pasteur.epa.gov · web viewa number of monitoring sites were considered throughout the us. table...

WA 40-3/ Near-Road Noise Barrier Air Quality StudyRevision No. 05/6/2023 of 38

Mobile and Stationary Monitoring to Characterize the Impact of Noise Barriers on Near-Road Air Quality

Quality Assurance Project Plan - Revision 0Category III / Measurement Project

NRMRL/APPCD/ECPBEPA Technical Lead: Richard Baldauf

6 May 2023

1


Mobile and Stationary Monitoring to Characterize the Impact of Noise Barriers on Near-Road Air Quality

Quality Assurance Project Plan

Table of Contents

1. Approvals............................................................................................................................. 32. Distribution List..................................................................................................................... 43. Project Description and Objectives........................................................................................5

3.1. Background.................................................................................................................... 53.2. Project Purpose and Objectives......................................................................................6

4. Organization and Responsibilities.........................................................................................64.1. Project Personnel...........................................................................................................64.2. Project Schedule............................................................................................................8

5. Scientific Approach...............................................................................................................85.1. Sampling Design............................................................................................................85.2. Process Measurements................................................................................................195.3. General Approach and Test Conditions for Each Experimental Phase..........................20

6. Sampling Procedures.........................................................................................................206.1. Site-Specific Considerations.........................................................................................206.2. Sampling Equipment and Procedures...........................................................................216.3. Quality Control in Sample Analysis...............................................................................316.4. Sample Preservation....................................................................................................316.5. Sample Numbering.......................................................................................................316.6. Sample Chain-of-Custody............................................................................................31

7. Measurement Procedures...................................................................................................317.1. Analytical Method.........................................................................................................327.2. Project Overview..........................................................................................................327.3. Calibration Procedures.................................................................................................35

8. Quality Metrics.................................................................................................................... 358.1. QC Checks................................................................................................................... 358.2. QA Objectives and Acceptance Criteria........................................................................36

9. Data Analysis, Interpretation, and Management..................................................................399.1. Data Reporting.............................................................................................................399.2. Data Validation.............................................................................................................399.3. Data Analysis...............................................................................................................399.4. Data Storage Requirements.........................................................................................41

10. Reporting............................................................................................................................ 4110.1. Deliverables.................................................................................................................41

2


10.2. Expected Final Products...............................................................................................4111. References.........................................................................................................................41

1. Approvals

______________________________________Richard Baldauf DateNational Risk Management Research LaboratoryU.S. Environmental Protection AgencyProject Leader/ARCADIS Work Assignment Manager

______________________________________Gayle Hagler DateNational Risk Management Research LaboratoryU.S. Environmental Protection AgencyProject Scientist/Alternate ARCADIS Work Assignment Manager

______________________________________Robert Wright DateNational Risk Management Research LaboratoryU.S. Environmental Protection AgencyQuality Assurance Representative

______________________________________Bobby Sharpe DateARCADISWork Assignment Leader

______________________________________Libby Nessley DateARCADISQuality Assurance Representative

3


2. Distribution List

Bob Wright – EPARich Baldauf – EPAGayle Hagler – EPARichard Shores – EPABill Mitchell – EPABill Squier – EPAJames Faircloth - EPAVlad Isakov – EPADavid Heist – EPASteven Perry – EPADavid Proffitt – ARCADISBobby Sharpe – ARCADISParikshit Deshmukh - ARCADISMichal Derlicki – ARCADISLibby Nessley - ARCADIS

4


3. Project Description and Objectives3.1. Background

EPA’s Office of Research and Development has a multi-faceted research program aiming to better understand the impact of traffic emissions on local air quality and human health. Past field studies have demonstrated that concentrations of traffic-related air pollutants exponentially increase with proximity to a major roadway (e.g., Hagler et al. 2009 provides results from multiple studies). EPA’s near-road research program incorporates a wide range of studies, such as dynamometer vehicle emissions characterization, ambient monitoring, health studies, and model development.

Recent modeling and field research has shown that the dispersion of traffic emissions can be significantly affected by the presence of roadside structures, such as noise barriers (Bowker et al. 2007, Baldauf et al. 2008, Heist et al. 2009). The altered dispersion of emissions is predicted to induce different near-road air pollutant patterns relative to the more widely studied flat terrain with no obstructions to air-flow case. Because of the impact of these features, ORD’s National Exposure Research Laboratory has designed an air dispersion modeling platform to account for these structures based on previous studies. In order to evaluate this model, a field study will conducted to provide data on the effects of noise barriers on the transport and dispersion of traffic emissions on an adjacent highway.

3.2. Project Purpose and Objectives

The purpose of this study is to measure the spatial patterns of traffic-related gas- and particle-phase pollutants at near-road sites experiencing consistent traffic activity with and without the presence of noise barriers.

The specific objectives of this research study are:

Characterize the effect of noise barriers on near-road air quality. Measure the variability of traffic-related pollutant concentrations in areas considered to

be representative of the urban background. Measure the variability of traffic-related pollutants on the highway and in near-road

neighborhoods (within 500 m of the highway) with and without noise barriers present.

4. Organization and Responsibilities4.1. Project Personnel

5


This QAPP addresses the measurement of spatial patterns of ambient air pollution nearby major roadways using air monitoring instruments onboard multiple mobile and fixed sampling units. The field measurements will be performed by ARCADIS personnel, with technical guidance from EPA/ORD/NRMRL personnel. Table 4-1 lists the personnel responsible for the oversight and QA review of this project. Additional team members that will be involved in this study are listed in Table 4-2.

Table 4-1. Key Points of Contact

Name Organization Affiliation Title Responsibilities Contact Information

Richard Baldauf EPA / NRMRL Technical Leader

Technical leadership for study, WAM for ARCADIS research support, data analysis for fixed and backpack monitoring

Phone: (919) 541-4386Email: [email protected]

Gayle Hagler EPA / NRMRL Project Scientist

Technical support for study, Alternate WAM for ARCADIS research support, data analysis for mobile monitoring


Bob Wright EPA/ NRMRL QA Manager

Quality Assurance through review of QAPP and presentation of results


David Proffitt ARCADIS Technical advisorTechnical guidance for ARCADIS research support activities

Phone: (919) 544-4535 Email: [email protected]

Bobby Sharpe ARCADIS Work Assignment Leader

Oversight of ARCADIS research support activities, field sampling technician.


Libby Nessley ARCADIS ARCADIS QA Manager

ARCADIS QA activities oversight


Table 4-2. Additional Project Team Members

Name Organization Affiliation Title Responsibilities Contact Information

Bill Mitchell EPA / NRMRL Project Engineer Data-logging set-up support


Bill Squier EPA/NRMRL Project Engineer Shop support for instrumentation


6


mounting.

