• Military

• WMD Menu                           

• Introduction
• Systems
• Facilities
• Agencies
• Operations
• Countries
• Hot Documents
• News
• Reports
• Policy
• Budget
• Congress
• Links

• Intelligence
• Homeland Security
• Space

APPENDIX B AIR QUALITY ANALYSIS

B.1 INTRODUCTION

This appendix presents more detailed information on the assessment of the air quality at and in the vicinity of Pantex Plant and the potential impacts associated with the alternatives being evaluated in this EIS.

B.2 OBJECTIVES

The overall objective of the air quality impact analysis is to determine the impact of air pollutant emissions resulting from construction and operation activities under the Proposed Action and Alternatives at Pantex Plant. Local and regional receptors include human beings, vegetation, and materials.

The impact and significance of ambient pollutant concentrations resulting from Pantex Plant emissions were determined by comparing the predicted concentrations with the appropriate Federal or State ambient air quality standards or guidelines. Ambient air quality standards represent the allowable pollutant concentrations at which public health and welfare are protected with a reasonable margin of safety.

Specific objectives include the following:

• Selection of an appropriate air quality dispersion model.
• Use of the model to evaluate maximum pollutant concentrations at the Pantex Plant boundary and residences near the plant boundary resulting from existing emissions and from emissions related to the Proposed Action and Alternatives.
• Estimation of air quality impacts at the plant boundary and at nearby residences.

B.3 METHODOLOGY

B.3.1 Model Selection

Model selection required consideration of the characteristics of Pantex Plant emission sources, terrain in the vicinity of the plant, and land use classification. Because of the large number of sources (over 100) and 90 pollutants, it was determined that a screening model could not be used for this air quality analysis. Screening models are most appropriate for assessing the air quality impacts of a single source with continuous, constant emission rates. Therefore, it was determined that a refined dispersion model would be required. The following attributes are required in the model:

• Accommodate multiple point, area, or volume, and fugitive sources.
• Allow for a sophisticated treatment of building downwash for point sources.
• Predict concentrations for flat and rolling terrain.
• Predict impacts on urban and rural land use classifications.
• Allow input of time-varying emissions rates (e.g., emission rates may vary by season, month, or hour-of-day).

A review of Environmental Protection Agency (EPA)-preferred air quality models listed in the Guidelines on Air Quality Models (Revised) (40 CFR, Part 51, Appendix W) indicated that the only model that would meet the above requirements was the Industrial Source Complex Model (ISC2). The model operates in both short-term (ISCST2) and long-term (ISCLT2) modes. ISCST2 can produce short-term concentrations averaged over periods of 1, 2, 3, 4, 6, 8, 12, or 24 hours. ISCLT2 can produce long-term concentrations averaged over a month, season, quarter, or year. ISCST2 can also produce annual concentrations and is often used in air analyses when both short-term and annual average concentrations are required.

B.3.2 Model Description and Application

B.3.2.1 Model Description

The ISC2 model is a steady-state Gaussian plume model that can be used to assess pollutant concentrations from a wide variety of sources associated with an industrial source complex. This model can account for the following:

• Settling and dry deposition of particles.
• Downwash.
• Point, area, line, and volume sources.
• Plume rise as a function of downwind distance.
• Multiple point sources.
• Limited terrain adjustment.

The ISC2 model is appropriate for the following regulatory uses:

• Industrial source complexes.
• Rural or urban areas.
• Flat or rolling terrain.
• Transport distance less than 50 kilometers(31 miles).
• One-hour to annual average times.
• Continuous toxic air emissions.

Input data requirements for the ISC2 model are as follows:

• Source data: location, emission rate, physical stack height, stack inside diameter, and stack gas temperature. Optional inputs include source elevation, building dimensions, particle size distribution with corresponding settling velocities, and surface reflection coefficients.
• Meteorological data: ISCST2 requires hourly surface weather data from the preprocessor program RAMMET, which provides hourly stability class, wind direction, wind speed, temperature, and mixing height. For ISCLT2, input includes stability wind rose (stability array [STAR] disk), average afternoon mixing height, average morning mixing height, and average air temperature.
• Receptor data: coordinates and operational ground elevation for each receptor.

