Estimation and Evaluation of Cancer Risks
Attributed to Air Pollution in Southwest Chicago

Final Summary Report
Prepared for EPA Region 5, Chicago, Illinois
April, 1993

Table of Contents:

Disclaimer
Executive Summary
Conclusion
1.2 Limitations
Chapter III: Study Methodology
III.1: Emissions Inventory
Appendix B: Midway Emissions Inventory and Dispersion Modeling Results
B.1 Introduction
B.2 Emission Inventory Preparation
B.2.1 Emissions From Aircraft Sources
THC Emissions Estimation
Particulate Emissions Estimation
B.2.2 Emissions From Automobile Sources
B.2.3 Emissions From Service Vehicles
B.2.4 Emissions From Other Non-Aircraft Sources
B.3 Emissions Inventory Results
B.3.3 Emissions From Service Vehicles
B.3.4 Emissions From Other Non-Aircraft Sources
B.4 Air Dispersion modeling And Risk Assessment
B.4.2 Air Dispersion Modeling Techniques
B.4.3 Risk Calculation
B.4.4 Unit Risk Factor
B.5 Risk Assessment Results
Exhibit B-1: Number And The Type of Aircraft at Midway Airport
Emission Trends

DISCLAIMER

This summary report was furnished to the U.S. Environmental Protection Agency in fulfillment of Work Assignment No. II-13, Contract No. 68-D0-0018. The opinions, findings, and conclusions expressed are those of the authors and not necessarily those of the U.S. Environment Protection Agency. Similarly, mention of company or product names should not be considered as an endorsement either by the U.S. Environment Protection Agency or by VIGYAN Inc.

ACKNOWLEDGEMENT

VIGYAN Inc. (VIGYAN) would like to acknowledge the contributions of the U.S. Environmental Protection Agency (U.S. EPA) for making this study possible. Funding for this work was provided jointly by the Air and Radiation Division of the U-S. EPA Region 5 and U.S. EPA's Office of Mobile Sources. Patricia Morris (U.S. EPA's Technical Representative), Pamela Blakley, John Summerhays, and Xuan-Mai Tran, all of Region 5's Air and Radiation Division, and Rich Cook and Rich Wilcox of the Office of Mobile Sources, supplied valuable suggestions, comments, and input in the development of this report. Technical assistance from other agencies such as the Illinois Environmental Protection Agency (Sam Long) and the Department of Aviation of the City of Chicago (David Suomi) is highly appreciated as well. VIGYAN would like to acknowledge the role of U.S. EPA Region 5's Southeast Chicago study and Transboundary study. The basic approach utilized by Region 5 for risk assessment in the Southeast Chicago study, including assumptions and emissions inventory, was used in this report. While the results discussed in this report are the work of VIGYAN, some parts of this report have been taken from Estimation and Evaluation of Cancer Risks Attributed to Air Pollution in Southeast Chicago (Summerhays, 1989) and The Transboundary Air Toxics Study (Blakley, 1990). Additionally, the Executive Summary and Chapter 1 of this report were prepared by the Region 5 Technical Representative.

EXECUTIVE SUMMARY

This report estimates the cancer risks associated with 30 air pollutants in the Southwest Chicago area. The study area, approximately 16 square miles including Midway Airport and neighboring suburbs, is bordered on the north by Pershing Road on the south by 70th Street, Harlem Avenue on the west, and Pulaski Avenue on the east. About 93,854 people live in the study area.

Significant uncertainties are associated with estimating risk. These are due to data limitations and assumptions inherent in our current risk assessment methodology. The numerical estimates presented in this report should be viewed only as rough indications of the potential for cancer risk caused by a limited group of pollutants found in the ambient air. A detailed discussion of uncertainties inherent in this study is contained in Chapter V.

The study's purpose is to estimate cancer risks that may be attributed to toxicants in the ambient air in Southwest Chicago. U.S. EPA estimated risks for 30 air pollutants, including 7 known human carcinogens, 21 probable human carcinogens and 2 possible human carcinogens.

According to statistics from the American Cancer Society, about one in three Americans will contract cancer over the course of an average lifetime. Of the approximately 31,000 cancer cases that can reasonably be projected for this population, the report finds that 20 (or about one case every three and a half years) may be caused by the air pollutants studied. There are known and suspected risk factors that can increase tile likelihood of contracting the disease (including both voluntary and involuntary exposures to carcinogens).

This cancer risk from toxic air pollution is of the magnitude of 2 chances in 10,000. This is consistent with other urban area studies that have estimated cancer risks from air pollution. These other studies reported asks ranging from a low of 1 in 10,000 to a high of 10 in 10,000.

Cars, trucks, buses, and trains are the major contributors of carcinogens accounting for about 25% of the total estimated cancer cases. Background concentrations of formaldehyde and carbon tetrachloride account for 19%. The third major contributor in the area is chrome plating operations accounting for about 16% of the total estimated cancer cases. Other significant contributors are aircraft engine emissions from Midway Airport and nonroad mobile sources (such as lawn mowers and snowblowers). Each of these two sources contributes approximately 11% of the total estimated cancer cases. These combined sources account for 81% of the estimated air-pollution-related cancer risks in the area.

Based on U.S. EPA unit risks used in this study, 1,3-butadiene is the most significant pollutant that contributes to cancer risk in the area. The second highest pollutant contributor is Polycyclic organic compounds (POM). Both these two pollutant contributors are emitted mostly from mobile sources such as automobile, aircraft and nonroad equipment engines. Other major pollutant contributors are hexavalent chromium (commonly used in the chrome-plating process) and formaldehyde (generated mostly by photochemical reactions).

CONCLUSION

The cancer risks from air pollutants estimated in this study are consistent with the risks found in other large urban areas. The study’s findings parallel similar studies in other major urban areas of the United States and are typical of highly industrialized communities.

