For Presentation at the Air & Waste Management Association's 91st Annual Meeting & Exhibition, June 14-18, 1998, San Diego, California

Use of PM10 Monitoring Data to Evaluate the Effectiveness of the Dust Control Program During the Construction of the Central Artery Tunnel Project

98-TAA.05P

James Dolan and Alex Kasprak
Bechtel/Parsons Brinckerhoff,
One South Station, Boston, MA 02110,
Phone: (617) 342-1135 Fax: (617) 426-1143

Guido Schattanek
Parsons Brinckerhoff, Inc.
One Penn Plaza, New York, NY 10119
Phone: (212) 465-5595 Fax: (212) 465-5595

Ping K. Wan
Bechtel Power Corporation
9801 Washingtonian Blvd., Gaithersburg, MD 20878
Phone: (301) 417-3144 Fax: (301) 670-0297

ABSTRACT

The Central Artery/Tunnel (CA/T) project in Boston has been under construction for over six years. The downtown portion of the project consists of the excavation and construction of an eight lane underground roadway over which the existing elevated highway continues to carry almost 200,000 vehicles a day. Currently, more than a 100 pieces of construction equipment operate within a mile stretch of this 300-foot wide urban corridor where residents, businesses, pedestrians, local street and elevated highway traffic will continue to exist during the next few years of construction.

The high density of so many construction activities in such a busy corridor required the implementation of a comprehensive dust control program. During the Spring, Summer, and Fall of 1997, a PM10 monitoring and dust inspection program was conducted along the CA/T project's alignment. Baseline PM10 levels were measured at the same locations during the Summer of 1994. This paper describes how the PM10 data was used to evaluate the effectiveness of the dust control program.

INTRODUCTION

The CA/T is a major highway expansion project, consisting of many individual construction contracts occurring simultaneously. When completed, in the year 2004, the CA/T project will include approximately 13 miles of highway including open tunnel and at grade sections.

During the construction of these roadways, surfaces previously covered will be exposed, numerous pieces of machinery will be in use simultaneously, and traffic flow in the city of Boston will be continually altered and restricted. All this occurs while residences, pedestrians and businesses must continue their way of life, and existing roadways remain open to local street and highway traffic.

The construction activities taking place along the alignment include: installation of slurry walls, temporary decking, fabrication of base, wall and roof slab, excavation of soil, underpinning of existing elevated artery, demolition of structures and surface restoration. Construction materials to build the project and excavated, demolition and other material are continually moved in and out of the construction areas. In all, 10 million yards of soil will be excavated during the construction of the CA/T project.

One major environmental concern of the project are air quality impacts, particularly from fugitive dust and emissions. To control fugitive dust and emissions the CA/T project requires every contractor to implement appropriate dust control measures. Implementation of these dust control measures helps to ensure that the Project will comply with the 24-hour PM10 National and State Ambient Air Quality Standards (AAQS) of 150 micrograms per cubic meter (µg/m3).

In 1992 per the agreement of the Construction Air Quality Committee, the CA/T project which up to that time had only modeled to determine air quality impacts decided to purchase air monitoring equipment for the purposes of collecting PM10 data. Then, in the years between 1992 and 1996 the CA/T project modeled and monitored PM10 levels before and during the utilities construction, monitored PM10 levels at various locations prior to mainline construction, and monitored various predicted and known hot spots.

In 1997 attempts were made to continue to use modeling as a prediction tool during the mainline construction activity, but because of the complexity, volume, proximity, and duration of work occurring concurrently, models were found not to be good indicators of true PM10 levels. Based on this information then, a conceptional plan to replace modeling with monitoring and compliance field inspections alone, was developed. The object was to monitor PM10 levels while simultaneously carry out field inspections to ensure the contractor(s) were in compliance with the dust specifications within their respective contracts. The duration for this activity June through October was accepted as the time of year PM10 levels would be highest, and visible dust most likely to cause a nuisance impact.

DUST CONTROL AND FIELD INSPECTION PROGRAM

Each contract has to comply with a "Dust control specification" originally derived from the Massachusetts Highway Department blue book1, that is continually upgraded as more information becomes available and modified to meet the more stringent requirements of an urban environment. The components of a typical dust control specification include:

• Using water or a wetting agent in solution to control dust on-site
• Using wind barriers and wind screens to prevent spreading of dust from the site
• Having a wheelwash station and/or a crushed stone apron at egress/ingress areas to prevent dirt being tracked onto public streets
• Using vacuum powered street sweepers to remove dirt tracked onto streets
• Covering all dump trucks leaving sites to prevent dirt spilling onto streets.
• Covering, wetting, or using a binding agent on all excavate stockpiles.

Through the field inspection program the contractor was contacted to regulate nuisance dust conditions, and also to implement additional controls at sites where elevated PM10 levels occurred. Among these supplementary mitigation measures implemented include, using a soil binding agent on exposed surfaces to control dust in the work zones, placing additional crushed stone at egress points to prevent dirt being tracked onto public streets, and increasing the frequency of vacuum powered street sweepers to remove dirt from public streets.

