Air Quality Assessment Techniques for Roadway Tunnel Projects

Guido Schattanek, Parsons Brinckerhoff Inc New York, New York
Ping K. Wan, Bechtel Power Corp. Gaithersburg, Maryland

AIR AND WASTE MANAGEMENT ASSOCIATION SINCE 1907

For Presentation at the 89th Annual Meeting & Exhibition Nashville, Tennessee June 23-28, 1996

INTRODUCTION

The Central Artery/Tunnel (CA/T) Project in Boston, Massachusetts is the largest roadway project underway in the United States. The project is funded by the Federal Highway Administration (FHWA), with construction under the jurisdiction of the Massachusetts Highway Department(MHD), and management and engineering assistance by the Bechtel/Parsons Brinckerhoff joint venture. It Includes the construction of thirteen miles of new tunnels within the City of Boston. The tunnel ventilation system as currently designed includes seven ventilation buildings to provide fresh air to the tunnels and to exhaust the mixture of tunnel air and vehicular emissions from cars traveling inside the tunnels. In addition, a less expensive ventilation system using "jet" fans that push air out of the tunnel portals rather than a system employing ventilation buildings is being evaluated to ventilate exit ramps. To comply with Federal, state, and local air quality regulations, an extensive air quality evaluation was performed to assess the impact of vehicular emissions from the ventilation buildings, tunnel portals, and at-grade open roadways which would be affected by the CA/T Project.

This paper describes the methods used to evaluate the air quality impact of these sources on ambient levels of carbon monoxide (GO), fine particulate matter (PM10), and nitrogen dioxide (NO2 ). The evaluation addresses the impact of the CA/T Project on regional levels of air pollutants, and the impact of emissions from ventilation buildings and tunnel portals on localized levels of air pollutants. The supporting air quality analyses employed both physical and analytical modeling techniques. The results of the analytical and physical modeling analyses are compared against monitoring data collected within the project area to determine precision in predicting future air quality levels.

Given the long duration of the project, a comparison of the results of air quality forecasts completed during the past seven years for different construction scenarios, using several versions of the same analytical models is also included in the paper.

BACKGROUND

A Final Environmental Impact Statement/Report (FEIS/R) prepared for the CA/T Project was published in August of 19851. Due to a number of major unresolved issues, the preparation of a Supplemental EIS/R was initiated in 1988, and culminated with the publication of a 14 volume Final Supplemental EIS/R in November of 1992.

The air quality evaluation included in the 1990 FSEIS/R provided a description of the existing air quality conditions in the area, an analysis of the impact of the project on regional levels of CO, hydrocarbons (HC), nitrogen oxides (NO,) and PM10, an evaluation of CO levels at 39 intersections affected by the project, an evaluation of the air quality impacts of six proposed ventilation buildings on CO, NO2 and PM10 levels, and an analysis of the eight motor-vehicle related toxic pollutants for which Massachusetts Department of Environmental Protection (MDEP) has established Acceptable Ambient Limits.

The tunnel ventilation system designed for this project and analyzed in the FEIS/R is a "Full Transverse System" in which supply air is introduced from under the roadway and exhaust air is extracted through the tunnel ceiling. This provides a curtain of air flow transverse to the flow of vehicular traffic. Both supply and exhaust air are pumped by large fans housed in each ventilation building. This is the most common type of ventilation system used in the United States for large roadway tunnels. Design of the systems based on the principle that dilution of motor vehicle emissions generated within the tunnel and release of them at a certain height and exit velocity, will achieve applicable ambient air quality standard (AAQS) in the area affected by the project. The system was also designed to maintain CO levels inside the tunnels within FHWA and Environmental Protection Agency (EPA) criteria during normal operating conditions, and to provide enough reserve capacity to control smoke and heat during a tunnel fire.

Given its long construction period, the 1991 FHWA Record of Decision for the project, included the creation of a Construction Air Quality Committee (CAQC) composed of representatives of FHWA, EPA, MDEP and MHD. The mandate of the CAQC is to evaluate, mitigate and minimize any air quality impacts that could occur during the project's construction period.