U.S. EPA recognizes the air quality in urban areas must be improved and is taking steps to address the air pollutants that adversely affect human health. The 1990 amendments to the Clean Air Act specifically address many of the sources of air pollution common to urban areas. Generally speaking, the Chicago metropolitan area as a whole should experience a dramatic and visible improvement in air quality as many specific provisions of the new Federal law are implemented.

U.S. EPA's actions include stricter regulation and enforcement of emissions of air toxicants including many of those studied in this report. Transportation control measures, use of cleaner fuels, vapor recovery at gas stations, and stricter controls on consumer solvents are only a few of the changes that will soon affect every person in the metropolitan area.

Finally, risk characterization involves deriving various measures of risk. The simplest measure is individual risk representing the risk attributable to air contaminants at a specific geographic location. An alternative and more appropriate measure of risk is the number of cancer cases estimated to occur among the population in the study area attributable to air contaminants. In addition to estimating these general measures of cancer risk, this study also investigated the origins of these risks and incidences. For example, which source types and which pollutants are the most probable causes of these individual and area-wide risks estimated to result from air pollution in the Southwest Chicago area.

Using the Agency's well-established dispersion models, such as the Industrial Source Complex - Long Term (ISCLT) and Climatological Dispersion Model (CDM), with carefully selected meteorological and emissions inventory data, air toxics concentrations in the receptor grid network from various point and aura sources can be predicted. Based on the estimated concentrations, environmental hazard indices such as lifetime individual risks and lifetime cancer incidences can be calculated at the receptors to support urban toxics and risk assessment studies for the designated study area.

It must be noted that the risk estimates presented in this report should be regarded as only rough approximations of total cancer cases and individual lifetime risks, and are best used in a relative sense. Estimates for individual pollutants are highly uncertain and should be used with particular caution.

1.2 LIMITATIONS

To put the air toxics risk in perspective, cancer risks due to other forms of environmental pollution must be considered. Other exposure routes include exposure through drinking water, skin contact, eating fish from or swimming in lakes that may contain contaminants, and exposure to indoor air contaminants including radon. This study however focused on air pollution risks and did not evaluate risks from other forms of exposure to environmental contamination. Also, other air pollutants that cannot be quantitatively evaluated may cause significant risks.

CHAPTER III
STUDY METHODOLOGY

>III.1 EMISSIONS INVENTORY

III.1.1 SOUTHEAST CHICAGO INVENTORY

The emissions inventory used in the Southeast Chicago Study was utilized as the primary emissions inventory in this study. The emissions inventory is described in detail in separate reports. A detailed description of the inventory is given in a July 1987 report entitled Air Toxics Emissions Inventory for the Southeast Chicago Area, authored by John Summerhays and Harriet Croke. This report documents emission estimates for a wide range of source types, including source types that are traditionally inventoried in air pollution studies as well as some source types that are not traditionally inventoried such as volatilization from wastewater at sewage treatment plants. An addendum to this report (dated August 1989) updates it by describing limited revisions to the previously described inventory. The addendum describes procedures and results of estimating air emissions from various waste handbag sources including facilities for the treatment, storage, and disposal of hazardous waste, abandoned hazardous waste sites, and landfills storing municipal waste. Further details on the estimation of air emissions from the handling of hazardous and nonhazardous waste are provided in two reports by the Midwest Research Institute: Estimation of Hazardous Air Emissions in Southeast Chicago Contributed by TSDFs covering air emissions from the treatment, storage, and disposal of hazardous waste facades and Estimation of Hazardous Air Emissions from Sanitary Landfills, covering air emissions from landfills for ordinary municipal solid waste. Further details for abandoned waste sites are given in a report by Alliance Technologies Corporation entitled Estimation of Air Emissions from Abandoned Waste Sites in the Southeast Chicago Area. The reader interested in more details on the procedures, data sources, and emissions should consult these separate reports.

There are 174 industrial point sources contained in the point source emissions inventory previously prepared for the Southeast Chicago study. These point sources were grouped under four major industry categories: Steel Mills, Wastewater Treatment Facilities, Chrome Platers, and Other Industrial Point Sources. The groupings were done to provide a basis for risk analysis by major industrial source types.

Emissions from the Treatment, Storage, and Disposal Facilities (TSDFs) were considered as well. Among all 75 TSDF point sources inventoried, three major source categories were assigned to distinguish their industrial apes. The three source categories were the Resource Conservation and Recovery Act (RCRA), regulated Hazardous Waste Sites, Municipal Waste Landfills, and Other Hazardous Waste TSDFs.

Because of the nature of the industry and emissions, some source categories included in the study can only be inventoried as area sources. These source categories are as follows:

Examples of per capita area sources inventoried are aerosol cans, paint stripping, and chlorinated drinking water.

III.1.2 ADDITIONAL POINT SOURCES

A search of the Toxic Release Inventory (TRI) database was conducted for new or additional sources to add to the inventory. Most of the sources listed in the TRI database were already included in the Southeast Chicago inventory. Sources of concern to the community were particularly scrutinized. Two proposed sources (Robbins Incinerator and Sun Chemical Incinerator) were added to the inventory to assess potential impact. In addition to the TRI database, information from RCRA permits and Illinois Environmental Protection Agency (IEPA) records were obtained. Six point sources were identified and added to the inventory. A detailed description of these additional sources is included in Appendix A. A list of the additional point sources and Midway Airport, as well as the emissions considered in the cancer risk assessment is provided in Table 3.

Mercury, considered non-carcinogenic in this study, from the proposed Robbins Incinerator was initially estimated at 2.2 tons per year. However, Robbins Incinerator is now committed to carbon injection to control Mercury. Pursuant to consent decree, emissions must not exceed 0.44 tons per year. The records on the General Electric PCB Reclamation facility were scrutinized for any actual or potential PO emissions. U.S. EPA PCB inspectors found no PCB emissions from this facility.

