GEORGIA DEPARTMENT OF NATURAL RESOURCES
ENVIRONMENTAL PROTECTION DIVISION
Air Protection Branch Ambient Monitoring Program
2006 Ambient Air Surveillance Report
2005-2006 Risk Assessment Discussion
This document is published annually by the Ambient Monitoring Program, in the Air Protection Branch of the Georgia Department of Natural Resources, Environmental Protection Division.
Date of Initial Publication November 1, 2007
Table of Contents
List of Figures.......................................................................................................................... iii List of Tables ............................................................................................................................v Executive Summary .................................................................................................................1 Glossary ...................................................................................................................................3 Introduction...............................................................................................................................5 Chemical Monitoring Activities..................................................................................................7
Carbon Monoxide (CO) .....................................................................................14 Oxides of Nitrogen (NO, NO2, NOx and NOy)....................................................16 Sulfur Dioxide (SO2)..........................................................................................20 Ozone (O3)........................................................................................................22 Lead ..................................................................................................................31 Particulate Matter..............................................................................................33 PM10 ..................................................................................................................34 PM2.5 .................................................................................................................39 PM2.5 Speciation ...............................................................................................47 Acid Precipitation ..............................................................................................55 Photochemical Assessment Monitoring Stations (PAMS) ......................................................61 Carbonyl Compounds .......................................................................................67 Air Toxics Monitoring ..............................................................................................................75 Metals ...............................................................................................................77 Hexavalent Chromium ......................................................................................85 Volatile Organic Compounds (TO-14/15) ..........................................................87 Semi-Volatile Organic Compounds ...................................................................93 Meteorological Report ............................................................................................................99 Summary of Meteorological Measurements (2006) ........................................101 Select Meteorological and Air Quality Case Studies for 2006 .........................105 Quality Assurance ................................................................................................................109 Quality Control and Quality Assessment.........................................................110 Gaseous Pollutants.........................................................................................111 Particulate Matter............................................................................................113 Air Toxics ........................................................................................................117 NATTS ............................................................................................................126 Photochemical Assessment Monitoring ..........................................................131 Meteorology ....................................................................................................132 Quality Control Reports...................................................................................134 Standards Laboratory .....................................................................................134 Laboratory and Field Standard Operating Procedures....................................134 Siting Evaluations ...........................................................................................134 2005-2006 Risk Assessment Discussion..............................................................................137 Introduction .....................................................................................................137 Results and Interpretation ...............................................................................137 Summary and Discussion ...............................................................................158 Outreach and Education.......................................................................................................163 Media Outreach ..............................................................................................166 Other Outreach Opportunities .........................................................................166 Appendix A: Additional Criteria Pollutant Data .....................................................................169 Carbon Monoxide (CO) ...................................................................................169
i
Nitrogen Dioxide (NO2)................................................................................... 169 Nitric Oxide (NO) ............................................................................................ 169 Oxides of Nitrogen (NOx) ............................................................................... 170 Reactive Oxides of Nitrogen (NOy) ................................................................ 170 Sulfur Dioxide (SO2) ....................................................................................... 171 Ozone (O3) ..................................................................................................... 172 Lead (Pb) ....................................................................................................... 174 Fine Particulate Matter (PM2.5) ....................................................................... 175 Particulate Matter (PM10)................................................................................ 178 Appendix B: Additional PM2.5 Particle Speciation Data........................................................ 181 Appendix C: Additional Meteorological Data........................................................................ 187 Appendix D: Additional PAMS Data..................................................................................... 195 PAMS Continuous Hydrocarbon Data (June- August 2006)........................... 195 PAMS 2006 24-hour Canister Hydrocarbons ................................................. 201 Appendix E: Additional Toxics Data..................................................................................... 207 2006 Metals.................................................................................................... 207 2006 Semi-Volatile Compounds ..................................................................... 212 2006 Volatile Organic Compounds................................................................. 219 2006 Carbonyl Compounds, 24-hour ............................................................. 238 2006 Carbonyl Compounds, 3-hour (June-August) ....................................... 239 Appendix F: Monitoring Network Survey.............................................................................. 241 Appendix G: Siting Criteria .................................................................................................. 247 Appendix H: Instrument and Sensor Control Limits ............................................................. 249 References .......................................................................................................................... 251
ii
List of Figures
Figure 1: North Georgia Air Monitoring Site Map....................................................................12 Figure 2: South Georgia Air Monitoring Site Map ...................................................................13 Figure 3: Carbon Monoxide Monitoring Site Map ...................................................................15 Figure 4: Oxides of Nitrogen Monitoring Site Map..................................................................19 Figure 5: Sulfur Dioxide Monitoring Site Map .........................................................................21 Figure 6: Typical Urban 1-Hour Ozone Diurnal Pattern ..........................................................22 Figure 7: Ozone Formation Process.......................................................................................23 Figure 8: Ozone Monitoring Site Map .....................................................................................25 Figure 9: Georgia's 8-Hour Ozone Nonattainment Area Map.................................................28 Figure 10: Metro Atlanta Ozone- Number of Violation Days per Year ....................................29 Figure 11: Metro Atlanta Ozone Exceedance Map.................................................................30 Figure 12: Lead Monitoring Site Map .....................................................................................32 Figure 13: PM10 Monitoring Site Map .....................................................................................35 Figure 14: PM10 Annual Arithmetic Mean Chart......................................................................37 Figure 15: PM10 24-Hour Averages ........................................................................................38 Figure 16: PM2.5 Federal Reference Method Monitoring Site Map..........................................41 Figure 17: PM2.5 Monitoring Site Map, Continuous and Speciation Monitors .........................42 Figure 18: Georgia's PM2.5 Nonattainment Area Map.............................................................44 Figure 19: PM2.5 Design Value, Annual Standard, By Site .....................................................45 Figure 20: PM2.5 Design Value, Daily Standard, By Site.........................................................46 Figure 21: 2003 PM2.5 Speciation ...........................................................................................48 Figure 22: 2004 PM2.5 Speciation ...........................................................................................48 Figure 23: 2005 PM2.5 Speciation ...........................................................................................49 Figure 24: 2006 PM2.5 Speciation ...........................................................................................49 Figure 25: PM2.5 Speciation, Trends in Ammonium Concentrations .......................................51 Figure 26: PM2.5 Speciation, Trends in Elemental Carbon Concentrations ............................51 Figure 27: PM2.5 Speciation, Trends in Organic Carbon Concentrations ................................52 Figure 28: PM2.5 Speciation, Trends in Sulfate Concentrations..............................................52 Figure 29: PM2.5 Speciation, Trends in Nitrate Concentrations ..............................................53 Figure 30: PM2.5 Speciation, Trends in Crustal Matter Concentrations...................................53 Figure 31: Acid Rain Monitoring Site Map ..............................................................................56 Figure 32: Acid Rain Trends, Statewide .................................................................................57 Figure 33: Acid Rain Trends, by Location...............................................................................58 Figure 34: Comparison of DNR and NADP Acid Rain Averages ............................................59 Figure 35: PAMS Monitoring Site Map ...................................................................................62 Figure 36: Isoprene Yearly Profile, 2003-2006 .......................................................................64 Figure 37: Toluene Yearly Profile, 2003-2006 ........................................................................65 Figure 38: Toluene & Isoprene, Typical Urban Daily Profile ...................................................66 Figure 39: Carbonyls Monitoring Site Map .............................................................................68 Figure 40: Average 24-Hour Carbonyls Concentration and Number of Detects, 2005 ...........69 Figure 41: Average 24-Hour Carbonyls Concentration and Number of Detects, 2006 ...........69 Figure 42: Average South DeKalb 3-Hour Carbonyls, June-August, 2005.............................70 Figure 43: Average South DeKalb 3-Hour Carbonyls, June-August, 2006.............................71 Figure 44: Average Tucker 3-Hour Carbonyls, June-August, 2005 ........................................72 Figure 45: Average Tucker 3-Hour Carbonyls, June-August, 2006 ........................................72 Figure 46: Average 24-Hour Carbonyls Concentration vs. Number of Detects, by Species,