While the control for nuisance dust ultimately lies with the contractor, the inspections ensured that the issue of dust control remained high on their agenda.

MONITORING PROGRAM

The baseline monitoring program performed during the Summer and Fall of 1994 had as its main emphasis the determination of PM10 levels before the commencement of construction, and also with aid of modeling to ascertain potential hot spots that might cause the NAAQS of 150 µg/m3 to be exceeded with the addition of construction activity. The monitoring program performed during 1997 on the other hand, replaced modeling, and was aimed at determining the actual PM10 levels over a sustained period of time with view to minimizing potential impacts through an extensive and on-going dust control program.

Site Selection

All sites selected for PM10 were 10 to 30 meters outside the construction fence lines, on sidewalks, at street intersections, or in parking lots adjacent to the CA/T alignment. The objective was to identify unbiased, representative sampling sites.

The 1997 monitored sites were chosen for three reasons. First, of the seven initial sites chosen, six had previously been monitored during the Summer and Fall of 1994. Second, the six sites were east of the CA/T alignment, where in Boston during the Summer and early Fall the prevalent wind is from the west, since the artery corridor runs south to north, the expected PM10 levels from the construction activity at these sites would be worst at this time of year. Third, sensitive abutters such as hospitals and residences were located within proximity to the CA/T alignment at these sites (see figures 1 and 2).

The 1997 monitoring program expanded from seven to twelve sites, however only the six sites for baseline data is available are considered here.

Monitoring Equipment and Procedures

24-hour average PM10 samples were collected two days per week at beginning and ending at midnight each day. Sample days were rotated each week to ensure that all week days were represented during the 5-month 1997 monitoring period2.

The monitoring equipment used to collect PM10 samples at the six sites were portable MiniVOL air samplers3. In addition, a Cahn C-33 microbalance for filter weighing, quartz 47 mm diameter micro fibre the filter medium, a Tapered Element Oscillating Microbalance (TEOM®) an EPA designated monitor (No. EQPM-1090-079) the reference station, and a RS-232 Data Logger to obtain and store data from the TEOM®, were used.

Portable MiniVOL samplers draw air through a pre-selector that allows only particles of 10 µm or less aerodynamic diameter to impact the sample filter. The control unit which forms the bulk of sampler ensures that a volumetric flow rate of five liters per minute at ambient conditions was maintained throughout the pre-programmed 24-hour sampling period. A rechargeable battery pak powers the unit.

MiniVOL samplers were mounted to brackets attached to utility poles approximately 10 feet above the ground. Each sampler remained in the field 2 days per week per the schedule described above. PM10 samples were not collected during the weekends.

The protocol used for the construction phase monitoring program included:

• Calibrating MiniVOL samplers prior to the start of the monitoring program and every two weeks thereafter.
• Programming MiniVOL samplers to collect 24 hour samples two days per week during the monitoring program.
• Exchanging sampler battery and filter holder assembly between sampling days.
• Equilibrating sample and control filters in a temperature and humidity controlled environment for 24-hours.
• Weighing sample and control filters on a calibrated Cahn C-33 microbalance scale.

For precision purposes, a collocated portable sampler was used to simultaneously obtain PM10 data at one of the six monitoring sites throughout the study period. In addition, for the entire study period, a portable air sampler was collocated at the CA/T project's TEOM® 1400 series air sampler. The TEOM®, which has been in operation since 1992, is a gravimetric method, that provides continuous 1-hour and 24-hour average PM10 data.

A Pearson linear correlation coefficient was calculated to compare TEOM® and a collocated MiniVOL sampler to determine the accuracy of the MiniVOL sampler. During construction (1997) the TEOM® and the MiniVOL PM10 levels were correlated (p<0.001, n = 26, r = 0.59).

RESULTS AND DATA ANALYSIS

During the 1994 baseline study, 81 valid out of a possible total of 84 PM10 samples were collected on two 14 weekday periods between July and September (see Table 1). During the 1997 construction phase study, a total of 223 valid out of a possible 264 PM10 samples were collected every two weekdays between June and September (see Table 2).

Data were subjected to the most appropriate test based on sample size and where appropriate, a normality test 4,5.

Data analysis was performed, first to ascertain whether without any construction activity had the PM10 levels changed in the Boston area. In order do this the PM10 levels from a Massachusetts Department of Environmental Protection (MDEP) Hi-VOL sampler located in Boston outside the CA/T alignment were compared during the months June through October, for 1994 vs. 1997. An independent two tailed t-test performed, indicated there was no difference between background PM10 levels for 1994 vs. 1997 (p = 0.871, n = 25).

A Mann-Whitney U test was then performed between the six sites monitored by the CA/T project on dates in 1994 and 1997 to determine if there was any consistent difference between the baseline and construction PM10 levels. PM10 levels at sites 1 and 6 during the construction phase monitoring were significantly higher than the baseline (Table 3). There was no significant difference between baseline and construction PM10 levels at the other 4 sites.

Finally, a Wilcoxon matched pairs test was performed to compare the construction phase PM10 levels recorded, during the first 10, and the last 10 days at site 6. The purpose of the test was to identify any increase or decrease in PM10 levels at the final days of the program once all the dust control measures were being applied for four months. PM10 levels recorded, were higher during the first 10, than the last 10 days (p = 0.01).