The first four point sources listed in Table 3 were classified as other Industrial Point Sources and appended to the Southeast Chicago point source emissions inventory. Sun Chemical Incinerator and Robbins Incinerator, two proposed facilities, were not considered when the aggregated cancer cases in the Southwest Chicago area were analyzed even though they were inventoried. Instead, these two facilities were singled out and the risk assessment for each of them was conducted separately. The results of the study are not significantly affected by either facility.

III.1.3 EMISSIONS AT MIDWAY AIRPORT

Midway Airport is located within the target 8 x 8 receptor grid network. An area source inventory was set up to scrutinize the estimated emissions from mobile sources at Midway. Based on the technical directives from OMS, the emissions inventory of the Midway Airport area source compasses a 1.7 kilometer x 1.7 kilometer square with the southwest corner of (UTMY,UTMX) = (4625.2,436.64). The main focus of the Midway emissions inventory is emissions from aircraft engines. Emissions by aviation category from all phases of the landing and takeoff (LTO) cycle (approach, taxi-idle, takeoff, and climbout) among all aircraft in 1990 were estimated. Twenty-five 340-meter by 340-meter emission grids were arbitrarily assigned covering the Midway Airport area for use in CDM modeling.

The last source listed in Table 3, the estimated vehicular emissions, occurred in three parking lots and the passenger pick-up and drop-off lane (Helen Mikols Drive) at Midway. The methodologies utilized to estimate air pollution from mobile sources at Midway in greater detail in Appendix B. Figure 4 provides a general layout of Midway Airport and assigned emission grids in the study. A detailed layout of Midway can be found in Appendix B as well.

III.1.4 NONROAD MOBILE SOURCE EMISSIONS INVENTORY

Nonroad mobile sources such as lawn mowers and snowblowers usually contribute significantly to air pollution in a highly populated area and, if feasible, must be accustomed for in an adequate risk assessment. Based on Nonroad Engine Emission Inventories for CO and Ozone Nonattainment Boundaries - Chicago CMSA (1992) provided by OMS, emissions from nonroad engines were estimated for the study. The source report was prepared for OMS in response to calls for the nonattainment area emissions inventory development. It provides the county emissions per person per criteria pollutant data computed using an equipment's population in a given region, the average load factor at which the equipment's engine is operated, the average annual hours of use and the horsepower rating of the engine, and the emission factor attributable to the engine. For this Southwest Chicago risk assessment study, the county emissions per person data were used in conjunction with the available population data.

APPENDIX B
MIDWAY EMISSIONS INVENTORY AND DISPERSION MODELING RESULTS

B.1 INTRODUCTION

An emissions inventory was developed for Midway Airport. The inventory includes emission estimates from two major mobile sources at the airport: aircraft and automobiles. These two mobile sources required diverse approaches to estimating the emissions, including computer database calculations and manual calculations. Source research was conducted in many areas, and several agencies were contacted regarding emissions and operations data.

Source data were collected on the number and type of aircraft that have used Midway Airport in the past ten years as well as the number of flights which were attributed to Midway Airlines before bankruptcy. This information is presented in Exhibit B-1. Additionally, source data were collected so that emissions amounts for total hydrocarbons (THC) and particulate matter could be calculated. When necessary, assumptions and default values were utilized due to the lack of actual data. These assumptions are identified in the report. Following calculation of the THC and particulate matter emission amounts, the THC data were then used to estimate emission amounts of three carcinogens: benzene, formaldehyde, and 1,3-butadiene. Particulate matter emissions were also estimated. Please note that it is particularly difficult to estimate emissions of particulate matter, since direct measurement of particulate matter emissions from aircraft engines are typically not available. Also, particulate matter emission rates that are available are likely to be overestimates.

Based on the prepared inventory, air dispersion modeling was conducted to estimate concentrations at targeted receptor grids. This provided the necessary information to assess carcinogenic risks to the area population attributed to air pollution from mobile sources at Midway Airport.

B.2 EMISSIONS INVENTORY PREPARATION

During the development of the Midway Airport emissions inventory, efforts were focused on four possible emission source types: aircraft automobiles, service vehicles, and other non-aircraft sources. In this section, we present the sources of data as well as the assumptions and approaches taken to estimate the emissions from these sources.

Emission estimates from the various types of sources were calculated using several methods and procedures, including two computer programs, the Emissions and Dispersion Modeling System (EDMS) and the FAA Aircraft Engine Emissions Database (FAEED), and spread sheet calculations. Both EDMS and FREED were released by the Federal Aviation Authority (FAA).

B.2.1 EMISSIONS FROM AIRCRAFT SOURCES

Aircraft are the largest source of pollutant emissions at Midway. However, there is limited emissions data on the various aspects of aircraft operations. Emissions for this source type were often calculated using default values and assumptions derived through research and discussions with various representatives in several agencies at the Federal, State, and local levels.

Four aircraft categories providing services at Midway in 1990 were identified: commercial air carriers, air taxi, general aviation, and military. For each category, the following steps were used to estimate particulate emissions and toxics emissions:

(1) Determine the mixing height to be applied to the landing and takeoff (LTO) cycle. A LTO cycle consists of 4 operational phases: approach, landing, taxi-idle, and climbout.

• The mixing height is important mainly for nitrogen oxide (NO2) emissions. Since this inventory only calculated THC and particulate mover emissions, a default value of 3,000 feet in FREED was used.

(2) Define the fleet make-up for each aircraft category using the airport. The make-up of an aircraft fleet contains all types in each category that either landed or took off from the airport during a given year. The 1990 Midway Airport fleet was used for this inventory.

