MITIGATING ROADWAY POLLUTION IN URBAN AREAS: LOCATING TRANSIT STOPS

Transcription

1 1 MITIGATING ROADWAY POLLUTION IN URBAN AREAS: LOCATING TRANSIT STOPS Suzanne Paulson Wonsik Choi, J.R. DeShazo, Dilhara Ranasinghe, Lisa Wong, Mario Gerla, Kathleen Kozawa,* Steve Mara* *UCLA Departments of Atmospheric and Oceanic Sciences, Institute of Environment and Sustainability, Public Policy, Luskin Center for Innovation, Computer Science & Environmental Health Sciences *California Air Resources Board By Frank Hank - Own work, CC BY-SA 3.0, /w/index.php?curid=

2 Aspects of the Built Environment that Influence Exposure The heights, size and layout of the buildings Where the people are relative to the traffic (land use) Barriers between the traffic and people Traffic Control Strategies Factors influencing transit user exposure 2

3 Minutes spent waiting for the bus/train each day (roundtrip) Boston, New York City, SF, LA: Brasil: 32-66; Colombia: 22 40; Germany, France: 20; UK: Spain: 16-20; Italy: Crowdsourced data from Moovit Realtime 3

4 4 METHODS Mobile measurements

5 Mobile Monitoring Platform Instrument CPC (TSI, Model 3007) FMPS (TSI, Model 3091) DisCMini (Testo) DustTrak (TSI, Model 8520) EcoChem PAS 2000 Measurement Parameter UFP number concentration (10 nm- 1mm) Particle size distribution ( nm) UFP number and average size PM 2.5 and PM 10 mass Particle bound PAHs California Air Resources Board Mobile Measurement platform (MMP) Toyota RAV4 electric vehicle LI-COR, Model LI-820 CO 2 Teledyne API Model 300E Teledyne API Model 200E Teledyne API Model 400A 3D-Sonic Anemometer (Campbell CSAT3) Garmin GPSMAP 76CS SmartTether TM KciVacs video CO NO X O 3 Temperature, Relative humidity, Wind speed/direction, Turbulence Characteristics GPS Vertical profiles of temperature, RH, wind speed/direction Video record for traffic and fleet composition

6 Processing Mobile Data Ranasinghe, D., W.S. Choi, A.M. Winer and S.E. Paulson (2016) Developing High Spatial Resolution Concentration Maps Using Mobile Air Quality Measurements. Aerosol and Air Qual. Res. 16 (8),

7 7 5 Meter Spatial Resolution Map for Downtown Los Angeles Ranasinghe, D., W.S. Choi, A.M. Winer and S.E. Paulson (2016) Developing High Spatial Resolution Concentration Maps Using Mobile Air Quality Measurements.Aerosol and Air Qual. Res. 16 (8),

8 8 Decay of pollutants around the intersections: the best place for the bus stop? Choi, W.S., D. Ranasinghe, J.R. DeShazo, J.J. Kim and S.E. Paulson (2017) Cross-Intersection Profiles of Ultrafine Particles in Different Built Environments: Implications for Pedestrian Exposure and Bus Transit Stops. Submitted.

9 How Far Should the Bus Stop be from the Intersection? Gary Larson s Far Side Cartoons 9 Near Side Far Side

10 10 Measurement Sites for Intersection Studies 10 Intersections 1,744 Profiles

11 Variety of Intersections; 1,744 Profiles Total 11 Street width Traffic flow rate (A.M.) Traffic flow rate (P.M.) Traffic density Distance between traffic lights Wilshire in Beverly Hills (5 intersections) Broadway & 7 th Downtown Los Angeles Olive & 12 th Downtown Los Angeles Vermont & 7 th Wilshire & Carondelet Temple City & Las Tunas m 22 & 26 m 17 & 28 m 25 & 30 m 17 & 37 m 24 & 30 m & & 4 39 & & & &20 8 & 3 38 & 12 2 & & 29 Long queues, WB in A.M., EB in P.M. Medium queues, slow vehicle speeds 330 m m Minimal queues (1) 180 m (2) 125 m Long queues, often for entire block Short queues (1) 224 m (1) 190 m (2) 174 m c (2) 100 m Long queues but queues dissipate rapidly (1) 200 m (2) 135 m