James Faircloth EPA / NRMRL Project TechnicianGMAP vehicle troubleshooting support


Parikshit Deshmukh ARCADIS Project Engineer Field study support


Jerry Faircloth ARCADIS Project Technician Field study support Phone: (919) 541-0314Email: [email protected]

Michal Derlicki ARCADIS Project Technician Field study supportPhone: (919) 544-4535Email: [email protected]

Much of the project team has extensive experience conducting similar research activities. Rich Baldauf and Gayle Hagler have performed a number of projects with the mobile monitoring vehicle and stationary sampling devices, and have published these results in peer reviewed journal articles. Bill Mitchell, and James Faircloth of the EPA, and Jerry Faircloth of Arcadis have also conducted similar research and co-authored publications on this subject. Parikshit Deshmukh has been the lead mobile monitoring operator of the electric vehicle and on-board instrumentation for multiple field studies across the United States. Michal Derlicki also has extensive experience regarding field work and air quality measurement instrumentation. Finally, Parik and Michal have all had hands-on experience with the instrumentation planned for use with the backpack sampling systems.

4.2. Project Schedule

The target project schedule is as follows: August/September, 2012: Preparation activities for field campaign October/November, 2012: Field campaign December, 2012: Preliminary data analysis for the campaign

5. Scientific Approach5.1. Sampling Design

In order to meet project objectives (Section 3.2), three series of field data will be collected:

7


1) High-resolution mobile monitoring campaign – EPA’s Geospatial Monitoring of Air Pollution (GMAP) electric vehicle will be deployed to map air pollutants on the highway and in near-road neighborhoods with and without a noise barrier.

2) Targeted fixed-point monitoring – Two types of portable air monitoring stations will be deployed. An SUV outfitted with real-time monitoring equipment will be located near the highway, collecting air quality data at a fixed location, as well as collocated data with the GMAP vehicle during a portion of time each day. Two additional fixed site locations will be established with hand-held air quality instruments at varying distances from the highway, both behind and in front of a noise barrier, to collect continuous air pollutant and wind data. The location of these two hand-held sampling stations may change depending on anticipated weather and traffic conditions.

The field study will be conducted over a 1-month period. The length of the field campaign was chosen to allow for repetitive fixed and mobile monitoring under a variety of meteorology conditions.

5.1.1.Site Selection

The monitoring sites are selected based on the following criteria, shown in Table 5-1.

Table 5-1. Site selection criteriaCriteria Prioritization

(1-3, 1 is high)

Accessible (by fixed point or mobile monitoring vehicle) near-road areas with and without noise barriers along the same stretch of highway 1

Near-road area at-grade with highway, to isolate effects of near-road buildings and vegetation 1

High roadway traffic volume 2

Co-located with other EPA or state near-road monitoring sites 3

Avoidance of other potentially confounding emission sources (e.g., industrial emissions, significant arterial road traffic) 1

Near-road population, overall and of air-pollution sensitive populations (e.g., children) 3

Noise barrier and clearing sections on both sides of road (not sensitive to wind direction) 2

A number of monitoring sites were considered throughout the US. Table 5-2 lists the final sites considered after a nationwide search. This search consisted of identifying highway

8


segments in the US with noise barriers using FHWA’s inventory of noise barriers (http://www.fhwa.dot.gov/environment/noise/noise_barriers/inventory/summary/index.cfm). Only highway segments were considered since the noise barrier and clearing sections need to be along the same stretch of limited access roadway so that traffic emissions along both the clearing and noise barrier portions of the road are consistent. After these road segments were identified, an online evaluation of sites was conducted using satellite maps and on-road pictures obtained from the Google MapsTM program. These resources were used to identify sites at-grade with the highway which did not have other obstacles (e.g. thick vegetation, buildings) that would affect air flow in either the clearing or noise barrier sections.

No roadway segments met all criteria; however, sites in Denver, Detroit, Las Vegas, and Salt Lake City were determined to meet minimum criteria. Details of these sites are shown in Table 5-2.

Table 5-2. Site selection criteriaDenver Detroit Las Vegas Salt Lake City

Site Location

Road Segment I-70 at W. 48th St and Frontage Rd.

I-94 at Tyler Rd. and Service Dr.

I-15 at S. Rancho Dr. I-15 at Wildcat Way

Estimated AADT (2010 HPMS) 150,000 100,000 200,000 150,000

Historical Annual Meteorology (2009-2012)

Accessible with and without noise barriers along same stretch of roadway

X X X X

At-grade with highway X X X X

High roadway traffic volume X X X

Co-located with other EPA or

9

http://www.fhwa.dot.gov/environment/noise/noise_barriers/inventory/summary/index.cfm

http://www.fhwa.dot.gov/environment/noise/noise_barriers/inventory/summary/index.cfm


state near-road monitoring sitesNo other emission sources X X X X

Near-road sensitive population

X X

Noise barrier and clearings on both sides of road

X X

Based on these criteria, Denver was selected as the priority site for assessing near-road spatial variability due to noise barriers.

5.1.2. Sampling Schedule

Sampling during the 0700-1000 time window appears to have a good chance of successfully achieving two objectives: (1) characterizing air quality concentrations downwind of the highway, and (2) measuring differences in air pollutant concentrations with and without noise barriers between the samplers and the highway. Sampling will be conducted for 6 days of the week, Monday through Saturday. A 24-day sampling schedule is anticipated for the study.

This sampling schedule is tentative, as the actual sampling days will be dependent on field conditions. For example, sampling during heavy rainfall risks damaging research instrumentation, thus a planned sampling day with significant rain forecasted will be rescheduled for a later date. The SUV and two portable, fixed-site monitoring platforms will collect a continuous time series of pollutant concentrations and local meteorology measurements. Based on available power for the mobile sampling vehicle, the SUV, and the backpack sampling configuration, the daily mobile sampling deployment is planned to be approximately 3-4 hours in duration, depending on the sampling system. An example daily sampling schedule is provided in Table 5-3.

Table 5-3 Daily Sampling Schedule

Time I-275 Sampling Day

0700 Monitoring initiates – GMAP vehicle on routeChecks/maintenance of fixed sites (SUV and portable)

1000 GMAP monitoring on route ends Conduct co-located monitoring of GMAP and SUVConduct co-located monitoring of portable units

1100 Discontinue co-located monitoring of GMAP and SUV (as power allows)1200 Discontinue co-located monitoring of portable units (as power allows)

10


5.1.3.Specific Driving Route and Fixed Point Locations

Wind speed and direction are important factors that guide the planned mobile monitoring routes and locations of fixed stations. Figure 5-3 shows historical wind patterns at the Denver monitoring site from 2009-2012. This figure shows annual averages, as well as historical winds for the months of October and November during the time period of 6-9AM (the anticipated dates of this study and time of monitoring). The prevailing wind direction in Denver is generally in the orientation of SW/NE for the years shown, although all wind directions can be experienced. During the morning hours in October and November, winds are much more pronounced from the SW, with relatively rare winds from the SE or NW.