(a) Commercial Air Carriers

• The fleet make up was taken from data supplied in Airport Activity Statistics of Certified Route Carriers, 1990. published by the FAA.

(b) Air Taxis

• No data regarding the fleet make up were available. After discussions with the U.S. EPA Office of Mobile Sources (OMS), it was decided to assume that 73% of air taxi operations were conducted by piston engine aircraft. The remaining 27% were assumed to be turbine engine operations.

(c) General Aviation

As with the air taxi fleet, no data regarding the make-up were available. OMS recommended an aircraft engine mix of 94% piston and 6% turbine.

(d) Military

• This category contained the least amount of available data. Several agencies were contacted to determine the fleet make-up.

(i) The U.S. Air Force Reserves stated no operations were conducted at Midway Airport, only at O'Hare. They recommended contacting the National Guard.

(ii) The National Guard does conduct operations at Midway, but does not maintain statistics. They did make some observations:

• There are almost no piston engines in operation; practically all of the air force aircraft use turboprop or turbojet engines.
• They estimated that approximately 95% of the operations are conducted by BELL UH-1 helicopters, with BEECH C-12 small cargo jets comprising most of the remaining operations.
• The National Guard approximations were discussed with the Midway Airport Control Tower. They generally concurred with the National Guard, but estimated that 95%-98% of operations were conducted by UH-I helicopters.

(3) Determine airport activity as the number of LTO cycles for each aircraft category.

• Airport operations data for commercial air carriers were taken from Airport Activity Statistics of Certified Route Carriers 1990.
• Airport operations data for all other aircraft categories were contained in FAA Air Traffic Activity. It supplied the number of operations (an operation being either a landing or takeoff), that was divided by two to determine the number of LTOs.

(4) Select emission indices for each category.

• Indices were based on tile type and number of engines on tile aircraft, and on type of pollutant being emitted.

THC Emission Estimation

(a) Commercial Air Carriers

• Aircraft types were identified in Airport Activity Statistics of Certified Route Carriers. 1990.
• Engine types and number of engines were identified in FAEED, with the following exceptions:

(i) Emission rates for the BEECH 18 aircraft engine (R-985-AN PW) were not listed in FAEED and could not be found elsewhere. Therefore, we used the emission rates from engine PT6A-41 (Piper PA-42 Cheyenne). This engine was used for general Air Taxi turboprop engines in the Illinois Environmental Protection Agency (IEPA) 1990 Base Year Emissions Inventory for Cook County.

(ii) Boeing B 737-100 and B 737-200 were grouped as B 737-100/200 in Airport Activity Statistics of Certified Route Carriers 1990. When the data were input to FAEED, the emission rates from Boeing B 737-100 aircraft were used.

• Emission rates were not available in FAEED for the JT8D-7D engine (one of the engine types of the B-737-100), so its market share (4%) was reassigned to engine JT8D-7B.

(iii) FAEED identified three engines for Boeing B 737-50o, each with a different emission rate and a 0% market share. To compute the inventory, each engine type was given a one third (33 1/3%) market share.

(b) Air Taxis

• No specific engine data were available. To estimate an emission rate, piston engine emissions were manually calculated using an emission rate that was the average of seven piston engine emission rates identified in AP-42. Emissions from turbine engines were calculated in FAEED using two turbojet engines: the PT6A-27 (32% of turbine operations) from the De Havilland DHC-6/300 aircraft and the PT6A-41 (68% of turbine operations) from the Piper PA-42 Cheyenne aircraft. The operation percents were based on percents relative to the number of U.S. registered aircraft as of December 31, 1989. Turbine data were obtained from Chapter 5 of tile U-S. EPA report number EPA-450/4-81-026d, Procedures for Emission Inventory Preparation -- Volume IV: Mobile Sources (Volume IV guidance).

(c) General Aviation

• The inventory for both piston and turbine engine aircraft was calculated using the same assumptions, data, and methods as Air Taxis for the appropriate engine type.

(d) Military

• Emissions from the Beech C-12 were calculated in FREED.
• Emissions from the Bell UH-I (engine T53-L-I ID) were calculated using data from the Volume IV guidance document.

Particulate Emission Estimation

Data were only available from AP-42. If the particulate emission rate for a corresponding engine type was available, the data were used. For all others, particulate emissions were calculated using an average emission rate.

(a) Commercial Air Carriers

• Specific particulate matter emissions data were available for only one engine type: the JT8D-17. This engine type comprised 81% of the DC-50 fleet in 1990; it was assumed to constitute 100% for the inventory.
• The average emission rate was determined from seven turbofan engine emission rates identified in AP-42. Five of the emission rates were specifically identified, the other two were assumed (in AP-42).

(b) Air Taxis

• Piston engine emissions were estimated by multiplying by 5% the total organic gases (TOG) emission factors for appropriate piston engines representative of the mix at Midway, as recommended by OMS. This percentage is approximately the percentage of particulate matter relative to TOG for non-catalyst light-duty gasoline vehicles (LDGVs).
• Turbine engine emissions were determined using the one available turboprop engine emission rate (from a Garrett AiResearch TPE 331-3 engine) identified in AP-42.

(c) General Aviation

• The inventory for both piston and turbine engine aircraft was calculated using the same assumptions, data and methods as Air Taxi for the appropriate engine type.

(d) Military

• Emissions from Beech C-12 aircraft were determined using the emission rate from the same turboprop engine (TPE 331-3) used for determining Air Taxi turbine engine emissions.
• No data could be found regarding particulate emissions from Bell UH-1 helicopters; since they are a relatively small part of the entire emissions inventory, they were considered negligible.

(5) Estimate a "Time-In-Mode" (TlM) for each aircraft category at the airport.