12 Cross-intersection profiles of UFPs for each traffic direction [UFP] ( 10 4 particles cm -3 ) (a) Beverly N=355 Early East-bound West-bound mornings [UFP] ( 10 4 particles cm -3 ) N= Afternoons East-b West- [UFP] ( 10 4 particles cm -3 ) North-bound (b) Broadway South-bound East-bound N=92 West-bound Averaged [UFP] ( 10 4 particles cm -3 ) N=76 North South East-b West- Avera [UFP] ( 10 4 particles cm -3 ) North-bound South-bound (c) Olive East-bound Averaged N=104 [UFP] ( 10 4 particles cm -3 ) N=107 North South East-b Avera

13 [UFP 2.0 [UFP Cross-intersection profiles of UFPs for-40 each -20 0traffic direction North-bound North-bo South-bound South-b N= East-bound East-bou 5.0 West-bo N=85 West-bound Average Averaged [UFP] ( 10 4 particles cm -3 ) (d) Vermont [UFP] ( 10 4 particles cm -3 ) [UFP] ( 10 4 particles cm -3 ) North-bound South-bound (e) Wilshire East-bound West-bound N=184 Averaged [UFP] ( 10 4 particles cm -3 ) N=181 North-bou South-bo East-boun West-bou Averaged [UFP] ( 10 4 particles cm -3 ) North-bound South-bound (f) Temple City N=79 East-bound West-bound Averaged [UFP] ( 10 4 particles cm -3 ) N=143 North-bou South-bo East-boun West-bou Averaged 1.0

14 [UFP] (particles cm -3 ) [UFP] (particles cm -3 ) 2.6e+4 2.5e+4 2.4e+4 2.3e+4 2.2e+4 2.1e+4 2.0e+4 1.9e+4 2.6e+4 2.5e+4 2.4e+4 2.3e+4 2.2e+4 2.1e+4 2.0e+4 1.9e+4 1.8e+4 N= % N= % Averaged [UFP] profile 1 Distance from the center of Intersection [UFP] (particles cm -3 ) 2.4e+4 2.3e+4 2.2e+4 Average Profiles 2.1e+4 Morning Afternoon 5e+4 4e+4 3e+4 2e+4 1e+4 0 1e+5 8e+4 6e+4 4e+4 2e+4 2.6e+4 Distance from the center of Intersection 7e+4 0 standard deviation (1 ) standard deviation (1 ) [UFP] (particles cm -3 ) 2.0e+4 1.9e+4 1.8e+4 2.6e+4 2.5e+4 2.4e+4 2.3e+4 2.2e+4 2.1e+4 2.0e+4 1.9e+4 1.8e+4 14 N= % Distance from the center of Intersection Distance from the center of Intersection Averaged [UFP] profile 1 All 8e+4 6e+4 4e+4 2e+4 0 7e+4 6e+4 5e+4 4e+4 3e+4 2e+4 standard deviation (1 ) standard deviation (1 ) 2.5e+4 N=1744

15 15 Cumulative distributions of UFPs at the peak and base locations of the profile [UFP] (particles cm -3 ) 10 5 [UFP] at peak location [UFP] at base location Data used for a linear fit at peak Data used for a linear fit at base Extended fit at peak Extended fit at base Cumulative Distribution

16 Exposure level of transit-users to UFP around intersections Simple time-duration model to simulate exposure reductions when the bus-stop is moved from 20 m to 40 m (or 60 m) from the intersection: 16 Set two UFP zones: within ± 20 m of the intersection (high UFP) vs. around (40 and 60 m) (low UFP). Transit-user s behavior includes disembarking, walking, crossing the intersection, waiting for a bus; assuming three pedestrian walk speeds: 0.5 (slow), 1.0 (comfortable), and 1.5 m/s (normal). Waits at the bus stop for only 10 minutes!