2005 .................................................................................................................................73
iii
Figure 47: Average 24-Hour Carbonyls Concentration vs. Number of Detects, by Species, 2006................................................................................................................................. 73
Figure 48: Metals Monitoring Site Map .................................................................................. 79 Figure 49: Total Detections of Metals, by Site, 2004-2006 .................................................... 80 Figure 50: Total Detections of Metals, by Species, 2004-2006.............................................. 83 Figure 51: Total Average Concentration of Metals, 2004-2006 ............................................. 83 Figure 52: Yearly Average Comparison of Zinc, by Site, 2004-2006 ..................................... 84 Figure 53: Hexavalent Chromium at South DeKalb ............................................................... 86 Figure 54: Total Volatile Organic Compounds Detected per Site, 2004-2006 ....................... 87 Figure 55: Number of Volatile Organic Compounds (TO-14/15) Detected, Select
Compounds, 20042006 ................................................................................................. 88 Figure 56: Total Average Concentration of Select Volatile Organic Compounds (TO-14/15),
20042006....................................................................................................................... 88 Figure 57: Total Volatile Organic Compound Loading all Species, by Site, 2004-2006 ......... 89 Figure 58: Volatile Organic Compounds, Seasonal Effects, 2004-2006 ................................ 91 Figure 59: VOC and SVOC Monitoring Site Map ................................................................... 92 Figure 60: Total Semi-Volatile Organic Compound Detections Per Site, 2004-2006 ............ 94 Figure 61: Number of Semi-Volatile Organic Compound Detections, by Compound, 2004-
2006................................................................................................................................. 95 Figure 62: Total Average Concentration of Semi-Volatile Organic Compounds, 2004-2006 . 95 Figure 63: Forecasted and Observed 8-hr Ozone for Metro Atlanta, May-July 2006........... 103 Figure 64: Forecasted and Observed 8-hr Ozone for Metro Atlanta, August-September 2006
....................................................................................................................................... 103 Figure 65: PM2.5 Observations For Atlanta Area By AQI Category, October 2003 May 2007
....................................................................................................................................... 104 Figure 66: Forecast Team Performance, Atlanta PM2.5, October 2003 May 2007 ............ 105 Figure 67: Gaseous Criteria Pollutants Accuracy Analysis .................................................. 112 Figure 68: Particulate Air Pollutants Accuracy Analysis....................................................... 114 Figure 69: Metals Monitoring, Collocated Precision............................................................. 121 Figure 70: VOC Monitoring, Collocated Precision ............................................................... 122 Figure 71: SVOC Monitoring, Collocated Precision ............................................................. 123 Figure 72: Yearly Summary of Sampling Flow Rate Accuracy, Air Toxics Network............. 125 Figure 73: VOC Monitoring Accuracy Analysis .................................................................... 132 Figure 74: Meteorological Measurements Accuracy Results ............................................... 133 Figure 75: Formulas For Calculating Risk and Hazard Quotient.......................................... 146 Figure 76: The AQI .............................................................................................................. 164 Figure 77: Sample AIRNOW Ozone Concentration Map ..................................................... 167
iv
List of Tables
Table 1: Georgia Ambient Air Standards Summary..................................................................9 Table 2: Georgia Air Sampling Station Locations for 2006 .....................................................11 Table 3: Common Oxides of Nitrogen Species and Terms.....................................................17 Table 4: Monthly Rainfall For 2006 and 30-Year Average, Selected Cities ............................99 Table 5: Temperature and Rainfall Statistics for 2006, Selected Cities................................100 Table 6: Meteorological Parameters Measured, 2006..........................................................101 Table 7: Audits Performed for Each Air Monitoring Program in 2006 ...................................110 Table 8: Results for Criteria Pollutants Performance Audits.................................................112 Table 9: Results for Particulate Sampler Performance Audits..............................................114 Table 10: Summary of Unexposed Filter Mass Replicates ...................................................116 Table 11: Summary of Exposed Filter Mass Replicates .......................................................116 Table 12: Total Precision Concentrations for the Georgia Air Toxics Network .....................118 Table 13: Yearly Summary of Flow Rate Accuracy Performance Audit, Air Toxics Network 124 Table 14: NATTS Sites with EPA Region Numbers and AQS Site Codes ...........................127 Table 15: Measurement Quality Objectives for the NATTS Program ...................................128 Table 16: MQO Data Sources for the Georgia NAATS Program..........................................128 Table 17: 23 Selected HAPs and Their AQS Parameter Codes...........................................129 Table 18: Percent Completeness of Georgia's 2006 AQS Data, Selected Compounds .......130 Table 19: Laboratory Analytical Precision Estimate .............................................................130 Table 20: Laboratory Speciated VOC Audit Results for PAMS Network ..............................131 Table 21: Compounds Monitored and Screening Values Used in Initial Assessment ..........139 Table 22: Summary of Chemicals Analyzed in 2005 ............................................................140 Table 23: Summary of Chemicals Analyzed in 2006 ............................................................141 Table 24: Site-Specific Detection Frequency and Mean Chemical Concentration, 2005......142 Table 25: Site-Specific Detection Frequency and Mean Chemical Concentration, 2006......144 Table 26: Cancer Risk and Hazard Quotient by Location and Chemical, 2005 ....................148 Table 27: Cancer Risk and Hazard Quotient by Location and Chemical, 2006 ....................150 Table 28: Aggregate Cancer Risk and Hazard Indices for Each Site, Excluding Carbonyls,
2005 ...............................................................................................................................152 Table 29: Aggregate Cancer Risk and Hazard Indices for Each Site, Excluding Carbonyls,
2006 ...............................................................................................................................153 Table 30: Summary Data for Select VOCs at PAMS Sites, 2005 .........................................154 Table 31: Summary Data for Select VOCs at PAMS Sites, 2006 .........................................155 Table 32: Summary Observations, Cancer Risk, and Hazard Quotient from Carbonyls, 2005
.......................................................................................................................................156 Table 33: Summary Observations, Cancer Risk, and Hazard Quotient from Carbonyls, 2006
.......................................................................................................................................157 Table 34: AQI Summary Data, 2006 ....................................................................................165 Table 35: AIRNOW Participation Evaluation Results ...........................................................168
v
vi
Executive Summary
The Ambient Monitoring Program of the Air Protection Branch of the Environmental Protection Division (EPD) has monitored air quality in the State of Georgia for more than thirty years. The list of compounds monitored has grown over the thirty years to more than 200 pollutants using several types of samplers at sites statewide. This monitoring is performed to protect public health and environmental quality. The resulting data is used for a broad range of regulatory and research purposes, as well as to inform the public. This report is the summary of the monitoring data from 2006, and is an assessment of the data in conjunction with previous years' findings.
The Chemical Monitoring Activities, Photochemical Assessment Monitoring (PAMS), and Air Toxics Monitoring sections provide an in-depth discussion of the chemicals that are monitored with maps identifying individual monitoring sites. These sections also contain discussions on health effects, measurement techniques, and attainment designations for the chemicals that are monitored. Additionally, these sections discuss trends and common sources for the monitored pollutants.
Six (6) pollutants fall within the criteria pollutant list. These pollutants are carbon monoxide, sulfur dioxide, lead, ozone, nitrogen dioxide, and particulate matter (now regulated in two size categories). The ambient concentrations of these pollutants must meet a regulatory standard. The regulatory standard is health-based. Concentrations above the standard are considered unhealthy for sensitive groups. After several years of favorable weather patterns, though, Georgia's 2005 and 2006 data show an uptick in ozone concentrations. It is possible that, given the long-term trend toward improving air quality in the state, that this change is a result of natural variation in weather patterns. At the same time, this recent data is less favorable to Georgia's regulatory status with respect to the national standards. Also, in December 2006 the national limit on 24-hour averaged concentrations of fine particulate matter was reduced. No exceedances of the stricter 24-hour standard occurred before the end of the year, but several sites recorded design values that violate the annual standard.
Another set of compounds called air toxics are monitored throughout the state in the Air Toxics Network. The sources of these emitted compounds include vehicle emissions, stationary source emissions, and natural sources. These air toxic compounds do not have ambient air regulatory standards. However, the compounds monitored in the Air Toxics Network are analyzed annually for theoretical lifetime cancer risk and potential non-cancer health effects. This analysis is presented in the Risk Assessment section of this report. Estimates of theoretical cancer risk posed by these compounds are primarily driven by a small number of chemicals in the metals, volatile organic compounds, and carbonyls groups of the air toxics. The estimates of theoretical lifetime cancer risk related to air toxic pollutants in the areas monitored across the state ranged from 7 in 100,000 to 1 in 10,000,000. The potential risk of non-cancer health effects from air toxic pollutants is estimated differently, and most chemicals were well below the `acceptable' hazard quotient of 1.
The Ambient Monitoring Program also operates an extensive network of meteorological stations. The Meteorological Report section discusses Georgia's climatology based on the meteorological data captured at the PAMS sites and the sites located statewide. The meteorological sites provide, at a minimum, wind speed and wind direction data. Some stations are very sophisticated and provide information on barometric pressure, relative
1
humidity, solar radiation, temperature, and precipitation. A discussion of the Georgia ozone and PM2.5 forecasting effort is also included in this section. The Quality Assurance section shows the Ambient Monitoring Program's effort in producing quality data. The data has to be collected and measured in a certain manner to meet requirements that are set forth by the EPA. The requirements for each monitored pollutant is provided, including field and laboratory techniques, as well as results of the quality assurance audits. The Outreach and Education section provides information concerning the efforts of the Clean Air Campaign to change the commuting habits of residents of Atlanta. The voluntary program partners with the public and private sector to reduce vehicle congestion and aid in reducing vehicle emissions. This section includes a description of educational and news media outreach activities, and explains how the Air Quality Index (AQI) is used to offer the public an easy to use indicator of air quality. The appendices of this document contain summary tables for the pollutants measured during 2006. Included in the summary tables is information on where air toxic compounds were detected, the number of samples collected, and average and maximum concentrations. Copies of this and previous annual reports are available in Adobe Acrobat format via the Ambient Monitoring Internet website at http://www.georgiaair.org/amp. A limited number of print copies are available and may be requested at 404-363-7006. Real time air monitoring information for the criteria pollutants may be found at the above website by selecting the pollutant of concern. In addition, the website also provides links to the Clean Air Campaign and the smog forecast.