• TIM values are necessary for each phase-of-operation: each phase contains its own characteristics, including fuel flow and emission rates. Therefore, modeling is different for each phase and source. The four phases-of-operation are: taxi/idle (TI), takeoff (TO), climbout (CO), and approach (AP). The TIM and phase-of-operation are important because they determine the position of a moving emission source.
• Table B.1 identifies all TIM values used in the emissions inventory. Unless otherwise noted, all numbers are default values given in FAEED and/or Volume IV guidance.

Table B.1 TIM VALUES

• The taxi/idle TIMs listed in Table B.1 and used in the inventory are the sums of the taxi/idle-in TIM and taxi/idle-out TIM identified in Volume IV guidance for each respective type of operation or aircraft category. The two TIMs were combined since FAEED computes emissions for only one taxi/idle mode; also, the emission tines for the respective phases am equal.
• Two inventories were prepared each with a different taxi/idle TIM. One utilized default TIMs found in FAEED or in Volume IV guidance document. The other utilized a 10 minute taxi/idle TIM. The 10 minute TIM is a de minimis value recommended by the Deputy Commissioner of Midway Airport, and confirmed by the Midway Control Tower. De minimis was defined as the minimum amount of time required for an aircraft to complete a phase-of-operation if it is the only aircraft in the queue. The de minimis assumption is valid for future emissions since Midway Airport operations were significantly reduced with the bankruptcy of Midway Airlines in 1991.
• No takeoff TIM was identified for UH-1 helicopters. We assumed the takeoff time negligible for the following reasons:

• The takeoff phase of a helicopter operation would take a short amount of time, less than one minute (based on the assumption that takeoff is defined as the time from the start of upward motion until the helicopter lifts off the ground).
• The identified emission rates are extremely low (less than 1 lb/hr).
• The small emission rate and short TIM would combine with the other factors to produce a calculated emission amount that was very small in comparison with the other phases of operation. Therefore, it can be considered negligible.

(6) Calculate an inventory based on airport activity, TIM, and aircraft emission factors.

• The inventory was calculated for THC and particulate matter emissions. FAEED was utilized for all THC emission calculations except for air taxi and general aviation turbine engines, and the UH-I helicopter engine. All particulate matter emissions were hand-calculated (or determined on a spread-sheet).

(7) Convert the THC data to determine the respective emission amounts of benzene, formaldehyde, and 1,3-butadiene using correction factors for the steps noted below:

(a) Convert the THC data to volatile organic compounds (VOC) data.
(b) Convert the VOC data to TOG data
(c) Use aircraft toxic fractions of TOG to determine required toxic pollutant emissions.

All correction factors, conversion formulae, and toxic fractions are contained in Volume IV guidance document or were supplied in a memo by OMS (Exhibit B-2).

B.2.2 EMISSIONS FROM AUTOMOBILE SOURCES

To evaluate the cancer risk attributed to Midway Airport, emissions caused by vehicles which traveled within the three parking lots located around Midway Airport for access by airline passengers and airport employees should be assessed wherever possible. Park Lot B shown on Figure B.1 was no longer in service prior to 1990 and therefore was not considered in the study. In addition, vehicular emissions generated by vehicles on the portion of Helen Mikols Drive (west of South Cicero Avenue) which is on Midway Airport property should also be taken into account.

The capacity of each parking lot was provided by the Chicago Department of Aviation (Chicago DOA) and is listed below.

• Main Parking Lot - 1,366 stalls (444 hourly spaces, 700 daily space, and 222 rental car spaces). We assumed that hourly spaces were occupied by different vehicles every hour during the 20-hour daily operating period from 5:00 am (exactly one hour earlier than the scheduled first departure) to 1;00 am (approximately one hour later than the scheduled last arrival). Any daily or rental car space was assumed to be accessed by only one vehicle per day.
• Employee Lot (East side of Cicero Avenue) - 300 stalls. We assumed that each space was used by one employee vehicle once a day.
• Economy Parking Lot (North side of 55th Street) - 2,200 stalls. We also assumed that each space was used only once as a daily parking space.

The Chicago DOA also provided the number of vehicles that traveled on Helen Mikols Drive in one summer week (ending June 17, 1990). During that week, 65,821 vehicles traveled on Helen Mikols Drive as counted by the Chicago DOA. All vehicles were counted travelling to the terminal, as the count was conducted past the main parking lot entrance. Therefore, it was assumed that no vehicles from the parking lots were included in the count. Based on this weekly number, annual traffic volume was estimated at 3,422,692 vehicles in 1990.

Distances travelled by vehicles in the parking lots or on Helen Mikols Drive were then estimated to compute the vehicle miles travelled (VMT). Table B.2 lists the estimated annual VMT. Please note that each parking lot was considered fully occupied year around.

The last step prior to estimating emissions was to determine the vehicle fleet in these four sources. We used the representative vehicle fleet mix averaged over all the 1990 fleet mixes of the 343 selected traffic zones in the Cook County motor vehicle emissions inventory. The representative fleet mix isas follows in Table B.3:

Based on the representative fleet mix and estimated annual VMT, toxic air pollution was estimated for each of the four sources. Emission factors selected were directly extracted from the 1990 IEPA Baseline Emissions data base and were used in conjunction with the OMS-suggested toxic fractions. The speed was assumed to be 10 mph in these four sources and Inspection/Maintenance (I/M) credits were in effect as well.

B.2.3 EMISSIONS FROM SERVICE VEHICLES

Service vehicles are those used in land-based airport operations (i.e. luggage transfer vehicles, aircraft refueling vehicles, emergency and support vehicles). Required data includes type of vehicle, hours of operation, and fuel use.