This CAQC established procedures, guidelines, and methodologies for the completion of project related technical air quality evaluations. The first technical evaluation performed in 1992 indicated the need for short term CO and PM10 monitoring programs due to predicted exceedances of the NAAQS. These monitoring programs established a baseline of the air quality conditions along the project area, to determine the effectiveness of required mitigation measures. To date, CO levels have been monitored at 32 locations and PM10 levels at 38 locations over a five year period.

During 1993 the MHD decided to evaluate alternative engineering designs that for the placement of the multilevel bridge over the Charles River with a combination of bridges and tunnels. A year later, the 1994 Charles River Crossing FSEIS/R3 was published updating the environmental assessment completed in 1990. The new air quality analysis evaluated the same issues as in the 1990 document with the incorporation of another ventilation building. This evaluation was performed with all the updated versions of the air quality models used in 1990.

In 1995, due to the positive results obtained from extensive in-tunnel fire tests performed at the West Virginia Memorial Tunnel, in which the effectiveness of alternative ventilation systems for the control of heat and smoke were tested, the FHWA authorized the evaluation of a more economical system to be used to ventilate exit and entrance ramps. This system consists of blowing air in a longitudinal pattern, the same as the flow of traffic, by using jet fans installed in the tunnel ceiling. In this type of system the exhaust air is released through the tunnel portals instead of through ventilation stacks. To evaluate the air quality impacts of these ground level horizontal releases, extensive wind tunnel testing is being performed for different physical configurations and traffic scenarios.

The following sections describe the techniques, methodologies, and results of these evaluations.

TECHNIQUES AND APPROACH

The air quality evaluation of a large scale highway/tunnel project commonly includes an assessment of the direct effects of emissions released through the tunnel ventilation system, and an evaluation of the effects that the project will have on the local street system at a localized (microscale) level, where traffic congestion could create violations of the CO standards, and at a regional (mesoscale) level to insure that the project will not increase the overall emissions for the region affected by the project.

In the case of the CA/T project all the air quality assessment techniques where developed through a long consultation period with the regulatory agencies involved (EPA Region I and MDEP), and many scenarios were tested before finalizing the assumptions for each modeling process. During this process a great effort was made to obtain localized project specific data, in terms of traffic, emissions, and meteorological conditions, instead of relying on averages provided by State or Federal agencies.

Regional Emissions (Mesoscale Analysis) A regional emission inventory of 5,600 traffic links was completed for an area shown in Figure 1. It included all major highways, arterials and local roads. The 1990 FSEIS/R predicted traffic and emission levels for 1987 (the base year), 1998 (the expected project completion year) and 2010 (the project design year). The traffic network was calibrated through extensive traffic counts, and the future traffic growth was based on detailed housing, construction and employment forecasts.

The main traffic effects of the project were to redistribute local traffic onto the new highway system, eliminating traffic bottle necks and reducing congestion. As a consequence, the future effects of the CA/T project were expected to result in a significant increase on the average network speed for the area.

Total vehicle miles traveled (VMT) for each roadway category for each neighborhood was divided by the corresponding vehicle hours travel (VHT) to obtain average network speeds by neighborhood and roadway type. These speeds were used to obtain the emission factors for each pollutant analyzed using the Mobile 4 Version of the EPA Emission Factor Model", for each one of the 24 neighborhoods affected by roadway category. Determination of the inputs to the model included the use of significant localized data; such as thermal state of the engine, vehicle classification, vehicle registration, ambient temperature, fuel volatility and other local area parameters.

In addition, emission estimates were developed on a link by link basis using the specific speed for each roadway link (instead of average network speeds), and summing up all the individual emissions. This approach resulted in total emission levels that were from 5% to 9% lower than the ones obtained through the neighborhood-type approach. However, since the percent of emissions reductions due to the CA/T project were equivalent for both approaches, and the neighborhood type method was less data intensive, the link by link approach was discontinued in the future assessments.