2
Aerosols AM APB AQCR Anthropogenic ARITH MEAN BAM CAA CFR CO CV EPA EPD FRM
GEO MEAN HAP HI HQ IUR LOD g/m3 m/s mean MSA
NAAQS NAMS NATTS NMHC NO2 NOx NOy NUM OBS NWS ODC O3 PAH PAMS Pb PM2.5 PM10 ppbC ppm Precursor PUF
2006 Georgia Annual Air Quality Report
Glossary
A gaseous suspension of fine solid or liquid particles Annual Mean Air Protection Branch Air Quality Control Region Resulting from human activity Arithmetic Mean Beta Attenuation Monitor Clean Air Act Code of Federal Regulations Carbon Monoxide Coefficient of Variation Environmental Protection Agency (federal agency) Environmental Protection Division (state agency) Federal Reference Method- the official measurement technique
for a given pollutant Geometric Mean Hazardous Air Pollutant Hazard Index Hazard Quotient Inhalation Unit Risk Limit of Detection Micrograms per cubic meter Meters per second Average Metropolitan Statistical Area,
as defined by the US Census Bureau National Ambient Air Quality Standard National Ambient Monitoring Site National Air Toxics Trends Station Non-Methane Hydrocarbons Nitrogen Dioxide Oxides of Nitrogen Reactive oxides of Nitrogen Number of Observations National Weather Service Ozone depleting Chemicals Ozone Polycyclic Aromatic Hydrocarbons Photochemical Assessment Monitoring Station Lead Particles with an aerodynamic diameter of 2.5 microns or less Particles with an aerodynamic diameter of 10 microns or less Parts per billion Carbon Parts per million A substance from which another substance is formed Polyurethane Foam
3
QTR Rawinsonde RfC Screening Value SLAMS SO2 SPMS TEOM TNMOC TRS TSP UV VOC w/m2
Calendar Quarter A source of meteorological data for the upper atmosphere Reference Concentration Initial level of air toxic compounds used in risk assessment State and Local Air Monitoring Site Sulfur Dioxide Special Purpose Monitoring Site Tapered Element Oscillating Microbalance Total Non-Methane Organic Compounds Total Reduced Sulfur Total Suspended Particulates Ultraviolet Volatile Organic Compound Watts per square meter
4
2006 Georgia Annual Air Quality Report
Introduction
This report summarizes the air quality data collected by the State of Georgia during calendar year 2006. The Air Protection Branch is a subdivision of the state's Department of Natural Resources, Environmental Protection Division. The United States Environmental Protection Agency (EPA) regulates air quality standards nationwide through authority granted by Congress in the Clean Air Act. Few people realize, though, that the air quality monitoring that is required by the Act is performed almost entirely by state and local governments. The Ambient Monitoring Program conducts monitoring in Georgia, both to satisfy Clean Air Act monitoring requirements and to exceed them in cases where additional monitoring proves beneficial to the citizens and industries of the State. Monitoring is performed to facilitate the protection of public health, as well as to protect our natural environment. The data is collected and quality assured using equipment and techniques specified by EPA. Once the data is ready, it is submitted to EPA's national air quality database, where it is available to a broad community of data users. Despite the technical nature of the information collected, every effort has been made to make the data relevant and useful to those who do not routinely study air quality data. To provide additional information for those who have interest in more detailed technical information, extensive Appendices are included. Further information about air quality in Georgia and nationwide is available from EPA.
5
6
2006 Georgia Annual Air Quality Report
Chemical Monitoring Activities
This section is a summary of the National Ambient Air Quality Standards (NAAQS), and the monitoring techniques used to measure ambient air quality for comparison with these standards.
The Clean Air Act (CAA) requires the EPA Administrator to identify pollutants that may reasonably be anticipated to endanger public health or welfare. The Act also requires the Administrator to issue air quality criteria that reflect the latest scientific knowledge useful in indicating the kind and extent of all identifiable effects on public health or welfare that may be expected from the presence of such pollutant in ambient air. Under the Clean Air Act, the EPA Administrator establishes NAAQS for each pollutant for which air quality criteria have been issued. The EPA is to set standards where "the attainment and maintenance are requisite to protect public health" with "an adequate margin of safety." In 1971, the EPA established standards for five "criteria" pollutants as required by the Clean Air Act. The standards and pollutants have changed over time to keep up with improvements in scientific knowledge. The current list is summarized in Table 1.
As shown in Table 1, there are primary and secondary ambient air quality standards. Primary standards are designed to protect the most sensitive individuals in a population. These sensitive individuals include children, the elderly, and those with chronic illnesses. The secondary standards are designed to protect public welfare or quality of life. This includes visibility protection, limiting economic damage, damage to man-made material, wildlife, or the climate.
The various averaging times are to address the health impacts of the pollutants. Short-term averages, such as the limit for 8-hour ozone averages, are to protect against acute effects. Long term averaging, such as the annual limit for fine particles, is to protect against chronic effects.
The Georgia ambient air monitoring network provides information on the measured concentrations of criteria and non-criteria pollutants at selected locations. The 2006 Georgia Air Sampling Network collects data at 65 locations in 37 counties. Monitoring takes place year-round, with the exception of ozone, which is sampled from March through October, and the continuous Photochemical Assessment Monitoring Stations (PAMS) volatile organic compounds that sample from June to August. For a list of all the sites in the monitoring network, detailing what pollutants are monitored at which sites, see Table 2. That information is followed by Figure 1 and Figure 2, which are overview maps of all the air monitoring locations in the state. Not all pollutants are monitored at all sites. Maps of the monitoring locations for individual pollutants are provided in each pollutant's respective section.
The number and location of the individual sites varies from year to year, depending on: availability of long-term space allocation, regulatory needs, etc. Once a site is established, the most common goal for its use is to monitor for long-term trends.
All official monitoring performed in support of the National Ambient Air Quality Standards (NAAQS) must use U.S. EPA-defined reference methods described in 40 CFR Part 53, Appendix A, or equivalent methods designated in accordance with Part 53 of that chapter. All
7
the data collected in the networks undergoes extensive quality assurance review and is then submitted to the Air Quality System (AQS) database maintained by EPA. In general, the basic monitoring objectives that govern the selection of sites are: 1) to measure the highest observable concentration; 2) to determine representative concentrations in areas of high population density; 3) to determine the impact of significant sources or source categories on ambient pollution levels; 4) to determine the general background concentration levels; and 5) to determine the concentration of a number of compounds which contribute to the formation of ground level ozone. Data from EPD's continuous monitors are published on EPD's web site at http://www.georgiaair.org/amp. The data is updated hourly. Specific annual summary data for 2006 may be found in Appendix A of this document.
8
2006 Georgia Annual Air Quality Report
Criteria Pollutants
Compound Sulfur Dioxide
Primary Standard
0.14 0.03
Particulate Matter (PM2.5)
Particulate Matter (PM10)
Carbon Monoxide
15.0
98th percentile: 65.01/35.0 50.02
150.0 2nd Maximum:
35.0 2nd Maximum:
9.0
Ozone
4th Maximum: 0.085
Secondary Standard
0.50
Same as Primary
Same as Primary Same as Primary Same as Primary
Same as Primary
Units
Time Interval
ppm
micrograms per cubic meter
3 Hour 24 Hour Annual Mean Annual Arithmetic Mean (3 years)
24 Hour
micrograms per cubic meter
Annual Arithmetic Mean
24 Hour
1 Hour ppm
8 Hour Average
ppm
8 Hour Average
Nitrogen Dioxide Lead
0.053 1.5
Same as Primary
Same as Primary
ppm
Annual Mean
micrograms per
Calendar
cubic meter Quarter Average
Table 1: Georgia Ambient Air Standards Summary
1 The 24-hour PM2.5 standard was strengthened by EPA effective December 17, 2006. 2 The annual average PM10 standard was revoked by EPA effective December 17, 2006.
9
Site ID
130090001 130210007
130210012
130210013 130510014 130510017 130510021 130510091
130511002
130550001 130590001 130590002 130630091 130670003
130670004
130690002
130730001 130770002
130850001
130890002 130890003 130892001 130893001 130950007 130970003 130970004
131110091
131130001 131150003 131150004 131150005
131210001
131210020 131210032 131210039
Common Name
Baldwin Co. Airport Allied Chemical
Georgia Forestry Comm.
Lake Tobesofkee
Shuman Middle School
Market St. E. President St.
Mercer Middle W. Lathrop & Augusta
Ave. DNR Fish Hatchery
UGA College Station Rd.
