Spatial and Temporal Characterization of Diesel Exhaust Fumes in School Bus Cabins

Maxwell Martin , Civil and Environmental Engineering Department

Clarkson University

Advisor: Dr. Andrea Ferro , Civil and Environmental Engineering Professor

Clarkson University

Introduction

The characterization of human exposure to diesel particulate matter (DPM) on school buses is an important step in understanding the risks to respiratory disease that children may face throughout the day, both acutely and chronically. Particulate Matter (PM) and especially DPM have been found to be contributors to children’s’ asthma and other respiratory illness (El-Fadel and Massoud, 2000).

Other studies have analyzed DPM on school buses (Behrentz et al., 2004; Wu et al., 1998; Marshall et al., 2005), but there has never been a spatial and temporal analysis of the school bus following normal bus routes. These previous studies were conducted in large cities where infiltration from other diesel-powered vehicles (and other particle sources) may affect the results. Potsdam, NY, provides an environment with a relatively clean background so that the self-pollution of the bus can be more easily isolated.

Methods

To characterize the potential exposures, DPM was measured by mass, particle count and composition in unoccupied buses. Two school buses from the Potsdam, NY school district were instrumented and monitored while the buses drove prescribed routes (see Table 1 for bus and route information). Self-pollution and spatial distribution of self-pollution was determined by adding sulfur hexafluoride (SF 6) directly to the exhaust and monitoring the SF 6 at 8 locations inside the bus. Composition and size distribution of the DPM was determined using a suite of semi-continuous instruments. The impact of idling and traveling at various speeds was determined by keeping records of bus operation and analyzing the semi-continuous exposure data. An aerosol mapping technique was used to provide the temporal and spatial relationship of DPM on the bus during various operational modes. The mapping is used to optimize the monitoring protocol for occupied bus scenarios to best estimate children’s exposures while they are riding, boarding and de-boarding the bus and to correlate these real-time exposures with acute health endpoints.

Spatial distribution of PM was determined using Personal Data Rams (Thermo Electron) with BGI sampling pumps in 6 locations throughout the bus. Paired monitors were housed on the right and left sides of the bus at 1/3 distance from front to back and 2/3 distance from front to back, while one monitor was used at the back of the bus and one other monitor was used at the front of the bus (see Figure 1 for instrument locations).

Table 1. Brief Description of Sampling Runs

Self-pollution and spatial distribution of self-pollution was determined by adding sulfur hexafluoride (SF 6) directly to the exhaust (e.g. Behrentz, 2004) and monitoring the SF 6 at 8 locations inside the bus using an Innova 1312 Multigas Analyzer and a Mark 3 8-Point Sampler.

Composition and size distribution of the DPM inside the bus was determined using a suite of semi-continuous instruments, including:

1) 6 Personal DataRams (PDRs) (PM 2.5);

2) 2-Optical Dust particle size distribution counters (Grimms) (15 size channels 0.3µm < D p < 20 µm);

3) 2- Ecochem PAS 2000CE PAH Monitors for particle-bound (PAH’s);

4) A Magee Scientific Aethalometer (BC);

5) A TSI 3007 Condensation Particle Counter (CPC, 10  nm < D p < 1000 nm);

6) 1-TSI Fast Mobility Particle Sizer ( 32 size channels 5.6 nm < D p < 560 nm);

All instruments were operated off of portable power supplies.

The impact of idling and traveling at various speeds was determined by keeping records of bus speed and analyzing the semi-continuous exposure data. The specific routes of the buses were determined by the volunteer bus drivers, and ranged from between 34 minutes and 75 minutes. A Global Positioning System (GPS; Garmin Etrex Vista C) was used to record the bus’ geographic route by tracking the route and marking positions of stops, restarts, turns, etc. The investigators rode the bus during the experiments to log events, bus velocity, and bus engine RPM, and to monitor instruments.

The impact of open windows was determined by including open window scenarios.

Results

A representative daily particle number concentration time series from the CPC is provided in Figure 2. Measurements were taken semi-continuously for the full experimental period, which included three bus runs (1A, 1B, and 1C). The instrument was started before the bus was running and was left on after the bus was turned off and between each run to determine background concentrations. Peaks are clearly defined and increase in frequency and magnitude during the bus runs. The particle concentration may also be influenced by changes in the bus environment (e.g., open windows or doors), changes in speed, or outside particle concentrations.

Figure 1. Location of Instruments

Figure 2. Time Series for CPC, First Sampling Run

Background concentrations were calculated for the periods between bus runs, for the periods before data collection, and for the periods after data collection.

Mean values were compared for the PM 2.5 data measured by the pDRs for each of the 12 runs, giving an idea of the spatial distribution of particles in the bus. As is seen in figure 2, PDR 2 is at 1/3 the length of the bus from front to back, PDR 4 is at 2/3 the length of the bus from front to back, PDR 5 is paired with PDR 2 on the side of the bus where the exhaust pipe discharges at 1/3 the length of the bus from front to back, and PDR 6 is at the rear of the bus. The means for these values for one run are shown in Table 2. The mean concentration for a collocation experiment where the monitors were placed together for one bus run is provided as a comparison.

Table 2. Summary of Spatial Analysis

Table 2 shows that the concentrations of particles are highest at the rear of the bus. This supports previous studies’ conclusions that the concentrations at the rear of the bus are higher than in other areas of the bus; however, the high concentrations of pDR 5, at 1/3 the length of the bus, may be due to some other factors (Solomon et al., 2001; Behrentz et al., 2004). This requires further investigation.

This work will contribute to the understanding of the self-pollution of the bus and provide improved estimates of children’s’ exposures when they are riding the bus. The exposure estimates are useful for relating exposure to health outcomes as well as for developing solutions to reduce exposure and related illness.

References

• Wu CC, Suarez AE, Lin Z, Kidwell CB, Borgoul PV, Caffrey PF, Ondov JM, Sattler B. Application of an Ir Tracer to determine soot exposure to students commuting to school on Baltimore Public Buses. Atmospheric Environment, 32: 1911-1919, 1998.
• Solomon GM, Campbell TR, Feuer GR, Masters J, Samkian A, Paul, KA, and Guzman JS. No breathing in the aisles: diesel exhaust inside school buses. N atural Resources Defense Council Coalition for Clean Ai, New York (www.nrdc.org) : report. Jan. 2001.
• Behrentz E, Fitz DR, Pankratz DV, Sabin LD, Colome SD, Fruin SA, Winer AM. 2004. Measuring self-pollution in school buses using a tracer gas technique. Atmospheric Environment 38: 3735-3746.

• Marshall, Julian D.; Behrentz, Eduardo. 2005. Vehicle Self - Pollution Intake Fraction: Children's Exposure to School Bus Emissions. Energy and Resources Group, University of California, Berkeley, CA, USA. Environmental Science and Technology 39(8), 2559-2563.  

• El-Fadel, M.; Massoud, M. Particulate matter in urban areas: health-based economic assessment. Science of the Total Environment (2000), 257(2-3), 133-146.