11


Figure 5-3. Wind direction and speed measured at Denver National Weather Service site for annual averages (2009-2012) and during the months of October/November for the hours 0600-0900 for the same period of time

A range of wind orientations will likely be encountered at the monitoring location, although the barrier section to the north of I-70 will likely be downwind during many of the morning hours. Since mobile monitoring will occur on both sides of the highway as noted below, both upwind and downwind concentrations during crosswind conditions should be experienced. Measurements collected on the highway will most likely be influenced by vehicle-induced turbulence rather than any wind conditions; thus “upwind/downwind” scenarios will not be applicable for this part of the mobile monitoring driving route.

Site Sampling Configuration

Sampling will include continuous sampling at 1 location with the SUV and 2 other locations using portable air monitoring stations, with 3-4 hour intensive sessions using mobile monitoring with the GMAP electric vehicle. The stationary sampling locations are shown in Figure 5-4. The mobile monitoring route is shown in Figure 5-5.

Stationary Monitoring Positions

The planned positions for the SUV and the two portable units are shown in Figure 5-4. Sampling will be conducted continuously for approximately 4 hours a day at these locations. The instruments planned for use in the portable stationary sampling platform are

12


listed in Table 5-5 (Section 5.2). The instruments will have 1 Hz time-resolution sampling, although averaging and reporting of data will be at 1-minute or more time averages.

Figure 5-4. Anticipated positions for fixed site SUV and portable, stationary (PS) monitoring systems

13


Figure 5-5. Planned driving route for GMAP vehicle at the I-70 site.

Portable Monitoring

During the time period from 0700 to 1000, the portable units will be at two separate, fixed locations to compare changes in air pollutant concentrations under different road configurations (e.g. behind the noise barrier, in front of the barrier). Operators may move the portable samplers to specific positions along the roadway segment each day to capture multiple changes in configuration (see Figure 5-4). This time period is of interest due to the high and steady traffic volumes anticipated during morning rush hour, as well as the amount of vertical mixing in the atmosphere is expected to be lower for the morning due to inversion conditions. Example sampling schedules for a morning deployment is provided in Table 5-4. At 0700, it is anticipated that daily recharged battery changes will occur at the and the portable, stationary sampling platforms will be put in place. The sampler position GPS location will be identified as part of the system. After approximately 3 hours of

14


sampling, the units will be placed next to the co-located GMAP and SUV vehicles for 1 – 1:30 hours of side-by-side measurements for quality assurance purposes.

SUV Monitoring

During the time period from 0700 to 1000, the SUV will be at fixed location near the highway, but in a clearing section with no obstructions to air flow from the road. This time period is of interest due to the high and steady traffic volumes anticipated during morning rush hour, as well as the amount of vertical mixing in the atmosphere is expected to be lower for the morning due to inversion conditions. Example sampling schedules for a morning deployment is provided in Table 5-4. SUV monitoring will be conducted on-board the vehicle with analyzers collecting data at <10 s time resolution, and samplers powered by an on-board battery pack system. At 0700, it is anticipated that the SUV will be driven into place after completion of all instrument checks and maintenance. Recharging of the SUV battery packs will occur during overnight, non-sampling times. The SUV position will be recorded by GPS each day. After approximately 3 hours of sampling, the GMAP and portable units will be placed near the SUV for 1 – 1:30 hours of side-by-side measurements for quality assurance purposes.

GMAP Monitoring

Mobile monitoring using an instrumented electric vehicle, collecting data at <10 s time resolution, will be conducted along the driving route shown in Figure 5-5. This driving route is selected to include parallel and transects on both sides of the highway with and without noise barriers affecting air flow. The driving route also has transects extending beyond 300 m in order to observe the extent of highway impact and distance at which the highway signal is no longer discernable from background concentrations. The planned route is approximately 20 miles in distance and anticipated to be completed in 35 minutes, allowing for 4-5 repeats within a 2-3 hour sampling period.

Documentation of Site Characteristics

The monitoring site will be characterized by the following means:

The noise barrier will have the physical dimensions estimated (height, thickness, length) including coordinates for the barrier ends, and any changes in height or distance from the road.

The highway and side road elevation characteristics will be quantified using the GPS elevation data.

15


Other roadside features (e.g., buildings, walls, dispersed trees) will be observed using imagery – the GMAP webcam, Google Streetview, and ArcGIS.

Co-located Sampling

During the sampling period, all air monitoring equipment will be placed together at an equal distance from the road to collect data for a minimum of one and one half hour period. Mobile monitoring with the GMAP vehicle will occur for approximately 2-3 hours; afterwards, the GMAP vehicle will be parked next to the SUV for a minimum of 1-hour (or as long as battery power is available). The portable air monitoring stations will also be placed nearby the GMAP and SUV for a minimum of 1-hour of collocated measurements. The portable samplers will continue collocated measurements for up to 2-hours prior to ending the daily sampling protocol.

Table 5-5. Measurements to be conducted during the monitoring campaignMethod Measurement Rate Instrument

GMAP Carbon monoxide (CO) 1 s Quantum cascade laser (Aerodyne Research, Inc.)

GMAP Nitrogen dioxide (NO2) 1 s Cavity Attenuation Phase Shift (Aerodyne Research, Inc.)

GMAPParticle number concentration (size range 5.6-560 nm, 32 channels)

1 s Engine Exhaust Particle Sizer (Model 3090, TSI, Inc.)

GMAPParticle number concentration (size range 0.5-20 µm, 52 channels)

1 s Aerodynamic Particle Sizer (Model 3321, TSI, Inc.)

GMAP Black carbon 1-5 s Single-channel Aethalometer (Magee Scientific, AE-42)

GMAP Longitude and latitude 1 s Global positioning system (Crescent R100, Hemisphere GPS)

GMAP Video of route <1 s WebcamSUV Carbon Monoxide (CO) 1 min Non-dispersive infrared (NDIR) – FRM

SUV Nitrogen Oxides (NO/NO2/NOx) 1 min Chemiluminescence - FRM

SUV Black Carbon 1-5 s Single-channel Aethalometer (Magee Scientific, AE-42)

PFSP Black carbon 1 min Micro-Aethalometer (Magee Scientific, Model AE-51)

PFSP Particle size and number 1 min Handheld PM sampler (HHPC-6, MetOne)

PFSP 3D wind speed and direction 1 s Ultrasonic anemometer (RM Young, Model )

BSP Black carbon 1 min Micro-Aethalometer (Magee Scientific, Model AE-51)

Note: GMAP = Geospatial Mapping of Air Pollutants electric vehicle16


SUV = stationary sampling Sport Utility VehiclePFSP = Portable Fixed-Site PlatformBSP = Backpack Sampling Platform

5.2. General Approach and Test Conditions for Each Experimental Phase

This is an ambient monitoring study that will not have experimental phases. The sampling details are described in other portions of Section 6.