Georgia DOT National Guard Macland Aquatic
Center General Coffee State
Park Riverside Park Univ. of West Georgia Georgia Forestry
Comm. South DeKalb
DMRC Police Dept. Idlewood Rd. Turner Elementary Beulah Pump Station
W. Strickland St. McCaysville Elementary
Georgia DOT Coosa Elementary Floyd Co. Health Dept.
Coosa High School Fulton Co. Health Dept. Utoy Creek E. Rivers School Fire Station #8*
City
Milledgeville Macon
Macon
Macon Savannah Savannah Savannah Savannah
Savannah
Summerville Athens Athens
Forest Park Kennesaw
Powder Springs
Douglas
Evans Newnan
Dawsonville
Decatur Decatur Doraville Tucker Albany Douglasville Douglasville
McCaysville
Fayetteville Rome Rome Rome
Atlanta
Atlanta Atlanta Atlanta
County O3 CO
Baldwin Bibb
PM2.5 24h FRM
X
PM2.5 Cont
PM2.5 Speciation
NO2
NOy
SO2
X X
TRS
Bibb X
X
X
X
Bibb X
Chatham
Chatham
X
Chatham X
Chatham
X
X X
Chatham
X
X
Chattooga X Clarke Clarke X Clayton Cobb X
X
X
X
X
X
Cobb
X
Coffee
Columbia X
Coweta X
X
Dawson X
DeKalb X X X
X
DeKalb
DeKalb
X
DeKalb X
Dougherty
X
Douglas
Douglas X
Fannin
Fayette X
Floyd
Floyd
Floyd
X
Fulton
Fulton
Fulton
X
Fulton
X
X X
X
X X
X X
X X X
Lead
PM10
Acid Rain
X
X
X X
X X
X X X
X X X X
PAMS VOCS
X X
VOC (TO14/15)
X X X
X X X
X X
SVOC X X X
X X
X X
Carbonyls
X
X X X
Trace Metals
X X X
X X X
X X
Site ID
131210048 131210055 131210099 131270004 131270006
Common Name
Georgia Tech Confederate Ave.
Roswell Road Arco Pump Station Risley Middle School
City
Atlanta Atlanta Atlanta Brunswick Brunswick
County O3 CO
Fulton
Fulton X
Fulton
X
Glynn
Glynn X
PM2.5 24h FRM
X
PM2.5 Cont
PM2.5 Speciation
NO2
NOy
SO2
X
X
X
X
TRS X
Lead
PM10
Acid Rain
X
X
PAMS VOCS
VOC (TO14/15)
SVOC
Carbonyls
Trace Metals
131273001 131350002
Brunswick College Brunswick Gwinnett Tech Lawrenceville
Glynn Gwinnett X
X
X
X
X
X
X
131390003
Fair St. Elementary Gainesville
Hall
X
X
X
X
131510002 131530001 131850003
County Extension Robins Air Base
Mason Elementary
McDonough Warner Robins
Valdosta
Henry X Houston Lowndes
X X X
X
X
X
X
X
X
131890001
DNR Fish Hatchery
Thomson McDuffie
X
132130003
Fort Mountain Chatsworth
Murray X
132150001
Health Dept. Columbus Muscogee
X
132150008
Columbus Airport Columbus Muscogee X
X
X
132150011
Cusseta Elementary Columbus Muscogee
X
X
X X
132151003 Columbus Crime Lab Columbus Muscogee X
132155000 Columbus State Univ. Columbus Muscogee
X
X
X
132230003
Yorkville
Yorkville Paulding X X X
X
X
X
X
X
X
132450005 132450091 132450092
Medical College of GA
Bungalow Rd. Elementary
Clara Jenkins School
Augusta Richmond Augusta Richmond X Augusta Richmond
X
X
X
X
X
X
X
X
132470001 132550002 132611001
Monastery
UGA Experiment Station
Union High School
Conyers Griffin Leslie
Rockdale X Spalding
Sumter X
X
X
X
132810001
Lake Burton Hiawassee
Towns
X
132950002
Health Dept.
Rossville
Walker
X
X
132970001
DNR Fish Hatchery Social Circle
Walton
X
133030001
Health Dept. Sandersville Washington
X
X
133190001
Police Dept.
Gordon Wilkinson
X
* The Fire Station #8 was shut down because it did not meet siting requirements. With EPA's permission, the PM10 monitoring was discontinued. PM2.5 FRM monitoring was moved to Georgia Tech.
Table 2: Georgia Air Sampling Station Locations for 2006
Hamilton
Catoosa
Walker
Whitfield
Murray
Floyd
Bartow
Pickens Cherokee
Dawson Hall
Forsyth
Haralson Carroll Heard
Paulding
Cobb
Gwinnett
Barrow
Douglas Coweta
Fulton
De Kalb
Fayette
Clayton Henry
Spalding
Rockdale
Walton Newton
Butts
Jasper
Madison
Clarke
Oconee
Oglethorpe
Meriwether
Pike Lamar
Monroe
Jones
Harris Muscogee
North Georgia Monitoring Sites
Urban BAibbreas
Crawford
MSAs Shown asTwSiggos lid Color
Houston
Figure 1: North Georgia Air Monitoring Site Map
12
2006 Georgia Annual Air Quality Report
Floyd Bartow
Pickens
Dawson
Cherokee
Hall Forsyth
Paulding Haralson
Cobb
Carroll
Douglas
Fulton
Clayton
Barrow
Gwinnett
De Kalb
Walton Rockdale
Newton
Coweta Heard
Fayette
Henry
Spalding
Butts
Jasper
Madison
Clarke Oconee
Oglethorpe
Meriwether
Pike
Lamar
Monroe
Jones
Harris
Muscogee
Chattahoochee Russell
Marion
Crawford
Bibb
Twiggs
Houston
Terrell
Lee
Dougherty Baker
Worth
Lanier
Brooks Lowndes
Echols
Edgefield
Columbia McDuffie
Richmond
Aiken
Burke
Effingham
Bryan
Long
Liberty
Chatham
McIntosh
Brantley
Glynn
South Georgia Monitoring Sites Urban Areas MSA's Shown as Solid Color
Figure 2: South Georgia Air Monitoring Site Map
13
Carbon Monoxide (CO)
General Information Carbon Monoxide (CO) is an odorless, colorless, poisonous gas that is a byproduct of the incomplete burning of fuels. The principal source of CO pollution in most large urban areas, including Metro Atlanta, is the automobile, which contributes approximately 60% of CO emissions nationwide. Other sources include fires, industrial processes, cigarettes, and other sources of incomplete burning in the indoor environment. High concentrations of ambient CO tend to occur in the colder months of the year. In cool weather, weather patterns called inversion layers occur more frequently. These inversions trap pollutants near the surface. CO is inhaled and enters the blood stream, where it binds chemically to hemoglobin. Hemoglobin is the component of blood that carries oxygen to the cells. When CO binds to hemoglobin, it reduces the ability of hemoglobin to do its job, which reduces the amount of oxygen delivered to all tissues of the body. The percentage of hemoglobin affected by CO depends on the amount of air inhaled, the concentration of CO in air, and length of exposure. At the levels usually found in ambient air, CO primarily affects people with cardiovascular disease. The Clean Air Act (CAA) requires that Metropolitan Statistical Areas (MSAs) with a population greater than 500,000, as determined by the last census (2000), have at least two CO National Air Monitoring Stations (NAMS). In Georgia, only the Atlanta MSA meets the population requirement. Currently, the NAMS site is located at Roswell Road (Figure 3). The Roswell Road site was established to monitor for CO at a microscale level. The purpose of microscale measurements is to measure peak concentrations in major urban traffic areas. A microscale site monitors an air mass that covers a distance of several meters to about 100 meters. In substitution for a second NAMS monitor, high sensitivity CO monitors have been installed at the Yorkville and South DeKalb sites. The purpose of these monitors is to detect very small concentrations of CO in order to gain a more complete understanding of the background levels of CO and its role in atmospheric chemistry.
14
Floyd
Bartow
2006 Georgia Annual Air Quality Report
Pickens
Dawson
Cherokee
Forsyth Hall
Yorkville
Paulding
Cobb
Gwinnett
Roswell Rd
Haralson
De Kalb
Douglas
Fulton
S DeKalb
Carroll
Clayton
Rockdale Newton
Fayette
Henry
Heard
Coweta
Meriwether Pike
Figure 3: Carbon Monoxide Monitoring Site Map
CO Sites
UrbaSnpalAdinrgeas
Butts
MSAs Shown as Solid Color
Lamar
15
Health Effects The health effects of CO include weakening the contractions of the heart, which reduces blood flow to various parts of the body, decreasing the oxygen available to the muscles and various organs. In a healthy person, this effect significantly reduces the ability to perform physical activities. In persons with chronic heart disease, these effects can threaten the overall quality of life, because their systems may be unable to compensate for the decrease in oxygen. CO pollution is also likely to cause such individuals to experience chest pain during activity. Adverse effects have also been observed in individuals with heart conditions who are exposed to CO pollution in heavy freeway traffic for one or more hours.