The 1994 FSEIS/R repeated the air quality analysis for an expanded area of coverage to the north and west of the area covered in 1990 as shown in Figure i. The overall traffic network was then updated to reflect current growth projections and ongoing transportation projects. The emission factors were updated based on the Mobile 5a version of the EPA Emission Factor Models. Some of the above mentioned parameters, were also updated to reflect the mandated changes in the 1990 Clean Air Act Amendments, and to be consistent with the State current emission inventory.

Localized CO Effects at Critical Intersections (Microscale Modeling) The second component of the evaluation was the prediction of CO levels at 39 critical intersections through a microscale modeling analysis (shown in Figure 2). The analysis included in the 1990 FSEIS/R was completed using the EPA 's CAL3QHC dispersion model", and also the Mobile 4 version of the EPA Emissions Factor Model. It predicted CO levels for 39 intersections for the years 1987, 1998 and 2010; and 17 intersections for the peak of the construction period estimated at the time to be the year 1994. These selected intersections represented the locations where the worst CO levels, and the highest project related impacts could be expected to occur.

At the time that the 1990 FSEIS/R was completed, the CA/T this was the first large scale project to apply the CAL3QHC model. (EPA's manual for CAL3QHC was released in September of 1990). The modeling effort used peak hour traffic as the basis to determine the maximum I-hour CO level and an average hourly traffic for the peak 8-hour period to obtain 8-hour CO levels. This approach was recommended by EPA Region I, and it was based on the concept that the average 8-hour traffic was more accurate than using peak hour traffic and a persistence factor to estimate 8-hour CO levels.

A few years later EPA recommended to change the method to use peak-hour traffic and persistence factor to obtain eight-hour CO levels, following the new 1992 "EPA Guideline for CO Modeling at Roadways Intersections". A comparison of both approaches performed at a few intersections, resulted in lower CO levels (ranging from 5% to 20%) when the later (peak-hour traffic and persistence factor) approach was used.

Due to high CO levels predicted for the peak of the construction period, the CAQC initiated a reevaluation of the construction related air quality impacts for alternative types of traffic management plans, which, in turn, resulted in the initiation of a short term saturation CO monitoring program. This short term monitoring program measured 8-hour CO levels at 32 locations along the project alignment, for 3 to 6 week periods during the last five winters (1992-1996).

Portable samplers were deployed to measure 8-hour CO concentrations. The selection of the monitoring locations was based on the previous CO modeling predictions and traffic forecasts. Each monitor sampled two 8-hour periods per day covering the AM- and PM-peak traffic periods. This monitoring data provided an excellent baseline to assess the need and extent of mitigation measures. A total of 2,640 8hour CO samples were collected.

The 1994 FSEIS/R re-evaluated future CO levels at 23 intersections for the years 2001 and 2010. This reevaluation was conducted using updated version of air quality models, i.e., Version 2 of the EPA CAL3QHC model8 (released in 1992), and the Mobile 5a version of the EPA Emission Factor Model.

Ventilation Building Analysis (Analytical Modeling and Wind Tunnel Testing) The third component of the evaluation included the assessment of localized effects of the project's ventilation buildings on CO, NO2 and PM10 levels for the area surrounding each ventilation building. The localized evaluation of NO2 was performed to determine if the project complied with the one-hour MDEP Policy Guideline for NO2 of 320 ug/m3. This guideline applies to stationary facilities, including tunnel ventilation systems, built within the Boston Metropolitan area.

In-Tunnel Air Quality. The tunnel ventilation system was designed with an installed supply air capacity of 65 cubic-feet-per-minute per lane-foot of tunnel. This system is capable to maintain in-tunnel CO levels between 20 and 60 PPM, and NO, levels between 1 and 5 PPM, during normal peak hour traffic conditions

FHWA-EPA in-tunnel criteria (based on time exposure) requires to maintain CO levels below 120 PPM, when the time exposure does not exceed 15 minutes during peak rush hour traffic, 65 PPM for exposure between 15 and 30 minutes, 45 PPM for 30 to 45 minutes, and 35 PPM when the motorists could remain 60 minutes inside the tunnels. Since the longest distance of any continuous tunnel for the CA/T project is 16,130 feet, it is estimated that an averaged motorist traveling inside the tunnel would spend less than 8.5 minutes during PM peak hour conditions, giving sufficient ventilation margin to meet the CO intunnel criteria.