B.2.4 EMISSIONS FROM OTHER NON-AIRCRAFT SOURCES

Covered under this source type are power plants/heating plants, incinerators, fuel storage tanks, and training fires; however, the Deputy Commissioner of Midway Airport identified fuel storage tanks as the only source of this type at Midway. Required information includes:

• Number and size of land;
• Type and amount of fuel stored in each tank; and
• The percent vapor recovery - both when filling the tank and emptying the tank into vehicles.

B.3. EMISSIONS INVENTORY RESULTS

B.3.1 EMISSIONS FROM AIRCRAFT

Tables B.4 - B.8 list the annual THC, particulate matter, VOC, TOG, benzene, formaldehyde, and 1,3-butadiene emissions breakdown by phase-of-operation. Please note that VOC and TOG are listed for referenced only. They Were not used to conduct the required air dispersion modeling.

Table B.9 lists the total emission amounts for each phase-of-operation as well as a combined total of all aircraft emissions at Midway Airport. Detailed emissions data may be found in Echibit B-3, which identifies annual THC and particulate matter emissions by aircraft and engine type. The exhibit also Identifies annual VOC, TOG, benzene, formaldehyde, and 1,3-butadiene emissions by aircraft category.

B.3.2 EMISSIONS FROM AUTOMOBILE SOURCES

The calculated emissions for automobile sources at Midway Airport are listed in Tables B.10-B.13.

Please note that no emissions generated by refueling loss were considered. This may underestimate the overall emissions. On the other hand, the assumed maximum usage of each parking lot may cause the over-estimation of emissions. Also, cadmium and asbestos emissions from automobile sources at Midway were not inventoried since the estimated emissions are negligible.

B.3.3 EMISSIONS FROM SERVICE VEHICLES

Limited information was received from Midway Airport regarding the service vehicles that the airport operates. The vehicles were identified by name and type, but no operation parameters were supplied. Generally, Midway Airport operates a rolling vehicle fleet of approximately 700 vehicles. This number is split relatively equally between heavy vehicles used predominantly for snow removal efforts during the winter and light vehicles utilized throughout the year. Comparing this number to the frequency of daily vehicle assess to Midway due to travelers’ activities (comprising a fleet of approximately 9,000 vehicles), emissions from the Midway service vehicles are negligible. In addition, emissions from utility vehicles (such as snowblowers, de-icers, and lawn mowers) at Midway were accounted for in the nonroad mobile source emissions inventory prepared for the entire study area and could not be easily singled out as a separate emission category. For these reasons, we excluded the service vehicle emissions from the Midway Airport inventory.

B.3.4 EMISSIONS FROM OTHER NON-AIRCRAFT SOURCES

Midway Airport informed us that our request for information regarding fuel storage tanks at the Airport was forwarded to the Chicago DOA for review and authorization. At the time of this report, we had not yet received this information.Due to time constraints, fugitive and other emissions from the fuel storage tanks at Midway are not included in this risk assessment.

B.4 AIR DISPERSION MODELING AND RISK ASSESSMENT

B.4.1 EMISSION GRIDS FOR AIRCRAFT EMISSIONS DISPERSION MODELING

Dispersion modeling was conducted based on the results of the Midway Airport emissions inventory. The entire airport area (including runways and terminals) was included in the emission modeling domain (Figure B.1). The emission quantities were identified and distributed by the location on an emission grid of a specific operational phase. The airport area, encompassing approximately a 1.7-kilometer by 1.7-kilometer square with a southwest corner of (UTM northing, UTM easting) = (4625.2, 436.64) in kilometers, was divided into 25 340-meter by 340-meter equal squares as displayed in Figure B.2. The emission rate for each toxic and particulate per phase was determined using the following formula:

Emission rate(per aircraft per foot traveled) = Annual Emissions/Total Aircraft Travelling Feet

Next, the annual emissions (by phase) from each emission grid were determined using the following formula:

Annual Emissions in Grid = (No. Aircraft x Distance Covered in Grid) x Emission rate(per aircraft per foot traveled)

Based upon discussions with the personnel at Midway Airport's Control Tower, the following assumptions were used in preparation for emission rates assigned among all emission grids:

The control tower stated that most takeoffs are to the northwest, and most approaches from the southeast. Therefore, the No. 31 runways are used most often, approximately 75% of the time. A majority of the remaining operations occur on the No. 4 runways (taking off to the northeast, and approaching from the southwest). Therefore, the modeling will be conducted assuming 75% of operations occur on the No. 31 runways, and the remaining 25% of operations occur on the No. 4 runways. In addition, runways No. 31C and No. 31R are for commercial carriers and turbine-engine aircraft only. Piston aircraft will land on runway No. 31L.

• All turbine-engine air taxis and general aviation aircraft will utilize the main terminals. These are the same terminals as those used by commercial air carriers.
  
  
• All piston-engine air taxis and general aviation aircraft will use the west facilities and the southwest corner of Midway's property.
  
  
• The military aircraft are located in the southwest corner of the airport. The emissions for dispersion modeling would only include helicopter emissions, since helicopters account for 95% of military aircraft emissions.
  
  
• Takeoffs and landings will each utilize approximately 2/3 of their respective runways.

Additional assumptions were also made to facilitate the estimation of emissions among all emission grids. These assumptions are as follows:

• Emissions from the taxi/idle phase for both before takeoff (out) and after landing (in) segments must be accounted for. In the emissions inventory, both segments were combined into one taxi/idle phase. For dispersion modeling, all pollutant emission quantities will be proportioned between the two segments according to the ratio of segment time over taxi/idle phase time. Therefore, the taxi/idle in segment will contain 19 minutes/26 minutes, or 19/26 of total pollutant emissions in the taxi/idle phase; the taxi/idle out segment will contain 7/26 of the pollutant emissions.
  