DISCUSSION OF RESULTS

In a urban environment trying to determine the causes of high PM10 concentrations (i.e., exceeding the NAAQS of 150 µg/m3) is a difficult task. This difficulty is attributed to the fact that there are multiple contributing PM10 sources occurring simultaneously. However, it is fair to say that construction activity generates a condition, or set of conditions, that exacerbate PM10 concentrations. Among these activities are demolition of structures, operation of construction equipment on haul roads, and movement of large amounts earth material.

As indicated by the data, PM10 levels at sites 1 and 6 were significantly higher than the baseline, during the construction phase monitoring. Coincidentally these sites were also the most active with respect to construction activity and street traffic, and therefore required the most extensive mitigation measures to control dust.

In the case of site 1, construction activity from two adjacent contracts at different stages of completion, at a heavily trafficked intersection resulted in elevated PM10 levels during the 1997 monitoring period. Mitigation measures used at this site included extensive use of a vacuum street sweeper on streets bordering contract areas, and maintenance of egress areas to ensure dirt was not tracked onto to streets.

Site 6, located west of both a temporary elevated artery and a construction contract whose activities were carried out on exposed earth, and during early August demolished old elevated artery parallel to the monitoring site. The cumulation of these activities and circumstance resulted in the most persistent elevated PM10 levels occurring at this site. As a result, this site required the most far-reaching dust mitigation, including the use of Soil Sement® as a binding agent on the exposed surface, extensive watering during artery demolition, and frequent use of a vacuum street sweeper. As demonstrated by the analysis of the data, the use of these measures to control dust, did reduce PM10 levels during the latter part of the construction phase monitoring.

The construction activities at site 2 included slurry wall work at two work areas that were not altered during the 1997 monitoring period. Because of this permanency, both the work zones and the egress areas were graded with crushed stone, and a vacuum street sweeper was in constant use at the intersection. However, because the amount of construction activity was prominent in this area, high PM10 levels did occur on isolated days, though the average concentration was not significantly higher than the baseline.

Construction activity in the vicinity of site 3 was the most progressed at the time of monitoring, all slurry walls were completed and roof slabs were being put in place. Field inspections at this site have indicated that the contractor maintained egress areas with crushed stone and constantly used a vacuum street sweeper on streets bordering the contract areas.

At site 4, construction activity commenced in mid-July 1997, and involved utility relocations that was the only work performed during the monitoring period. Relocating utilities did not generate large amounts of dust. Field inspections indicated that a vacuum street sweeper was in constant use on public streets in this area throughout the monitoring period.

Finally, site 5 was located in the center of an half acre parking lot, west of a construction area that contained about 20 excavate stockpiles. Field inspections have indicated that the stockpiles remained covered, and/or were wetted down throughout the monitoring period.

Because of extensive mitigation and/or characteristic of the locations, PM10 levels on average remained relatively low at sites 3, 4, and 5 throughout the monitoring period.

CONCLUSION

Dust control programs can only be effective if the contractors are committed to implement them.

When the construction phase monitoring program was established, its intended objective was to use the PM10 results as a indicator of nuisance dust conditions, and as a tool to regulate the level of mitigation required. This paper clearly indicates that PM10 levels can be reduced using dust mitigation methods, such as vacuum street sweepers, crushed stone aprons, and dust binding agents.

The combined effect of the monitoring and field inspection programs achieved the objective of keeping contractors aware of the permanent need of mitigation measures. Even though, we did not have sites without mitigation (where we could have observed how high PM10 levels could have been), the reduction in PM10 levels during the length of the program at site 6, gives an indication of the beneficial effects of the program in keeping the average PM10 levels close to the baseline with the exception of a few isolated instances.

ACKNOWLEDGMENTS

We thank Ralph DeGregorio of the joint venture of Bechtel Corporation and Parsons Brinckerhoff for statistical analyses.

REFERENCES

1. Massachusetts Highway Department,Standard Specifications for Highways and Bridges: pp 118 - 119, 1988, Boston.

2. Lin Y.J.; Kasprak, A. Portable PM10 monitoring for a large Roadway Tunnel Project: A & WMA 90th Annual Meeting, 1997, 97-MP 6.07.

3. AIRmetrics Inc. MinVOL Portable Air Sampler Operation Manual: Springfield, Oregon 1997.

4. Gomez, K.; Gomez, A. Statistical Procedures for Agricultural Research: New York, 1983.

5. Statsoft. Statistica: Tulsa, 1991

Table 1 Summary of 1994 background PM10 Concentrations
Site ID #	1	2	3	4	5	6
# of sample days	14	12	13	14	14	13
Average 24hr (µg/m3)	64	62	56	51	25	38
Maximum 24hr (µg/m3)	84	83	76	77	39	45
Std. Deviation (µg/m3)	13	13	14	16	8	10

Note:
PM10 24 hour NAAQS = 150 µg/m3

This manuscript contains graphics that must be viewed and printed by the Adobe Acrobat (TM) plug-in, click to download the graphics.