The quantities of air flows for each tunnel section, were determined using the Subway Environmental Simulation model (SES). This model was developed by Parsons Brinckerhoff(it is also proprietary) to evaluate air flows within tunnels, and has the capability to include the effects of the piston action created by the moving vehicles. The emission sources within the tunnel segments were calculated using the future traffic forecasts for the peak and off-peak hour scenarios, and the Mobile 4 and 5a versions of the EPA Emission Factor Model.

The tunnel emission calculations assume that air within the tunnel is well mixed with pollutant concentrations in the tunnel at the same levels as those within the ventilation stacks. One important consideration is that different traffic scenarios produce different sources, and different ventilation scenarios change pollutant concentrations inside the tunnels. Serving as a " reality check", these predicted future in-tunnel emissions were compared with the results of short term CO monitoring inside the existing Sumner and Callahan tunnels in Boston during the summer of 1991. Monitored levels correlated well with the predicted future emissions for the proposed tunnels.

Analytical Modeling - Stack Height Determination. Given the highly urbanized area that surrounds these ventilation buildings, low pressure regions, referred to as cavity and wake regions, can be formed immediately downwind of these buildings. To avoid aerodynamic downwash of the exhaust plume caused by adjacent buildings, the first level of analysis was to calculate the minimum stack height through the Good Engineering Practice (GEP) using the EPA Building Profile Input Program (BPIP 1985) 9.

These evaluations resulted on GEP stack heights ranging between 900 and 1,100 feet. Since the construction of stacks of this height found very impractical in the urban areas, the air quality analysis focused on the determination of a more practical stack height that would comply with the federal and local air quality regulations.

This analysis was completed in two steps:

Confirmation of minimum stack height using a Cavity Analysis, using the SCREEN-2 model'0 (EPA, 1988) to estimate pollutant levels at receptors close to the ventilation buildings (normally three buildings height downwind of a structure).

A dispersion analysis using the Industrial Source Complex Short Term model (ISCST-I for the 1990 SFEIS/R and ISCST-211 for the 1994 document), with the incorporation of 5 years of hourly surface meteorological data from Logan Airport National Weather Station (NWS) station, and upper-air data from Portland, Maine NSW station located 100 miles north on the Atlantic Coast. The Portland station was selected because it is located along the Atlantic coast, with similar terrain features as Boston, and no significant climatic differences between the two cities.

In the case of the i-hour NO2 evaluation, the Ozone Limiting Method (OLM) was applied to account for the conversion of nitric oxide (NO) to NO2. Most of NO, emitted by motor vehicles is in the form of NO. Once it is emitted, NO is converted to NO2 through atmospheric reactions with other oxidants, mainly Ozone (03)·

The OLM method requires concurrent background NO2 and 03 hourly data, with the percentage of conversion limited by the amount of 03 available for the period. These analyses assumed that 5% of NO, vehicular emissions are emitted as NO2 and 95% as NO; and that the maximum conversion possible, given the distance to the receptors analyzed, was 76% for the cases where 03 was not limiting the conversion.

The OLM method was applied as follows:

In the cases of the other pollutants analyzed (GO, PM10 and annual NO2 ) the impacts of the sources were added to the appropriate background concentrations.

Wind Tunnel Tests. Wind tunnel testing for each of the seven ventilation buildings, was applied due to the fact that the proposed ventilation buildings will be positioned among numerous tall and massive existing structures, which would induce intricate flow patterns and complex mechanical turbulence, beyond the limits of accurate simulation by analytical models.