  
• The climbout phase to be considered for the dispersion modeling will include only the distance from point of takeoff to the edge of the airport boundary. This will include 1/3 of the relevant runway plus any ground or other area to the airport border. Although aircraft are typically only at 300 feet when leaving the airport, we will assume that all climbout phase emissions occur within the airport boundary only.
  
  
• The approach phase to be considered for dispersion modeling will include only the distance from the edge of the airport to a respective runway, plus 2/3 of that runway. Although aircraft begin their approach some distance from the airport, we will assume that all approach phase emissions occur within the airport boundary only.

A sample calculation of emissions at an individual emission grid is included in Exhibit B-4.

B.4.2 AIR DISPERSION MODELING TECHNIQUES

Emissions from each phase of a LTO cycle distributed to each of the 25 emission grids were modeled via the Complex Dispersion Model (CDM). For approach and climbout phases, the emission height was assumed to be 50 meters for certain emission grids located at the boundary of Midway Airport, specifically grids Al, A2, A4, A5, Dl, D5, El, or E5 (see Figure B-2). This simulates the emissions originated in the air. Emissions generated from taxi/idle and landing at each emission grid were assumed to be released at the ground level (5 meters). Emissions due to automobiles traveled in the parking lots and passenger pick-up and drop-off lanes (Helen Mikols Drive) were assigned to one of the four 340-meter by 340-meter emission grids corresponding to the locations of the sources. These emissions were then modeled via CDM using the same approach. The emission height selected to model vehicular emissions was also set to be 5 meters.

Thirty-three modeling runs were conducted to generate the necessary concentration estimates from aircraft engines at Midway. One run was conducted for each emission grid (32 total) with the assumption of ground-level release height. One additional run was conducted to model the emissions released in the air for each of the identified eight emission grids. One thousand metric tons per year was used as the input to CDM for each modeling run. We then stored the CDM-modeled concentrations computed at the 8 x 8 receptor grids in the matrix format as unit concentration profiles. For each pollutant, combining the CDM concentration profile, based on the phase of LTO cycle, with its actual emission rate divided by a thousand, one can come up with the estimated annual concentration profile at the 8 x 8 receptor grids.

B.4.3 RISK CALCULATION

Once we have the annual concentration estimates at the receptor grids, determining the individual risk estimates is a straightforward procedure. Two assumptions were made. First, we assume a linear relationship between annual concentration and cancer risk. Second, we assume a zero threshold value for computing cancer risk. Based on these two assumptions, we can compute the individual lifetime cancer risk by multiplying the annual concentration value at a receptor grid by the unit risk factors of those pollutants emitted from the Midway mobile sources.

R^i,j,k = C^i,j,k x F^j

Figure B.3 shows the population profile for the targeted 8 x 8 receptor network. With the population data available for each receptor grid, cancer cases over a 70-year period can be computed using the following equation.

I^i,j,k = Ri,j,k x Pk

B.4.4 UNIT RISK FACTOR

Table B.14 lists the unit risk factors used in the study to estimate cancer risks attributed to emissions from the Midway mobile sources. The unit risk factor is defined as an estimate of the probability that an individual would contract cancer when exposed to a pollutant at an ambient concentration of one microgram per cubic meter (µg/m3) for 70 years (the average lifetime).

B.5 RISK ASSESSMENT RESULTS

In this section, we present the cancer risk estimates computed using the modeled concentrations at receptor grids. The total number of cancer cases attributed to estimated air pollution emitted from the Midway mobile sources is approximately 2 cases over a 70-year period, or one case every 30 years. This includes cancer risks attributed to emissions from both aircraft and vehicles traveled at Midway in 1990. The population residing at receptor grids were estimated at 93,854 people. This suggests that the average risk across the area due to the emissions from Midway is approximately 2.3 x 10^-5. By comparing to the average cancer risk of 1.9 x 10^-4 assessed for all identified sources in the Southwest Chicago area, tills average risk is less by roughly 10-fold. Figure B.4 displays the lifetime cancer cases caused by mobile sources' air pollution at Midway at the receptor grid network. Not surprisingly, receptors located around Midway Airport have higher cancer cases than others. Table B.15 lists the hazard indices at each grid with the percentage of cancer contribution over the estimated total cancer cases of 2, which are attributable solely to air pollution at Midway.

Cancer cases attributd to the Midway mobile sources were also studied by refined source types and by pollutant. Table B.16 provides a cross reference list of cancer contribution by mobile source origin and by pollutant.

From Table B.16, we found that 1,3-butadiene is the most significant contributor to cancer risk in the area. Approximately one case, or 57% of the total cancer cases attributed to the identified Midway air pollution is caused by 1,3-butadiene. Formaldehyde and particulate emissions each contributes roughly 20% of the total cancer cases (about a half case respectively). Cancer cases due to benzene emissions from Midway, on the other hand, are negligible in comparison to the total cancer cases of 2.

Overall, emissions from aircraft operated at Midway in 1990 contribute up to 99% of the total cancer cases. This was expected since the vehicular emissions estimated at Midway are insignificant compared to the aircraft emissions at Midway. Figures B.5 - B.10 portray the cancer cases at the receptor grid network by pollutant and by emission source.

EXHIBIT B-1
NUMBER AND TYPE OF AIRCRAFT AT MIDWAY AIRPORT

The data used in this exhibit were obtained at the FAA Library from the FAA Airport Activity Statistics of Certified Route Carriers reports for the years 1981 through 1991. Specifically the data was contained in Table 7 - Aircraft Departures Scheduled and Aircraft Departure Performed By Type of Operation, By Aircraft Type, By Community, and By Carrier. This source contained a listing of the air carriers and types of aircraft that used Midway Airport between 1981 and 1991. In addition, it was possible to quantify the number of flights (departures) attributed to Midway Airlines. This data contains only information for departures.