For more detailed information concerning the ISC2 model refer to the EPA Guidelines on Air Quality Models (Revised) (40 CFR, Part 51, Appendix W).

B.3.2.2 Model Application

The air quality dispersion modeling was performed in accordance with the guidance provided in the EPA document Guidelines on Air Quality Models (Revised) and the Texas Natural Resources Conservation Commission (TNRCC) document Air Quality Modeling Guidelines (40 CFR, Part 51, Appendix W; TNRCC 1993).

The Industrial Source Complex (ISC) dispersion model (Version 93109) was used to estimate all pollutant concentrations.

B.3.3 Land Use Analysis

In this analysis rural dispersion coefficients were used. The selection of rural coefficients was based on the Auer land-use method (Auer 1978). This procedure involves classifying the land use within a 3,000-meter (9,843-foot) radius around the emission source. If urban land use types account for 50 percent or more of the total area, urban dispersion coefficients should be used; otherwise, rural dispersion coefficients should be used. In an earlier land-use analysis performed by the Radian Corporation for Pantex Plant, it was determined that the percentage of urban land-use within 3,000 meters (9,843 feet) was less than 5 percent (Radian 1993:3-1, 5-1). Since the percent of urban land use is less than 50 percent, rural dispersion coefficients were selected.

B.3.4 Meteorological Data

Meteorological data for the ISCST2 model consisted of hourly surface observations and mixing heights obtained from meteorological data recorded at the Amarillo International Airport for calendar year 1988. Meteorological data for the ISCLT2 model consisted of a joint frequency distribution of wind speed, wind direction, and stability class (STAR data). The STAR data were based on 1985 through 1989 surface observations recorded at the Amarillo International Airport. The meteorological data were obtained from TNRCC's bulletin board.

This set of meteorological data was selected because it was specified by TNRCC and was used by the Radian Corporation in modeling emissions from the Pantex Plant Burning Ground. Also, the use of these data would make the site-wide modeling performed for this EIS comparable with the Radian analysis from a meteorological standpoint (Radian 1994).

B.3.5 Emission Source Characteristics

Emission sources are located in several functional areas (some of these are referred to as numbered zones). These areas include a weapons assembly/disassembly zone (Zone 12), a weapons staging area (Zone 4), and an area for experimental explosives development (Zone 11), a drinking water treatment plant, a sanitary wastewater treatment facility, and a vehicle maintenance and administration area (see Figure B.3.5-1).

Figure B.3.5-1. Air Quality Monitoring Stations, Emission Sources, and Receptor Sites at Pantex Plant. (.pdf)

Other emission sources included an explosive test-firing facility, an open burning ground to burn high explosive(s) (HE) materials, and a Burning Ground Upgrade (BGU) facility for processing explosive-contaminated materials or components. Emission sources also included welding and cutting operations, standby diesel and gasoline engines, and the container storage area where drum sampling and bulk handling of chemicals are performed.

Most of the emissions from Pantex Plant are fugitive emissions, except emissions from Building 16-13. This building was a point source (stack) with the following parameters:

• Height of stack: 19.81 meters (65.0 feet).
• Diameter of the stack: 0.91 meters (3.0 feet).
• Exit Speed: 5.03 meters per second (16.5 feet per second).
• Exit Temperature: 418.76 K (294.1 °F).

Building 16-13 was the only source at Pantex Plant that required the application of building downwash effects.

As indicated in Tables B.3.6-1 through B.3.6-9, facilities from which fugitive emissions were emitted were treated as volume sources (all tables are presented at the end of this appendix). The emission rates were expressed in grams per second. The release height of the emissions was assumed to be located at one half the building height. Groups of point sources with small emission rates were treated as an area source.