In addition, fetuses, young infants, pregnant women, elderly people, and individuals with anemia or emphysema are likely to be more susceptible to the effects of CO. For these individuals, the effects are more pronounced when exposure takes place at high altitude locations, where oxygen concentration is lower. CO can also affect mental function, visual acuity, and the alertness of healthy individuals, even at relatively low concentrations.
Measurement Techniques CO is monitored using specialized analyzers made for that specific purpose. The analyzers continuously measure the concentration of CO in ambient air using the non-dispersive infrared analysis and gas filter correlation methods.
Attainment Designation Data collected from the continuous monitors is used to determine compliance with the Clean Air Act (CAA) 8-hour and 1-hour standard for CO. This standard requires that, for 8-hour averages, no concentration greater than 9 ppm may be observed more than once per year. For 1-hour averages, no concentration greater than 35 ppm may be observed more than once a year. If the data shows that these criteria are met, then the area is considered to be in attainment of the standard.
All of Georgia is in attainment of both the 8-hour and 1-hour standards for carbon monoxide. For additional summary data on this topic, see Appendix A.
Oxides of Nitrogen (NO, NO2, NOx and NOy)
General Information Oxides of Nitrogen (see Table 3) exist in various forms in the atmosphere. The most common is nitric oxide (NO), but other forms such as NO2, HNO3 and N2O5 also occur. High temperature combustion and lightning produce the bulk of these compounds in the atmosphere. Nitrogen is a very stable molecule and is essentially inert unless subjected to extreme conditions. The oxides of nitrogen are less stable, however, and are key participants in atmospheric chemistry, reacting back and forth between numerous states depending on conditions. Many of these reactions involve the conversion of oxygen atoms between their atomic (O2) and ozone (O3) forms. As such oxides of nitrogen are studied more intensely than their direct health impacts would imply; they are precursors of (and alternately byproducts of) ozone formation. The many forms of oxides of nitrogen in the atmosphere are the reason that they are sometimes referred to using the generic terms NOx or NOy.
16
2006 Georgia Annual Air Quality Report
NO is changed to NO2 in very rapid atmospheric reactions. During daylight hours, UV radiation from the sun breaks apart NO2 into NO and free oxygen (O). The free oxygen atom will attach itself to molecular oxygen (O2) creating an ozone (O3) molecule. This is the origin of all ground level ozone. Daytime levels of NO2 and N2O5 are low but their concentration rises rapidly in overnight. When the sun rises again in the morning, they are converted back to NO and ozone. Nitric acid (HNO3) is the most oxidized form of nitrogen in the atmosphere. This species is water-soluble and is removed from the atmosphere in the form of acidic raindrops.
Abbreviation Full Name
Creation Processes
Elimination Processes
NO
NO2
HNO3 PAN NOx NOy
Nitrous Oxide
Result of ozone photochemistry
Reacts with ozone to form NO2 and oxygen
High-temperature
combustion
Nitrogen Dioxide
High-temperature
Reacts with
combustion
oxygen in strong
Reaction of NO and ozone
sun to form ozone plus NO
"Washes out" in
rain
Nitric Acid
NO2 + H2O
"Washes out" in rain
Peroxyacetyl Nitrate Oxidation of hydrocarbons Slow devolution to NO2 in sunlight
Catch-all name for NO + NO2
Catch-all name for all atmospheric oxides of nitrogen- mostly NO, NO2, HNO3, N2O5, and PAN
Table 3: Common Oxides of Nitrogen Species and Terms
Nitrogen dioxide (NO2), is one of the important oxides of nitrogen. It is a light brown gas, and can be an important component of urban haze. Nitrogen oxides usually enter the air as the result of high-temperature combustion processes, such as those occurring in automobiles and power plants. Home heaters and gas stoves also produce substantial amounts of NO2. NO2 is formed from the oxidation of nitric oxide (NO), which has a pungent odor at high concentrations and a bleach smell at lower concentrations. NO2 is a precursor to ozone formation and can be oxidized to form nitric acid (HNO3), one of the compounds that contribute to acid rain. NO2 and sulfur dioxide (SO2) can react with other substances in the atmosphere to form acidic products that can be deposited in rain, fog, snow, or as particle pollution. Nitrate particles and NO2 can block the transmission of light, reducing visibility.
Health Impacts Individuals experience respiratory problems when exposed to high levels of NO2 for short durations (less than three hours). Asthmatics are especially sensitive to NO2. Changes in airway responsiveness have been observed in some studies of exercising asthmatics exposed to relatively low levels of NO2. Studies also indicate a relationship between indoor
17
NO2 exposures and increased respiratory illness rates in young children, but definitive results are still lacking. Many animal studies suggest that NO2 impairs respiratory defense mechanisms and increases susceptibility to infection.
Several studies also show that chronic exposure to relatively low NO2 pollution levels may cause structural changes in the lungs of animals. These studies suggest that chronic exposure to NO2 could lead to adverse health effects in humans, but specific levels and durations likely to cause such effects have not yet been determined.
Measurement Techniques Oxides of Nitrogen, and in particular NO2, are monitored using specialized analyzers made for that specific purpose. The analyzers continuously measure the concentration of oxides of nitrogen in ambient air using the ozone-phase chemiluminescent method. There are two major instrument designs. While they are closely related, they do not monitor the same species. NOx analyzers measure NO, NO2, and NOx. NOy analyzers measure NO and NOy, but cannot measure NO2. The NOy analyzers are also specialized for measuring trace-level concentrations; as such they cannot measure higher concentrations. Because of these tradeoffs, it is necessary to operate a network of both instrument types to get a complete picture of local conditions.
Of the oxides of nitrogen, only NO2 is regulated under the NAAQS. As such only the NOx type analyzers produce data directly relevant to the standard. NO2 monitoring is required in urban areas with populations greater than 1,000,000. Atlanta is the only urban area in Georgia that meets that population requirement. Atlanta metro area has five NO2 sites. They are located at the South DeKalb, Georgia Tech, Conyers, Yorkville, and Tucker sites. The complete oxides of nitrogen monitoring network, including NOx and NOy monitor locations, can be found in Figure 4.
Attainment Designation Data collected from the continuous monitors is used to determine compliance with the NAAQS primary and secondary annual standards for NO2. This standard requires that a site's annual average concentration exceed 0.053 ppm no more than an average of once a year over a three-year period. The Atlanta MSA, as well as the rest of the nation, is in attainment of the NO2 standard. Los Angeles was the only urban area nationwide that ever recorded violations of the NAAQS for NO2. In July 1998, EPA announced the redesignation of that area to attainment. For additional summary data on this topic, see Appendix A.
18
2006 Georgia Annual Air Quality Report
Catoosa Whitfield Murray
Walker
Floyd
Bartow
Pickens Cherokee
Dawson
Hall Forsyth
Paulding
Yorkville
Haralson
Cobb
Carroll
Douglas
Fulton
Barrow
Gwinnett
Tucker GA TechDe Kalb
Rockdale
S DeKalb
Walton
Clayton
ConyersNewton
Heard
Coweta
Fayette
Henry Spalding
Butts
Jasper
Madison
Clarke Oconee
Oglethorpe
Meriwether
Pike
Lamar
Monroe
Jones
Harris Muscogee
Chattahoochee Russell
Marion
BibbLake Tobesofkee Twiggs
Crawford Houston
Edgefield
McDuffie
Evans
Columbia
Aiken
Richmond
Burke
Terrell
Lee
Long
Dougherty Baker
Worth
Brooks
Lanier Lowndes
NOx/NOy
Urban Areas
Brantley
MSAs Shown as Solid Color
Figure 4: Oxides of Nitrogen Monitoring Site Map
19
Sulfur Dioxide (SO2)
General Information Sulfur dioxide (SO2) is a colorless reactive gas that is formed by burning of sulfur-containing material, such as coal, or by processing sulfur-containing ores. It is odorless at low concentrations, but pungent at very high concentrations. It can be oxidized in the atmosphere into sulfuric acid. When sulfur-bearing fuel is burned or ores that contain sulfur are processed, the sulfur is oxidized to form SO2, which can react with other pollutants to form aerosols. In liquid form, SO2 may be found in clouds, fog, rain, aerosol particles, and in surface liquid films on these particles. Both SO2 and NO2 are precursors to the formation of acid rain and lead to acidic deposition. SO2 can also be a precursor for sulfate particles. Major sources of SO2 are fossil fuel-burning power plants and industrial boilers.
Health Impacts Exposure to SO2 can cause impairment of respiratory function, aggravation of existing respiratory disease (especially bronchitis), and a decrease in the ability of the lungs to clear foreign particles. It can also lead to increased mortality, especially if elevated levels of particulate matter (PM) are also present. Groups that appear most sensitive to the effects of SO2 include asthmatics and other individuals with hyperactive airways, and individuals with chronic obstructive lung or cardiovascular disease. Elderly people and children are also likely to be sensitive to SO2.
