uptake Health Effects

Bicyclists' Pollution Uptake

Background

3Bicyclists' Pollution Uptake

Bicyclists’ Exposures

4

0 5 10 15 20

Ultrafine PM

Fine PM

Coarse PM

Black Carbon

CO

VOC

NO2

# studies measuring on-road bicyclists’ exposure concentrations

Pollutant 42 studies


Modal Comparisons of Exposure

– Is this actionable information?

Context-dependent results– Bicyclists lower if separated


vsvs

High-Traffic/Low-Traffic Routes


-50%

0%

50%

100%

150%

200%

250%

CO VOC UFP PM2.5 PM10 BC

Expo

sure

Inc

reas

e on

H

igh-

Traf

fic R

oute

s

N=6 N=11 N=8 N=6 N=3 N=5

Ventilation and Exercise


50 100 150 200 250 Watts

Vent

ilatio

n (

liter

/min

) 75

50

25

Ventilation & Bicycle Studies

8

57 studies assessbicyclists’ exposure

Ignored38

Constant16

Assumed/Modeled

15Measured

1

Variable3

Modeled 2

Measured1

Ven

tila

tio

n:

Urban Bicyclists' Pollution Uptake

Bicyclist Uptake Studies

2 studies of biomarkers:

• VOC: blood & urine– Urban bikers > rural bikers

• BC: induced sputum– Bicyclists > transit riders


Research Questions

1. How does uptake vary with roadway and travel conditions?

2. Can transport strategies reduce bicyclist uptake?


Data & Methods


Breath BiomarkersExhaled breath is a good proxy for

blood concentrations of VOC


alphaszenszor.com

VOC in blood

Sampling Equipment


1) Local roads

Pre Breath

Post Breath

On-Road Sampling Segments

20-30 minutes, 3-5 milesExposure & breath VOC

Paired subjects


Pre Breath

Post Breath

2) Major arterials

Exposure Data coverage


74 breath samples

3 subjects

9 days

~40 hours of data

Model Development

• Exposure concentrations

• Ventilation

• Breath concentrations~ exposure + ventilation~ weather + roadway + travel conditions


Results


Results: VOC Exposure Models

• +2% per 1,000 ADT

• +20-30% in stop-and-go riding

• Off-street path +300% in industrial

corridor


19

Industrial Corridor


20

Parallel Path Comparison


E Burnside St. SE Ankeney St.

N Williams Ave. NE Rodney Ave.

Naito Pkwy. Riverside Path

Results: Ventilation Models

• 4-8% increase in ventilation per 10 W

• Mean lag ~50 sec


0 50 100 150

0.0

0.2

0.4

0.6

0.8

Seconds workload lagged

Cum

ulat

ive im

pact

on ve

ntila

tion

(%

SubjecABCPooled

% Δ

vent

ilatio

n pe

r W

Seconds lag

Breath and Exposure Concentrations

0 1 2 3

Breath

Exposure

Concentration (normalized to Park)

Toluene

Major arterials

Local roads


Results: Breath Biomarkers

• 8 aromatic hydrocarbons were biomarkers of traffic exposure

• Issues– Background concentrations– Endogenous production– High water solubility


Explained variance in breath BTEX


0%

10%

20%

30%

40%

50%

Measured variables Modeled variables

Exposure

Ventilation

Exposure variability

Weather variables

Road & traffic

variables

Applications


Steady-state biking work


0

50

100

150

200

250

0 2 3 5 6 8 9 11 13 14 16

Wor

kloa

d (W

)

Speed (mph)

Rolling Resistance Drag 1% Grade

Inhalation and Speed


0

50

100

150

200

250

300

0 4 8 12 16

Ven

tila

tion

(l/

km)

Speed (mph)

0% Grade2% Grade

Minimum-Inhalation Speed


0

5

10

15

0% 2% 4% 6% 8% 10%

Min

imu

m-v

enti

lati

on

spee

d (m

ph)

Grade

Inhalation and Stops

29Urban Bicyclists' Pollution Uptake

0

100

200

300

400

500

600

6 8 10 12 14 16 18

Cru

ise-

equ

ival

ent

exce

ss

ven

tila

tio

n (

ft)

Cruise speed (mph)

Route Choice

• Detour 1 block to a bikeway vs.– 1.6 blocks on a major arterial– 4.3 blocks on a minor arterial


?

Comparison with Preferences


Will bicyclists naturally minimize inhaled dose over a trip?

Bike boulevard

or

neighborhood greenway

Bike lane

Minor arterial (no bike lane)

Major arterial (no bike lane)

• Slightly over-avoid

• Balance on collectors (6-10k ADT) • Under-avoid arterials

• Greatly over-avoid

vs.

vs.

vs.

Exercise and Uptake

• Inhalation rate: 2-5x higher

• PM uptake: ≥2-5x higher

• VOC uptake: 1.5-2x higher– Limited by blood/air equilibrium


Conclusions


Findings

Determining factors

• ADT• Stop-and-go riding• Industrial

corridors• Speed & grade• Exposure &

ventilation

Mitigation

• Low-volume streets

• Travel speed choice

• Reducing stops• Separated

facilities


Take-Away Principles 1. Bicyclist Exposure

a) Many different pollutantsb) Traffic, weather, and land-use all important c) Benefits of separation from traffic

2. Bicyclist Inhalationa) Varies greatly with workload (speed, grade)b) Breath response spread out over 1-2 min

3. Bicyclist Uptakea) For particles, highly sensitive to breathingb) For some gases, more sensitive to duration


Bikeway Design ConsiderationsBike lane • High-traffic streets

• Some lateral separation• Dedicated lane reduces duration in

congestion

Bike boulevard

• Low-traffic streets• Additional benefits from traffic calming• Fewer stops reduces uptake

Cycle track • Lateral separation• Fewer stops reduces uptake

Off-street path

• Low exposure (nearby industry?)• Fewer stops reduces uptake


Next Steps

• Abstraction for HIA & CBA• Additional biomarkers • Characterizations of urban bicyclists • Similar study for pedestrians• Crowd-source pollution data


Crowd-Sourced Pollution Data


Questions?

[email protected]


Acknowledgments• Dissertation committee: Miguel

Figliozzi, Jim Pankow, Robert Bertini, Jennifer Dill

• NITC research project with support from City of Portland and Metro

• NSF and OTREC fellowships

Bonus Slides!


ADT


Parallel Paths

• Burnside > Ankeny– 51% TVOC, 201% CO, 9% PM2.5

• Williams > Rodney– 329% TVOC, 221% CO

• Naito > McCall path– 112% TVOC, 30% CO, 4% PM2.5


Explained Variance


0%

10%

20%

30%

40%

50%

60%

BTEX Exposure Exposure variability Ventilation

Road & traffic variables Weather variables

Inhalation and Grade


0

50

100

150

200

250

300

-2% -1% 0% 1% 2% 3% 4% 5%

Ven

tila

tion

(l/

km)

Grade

11 kph18 kph

Minimum-ventilation envelope


0

5

10

15

20

25

30

0% 2% 4% 6% 8% 10%

Min

-ven

tila

tio

n v

enti

lati

on

(l

pm

)

Grade

Results: Uptake Models

• Elasticity of breath to exposure ~0.5– Δ Breath ~ ½ Δ Exposure– No significant difference by subject

• Ventilation effect small but significant for some compounds– Elasticity of 0.1-0.2