6. Sampling Procedures6.1. Site-Specific Considerations

The primary objective of this monitoring study is to measure in situ the dispersion of traffic emissions from major roadways to surrounding near-road areas. The measurement results are understood to be site-specific in nature due to the unique building, road, and topography of the sites. Another site-specific feature are the highway traffic characteristics - parameters such as vehicle speed, total traffic volume, vehicle age and fleet mix, and fraction of heavy-duty vehicles all play a role in the resulting spatial distribution and chemical composition of emissions. All of these factors will be considered in the interpretation of results.

The primary site-specific factors that will affect monitoring procedures are local meteorology, traffic volume, and side-road traffic. As discussed in Section 5, the monitoring sites and schedule are determined based upon an analysis of local wind trends, with the sampling locations selected to have areas of interest predominantly downwind of the highway. In addition, GMAP driving routes were selected to include areas on both sides of the highway, allowing for downwind/upwind air pollutant concentrations to be observed for multiple wind directions.

Side-road traffic is an important consideration affecting both the monitoring procedures as well as the ensuing data analysis. Emissions from local traffic can lead to biases in the monitoring data and obscure the characterization of the highway traffic emissions impact on near-road areas. Several methods are used to minimize this potential bias in the data set. First, the portable, fixed-site platforms will be located away from other air pollutant sources besides the highway. The SUV and portable fixed sites will be located away from other known air pollutant sources beside the highway. In addition, the highway will be the primary source within 300 meters of the sites in the predominantly upwind direction from where monitoring will occur. For the mobile and fixed monitoring, the degree of side road traffic is a site selection criterion (Section 5) for the near-road sections of the mobile monitoring driving route and the fixed site locations. During sampling, webcams will be used onboard the GMAP to record local vehicle exhaust episodes. Also, the GMAP vehicle will be driven as to provide maximum distance

17


behind other vehicles on the road. Finally, post-processing will be completed of the electric vehicle data set - an algorithm will be applied that detects and flags time periods with apparent local exhaust impact, characterized by a sudden sharp spike in carbon monoxide concentrations. This algorithm will be applied equally to all data recorded by the GMAP vehicle. This algorithm was developed by EPA based upon multiple near-road field studies, and found to successfully remove incidences of local exhaust, with a total loss of data around 2-3% for driving routes with relatively low side road traffic similar to the side roads targeted for this study. The MATLAB code for this algorithm is provided in Appendix A. This algorithm will be tested again for the field study by comparing the flagged time periods for one complete field sampling deployment with a webcam video record from the dashboard of the GMAP vehicle. If found to be insufficient in detecting biases from local vehicle exhaust, the algorithm will be modified with the rationale for any changes documented.

6.2. Sampling Equipment and Procedures

The sampling equipment used in this study includes one mobile monitoring vehicle equipped with air monitoring analyzers and webcams, and SUV equipped with air monitoring sensors, and two portable, fixed-site platforms equipped with air quality measurement devices. The SUV fixed-site platform will also have meteorological monitors (i.e., ultrasonic anemometer). The GMAP vehicle and the portable fixed sites also have global positioning systems (GPS). Finally, supporting equipment includes a truck equipped with a car-hauling trailer to transport the electric vehicle as needed.

The following general tasks will be completed as part of the daily instrument deployment process. Specific details regarding the operation of the GMAP vehicle are found in Appendix B.

1. GMAP electric vehicle will be fully charged (refer to Appendix B, Section 3, GMAP MOP #1) and equipped with a GPS, webcam, and high time resolution air monitoring instruments. Time of response will be tested for the various instruments to exactly time-align data.

2. QC checks on all GMAP analyzers will be performed prior to sampling each day, either in the laboratory or in the field immediately prior to sampling. The multiple computers used for data logging will be time-aligned with the GPS-derived “true” timestamp.

3. The SUV battery packs will be fully charged and equipped with a webcam and high time resolution air monitoring instruments. Time of response will be tested for the various instruments to exactly time-align data.

4. QC checks on all SUV analyzers will be performed prior to sampling each day, either in the laboratory or in the field immediately prior to sampling. The multiple computers used for data logging will be time-aligned with the GMAP GPS-derived “true” timestamp.

5. The batteries for the portable, fixed-site sampling equipment will also be fully charged at the same location as the GMAP and SUV battery packs.

18


6. The GMAP electric vehicle will be transported to the field site on its own power, using the SUV, or using a vehicle equipped with a car-hauling trailer depending on where a suitable storage and charging area is found near the sampling site. The portable, fixed-site sampling boxes and/or charged batteries will be driven to the field site using these or a separate vehicle.

7. The SUV will be parked in the pre-determined location described in Section 5 for approximately 4-5 hours of sampling, depending on battery charge.

8. The portable, stationary sampling equipment will be replaced at the field site within the weatherproof boxes equipped with air monitoring instruments, ultrasonic anemometers, and data logger to perform data acquisition of air quality instruments and ultrasonic anemometer meteorology data. If the boxes cannot be left in the same location overnight, the boxes will be placed in their proper sampling location. The inside temperature of the weatherproof boxes will be logged to detect and mitigate any potential temperature-induced complications with sampling instruments.

9. Ambient mobile monitoring sampling will take place for approximately 3 hours. The GMAP electric vehicle will be driven repeatedly around an assigned route.

10. Three hours of sampling will be conducted with the portable, fixed-site sampling system, GPS locations of the portable and SUV sampling locations will be confirmed each day.

11. After approximately 3 hours of sampling, the GMAP air monitoring instrumentation will be parked next to the SUV for collocated sampling for a minimum of 30 minutes (1-hour desired if battery charge allows).

12. The two portable, fixed-site samplers will be re-located to the SUV/GMAP sampling location for co-located sampling with these two platforms for a minimum of 30 minutes (1-hour desired if battery charge allows)

13. The GMAP/SUV analyzers will be shut down, with data downloading occurring in the field or at an off-site location

14. The two portable, fixed-site sampling platforms will continue side-by-side sampling after the GMAP/SUV analyzers have been shut down. Co-located sampling of the portable, fixed-site samplers should occur for a minimum of 2-hours.

15. All vehicles and sampling equipment will be relocated to a secure and temperature-controlled environment for recharging and overnight storage. Data downloading from these analyzers can occur either at the sampling or the storage location. The weatherproof boxes may remain overnight at the field location if permission is obtained from applicable property owners.