Table B.3.6-1. Estimated Emission Rates from Facility-Wide Criteria Pollutants for 2,000, 1,000, and 500 Weapons Scenarios (.pdf)
Table B.3.6-2. Estimated Emission Rates from Facility-Wide Hazardous Air Pollutants for 2,000, 1,000, and 500 Weapons Scenarios (.pdf)
Table B.3.6-3. Estimated Emission Rates for Hazardous Air Pollutants from Open Burning for 2,000, 1,000, and 500 Weapons Scenarios (.pdf)
Table B.3.6-4. Estimated Emission Rates for Criteria Pollutants from the Burning Ground Upgrade for 2,000, 1,000 and 500 Weapons Scenarios (.pdf)
Table B.3.6-5. Estimated Emission Rates for Hazardous Air Pollutants from the Burning Ground Upgrade for 2,000, 1,000, and 500 Weapons Scenarios (.pdf)
Table B.3.6-6. High Explosive Emission Factors (.pdf)
Table B.3.6-7. Calculated Rates for Pantex Affected Environment (.pdf)
Table B.3.6-8. Calculated Emission Rates for 2,000, 1,000, and 500 Weapons (.pdf)
Table B.3.6-9. Estimate Pollutant Emission Rates for Critera and Hazardous Air Pollutants from Open Burning for 2,000, 1,000, and 500 Weapons Scenarios. (.pdf)

The Burning Ground is an onsite facility used to demilitarize and sanitize explosives components and treat materials contaminated with explosives. The Burning Ground covers approximately 23.5 hectares (58 acres) in the north-central portion of Pantex Plant (see Figure B.3.5-1), and currently includes the firing sites, nine burn trays, three burn pans, two burn cages, and three burn pits. There are no structures at the Burning Ground that could affect the normal dispersion of air pollutants during burning operations (Radian 1993:6-1).

The emissions from the open burning of HE in the trays were modeled using a "flare" methodology developed by the Radian Corporation and approved by TNRCC (Radian 1993). The input parameters used for the methodology were as follows:

• Flare release height above the ground: 2 meters (6.6 feet).
• Flame temperature: 1900 K (2961 °F).
• Exit speed: 1.0 meter per second (3.3 feet per second).
• Effective Diameter of flare:
• 45.4 kilograms (100 pounds) HE: 1.92 meters (6.3 feet).
• 363 kilograms (800 pounds) HE: 8.66 meters (28.4 feet).

It was assumed that the burning of HE that did not contain fluorides would occur between the hours of 7:00 a.m. and 6:00 p.m. Explosives that contained fluoride would only be burned twice a day, at 11:00 a.m. and 3:00 p.m. It should be noted that not more than three burns can be conducted in a day. Also, there is at least a 3-hour waiting period between any two burns for explosives emitting hydrogen fluoride (HF). Two HE scenarios were modeled: one assumed that 45.4 kilograms (100 pounds) of HE were burned, while the other assumed a burn of 363 kilograms (800 pounds) of HE.

Pantex Plant has a commitment to construct a project called the Burning Ground Upgrade (BGU) to replace the current pit burn operation. This project should be completed in the timeframe analyzed in this EIS. This upgrade will consist of a covered three-sided structure with a fan to exhaust emissions through an elevated stack. The wood currently used as an auxiliary heat source for the pit burn will be replaced by natural gas in the BGU. These changes will improve the dispersion characteristics of emissions from this operation and eliminate the toxic emissions common to wood burning.

The BGU operation was assumed to be a point source (stack) with the following stack parameters:

• Stack height: 12.2 meters (40.0 feet).
• Stack diameter: 0.61 meters (2.0 feet).
• Exit speed: 16.2 meters per second (53.2 feet per second).
• Exit temperature: 388.8 K (240.0 °F).

Building downwash for the BGU was not required for emissions released through the BGU stack (Radian 1994:6-1). Downwash from the BGU structure was also not a consideration with regard to tray burns (Radian 1994:6-1).