Two other sources of airport operations data for Chicago's Midway Airport were obtained. The first set of data includes excerpts from Table 4 - Airport Operations at Airports with FAA operated Traffic Control Towers By Region and By State and Aviation Category, of the FAA Air Traffic Activity report for fiscal years 1982 through 1991. The final source of data is the Illinois Depart of Transportation (DOT) Illinois Airport Inventory Report, 1992. This data contains counts of all operations, not just departures.

Based on the data received, and telephone calls to SOT and Mr. David Sourni, Deputy Commissioner at Midway Airport, the following information on the Chicago Midway Airport was obtained:

Following deregulation in 1978, Midway Airlines was formed. Midway Airlines was based out of Midway Airport. As Midway Airlines grew, the air traffic at Midway Airport increased. Once other airlines recognized the market that Midway Airlines had discovered, they also began to fly out of Midway Airport increasing air traffic accordingly. This growth in air traffic at Midway Airport went through a dip in the early 1980's as a result of the air traffic controller strike of August 3, 1981 and the recession that affected the entire airline industry in the early 1980's. Air traffic at Midway Airport was also significantly affected by the cancellation of all Midway Airlines flights when they filed for Chapter 7 (bankruptcy) at midnight on Wednesday, November 13th, 1991. At the time of Midway Airlines collapse they represented 70% of the airports traffic. Following the Midway Airlines collapse, some but not all of their gates and corresponding air traffic have been assumed by other airlines. Southwest Airlines alone acquired 17 gates from Midway Airlines.

The types of aircraft that used Midway Airport between 1981 and 1991 are:

There were several Air Carriers who used Midway Airport during the 1981 to 1991 time period. The Air Carriers and their associated FAA codes are listed below:

The total number of flights at Midway Airport and the number associated with Midway Airlines by year (obtained from Table 7 of FAA Airport Activity Statistics of Certified Route Carriers) are depicted in the Table 1 and Figure 1, Midway Airlines and Airport Total Departures.

Table 1

The number of flights by aircraft type and year are depicted in the Table 2 and Figure 2, Departures By Aircraft Type. Please note that several types of aircraft are grouped together under miscellaneous because the number of departures for those aircraft types were too small to appear on the chart.

Table 2

The events described above regarding the fluctuations in air traffic at Midway Airport are depicted in Figure 3, FAA and IDOT Annual Totals. This chart represents the data received from FAA Airport Traffic Activity and Illinois Airport Inventory Report. The next figure (Figure 4), Total Operations By Category, represents the FAA data broken out by aviation category. Following the figures are two tables containing the data which were used to create Figures 3 and 4.

Table 5.2 illustrates that for the St. Louis airport, substantial reductions in CO and HC emissions can be effected by the modification of ground operations. The taxi and idle modes account for 89.6 and 94.1% of the CO and HC emissions, respectively. A reduction of 502 in taxi-idle time could, therefore, decrease the total aircraft co and HC emissions by about 452 and 47X, respectively.' Techniques for reducing this taxi-idle time, such as towing aircraft into position rather than having them move under their own power, must, therefore, be given major consideration in the control of emissions. (There are, unfortunately, other considerations which can greatly change the viability of this technique. The safety aspects of such a modification have not yet been fully evaluated. As with almost all engineering systems, benefit is not gained without paying a price for it).

Table 5.2
Aircraft Emissions by Mode of LTO Cycle
St. Louis Airport
1990 - Speas Study

*The contribution to particulate emissions by each mode varies over the forecast peroid 1975-2000. The 1990 value and the range of variation are given.

Table 5.2 also shows that modification of ground operations will do little to improve the NOx emission picture. Only 8% of the total NOx emitted comes from ground operations. It is highly unlikely that substantial modifications for the sake of reducing NOx emissions can be made in the flight modes without adversely affecting the safety of the aircraft. Hence, it appears that any required NOx emission reduction will have to come from controls placed on the aircraft engines themselves. NOx control, therefore, is out of the reduction ability of the local airport operator.

Particulate emissions can be partially controlled by the changes in ground operation but not nearly as dramatically as CO and HC. In 1990, a 50% reduction in taxi-idle time would result in about a 13% reduction in particulate emissions. By 2000, this could be as much as 20%.

Thus, the simple task of displaying the emissions by mode of aircraft operation has provided several starting points from which to approach the evaluation of alternative aircraft emission controls.

Emission Trends

From the standpoint of regional air pollutant emission management, an item of concern would be the projected trend in aircraft emissions. One index of this trend is the average pollutant emission rate per aircraft LTO at the study airport, this is obtained by dividing the annual amount of pollutants emitted from all aircraft by the annual number of aircraft LTOs. Figure 5.3 depicts this computation for the Speas forecast for the St. Louis airport. It can be seen that the trend is toward higher aircraft emission rates in CO and NOx, somewhat higher rates for HC and lower rates for particulates. The implication is that on a per airplane basis, the emission situation is worsening for CO, HC, and NOx and improving for particulates.

This does not, however, complete the picture. consideration of only the above information might indicate that from an emission standpoint, it would be more profitable to have passengers travel in many smaller, lower- emitting aircraft rather than a few larger, higher-emitting ones. This preliminary conclusion could have important implications if it were decided to control airport air pollution by regulating the aircraft mix. The conclusion does not stand up under further investigation. A more revealing index of emission trend is the average pollutant emission rate per enplaned passenger. Figure 5.4 is a plot of this rate for the Speas forecast. The curves show a decrease in emissions per passenger for all pollutant species with the exception of a slight increase in NOx for the 1975-1985 period. The salient feature of this display is that the trend toward the wide body jets is resulting in the movement of passengers at an improved air pollutant emission level. The air transportation system is improving its position in the transportation/air pollution picture by moving people in and out of airports with less emissions per person.

Chicago Midway Air Traffic Data