The purpose of the wind tunnel tests was to verify the analytical modeling results, and to refine the minimum stack height that will comply with the federal and local air quality regulations. A 1:400 scale model for each building was constructed at the RWDI Wind Tunnel Testing Facility in Guelph, Ontario. These models included all nearby existing structures, and future approved developments scenarios within a 1,600 feet radius of each ventilation building.

A flow visualization test was performed for each building to identify the receptor locations where the highest impacts would be expected to occur. The second step was to perform a tracer gas test, using a known concentration of a tracer gas released from the stacks. These tests were performed for several stack height scenarios (including future proposed developments surrounding each site), under a broad range of wind speeds, wind directions and stack exit velocities.

The results from the wind tunnel tests provided dilution ratios for each studied scenario. These results were then combined with five years of local meteorological data to obtain their probability of occurrence. The final results were compared to the stack heights determined through analytical modeling.

Jet Fan Ventilation Analysis. This air quality assessment investigated the effects of the use of Jet Fan Tunnel Technology (longitudinal ventilation) in lieu of, or in combination with, a full transverse ventilation system.

The evaluation analyzed two underground tunnels for two exit ramps approximately 2, 150 and 1,300 feet long. The use of jet fan ventilation system, if implemented, would eliminate the exhaust emissions through Ventilation Building #8, and result in significant cost savings.

Due to the same complexity of the flow patterns in Boston, a wind tunnel test simulation study was deemed more appropriate than the use of analytical modeling. Two 1:100 scale models were built at the RWDI wind tunnel testing facility. These models included all major buildings within 500 feet of the portals. The effects of the moving vehicles were simulated using moving belts with hemisphere scaled objects representing the aerodynamic characteristics of the predicted traffic speed and density.

Flow visualization tests were performed to facilitate the determination of critical receptor locations, and tracer gas tests were performed for three traffic scenarios. The results of these tests were then added to the appropriate background concentrations, and surface street levels estimated through microscale CO modeling.

In the case of the NO2 evaluation, the assumed 76% maximum conversion factor from NO to NO2 (as used in the ventilation buildings analyses) was believed to be extremely conservative and unrealistic, given the proximity of receptor locations to the tunnel portal. Consequently, a monitoring program for NO and NO, was conducted to obtain a more realistic conversion factor that could be used for receptor locations within a few hundred feet to the source. This monitoring program consisted of the placement of six NO/ NO, monitors and a meteorological tower along a 600 foot fence in the vicinity of North Station in Boston. The diesel emissions from the trains operating at the Station were used as a source of NO,, and the distance to the train sources was used to estimate an average conversion rate. The results of this evaluation are presented in a separate paper in this conference under the title of "Atmospheric Transformation of Vehicular Emissions from Nearby Tunnel Portals."

DISCUSSIONS OF RESULTS

Regional Emissions.
The overall vehicle miles traveled (VMT) for the year 2010 was predicted to be 5.180 million VMT per day, with a 17% growth from 1987 to 1998, and a 33% growth from 1987 to 2010. Given the highly urbanized level of the Boston area, the VMT induced growth attributed to the CA/T project was in the order of 7%, when compared to the future No Build scenario. Significant emissions reductions (see Tables 1 and 2) of CO, HC, and NO, over 20 years were forecast (despite the 33% increase in VMT) due to the effects of cleaner vehicles mandated under federal and state emission control programs.

Comparison of future forecasts based on the Mobile 5a emission model with the earlier predictions using the Mobile 4 version of the model, indicated that the most recent predictions are less optimistic about emission reductions for the future. As shown in Table 2, CO reductions over 20 years decrease from 67% to 13%, HC reductions decreases from 62% to 34%, and NO, reductions decreases from 35% to 18%.

In the case of PM10, where the emission rates are proportional to VMT by roadway type, the increase in emissions over the 20 year period closely follows the traffic increase for each period, i.e., 22% from 1987-2010 and 19% from 1990-2010.

The percentage of emissions generated by traffic on the local streets for the 2010 CA/T scenario amounts to: 67% for CO, 59% for HC, 55% for NO, and 87% for PM10. Since the traffic in the local streets accounts for 42% of the total VMT, this clearly indicates, that on a VMT basis, the emissions generated by traffic in the local streets are higher than those generated on the highway system.