Effects of short-term peak exposures have been evaluated in controlled human exposure studies. These studies show that SO2 generally increases airway resistance in the lungs, and can cause significant constriction of air passages in sensitive asthmatics. These impacts have been observed in subjects engaged in moderate to heavy exercise while exposed to relatively high peak concentrations. These changes in lung function are accompanied by perceptible symptoms such as wheezing, shortness of breath, and coughing in these sensitive groups.
The presence of particle pollution appears to aggravate the impact of SO2 pollution. Several studies of chronic effects have found that people living in areas with high PM and SO2 levels have a higher incidence of respiratory illnesses and symptoms than people living in areas without such a synergistic combination of pollutants.
Figure 5 shows the locations of the Georgia SO2 monitoring stations for 2006.
Measurement Techniques Sulfur dioxide is measured in the ambient air using EPA-approved "equivalent method" instruments as defined in 40 CFR Part 53, Appendix A. Georgia's network consists of instruments using a pulsed UV fluorescence technique. In Brunswick, a variation of this instrument is configured to monitor for total reduced sulfur (TRS), which monitors for other sulfur-bearing compounds such as hydrogen sulfide.
20
2006 Georgia Annual Air Quality Report
Figure 5: Sulfur Dioxide Monitoring Site Map 21
Attainment Designation To demonstrate attainment, the annual arithmetic mean and the second-highest 24-hour average must be based upon hourly data that is at least 75 percent complete in each calendar quarter. A 24-hour block average is considered valid if at least 75 percent of the hourly averages for the 24-hour period are available [61 FR 25579, May 22, 1996]. To be considered in attainment, a site must have an annual mean less than 0.03 parts per million (ppm), no more than one 24-hour average exceeding 0.14 ppm, and no more than one 3hour average exceeding 0.50 ppm.
All of Georgia is in attainment of the sulfur dioxide standard. For additional summary data on this topic, see Appendix A.
Ozone (O3)
General Information Ozone is a colorless gas. Ground level ozone, unlike the other gaseous air pollutants previously discussed, is not a primary pollutant. This means that ozone is not directly emitted by any sources, mobile or stationary. Ozone forms through a complex series of chemical reactions, which take place in the presence of strong sunlight (photochemical reactions). For these reactions to take place, certain ingredients (precursors) must be available. Since the reactions must take place in the presence of strong sunlight, ozone concentrations have a strong diurnal pattern (Figure 6).
Figure 6: Typical Urban 1-Hour Ozone Diurnal Pattern The precursors3 to ozone formation are oxides of nitrogen (NOx) and reactive organic substances (sometimes referred to as VOCs or hydrocarbons). Examples of such reactive organic substances include hydrocarbons found in automobile exhaust (like benzene and propane), vapors from cleaning solvents (like toluene), and biogenic emissions (like isoprene). Ozone, when mixed with particles and other pollutants, such as NO2, forms smog, a brownish, acrid mixture (see Figure 7). This type of pollution first gained attention in the 1940's as Los Angeles photochemical "smog". Since then, photochemical smog has been observed frequently in many other cities.
3 For a more complete discussion on ozone precursors, please see the NO2 section and the PAMS section of this report.
22
2006 Georgia Annual Air Quality Report As indicated above, ozone is formed when its precursors come together in the presence of strong sunlight. This reaction can only occur as long as both precursors are present, and the reaction itself consumes the precursors as it produces ozone. The amount of ozone produced, assuming sufficient sunlight, is controlled by what is known as the "limiting reactant." This limiting reactant can be thought of in terms of household baking. You can only bake cookies until you run out of any one of the ingredients you need. If you run out of flour, it doesn't matter how much milk and sugar you have on hand; you can't make any more cookies without more flour. In the same way, ozone production can only occur until the process has consumed all of any one of the required ingredients. As it turns out, natural background hydrocarbon levels are quite low in Los Angeles, so in that area hydrocarbons are typically the reactant that limits how much ozone can be produced. The control measures that proved effective in reducing smog there involved reducing hydrocarbon emissions. These control measures and the science behind them have become relatively advanced because the Los Angeles ozone problem was so severe and developed so long ago. But many of the fundamental lessons learned about smog formation in Los Angeles over many years of research have proven not to apply in the same way in Georgia.
Figure 7: Ozone Formation Process As air quality control measures were first implemented in Georgia, the then-standard assumption was made that Georgia was also hydrocarbon limited. However, the initial control measures seemed ineffective in actually reducing ozone levels. In time, researchers discovered that trees naturally emit large quantities of hydrocarbons. The quantity of hydrocarbons emitted by the abundant trees in this region is sufficient that Atlanta could theoretically violate ambient ozone standards even if humans reduced their hydrocarbon emissions to zero4. Even in that impossible case, there would still be plenty of natural
4 Note that this is not to say that trees cause "pollution"; in the absence of emissions of oxides of nitrogen caused by humans, these hydrocarbons would not react to produce ozone.
23
hydrocarbons around to react with any oxides of nitrogen that human activities were to produce, so virtually the same amount of ozone would be produced. The solution to ozone control in Georgia, then, would have to focus on a different limiting reactant. Since there will always be strong sunshine in the summer, and there will always be oxygen, the only effective way left to control ozone production is to reduce emissions of oxides of nitrogen.
Air quality science had not, and still has not, had time to fully catch up with this discovery. The control technologies that reduce hydrocarbon emissions are generally not effective on oxides of nitrogen, so a whole new set of control technologies had to be developed. This area has been in some ways unable to take full advantage of the technologies developed for Los Angeles, then, because those technologies were not suited to local conditions. With respect to reducing emissions from automobile engines, for example, the addition of relatively simple and inexpensive catalytic converters to existing engine designs was a great leap forward in reducing hydrocarbon emissions. Catalytic converters have been used with great success since the early 1970s. Thus far, emissions of oxides of nitrogen have proven more difficult to control than hydrocarbon emissions, especially given that the control measures have not had forty years to mature. Research on the topic continues, and new emissions control equipment is always under development. For example, where catalytic converters could be added to existing engine designs to greatly reduce hydrocarbon emissions, solutions for reducing emissions of oxides of nitrogen have generally required far more reengineering of the engines themselves. There is no easy "bolt on" solution.
Ozone in Georgia, unlike other pollutants previously discussed, is only monitored during the "summer" months (March through October). The reason for the shorter season is that long periods of strong sunlight and warm temperatures are needed for ozone formation. Many urban areas experience high levels of ground level ozone during the summer months. We also see high ozone levels in rural and mountainous areas. This is often caused by ozone and/or its precursors being transported by wind for many hundreds of miles.
A final difference between ozone and the other pollutants is that ozone is sometimes good. While ground level ozone is considered a hazardous pollutant, the ozone in the upper atmosphere, approximately 10-22 miles above the earth's surface, protects life on earth from the sun's harmful ultraviolet (UV) rays. This ozone is gradually being depleted due to manmade products called ozone depleting chemicals, including chloroflourocarbons (CFC), which when released naturally migrate to the upper atmosphere. Once in the upper atmosphere, the CFCs break down due to the intensity of the sun's UV rays, releasing chlorine and bromine atoms. These atoms react with the ozone and destroy it. Scientists say that one chlorine atom can destroy as many as 100,000 "good" ozone molecules. The destruction of this ozone may lead to more harmful ultraviolet rays reaching the earth's surface, causing increased skin cancer rates. This reduction in the protection provided by ozone in the upper atmosphere is usually referred to as the "ozone hole" and is most pronounced in polar regions.
The Georgia Environmental Protection Division monitors ground level ozone at 24 sites throughout the state (Figure 8).
24
2006 Georgia Annual Air Quality Report
Sequatchie
Marion
Hamilton
Dade
Catoosa
Walker
Whitfield Murray
Fort Mtn
Summerville
Pickens
Dawson
Dawsonville
Hall
Floyd
Bartow
Cherokee
Forsyth
Madison
Kennesaw Gwinnett Tech
Yorkville Paulding
Cobb
Gwinnett
HaralsonDouglasville
Tucker
DougClasonfeSdeDFruealtoKtnealb DeKalb
Rockdale
Barrow
Clarke
College Station
Oconee
Walton
Carroll
ClayCtoon nyers
Newton
Fayetteville
Henry
Newnan
Coweta Heard
Fayette
McDonough
Spalding
Butts
Jasper
Oglethorpe
Meriwether
Pike
Lamar
Monroe
Jones
Harris
Crime Lab Columbus Airport Muscogee
Russell
Chattahoochee Marion
Lake Tobesofkee
Crawford
GA Forestry
Bibb
Twiggs
Houston
Leslie
Terrell
Lee
Dougherty Baker
Worth
Brooks
Lanier
Lowndes
Echols
Ozone Monitors Urban Areas MSAs Shown as Solid Colors
Edgefield
Evans
Columbia McDuffie
Bungalow ES
Richmond
Aiken
Burke
Effingham
E President
Bryan
Chatham
Long
Liberty
McIntosh
Glynn Brantley
Brunswick
Figure 8: Ozone Monitoring Site Map
25
Health Impacts Ozone and other photochemical oxidants such as peroxyacetyl nitrate (PAN) and aldehydes are associated with adverse health effects in humans. Peroxyacetyl nitrate and aldehydes cause irritation that is characteristic of photochemical pollution. Ozone has a greater impact on the respiratory system, where it irritates the mucous membranes of the nose, throat, and airways; ninety percent of the ozone inhaled into the lungs is never exhaled. Symptoms associated with exposure include cough, chest pain, and throat irritation. Ozone can also increase susceptibility to respiratory infections. In addition, ozone impairs normal functioning of the lungs and reduces the ability to perform physical exercise. Recent studies also suggest that even at lower ozone concentrations some healthy individuals engaged in moderate exercise for 6 to 8 hours may experience symptoms. All of these effects are more severe in individuals with sensitive respiratory systems, and studies show that moderate levels may impair the ability of individuals with asthma or respiratory disease to engage in normal daily activities.