The manuals providing procedures for operating the electric vehicle, air monitoring instruments, and supporting instrumentation are provided in the Appendices B-J as follows:

Appendix B: GMAP manual Appendix C: Hemisphere GPS manual Appendix D: EEPS manual

19


Appendix E: APS manual Appendix F: Aethalometer manual (Model AE42) Appendix G: QC Laser system software operating instructions Appendix H: HHPC-6 analyzer manual Appendix I: micro-Aethalometer manual Appendix J: Ultrasonic anemometer manual Appendix K: Leaf area index monitor manual Appendix L: Aethalometer manual (Model AE51) Appendix M: CO NDIR manual Appendix N: NOx Chemiluminescence manual Appendix O: CAPS manual

Log sheets will be kept to record the daily sampling events and QA checks – example log sheets for mobile, stationary, and portable monitoring are below.

20


Example:GMAP Daily Log Sheet

Date: 11/5/12 Sampling Day: 12Site: Denver – 48th Street Operator: ParikSync computer clocks with GPS

Fill QCL with liquid N2

PM instrument zero checks: QCL Check:

ADD CAPS here or separate?

Start time: End time:

Route 8:15 AM 11:13 AM

Webcam 8:13 AM 12:18 PM

Inter-comparison 11:13 AM 12:15 PM

QC checks acceptable?

Aethalometer EEPS APS CO NO2

Total number of laps for the run: 5Observed weather: Light winds from the SWComments: For QCL CO Check, using N2 cylinder with 0.18 ppm CO

21

Time 7:05 AM 2:05 PM

CO 176 ppb 164 ppb

N2O 1.463 ppb 0.311 ppb

Aethalometer 6:49 AM

EEPS 6:48 AM

APS 6:47 AM


________________________________________________________________________

Example:SUV Daily Log Sheet

Date: 11/5/12 Sampling Day: 12Site: Denver – 48th Street Operator: ParikSync computer clocks with GPS

PM instrument zero checks: Gaseous Analyzer Checks:

Start time: End time:

Sampling 8:15 AM 11:13 AM

Webcam 8:13 AM 12:18 PM

Inter-comparison 11:13 AM 12:15 PM

QC checks acceptable?

Aethalometer CO NO2

Total minutes of co-located measurements with GMAP: 58Observed weather: Light winds from the SWComments: ________________________________________________________________________

22

Time 7:15 AM 2:15 PM

CO 172 ppb 174 ppb

NO2 146 ppb 141 ppb

Aethalometer 6:52 AM


Example:Portable, Stationary Daily Log Sheet

Date: 11/5/12 Sampling Day: 12Site: Denver 48th Street Operator: JamesUltrasonic anemometer direction aligned

Monitoring Station 1: Location: Latitude: 42.284324° Longitude: -83.441834° Comments: Behind barrier, 50 m from barrier edge, 40 m from highway

Start time: 8:04End time: 11:06

Aethalometer: HHPC:Sync GPS/computer/instrument clocks Sync GPS/computer/instrument clocks Batteries charged Batteries charged Data downloaded Data downloaded Instrument zero checked Instrument zero checked

Monitoring Station 2: Location: Latitude: 42.285342° Longitude: -83.441980° Comments: Behind barrier, 150 m from barrier edge, 40 m from highway

Start time: 8:05End time: 11:07

Aethalometer: HHPC:Sync GPS/computer/instrument clocks Sync GPS/computer/instrument clocks Batteries charged Batteries charged Data downloaded Data downloaded Instrument zero checked Instrument zero checked

Comments:Sampling ended early due to precipitation.

23


Co-located Sampling Start time: 11:14End time: 1:16

24


6.3. Quality Control in Sample Analysis

No physical samples will be collected and analyzed – data will be collected using air monitoring instruments. The Quality Control procedures for the air monitoring and supporting measurements are described in Section 8.

6.4. Sample Preservation

No physical samples will be collected in this study – data files will be stored as described in Section 6.5 and 6.6.

6.5. Sample Numbering

Air monitoring data will be time-stamped and will not have specific sample numbers assigned. For the GMAP and SUV samplers, the individual data files will be uniquely labeled by sampling vehicle, location, instrument, and date – PlatformName_Location_ Instrument_YYYYMMDD. For example, UFPs measured using the EEPS on-board the GMAP electric vehicle on Nov. 10, 2012 during the Denver Study at 48th Street would be labeled as – GMAP_DENVER48ST_ EEPS_20121110. For the portable, fixed-site platform (PFSP) samplers, for example, PM measured using the HHPC at Site 1 on Nov. 10, 2012 during the Denver Study would be labeled as – PFSP-1_DENVER48ST_ HHPS_20121110. The raw data files will maintain this naming scheme in a master database stored by the EPA Project Leader. The EPA Project Leader will maintain a project folder labeled as DENVER-BARRIERS and nested folders will be labeled with the parent-level label plus a descriptive label, such as DENVER-BARRIERS-RawData, DENVER-BARRIERs-DataAnalysis, etc. Any periods of missing data due to equipment malfunction, severe weather, or unacceptable quality of data will be documented in the project notebook.

6.6. Sample Chain-of-Custody

The original data files will be collected and maintained by ARCADIS personnel on a hard drive for a period of 5 years. At the completion of the entire field campaign and data post-processing, final data files and site notes will be sent to the EPA Project Leader, Rich Baldauf, for final storage on an EPA server, which is backed-up nightly. An overview of the raw data collection and storage is provided in Figure 6-1. Prior to sampling with the GMAP vehicle, the on-board data logging computers are manually time-synchronized to the satellite-based time recorded by the on-board GPS. The fixed point monitoring stations will be time-synchronized to be within 10 s during daily checks. The instruments log data using either generic programs (e.g., WinWedge or HyperTerminal) or instrument-specific programs (e.g., Aerosol Instrument

25


Manager for the EEPS). Prior experience using these instruments guides the number of external computers needed to simultaneously log all data streams or reliance upon internal memory for certain instruments. For this project, two instrument models (HHPC-6 and AE51) may log internally after time-synchronization to the GPS. The remaining instruments (GPS, EEPS, APS, AE42, QC laser, ultrasonic anemometer, and webcam) will log to an external onboard computer.

After data is recorded and downloaded from the instruments or external data-logging computers, the data is transferred using a USB drive and a copy is retained by ARCADIS personnel. Along with field notes recorded electronically, the raw data is later transferred to the EPA network, with the exception of large video files that are stored to an external hard drive (200+ GB) and maintained by the EPA TL. The raw data files are stored in a folder labeled “raw data” and remain unchanged, with copies of these files made for post-processing activities. Secondary processing of data, for purposes of aligning real-time concentrations and location data as well as analysis of trends, is described in detail in Section 9.