B.3.6 Emission Inventory Scenarios

The basic emission inventory used to derive emission rates for Pantex Plant air quality analysis was developed by the Radian Corporation. This emission inventory included emissions from furnaces, boilers, water heaters, diesel generators and engines, the BGU, tray burns (hemispheres, HE, and wet HE), and the solvent area (sampling and bulk handling). The emission sources used for this site-wide analysis are presented in Tables B.3.6-1 through B.3.6-5.

Tables B.3.6-1 and B.3.6-4 present the estimated emission rates for the criteria pollutants. The estimated emission rates in Table B.3.6-1 were used in modeling the current emissions (Affected Environment), which were assumed to be related to the 2,000 weapons level at Pantex Plant. In addition, emission rates for two other weapons levels (1,000 and 500) were developed. Table B.3.6-4 shows the estimated criteria pollutant emission rates for the BGU.

Pantex Plant personnel provided information regarding facilities whose emissions would change or would remain the same if weapons operations were reduced. The facilities whose emissions would change are indicated with a "W" in the Status column while those remaining the same are indicated with an "S". For those facilities whose emissions would be reduced, it was assumed that the reduction would be in proportion to the weapons operation reduction.

Table B.3.6-2 presents the estimated emission rates for 33 hazardous air pollutants (HAPs), as listed in the Clean Air Act, as amended, November 1990 (42 U.S.C. 7401). The emission sources and pollutant emission rates are presented in the same manner as described for Table B.3.6-1.

Table B.3.6-3 presents estimated emission rates for HAPs that are emitted from tray burns of hemispheres, HE, and wet HE. These emission rates are presented in the same manner as described for Table B.3.6-1.

Tables B.3.6-4 and B.3.6-5 present the estimated emission rates for criteria pollutants and HAPs from the BGU. These emissions were not used in modeling the current emission rates (Affected Environment) because the BGU project has not been started but should be completed in the next few years. Therefore, these emissions were included in modeling the emissions for the three weapons levels (2,000, 1,000, and 500) for future operations.

Two scenarios were assumed for modeling the emissions from the trays. One scenario assumed 45.4 kilograms (100 pounds) of HE were burned while the other assumed a burn of 363 kilograms (800 pounds) of HE. For modeling the affected environment, it was assumed that an HE mixture of PBX-9404 and LX-17 was burned. For modeling the emissions from 2,000, 1,000, and 500 weapons levels, it was assumed that a mixture of PBX-9404 and LX-04 was burned. The LX-17 and LX-04 HE were selected because they both emit HF. Although LX-17 is currently burned, it will not be burned during the 10-year period (1997-2007) covered by the EIS. LX-04, however, may continue to be burned during this period.

The emission factors that were used to calculate the emission rates for carbon monoxide (CO), nitrogen dioxide (NO2), hydrogen chloride (HCl), and HF for these HE mixtures are presented in Table B.3.6-6. The different kinds of HE are mixed to maximize the heat produced by the burning and keep the amount of HF emitted below the regulatory limits. In order to be conservative, the HE mixes modeled in this EIS were chosen for the lowest heat output and highest emission of HF allowed under the regulatory limits. The emission rates for these HE mixtures are presented in Tables B.3.6-7 and B.3.6-8.

A summary of the estimated emission rates for CO, NO_2, PM₁₀, HCl, and HF for the affected environment and the three levels of future operations (2,000, 1,000, and 500 weapons per year) is presented in Table B.3.6-9.

B.3.7 Receptor Locations

A set of discrete receptors was placed at 100-meter (328-foot) intervals at the Pantex Plant boundary. Another set of 11 discrete receptors was established at the locations of the 11 residences which are situated near the Pantex Plant boundary. A grid of offsite receptors was not used since most of the emissions from Pantex Plant sources were released at relatively low heights above the surface. Surface concentrations from low-level releases will be highest near the source and decrease with increasing distance from the source. This concentration decrease with distance is the result of the pollutant cloud being continuously dispersed with increasing distance by atmospheric turbulence.

B.4 MODELING RESULTS

This section presents the results of a conservative modeling of the current environment at Pantex Plant and the impacts associated with activity levels of 2,000, 1,000, and 500 weapons per year.