The potential chronic effects of repeated exposure to ozone are of even greater concern. Laboratory studies show that people exposed over a six to eight hour period to relatively low ozone levels develop lung inflammation. Animal studies suggest that if exposures are repeated over a long period (e.g. months, years, lifetime), inflammation of this type may lead to permanent scarring of lung tissue, loss of lung function, and reduced lung elasticity.
Measurement Techniques Ozone is monitored using specialized commercial instruments made for that specific purpose. The analyzers continuously measure the concentration of ozone in ambient air using the U.V. photometric method and are EPA-approved for regulatory air monitoring programs. Data gained from the continuous monitors is used to determine compliance with the NAAQS ozone.
Attainment Designation Ozone monitoring has been in place in the Atlanta area since 1980. The 1980 network consisted of two monitors located in DeKalb and Rockdale Counties. Currently the metro Atlanta ozone network includes eleven monitors located in ten counties.
In July 1997 the US EPA issued a new 8-hour ozone standard intended to eventually replace the older 1-hour standard. This 8-hour standard is attained when the average of the fourth highest concentration measured is equal to or below 0.08 ppm (0.085 ppm with the EPA rounding convention) averaged over three years (see Table 1; 62 FR 38894, July 18, 1997). Areas EPA has declared in attainment of 1-hour standard are immediately exempt from that standard, but thereafter are subject to the 8-hour standard. In the summer of 2005, metro Atlanta was declared in attainment of the 1-hour standard. As of the printing of this report, then, only the 8-hour ozone standard is applicable in Georgia.
The Atlanta ozone nonattainment area was officially expanded in 2004. Previously Rockdale, Coweta, Fulton, Cherokee, Henry, Clayton, Fayette, Gwinnett, Paulding, Forsyth, Cobb, Douglas, and DeKalb Counties were included. With new monitoring data, implementation of the 8-hour ozone standard, and the results of the 2000 Census, the following counties were added to the nonattainment area: Barrow, Bartow, Carroll, Douglas, Hall, Newton, Spalding, and Walton. Catoosa County is part of the Chattanooga Early Action Compact area. New basic nonattainment areas were also declared in 2004. The Macon metro area was declared
26
2006 Georgia Annual Air Quality Report a new nonattainment area. It includes Bibb County and part of Monroe County. Finally, portions of Murray County were added to a new Chattahoochee National Forest nonattainment area. Figure 9 shows the boundaries of these nonattainment areas. A number of activities to aid in controlling the precursors to ozone formation have been implemented. As new areas are declared in nonattainment, these control measures may be expanded to include them. These activities could include a strict vehicle inspection program, controls on stationary emission sources, and the establishment of a voluntary mobile emissions reduction program. An example of such a program in metro Atlanta is called The Clean Air Campaign (CAC). Activities of The Clean Air Campaign include distributing daily ozone forecasts (as well as PM2.5 forecasts produced by EPD) during the ozone season to enable citizens in the sensitive group category as well as industries to alter activities on days that are forecasted to be conducive to ozone formation. This is also done for the Macon area. In addition to the daily forecasts, citizens have access to forecast and monitoring data on an as needed basis by either calling 1-800-427-9605 or by accessing our website at http://www.georgiaair.org/amp. For a more detailed discussion concerning the CAC, see the section titled "Outreach and Education".
27
DADE
CATOOSA
WALKER
WHITFIELD MURRAY
CHATTOOGA
GORDON
FLOYD
BARTOW
FANNIN
UNION
TOWNS
RABUN
ChGIaLMttEaRhoochee N.F.
LUMPKIN
PICKENS
DAWSON
CHEROKEE
FORSYTH
WHITE
HABERSHAM STEPHENS
HALL
BANKS
FRANKLIN
JACKSON
MADISON
HART ELBERT
POLK HARALSON
CARROLL HEARD
PAULDING DOUGLAS COWETA
COBB
GWINNETT
BARROW
CLARKE
DEKALB
WALTON
OCONEE
FULTON CLAYTON
ROCKDALE NEWTON
MORGAN
FAYETTE
HENRY
OGLETHORPE
WILKES
LINCOLN
GREENE
TALIAFERRO
COLUMBIA MCDUFFIE
WARREN
SPALDING
BUTTS
JASPER
PUTNAM
HANCOCK
GLASCOCK
TROUP
MERIWETHER
PIKE
LAMAR
Monroe partial MONROE
JONES
BALDWIN
HARRIS
UPSON
TALBOT
CRAWFORD
BIBB
TWIGGS
WILKINSON
JEFFERSON WASHINGTON
JOHNSON
MUSCOGEE
TAYLOR
CHATTAHOOCHEE
MARION
SCHLEY
MACON
STEWART
WEBSTER
SUMTER
PEACH HOUSTON
BLECKLEY
LAURENS
EMANUEL TREUTLEN
DOOLY
PULASKI WILCOX
DODGE
MONTGOMERY
WHEELER
TOOMBS
TELFAIR
Figure 9: Georgia's 8-Hour Ozone Nonattainment Area Map
28
2006 Georgia Annual Air Quality Report
Figure 10 shows how past air quality would relate to the new standard and how current air quality relates to the old standard. This chart was produced by comparing measurement data against both ambient standards. This demonstrates the relative strictness of each standard and shows how our air quality has changed over time. Despite a great deal of fluctuation, over the course of the past twenty years we have seen a gradual reduction in the number of days exceeding either ozone standard. The trendline, produced by regression analysis, shows that the number of days that exceed the current 8-hour ozone standard has fallen by about a half day each year over this time period. The most recent years of data show an upturn in the number of exceedance days. This is believed to be a result of natural variation in weather patterns, not a sign of a reversal of the long-term trend.
Metro Atlanta OzoneNumber of Days Exceeding NAAQS per Year
80 70 60 50 40 30 20 10
0
Exceedance Days 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1-Hour
Year 8-Hour Trendline of 8-Hour
Figure 10: Metro Atlanta Ozone- Number of Violation Days per Year
29
Figure 11 maps each Metro Atlanta ozone monitor that exceeded the 8-hour ozone standard in 2006, and also indicates the monthly breakdown of the exceedances.
Figure 11: Metro Atlanta Ozone Exceedance Map 30
2006 Georgia Annual Air Quality Report
Lead
General Information In the past the Clean Air Act required extensive lead monitoring in order to detect the high levels of airborne lead that resulted from the use of leaded gasoline. With the phase-out of leaded gasoline, lead concentrations decreased to nearly zero by the late 1980s. Since then, the concentrations have hovered just above zero. Today, metals processing is the major source of lead emissions to the atmosphere. Other sources of lead emissions include the combustion of solid waste, coal, oils, emissions from iron and steel production, and the two remaining uses for lead in gasoline (aviation gasoline and racing gasoline).
Based on EPA guidance and the very small amounts of lead pollution remaining in Georgia, the lead monitoring network shrunk in 2005. Two lead monitoring sites in Columbus were closed. There are now two dedicated lead monitors remaining in Georgia for comparison to the NAAQS lead standard. One is in the Atlanta area for monitoring long-term trends in ambient lead levels as part of the National Air Toxics Trends Stations (NATTS) network. The other is in Columbus for industrial source monitoring, given the historical issues with lead pollution in the area.
The current criteria lead monitoring network is as indicated in Figure 12. For more information on criteria lead monitoring, see Appendix A. In addition to the criteria network sites, lead is also being monitored at 15 sites throughout Georgia as a trace metal in the Georgia Air Toxics Monitoring Network. The equipment used at those sites can detect far smaller concentrations. For additional summary data on lead as collected as an Air Toxics trace metal, see Appendix E.
Health Impacts Exposure to lead occurs mainly through inhalation and ingestion of lead in food, water, soil, or dust. It accumulates in the blood, bones, and soft tissues. Lead can adversely affect the kidneys, liver, nervous system, and other organs. Excessive exposure to lead may cause neurological impairments, such as seizures, mental retardation, and behavioral disorders. Even at low doses, lead exposure is associated with damage to the nervous systems of fetuses and young children, resulting in learning deficits and lowered IQ. Recent studies also show that lead may be a factor in high blood pressure and subsequent heart disease. Lead can also be deposited on the leaves of plants, presenting a hazard to grazing animals. Lead deposition in soil puts children at particular risk exposure since they commonly put hands, toys, and other items in their mouths, which may come in contact with the lead-containing dust and dirt.