Figure 6-1. Data collection and storage process

7. Measurement Procedures7.1. Analytical Method

No analytical methods will be used in this project.

7.2. Project Overview26


7.2.1.GMAP monitoring

This study will utilize a mobile monitoring vehicle (GMAP) operating in driving-mode sampling. An image of the GMAP vehicle is provided in Figure 7-1.

Figure 7-1. Mobile monitoring vehicle planned for use in this study.

While most sampling approaches used in typical stationary fixed site sampling studies directly translate to this project, such as the use of electrically conductive tubing to minimize particle loss and careful time-alignment of data-logging laptops, some unique considerations need to be made for the GMAP vehicle which operates in driving mode. The two primary additional considerations made are the sampling inlet and the determination of lag time for sampling instruments.

The GMAP sampling inlet is designed to provide isokinetic conditions while the vehicle is in motion. Isokinetic conditions are most important for the larger particle sizes (e.g., PM10) and are generally negligible for gases and ultrafine particles. Two air velocity parameters are taken into consideration in the design of the inlet – the combined inlet volume flow of the instruments on-board the electric car and the air flow rate as the vehicle is in motion. The vehicle speed is expected to vary over the course of a driving route from approximately 0-60 mph. Thus, the inlet design will assume an air flow rate for a vehicle driving at approximately 30 mph. In order to determine whether speed-based correction will be needed for higher driving speeds, preliminary field tests will be conducted at RTP driving the GMAP vehicle over a range of speeds (0-60 mph) on roads with minimal traffic (refer to Appendix A for more information).

In order to precisely align position and air concentration data for the GMAP vehicle, another important factor is characterizing the amount of lag time associated with an air sample transporting through the sample line and measurement by a given air monitoring

27


instrument. This lag time will be experimentally determined by inducing a sudden concentration change for the analyte of interest, such as using a HEPA filter for the particulate instruments, and observing the amount of time before the concentration is recorded by the monitoring instrument. Further details are provided in Appendix B.

7.2.2.Portable Stationary Monitoring

This study will utilize portable stationary sampling systems. The use of portable, stationary sampling systems has been utilized in previous near-road studies, including the ORD study in Raleigh, NC in 2006. Key components of conducting a successful portable, stationary sampling scheme include proper charging and handling of batteries, timely downloading of data from these systems, careful mapping of monitoring sites, and extensive documentation of instrumentation or environmental issues encountered during sampling. A minimum of three days of testing in RTP will be conducted with fully equipped portable, stationary sampling platforms to ensure that rechargeable batteries are used capable of supporting over 24 hours of sampler operation, data logging, cooling of the box interior, and temperature monitoring. To limit the potential for data loss due to insufficient battery power, data downloading will occur each day from this equipment. A handheld GPS unit will be used to ensure that monitoring locations are consistent each day. In addition, the site operator will maintain a daily log, noting the time at each location, any instrument errors or problems, and local environmental and traffic conditions. Stationary and backpack sampling units will be time synchronized with the GMAP analyzers daily. The GPS and diary will ensure consistent placement of the sampling systems and verification of weather conditions each day. The diary will also be used to identify any environmental or traffic conditions that may also influence monitoring results. The clock used for identifying sampling times will be synchronized to the SUV and GMAP analyzers. For PM sampling, conductive tubing materials and lengths will be the same for each pair of samplers located in the portable fixed site platforms. Thus, both micro-aethalometers will use the same type and length of tubing, and both HHPC samplers will use the same type and length of tubing.

7.2.3.SUV Monitoring

This study will utilize the SUV to provide a moveable stationary sampling platform of continuous pollutant measurements. The SUV has been utilized in previous near-road studies, including the ORD studies in Raleigh, NC in 2006 and Las Vegas, NV in 2009. Key components of conducting a successful sampling scheme with the SUV platform include proper charging and handling of batteries, timely downloading of data from these systems, careful quality assurance, and extensive documentation of instrumentation or environmental issues encountered during sampling. A minimum of three days of testing in RTP will be conducted with the fully equipped SUV sampling platform to ensure that

28


rechargeable batteries are used capable of supporting sampler operation, data logging, and temperature monitoring. To limit the potential for data loss due to insufficient battery power, data downloading will occur each day from this equipment. A handheld GPS unit will be used to ensure that monitoring locations are consistent each day. In addition, the site operator will maintain a daily log, noting the time at each location, any instrument errors or problems, and local environmental and traffic conditions. All SUV sampling units will be time synchronized with the GMAP analyzers daily. The GPS and diary will ensure consistent placement of the sampling systems and verification of weather conditions each day. The diary will also be used to identify any environmental or traffic conditions that may also influence monitoring results. The clock used for identifying sampling times will be synchronized to the SUV and GMAP analyzers. For PM sampling, conductive tubing materials and lengths will be minimized.

7.2.4.Critical measurements

The critical measurements for this project are listed in Table 5-5.

7.3. Calibration Procedures

All equipment will be calibrated annually and/or cal-checked as part of standard operating procedures. Calibration records are kept on file. Maintenance records are kept for any equipment adjustments or repairs in project logbooks that include the date and description of maintenance performed. Details on the instrument-specific calibration and cal-check procedures are available in the Appendices and Section 8.1.

8. Quality Metrics8.1. QC Checks

The QC checks used in the field to assess the data quality indicator goals (section 8.2) are provided in Table 8-1. All mobile and fixed-site sampling equipment will also be collocated at the end of each sampling day as described in Section 3, which will also serve as a secondary QC check with acceptance criteria listed in Table 8-2. Finally, an extended inter-comparison (>4 hours) will be performed at RTP prior to shipment of the equipment and vehicles to Denver in order to detect and troubleshoot any inconsistencies between duplicated measurements.

Table 8-1. Procedures Used to Assess QA ObjectivesMeasurement

Parameter Analysis Method Assessment Method

29


Particulate size and number concentration EEPS 3090

Single point flow check prior to study initiation;Zero-check before and after daily field

deployment

Particulate size and number concentration APS Model 3321


deployment

Particle count and diameter HHPC


deployment

Black carbon AethalometerSingle point flow check prior to study initiation;

Zero-check before and after daily field deployment

NO2Cavity Attenuation Phase Shift

(CAPS), Aerodyne

Multi-point calibration check prior to study initiation; single-point zero/span verification

before and after daily deployment

CO QC Laser, AerodyneMulti-point calibration check prior to study

initiation; single-point zero/span verification before and after daily deployment

CO Non-dispersive infrared detector (NDIR), API

Multi-point calibration check prior to study initiation; single-point zero/span verification


NO/NO2/NOx Chemiluminescence, APIMulti-point calibration check prior to study

initiation; single-point zero/span verification before and after daily deployment

Ambient wind speed and wind direction

RM Young Ultrasonic Anemometer Model 81000

Calibration by Metrology Lab prior to study initiation; single-point zero/span verification


Location (longitude, latitude, elevation) Hemisphere GPS Pre-deployment comparison of measured GPS

data with known reference location

Leaf area index LAI2000Comparison of LAI data with known range of

historically observed values for similar vegetation types.