17 Summary Management Suggested Direction Approx. Size of Effect Bus/Transit Stop Siting Further from the intersection is better, but improvements diminish within several tens of meters, depending on built environment (block length, queue lengths, etc.) 17 Up to approximately a factor of 3 Some Other Options:

18 Traffic Management Management Suggested Direction Approx. Size of Effect Traffic Management Fewer stops and smaller queues reduce emissions and elevated concentrations around intersections Factor of

19 Normalized [UFP] 19 Plumes around Roadways: ~150 m during daytime, ~1500 m during Early Morning SM winter (Hu, 2009) 1 SM summer (Hu, 2009) 0.8 Upwind Downwind WLA day (Zhu, 2002) DoLA (this study) Paramount (this study) 0.6 Carson (this study) Claremont (this study) Distance from freeway (m) [ UFP ] [ UFP] -[ UFP] bkgnd [ UFP( x)] Normalized [ UFP( x)] [ UFP] [Choi et al., Atmos. Environ., 62, , 2012] peak

20 Land Use Around Heavily Travelled Roadways Management Suggested Direction Approx. Size of Effect Sensitive uses near highways: Daytime downwind Further is better, but under normal daytime conditions 150 meters is sufficient. Up to a factor of four or more. 20 Sensitive uses near highways: Night/Morning downwind 1500 meters is desirable. Other mitigation strategies: Up to a factor of four or more. Other Mitigation Options: Build solid barriers (quite effective); Grow trees (less effective but worthwhile), move physical education classes later in the day; filter indoor spaces

21 Beyond the street canyon: block scale characteristics influencing concentrations Management Suggested Direction Approx. Size of Areal aspect ratio (A area ); combines building area-weighted height, building footprint, and the amount of open space. Building Heterogeneity Lower building volumes and more open space lower pollutant concentrations. Isolated tall buildings lower concentrations compared to homogeneous shorter or higher buildings with similar volume. Effect Up to ~ a factor of 3. Up to ~ a factor of two. 21 Atmospheric Conditions & Notes Important under calm conditions. Important under unstable conditions with moderate winds.

22 Latitude ( o ) 22 Site 1: Street Canyon Building height 600 (Ft.) Longitude ( o ) 0 Broadway & 7 th Site (Street view: heading South)

23 Thank you for your attention 23

24 [UFP] (particles cm -3 ) / Traffic flow rate (veh. min -1 ) Best Explanatory Factor in the Morning: The Areal Aspect Ratio = Length scale of buildings over length scale of open space Ar 1400 area L diag Hbldg Hbldg / bldg 1 - S bldg / Asite Ldiag Aopen Asite Lopen H Site 1 Site 2 Site 3 Site 4 Site 5 H bldg : Mean area-weighted building height L diag : Diagonal length of block S bldg : Building surface area A site : Area of the sampling site A open : Area of the open space in sampling site Areal aspect ratio (Ar area ) Choi et al., 2016

25 [UFP] ( 10 3 particles cm -3 ) Best Explanatory Factor in the Afternoon: Turbulence strength (vertical fluctuations of surface winds, w ) (a) n -1 ) Site 1 Site 2 Site 3 Site 4 Site w (m s -1 ) 1800

26 [UFP] (particles cm -3 ) / Traffic flow rate (veh. min -1 ) [UFP] ( 15 Site 1 Site 2 Best Explanatory Site 3 Factor in the Afternoon: 10 Site 4 Turbulence strength (vertical fluctuations of surface winds, w ) Site 5 Appears to be from non-local emissions w (m s -1 ) (b) Fitting curves Site 1 Site 2 Site 3 Site 4 Site ( w m s -1 )