The overall emission benefits of the project when compared to the future no build scenario (shown in Table 2), are moderate for CO, and very small for HC, NO, and PM10, and result mainly from the expected reduction of traffic congestion in the downtown area, due to the expanded capacity of the highway. The completion of the CA/T project is expected to divert almost 10% of the local street traffic to the highway system.

Localized CO Levels at Intersections When examining the results of the microscale analysis for the worst five intersections for the year 2010 (shown in Table 3), the use of the Mobile 5a Emission Factors for the project 1994 document resulted in predicted future CO levels of approximately 25% higher than the levels predicted in the 1990 document with the Mobile 4 version of the program. This is consistent with the results of the regional analysis.

When the effects of the CA/T project are compared to the future no build, the CO levels at the five worst intersections are 10 to 15% lower with the CA/T project. These expected lower CO levels with the CA/T project are mainly the result of better expected traffic flows, and the reduction of traffic bottlenecks in the most congested areas.

A comparison of the year 2010 future predictions with current monitoring data indicates that the future predicted levels are higher than monitored levels. Given the expected reductions in future area wide emissions, the opposite effect (higher monitored values versus future predicted levels) should be expected. However, this counter intuitive result is an effect of the modeling analysis performed to represent worst case scenarios for regulatory compliance purposes.

Ventilation Buildings Analysis For all the ventilation buildings the limiting factor, in terms of minimum stack height determination, was compliance with the one-hour MDEP Policy Guideline for NO2. In all the cases the highest NO;! levels were predicted to occur at elevated receptors in the surrounding area of each ventilation building.

Due to engineering reasons, some of the ventilation building's design (as shown in Table 4), were changed prior to the wind tunnel test. These changes make the comparison between the results for wind tunnel testing and analytical modeling difficult. However, in the case of Ventilation Buildings #6 and #7, where the same physical ventilation building configurations remained through the whole process, it can be seen that the wind tunnel tests resulted in compliance stack heights that were 10 to 15% lower than the ones predicted using analytical modeling.

CONCLUSIONS

In the air quality evaluation of a large highway/tunnel project, it is essential to analyze both, localized effects, and regional long term effects.

In the case of the CA/T project, the long term air quality effects of the project are beneficial at a regional level, with some expected localized deterioration in a limited number of areas. Consequently, detailed evaluation of localized impacts was initiated and resulted in several design changes and project commitments, to ensure compliance with all federal, state and local air quality regulations.

Although the air quality impacts associated with the CA/T Project were re-analyzed in 1994 resulting in higher specific pollutant estimates, the same conclusions arrived at in the 1990 document in terms of overall compliance, are still valid. In addition, the monitoring data gathered by the project during those four years, provided a higher level of confidence for the project future air quality predictions.

ACKNOWLEDGMENTS

The authors would like to acknowledge Alex Kasprak from the Bechtel Corporation, Helen Ginzburg and Dal Fenton from Parsons Brinckerhoff, and Mark Vanderheyden and Mike Lepage from RWDI.

REFERENCES

1. Federal Highway Administration, 1985, Third Harbor Tunnel (190) Central Artery (193) Project. Final Environmental Impact Statement and Final 4(f) Evaluation, Massachusetts Department of Public Works, EOEA #4325, Boston, August.

2. Bechtel/Parsons Brinckerhoff (B/PB), 1990, Supplemental Environmental Impact Statement/Report, EOEA #4325, Boston, November.

Central Artery (193)/ Tunnel (190) Project, Final Massachusetts Department of Public Works,

3. Bechtel/Parsons Brinckerhoff (B/PB), 1993, Central Artery (193)/ Tunnel (190) Project. Charles River Crossing, Final Supplemental Environmental Impact Statement/Report, Massachusetts Highway Department/ Federal Highway Administration, EOEA #4325/ FHWA-MA-EIS-82-02-FS3, Boston, December.