8.1.1.Particle Measurement Instrument Assessment

The EEPS, APS, Aethalometers, and HHPC measure particulate components based on manufacturer calibration and are run using default manufacturer calibration settings. These instruments have all received manufacturer’s calibrations within the past three years and will be assessed prior to the study for performance by observing the instrument response to zero check. The zero check is conducted by attaching a high efficiency particulate air (HEPA) filter to the sampling inlet, which removes >99% of particulates of diameter >0.3 m. While the HEPA filter is in place, downstream particulate concentrations should read near zero for the instrument to be deemed acceptable. In addition, a single point flow verification will be conducted using a calibrated flow meter and flows should be within 10% of the set point.

To ensure the PM instruments are continuing to operate well during the field study, a zero check will be performed daily during the field study. Given a failure in meeting this data quality indicator, response actions include, but are not limited to, (1) performing cleaning

30


maintenance, (2) changing sampling inlet, and (3) seeking technical support from the instrument manufacturer.

8.1.2.Global Positioning System Assessment

The GPS system will be verified by driving along a specific route and comparing reported longitude/latitude against mapping data. Several software or internet-based programs are available to determine whether reported data matches the actual route, including ArcGIS, MATLAB, and Google Earth Pro. Location differences of more than 10 meters will be investigated using the high-resolution GPS.

8.1.3.Ultrasonic Anemometer Assessment

The ultrasonic anemometer DQIs are checked annually while in operation as part of a routine calibration procedure. Since the calibration procedure was previously conducted more than one year before the start of this field study, the procedure will be repeated again prior to the initiation of the field study. The calibration for wind speed and wind direction was conducted using the wind tunnel at the EPA Page Road Annex facility. The ultrasonic anemometer reported wind speed was compared against a factory-calibrated Shortridge airfoil and wind direction was verified at precise wind angles.

8.1.4.Quantum Cascade Laser Assessment

The QCL will be assessed by zero and span measures using gas standards (N2, CO). Prior to the field study, a multi-point check will be performed using a gas mixing system (e.g., Environics) over the anticipated measurement range: 100 ppb – 2 ppm for CO. During the field study, the QCL system will be verified using a one point zero and span check with a gas standard within the measurement range mentioned above.

8.1.5.Cavity Attenuation Phase Shift Assessment

The CAPS will be assessed by zero and span measures using gas standards (NO2). Prior to the field study, a multi-point check will be performed over the anticipated measurement range: 10 - 500 ppb NO2. During the field study, the CAPS system will be verified using a one point zero and span check with a gas standard within the measurement range mentioned above.

8.1.6.Non-dispersive Infrared Detector Assessment

31


The NDIR system will be assessed by zero and span measures using gas standards (CO). Prior to the field study, a multi-point check will be performed over the anticipated measurement range of 100 ppb – 2 ppm CO. During the field study, the NDIR system will be verified using a one point zero and span check with a gas standard within the measurement range mentioned above.

8.1.7.Chemiluminescence Assessment

The NO/NO2/NOx sampler will be assessed by zero and span measures using gas standards (NOx). Prior to the field study, a multi-point check will be performed using a gas mixing system (e.g., Environics) over the anticipated measurement range of 100 ppb – 2 ppm for NOx. During the field study, the system will be verified using a one point zero and span check with a gas standard within the measurement range mentioned above.

8.2. QA Objectives and Acceptance Criteria

The Data Quality Indicator goals for accuracy, precision, and completeness for this project are listed in Table 8-2. For this project, precision will be evaluated through collocated sampling of like instruments at the end of each sampling day. Any failure of the instrumentation to meet the DQI goals will be reported to the EPA TLP. Data collected during time periods in non-attainment with DQI goals will be flagged as questionable, but not necessarily considered invalid. Corrective action to be taken depends on the nature of the problem encountered.

Table 8-2. Data Quality Indicator Goals for the Project

Parameter

MeasuredAnalysis Method Assessment Criteria Completeness Precision

Corrective Actions Given Failure to meet

CriteriaBlack

carbon Aethalometers (1) Single point flow

(1) +/- 10% of 90% +/- 10% (1) Sampling inlet

will be checked for

32


check prior to study(2) Zero-check prior to and after field deployment

set-point(2) 5-min average at <20% of ambienta

obstructions. If flow errors continue, instrument troubleshooting and/or flow recalibration will take place.(2) Instrument connections will be checked and zero-check repeated. Data collection will continue given repeat failure, but data will be flagged.

Particulate size and number

concentration

HHPC-6

(1) Single point flow check prior to study(2) Zero-check prior to and after field deployment

(1) +/- 10% of set-point(2) 5-min average at <20% of ambienta

80% +/- 15%

(1) Sampling inlet will be checked for obstructions. If flow errors continue, instrument troubleshooting and/or flow recalibration will take place.(2) Instrument connections will be checked and zero-check repeated. Data collection will continue given repeat failure, but data will be flagged.


concentration

EEPS 3090



90% n/a

(1) Sampling inlet will be checked for obstructions. If flow errors continue, instrument troubleshooting and/or flow recalibration will take place.(2) Instrument connections will be checked and zero-check repeated. Data collection will continue given repeat failure, but data will be flagged.


concentration

APS Model

3321



90% n/a (1) Sampling inlet will be checked for obstructions. If flow errors continue, instrument troubleshooting and/or flow recalibration will take place.(2) Instrument connections will be checked and zero-

33


check repeated. Data collection will continue given repeat failure, but data will be flagged.

Carbon monoxide

Quantum cascade laser

(1) Multi-point calibration check prior to study(2) Zero/span one point verification check prior to and after daily field deployment

+/- 10% of set-point

90% +/- 10%

Instrument laser temperature, laser power, and spectral fit will be assessed. If unable to improve the agreement, instrument troubleshooting with manufacturer assistance will take place.

Nitrogen dioxide

Cavity Attenuation Phase Shift



90% +/- 10%

Instrument temperature and power will be assessed. If unable to improve the agreement, instrument troubleshooting with manufacturer assistance will take place.

Carbon monoxide

Non-dispersive infrared



90% +/- 10%


Nitrogen oxides

Chemiluminescence



90% +/- 10%


Wind speed and

direction

Ultrasonic anemometer

Data will be compared with field operator observations on log sheet.