The model calculates the concentrations at each receptor resulting from the weather conditions for each hour within the year's worth of data (i.e., 8,760 hourly values for each receptor). The model uses meteorological data values that were recorded hourly at the Amarillo airport for an entire year (data was validated by TNRCC). The model returns the single maximum concentration seen by each receptor.

For example, for hourly results this maximum concentration value is associated with a certain hour's weather conditions (e.g., 9:00 a.m. December 3). A different receptor (especially one in a different direction from the source) would have a maximum concentration value returned that might be associated with a different set of meteorological conditions (e.g., 7:00 p.m. July 8). Per the example, the maximum concentration value for residence R5 occurs when a gentle wind blows from the source towards R5, and the maximum concentration value for residence R3 occurs when a gentle wind blows towards R3 (see Figure B.3.5-1). Therefore, it is difficult to discuss the relation between the values for different receptors solely in terms of distance from the source. For calculation of health effects, the maximum concentration hourly value is assumed (for conservatism) to exist at all receptors 24 hours a day 365 days a year.

B.4.1 Affected Environment

The maximum concentration found on the site boundary for each modeled criteria pollutant and each HAP is presented in volume I, Tables 4.7.1.3-4 and 4.7.1.3-5, respectively. The appropriate National Ambient Air Quality Standards (NAAQS) or Effects Screening Levels (ESLs) are also shown in the tables for comparison with the calculated concentration.

As shown in Table 4.7.1.3-4, none of the criteria pollutants are expected to exceed their ambient air quality standards. The emission inventory specified an emission rate for a group of pollutants called alcohols. The alcohol species included in the group were not speciated in the inventory. TNRCC in a modeling analysis of Pantex Plant developed a 1-hour ESL for the group of alcohols (TNRCC 1995b). The calculated maximum 1-hour average concentration at the Pantex Plant boundary for this group of alcohols is the only pollutant concentration that exceeded the guideline ESL (see Table 4.7.1.3-5 in volume I).

The maximum concentrations that were calculated for the 11 residences located near the plant boundary are presented in Table B.4.1-1 (see Figure B.3.5-1 for residence location). Alcohols, modeled as a group, were the only pollutants which exceeded the ESL at a residence. A subsequent review of the inventory of the amounts of the individual alcohols present at the Plant showed that the ESL used in the modeling was excessively conservative. Since the emission rates for the individual alcohols in the group were unknown, an estimation of the maximum fence line concentration for each of the alcohols was calculated based on the ratio of the amount of each alcohol to the total inventory of the group of alcohols present at Pantex Plant (see Table B.4.1-2). None of the individual alcohols were calculated to exceed their respective ESLs at or near the Plant boundary.

Table B.4.1-1. Estimated Maximum Concentrations of Hazardous Air and Criteria Pollutants at Local Residences Surrounding the Pantex Plant
Table B.4.1-2. Calculated Concentrations of Individual Alcohols Based on Percentage of Inventory.

It should be noted that the maximum concentration at the residences shown in Table B.4.1-1 may not occur at the same time (day and hour). Since meteorological conditions also vary with time, the concentration values shown in this table cannot be used to determine the spatial variability of the concentrations. In order to determine the spatial variability the concentration would have to be determined simultaneously at all locations.

The modeling results for the current estimated pollutant emissions from Pantex Plant facilities indicate that the air quality at the plant boundary and beyond is not adversely impacted by Pantex Plant operations. The results also support the attainment designations for all criteria pollutants in the Pantex Plant area. The onsite air quality is well within the Threshold Limit Values for the pollutants. The impacts of the onsite air quality to workers are discussed in section 4.14.1.2 of volume I.

B.4.2 Air Quality Impacts

Air quality impacts were developed for three different levels of weapons which were assumed to occur during the analysis period of this EIS. The estimated maximum pollutant concentrations that would occur for each of the three levels of weapons are presented for criteria pollutants and HAPs in volume I, Tables 4.7.2.1-1 and 4.7.2.1-2, respectively. As shown in Table 4.7.2.1-1, concentrations of all the criteria pollutants are expected to be below their respective NAAQS.