4. U. S. Environmental Protection Agency (EPA), 1989, MOBILE 4 Mobile Emission Factor Model, EPA-AA-TEB-89-OI, Am Arbor, Michigan.

5. U. S. Environmental Protection Agency (EPA), 1992, MOBILE 5a Mobile Emission Factor Model. EPA-AA-TEB-92, Research Triangle Park.

6. U. S. Environmental Protection Agency (EPA), 1990, User's Guide to CAL3QHC, A Modeling Methodology for Predicting Pollutant Concentrations Near Roadway Intersections, Research Triangle Park, September.

7. U. S. Environmental Protection Agency (EPA), 1992, Guideline for Modeling Carbon Monoxide from Roadway Intersections. EPA-454/R-92-005, U.S. Research Triangle Park.

8. U. S. Environmental Protection Agency (EPA), 1992, User's Guide to CAL3OHC Version 2.0, A Modeling Methodology for Predicting Pollutant Concentrations Near Roadway Intersections. EPA454/12-92-006, Research Triangle Park.

9. U. S. Environmental Protection Agency (EPA), 1985, Guideline for Determination of Good Engineering Practice Stack Height, EPA-5-80-023R, June.

10. U. S. Environmental Protection Agency (EPA), 1988, Screening Procedures for Estimating the Air Quality Impacts of Stationary Sources. EPA-450/4-88-010, August.

11. U. S. Environmental Protection Agency (EPA), 1992, User's Guide for Industrial Source Complex Model (ICS2), Vols. I and II, EPA-450/4-92-008, March.

 

Table 1. Regional Analysis Traffic Forecasts 1990 SFEIS/R
Year	1987	2010		Change 1987-2010
		future no build	w/project	future no build	w/project
Total VMT (million)	3.908	5.180	5.546	33%	42%
Avg. Speed (mph)	14.1	10.7	13.3	-32%	-6%
% VMT on Local Streets	51%	52%	43%	-	-
% VMT on Tunnels	5%	5%	16%	-	-

1994 CRC SFEIS/R
Year	1990	2010 w/project	% change 1990-2010
Total Daily VMT (1)	3.752	5.446	45%
Total Daily VMT (2)	5.881	8.062	38%
Average Network Speed (1)	18.1	18.1	0%
Average Network Speed(2)	18.8	17.7	-7%
% VMT on Local Streets	50%	41%	-
% VMT on Tunnels	5%	17%	-
1. Same study area as 1990 SFEIS/R 2. Expanded study area * VMT = Daily Vehicle Miles Travel VHT = Daily Vehicle Hours Travel Average Speed = Average Network Speed = VhTNHT   Table 2. Regional Analysis Emissions Forecasts   Percentage of Emissions Reductions Scenario CO VOC NOx PM10 1990 SFEIS/R         1987-2010 Future No Build 67% 61% 34% 31% (increase) 1987-2010 with CA/T Project 71% 62% 35% 22% (increase) 2010-Project Benefits 12% 4% 2% 7% 1994 CRC SFEIS/R         1990-2010 with CA/T Project 13% 34% 18% 19% (increase)   Scenario CO VOC NOx PM10 1990 SFEIS/R         1987 71% 59% 55% 88% 2010 Future No Build 68% 65% 61% 90% 2010 with CA/T Project 67% 59% 55% 87% 1994 CRC SFEIS/R         2010 with Project 61% 65% 49% 82%

Table 4. Ventilation Building Analysis - Minimum Stack Height
Ventilation Building #	Analytical Modeling		Wind Tunnel Test
	Building Height	Stack Height	Building Height	Stack Height
1	90	135	80	120
3	55	240	55	240
4	78	125	78	125
5	85	181	121	181
6	61	91	61	75
7	72	107	72	95
8	62	155	45	112
Only Ventilation Buildings 6 and 7 retained the same configuration between the Analytical Modeling for the SFEIS/R and the Wind Tunnel Tests. In all cases the limiting factor is compliance with the I-hour MDEP Policy Guideline for NO2.