General matching of wind direction and speed

90% +/- 5%

(1) Orientation of sonic anemometer will be checked and corrected if found to be out of alignment.

Location Hemisphere GPS

Status lights indicate

Status lights 95% +/- 20 m Sampling will

discontinue until

34


collected signal

indicate collected signal

GPS is determined to be functioning properly.

aThe HEPA filter removes >99% of particulates of diameter >0.3 m, however the particle instruments also measure particles under 0.3 m, which may have a higher penetration efficiency through the HEPA filter.

9. Data Analysis, Interpretation, and Management9.1. Data Reporting

Research results are intended for publication in scientific journals, thus no writing of internal EPA reports is expected. The EPA Technical Lead, Rich Baldauf, will be responsible for generating a data report for internal use among EPA scientists. This report will include information on the sampling collection times, field notes, and preliminary data review (completeness, QC checks).

9.2. Data Validation

Verification and validation of the procedures used to collect and analyze data are critical to achieving project objectives. Study personnel will be responsible for ensuring that the sampling methods, quality control procedures, and validation methods described in earlier sections of this document are followed. ARCADIS will be responsible for the operation of all field instrumentation. The EPA Technical Lead, Rich Baldauf, will be responsible for the data review for the SUV and portable air monitoring data. An EPA Project Scientist, Gayle Hagler, will be responsible for reviewing the GMAP vehicle data.

9.3. Data Analysis

9.3.1.GMAP vehicle data

Following the collection of raw data, as described in Section 6, the GMAP data are processed using several standard algorithms developed in MATLAB, which is described in Figure 9-1 – (1) Adjustment for lag time, (2) Combining GPS location data and air monitoring data into a joint matrix that is now time and spatially-resolved concentration data, and (3) Spatially consolidating data from repeat drives into spatial increments of interest (user-defined) for purpose of calculating averages or other statistics. Data analysis can take place at various levels of post-processing. For example, inter-comparing data for the same variable (e.g., black carbon) may only require that the data be time-aligned (step 1). Observing concentration changes in both time and location would require steps 1 and 2 to be conducted. Finally, looking at 2-hour average concentration maps would require steps 1-3 to be completed. While the algorithms used to process steps 1-3 are customized

35


based upon each specific instrument’s data, the common algorithm elements are provided in Appendix A.

Figure 9-1. Post-processing of mobile monitoring raw data for use in analysis.

For this study, data analysis after the above processing steps will include parallel time series and correlation analysis of the air pollutant measurements (following step 1) as well as geospatial and temporal analysis (following step 2 and/or step 3) of the driving-mode mobile monitoring vehicle data. These and other analyses may lead to further post-processing of data, dependent on project needs. Additional data used for interpretation will include regional meteorology data and other air pollutants measured simultaneously on-board the electric vehicle. Gayle Hagler will be the main individual responsible for the analysis of the GMAP data.

9.3.2.Portable, Stationary Monitoring

Following the collection of raw data, as described in Section 6, the portable stationary sampling data will be processed using Microsoft Excel spreadsheet software programs. The 1-min average data, obtained from the raw 1-sec data, will be time-synchronized to allow for direct comparisons among measured values at the multiple monitoring locations. Collocated measurements from like monitors will be directly compared to determine averages, coefficients of variation, and correlation between the like monitoring pairs to identify precision and variability among the monitoring pairs. Time series measurements will then be evaluated at the multiple monitoring sites to determine normality of the data, whether there are statistically significant differences between monitoring site measurements, and then quantifying percentage differences between measurements at different locations, if applicable. Pending these results, regressions will be run to determine

36


associations among air quality measurements at select sites with other meteorological, time, and air quality data. Based on this regression analysis, specific parameters explaining a high proportion of the measurement variability will be further explored to identify associations and potential percent differences in measurements at multiple locations. All spreadsheets containing project data will be documented to include project identifiers including project title, dataset, responsible parties, and data qualifiers/comments. The spreadsheets willl also include the identification of any macros, equations, field/cell definitions, QA data, and individual sample comments.

9.3.3.SUV Monitoring

Following the collection of raw data, as described in Section 6, the SUV sampling data will be processed using Microsoft Excel spreadsheet software programs. The 1-min average data, obtained from the raw 1-sec data, will be time-synchronized to allow for direct comparisons among measured values at the multiple monitoring locations. Collocated measurements from like monitors will be directly compared to determine averages, coefficients of variation, and correlation between the like monitoring pairs to identify precision and variability among the monitoring pairs. Time series measurements will then be evaluated at the multiple monitoring sites to determine normality of the data, whether there are statistically significant differences between monitoring site measurements, and then quantifying percentage differences between measurements at different locations, if applicable. Pending these results, regressions will be run to determine associations among air quality measurements at select sites with other meteorological, time, and air quality data. Based on this regression analysis, specific parameters explaining a high proportion of the measurement variability will be further explored to identify associations and potential percent differences in measurements at multiple locations. All spreadsheets containing project data will be documented to include project identifiers including project title, dataset, responsible parties, and data qualifiers/comments. The spreadsheets willl also include the identification of any macros, equations, field/cell definitions, QA data, and individual sample comments.

9.4. Data Storage Requirements

No physical samples will be collected or require storage. Section 6.5 and 6.6 discuss the chain-of-custody and storage for the mobile monitoring data.

10.Reporting10.1. Deliverables

37


Deliverables from this study include final quality-assured field data and manuscripts for publication in scientific journals.

10.2. Expected Final Products

Anticipated final products for this study are peer-reviewed, published research papers in science journals and presentations at scientific conferences.

11. References Baldauf, R., E. Thoma, A. Khlystov, V. Isakov, G. Bowker, T. Long, and R. Snow. 2008. Impacts of

noise barriers on near-road air quality. Atmospheric Environment 42:7502-7507.Bowker, G. E., R. Baldauf, V. Isakov, A. Khlystov, and W. Petersen. 2007. The effects of roadside

structures on the transport and dispersion of ultrafine particles from highways. Atmospheric Environment 41:8128-8139.

Hagler, G. S. W., R. W. Baldauf, E. D. Thoma, T. R. Long, R. F. Snow, J. S. Kinsey, L. Oudejans, and B. K. Gullett. 2009. Ultrafine particles near a major roadway in Raleigh, North Carolina: Downwind attenuation and correlation with traffic-related pollutants. Atmospheric Environment 43:1229-1234.

Heist, D. K., S. G. Perry, and L. A. Brixey. 2009. A wind tunnel study of the effect of roadway configurations on the dispersion of traffic-related pollution. Atmospheric Environment In Press, Accepted Manuscript.

38

pasteur.epa.gov · web viewa number of monitoring sites were considered throughout the us. table...

Documents