As described in the Affected Environment (section B.4.1), only alcohols as a group exceeded the TNRCC ESL (see Table 4.7.2.1-2) for all three weapons levels.

Table B.4.2-1 presents the estimated maximum pollutant concentration on the Pantex Plant boundary for the 2,000 weapons level as a percentage of the NAAQS. The percentages for the 1,000 and 500 weapons levels would be the same or less than those shown in the table. The greatest percentages of the NAAQS were for the 24-hour and annual PM₁₀. The emission resulting from the burning of 45.4 kilograms (100 pounds) of HE resulted in maximum boundary concentrations that represented about 59 percent of the 24-hour average standard and 18 percent of the annual standard. Maximum concentrations from other criteria pollutants represented less than 7 percent of their respective standards.

Table B.4.2-1. Percentage of the NAAQS Consumed by the Estimated Maximum Criteria Pollutant Concentration at Pantex Plant Boundary (2,000 Weapons)

Although the amount of pollutants released from the 363-kilogram (800-pound) HE burn is greater than the amount released from the 45.4-kilogram (100-pound) HE burn, the concentration of pollutants at the surface produced from the emissions of a 45.5-kilogram (100-pound) HE burn was greater than those produced by a 363-kilogram (800-pound) HE burn. The lower surface concentrations for the 363-kilogram (800-pound) burn are the result of a greater plume rise than occurs for the smaller burn. The higher plume rise is the result of a larger amount of the hot air than the smaller burn. The larger amount of hot air results in increased buoyancy or vertical lift. The dispersion of the pollutants from the larger burn is therefore greater due to this increased buoyancy. The greater dispersion for the larger burn results in lower surface concentrations than the smaller burn.

Table B.4.2-2 contains information for HAPs similar to what was contained in Table B.4.2-1 for criteria pollutants. As shown in Table B.4.2-2, alcohols were estimated to be 95 percent over the 30-minute average ESL. All other HAPs are expected to be below their ESLs. The next highest concentration was produced by emission of HF resulting from the burning of 45.4 kilograms (100 pounds) of HE. It represented about 85 percent of the 3-hour average HF standard. Other HAPs which were more than 50 percent of the respective ESLs were nickel, HCl, methylene chloride, carbon disulfide, and silver.

Table B.4.2-2. Percentage of the TNRCC ESL Consumed by the Estimated Maximum Concentration of Air Pollutants at Pantex Plant Boundary (2000 Weapons)

Table B.4.2-3 presents the estimated maximum pollutant concentrations that could occur at the 11 residences near the Pantex Plant boundary. Maximum pollutant concentrations for each of the three weapons levels (2,000, 1,000, and 500) are shown. Alcohols, modeled as a group, exceeded the 30-minute average ESL at the Pantex Plant boundary, and at residence R10 for all three levels of operations. All of the other pollutant concentrations at the 11 residences were estimated to be below their respective ESLs. Since the maximum concentrations for air pollutants listed in the TNRCC's list (volume I, Tables 4.7.2.1-1 and 4.7.2.1-2) are all below their respective ESLs at the fence line for the 2,000 weapons level, they will not have any impact at the fence line or residences for the 1,000 and 500 weapons levels.

Table B.4.2-3. Estimated Maximum Concentrations of Hazardous Air and Criteria Pollutants for 2,000, 1,000 and 500 Weapons at Local Residences Surrounding the Pantex Plant (mg/m³)

Since violations of the NAAQS or the TNRCC ESLs would not occur at any of the 11 residences (sensitive receptors) near the plant boundary for any of the proposed levels of weapons operations, local air quality impacts would be negligible. In addition, since the emissions from Pantex Plant would contribute only a very small amount to the overall pollution burden in Carson County and the surrounding counties, regional air quality impacts would also be expected to be negligible.

REFERENCES