31
Sequatchie
Marion
Hamilton
Dade
Catoosa
Walker
Murray Whitfield
Floyd Bartow
Pickens Dawson Hall
Cherokee Forsyth
Haralson Carroll Heard
Paulding Cobb
Gwinnett
Barrow
De Kalb
Walton
Douglas
Rockdale
Fulton
DMRC
Coweta
Clayton Fayette
Newton Henry
Spalding
Jasper Butts
Madison
Clarke Oconee
Oglethorpe
Meriwether
Pike Lamar Monroe
Jones
Russell
Harris
Muscogee
Cusseta ES
Chattahoochee Marion
Crawford
Bibb
Twiggs
Houston
Terrell
Lee
Dougherty Baker
Worth
Brooks
Lanier Lowndes
Echols
Edgefield
Columbia McDuffie
Richmond
Aiken
Burke
Effingham
Bryan
Liberty Long
Chatham
McIntosh
Brantley
Glynn
Lead Sites Urban Areas MSAs Shown as Solid Colors
Figure 12: Lead Monitoring Site Map
32
2006 Georgia Annual Air Quality Report
Measurement Techniques Measurement for ambient air lead concentrations is performed using a manual method, unlike measurements for ozone, SO2, NO2 and CO. Samples are collected on 8" x 10" preweighed fiberglass filters with a high-volume sampler for 24 hours. The filter sample is shipped to a laboratory for analysis using inductively coupled plasma mass spectroscopy (commonly known as ICP-MS). Data gained from the lead sampler is used to determine compliance with the National Ambient Air Quality Standards for lead.
Attainment Designation The compliance with the national primary and secondary ambient air quality standards for lead and its compounds is determined based on the assumption that all lead is elemental lead. In order to comply with both the primary and secondary standard, the concentration of lead in the air must have an arithmetic mean no higher than 1.5 micrograms per cubic meter averaged over a calendar quarter. (See Secs. 109, 301(a) Clean Air Act as amended (42 U.S.C. 7409, 7601(a)) [43 FR 46258, Oct. 5, 1978].
All of Georgia is in attainment of the lead standard. For additional summary data on this topic, see Appendix A.
Particulate Matter
General Information Particulate matter is a broad range of material that consists of solid particles, fine liquid droplets, or condensed liquids absorbed onto solid particles. Airborne particulates are not a single pollutant as discussed for the other criteria pollutants, but rather a mixture of many different air pollutants. Primary sources that emit particles include combustion, incineration, construction, mining, metals smelting, metal processing, and grinding. Other sources include motor vehicle exhaust, road dust, wind blown soil, forest fires, open burning of vegetation for land clearing or waste removal, ocean spray, and volcanic activity.
There are two ways that particulate matter is formed. Primary particulate is emitted directly from a source, like a vehicle's tailpipe or a factory's smokestack. But a great deal of particulate matter is not directly emitted from such sources. In fact, the vast majority of primary air pollution is in the form of gases. But those gaseous air pollutants readily react in the atmosphere- with oxygen, each other, and the like. While many of those reactions produce other gases, they frequently produce particles. Particles formed through this process are known as secondary particulate matter. Examples of secondary particulates include:
Atmospheric sulfate particles, formed from the oxidation of gaseous SO2. Atmospheric nitrate particles, such as ammonium nitrate, formed from a complex
series of reactions that transform gaseous NOx. Atmospheric calcium nitrate or sodium nitrate particulates formed from a series of
atmospheric reactions involving gaseous nitric acid (HNO3) reacting with sodium chloride/calcium carbonate.
Particulate pollution may also be categorized by size since there are different health impacts associated with the different sizes. The Ambient Air Monitoring Program monitors for two sizes of particles: PM10 (up to 10 microns in diameter) and PM2.5 (up to 2.5 microns in
33
diameter). Both of these particles are very small in size. For example, approximately ten (10) PM10 particles can fit on a cross section of a human hair, and approximately thirty (30) PM2.5 particles would fit on a cross section of a hair.
Figure 13 shows the location of Georgia's PM10 monitoring sites, Figure 16 shows the location of the PM2.5 reference method monitoring sites, and Figure 17 shows the location of the PM2.5 continuous monitoring sites. For information on the makeup of these particles, see the following section (PM2.5 Speciation).
PM10
Particulate matter (PM) less than or equal to 10 microns in diameter is defined as PM10. These particles can be solid matter or liquid droplets from smoke, dust, fly ash, or condensing vapors that can be suspended in the air for long periods of time. PM10 represents part of a broad class of chemically diverse particles that range in size from molecular clusters of 0.005 microns in diameter to coarse particles of 10 microns in diameter (for comparison, an average human hair is 70-100 microns in diameter). PM results from all types of combustion. The carbon-based particles that result from incomplete burning of diesel fuel in buses, trucks, and cars are of particular concern. Another important combustion source is the burning of wood in stoves and fireplaces in residential settings. Also of concern are the sulfate and nitrate particles that are formed as a byproduct of SO2 and NO2 emissions, primarily from fossil fuel-burning power plants and vehicular exhausts.
The U.S. national ambient air quality standard was originally based on particles up to 25-45 microns in size, termed "total suspended particles" (TSP). In 1987, EPA replaced TSP with an indicator that includes only those particles smaller than 10 microns, termed PM10. These smaller particles cause most of the adverse health effects because of their ability to penetrate deeply into the lungs. The observed human health effects of PM include breathing and respiratory problems, aggravation of existing respiratory and cardiovascular disease, alterations in the body's defense system against inhaled materials and organisms, and damage to lung tissue. Groups that appear to be most sensitive to the effects of PM include individuals with chronic lung or cardiovascular disease, individuals with influenza, asthmatics, elderly people, and children.
Health Impacts Marked increases in daily mortality have been statistically associated with very high 24-hour concentrations of PM10, with some increased risk of mortality at lower concentrations. Small increases in mortality appear to exist at even lower levels. Risks to sensitive individuals increase with consecutive, multi-day exposures to elevated PM10 concentrations. The research also indicates that aggravation of bronchitis occurs with elevated 24-hour PM10 levels, and small decreases in lung function take place when children are exposed to lower 24-hour peak PM10 levels. Lung function impairment lasts for 2-3 weeks following exposure to PM10.
34
2006 Georgia Annual Air Quality Report
Sequatchie
Hamilton Marion
Dade Rossville Catoosa
Walker
Whitfield
Murray
Summerville
Pickens
Dawson
Hall
Floyd Coosa HS Bartow
Cherokee
Forsyth
Haralson Carroll Heard
Paulding
CobbDoraville Gwinnett
Fire Station 8 GA Tech
E Rivers Sch De Kalb
Douglas
Fulton FCHD Rockdale
Barrow Walton
Clayton
Newton
Coweta
Fayette
Henry
Griffin Butts
Spalding
Jasper
Madison
Clarke Oconee
Oglethorpe
Meriwether
Pike
Lamar
Monroe
Jones
Harris
Russell
Muscogee
Cusseta ES
Chattahoochee Marion
Bibb Allied Chem
Crawford
Twiggs
Houston
Terrell
Lee
Dougherty
Albany
Worth
Baker
Lanier
Bartow
Cherokee
Forsyth
Paulding
Cobb
Gwinnett
Doraville
Fire Station 8 E Rivers Sch
Douglasville
Douglas
Fulton
GA Tech De Kalb FCHD
Clayton
Coweta
Fayette Edgefield
McDuffie
Columbia
Aiken
Bungalow ES
Richmond
Rockdale Henry
Burke
Sandersville
Lathrop Augusta
Shuman MS
Long
Liberty
McIntosh
Brantley
Glynn
Brunswick
Brooks
Lowndes
Echols
PM 10 Sites Urban Areas MSAs Shown as Solid Colors
Figure 13: PM10 Monitoring Site Map
35
Measurement Techniques The Georgia PM10 monitoring network consists of two types of monitors. The first is an event monitor in which samples are collected for 24 hours on a quartz microfiber filter. A specialized sample-sorting device is used so that the filter collects only particles 10 microns in size and smaller. The filters are weighed in a laboratory before and after the sampling period. The change in the filter weight corresponds to the mass of PM10 particles collected. That mass, divided by the total volume of air sampled, corresponds to the mass concentration of the particles in the air. Because of the need for manual filter loading and unloading, shipping back to the laboratory, and so forth, there is significant time lag between taking the measurement and obtaining data. The other monitor is fundamentally similar, but has been greatly modernized. It draws particle-laden air through a filter and analyzes how the mass of the filter changes on an hourly or nearly continuous basis. This monitor gives much more information about how PM10 concentrations vary over time, is less labor-intensive, and produces results almost instantly.
Attainment Designation The primary and secondary standards for PM10 are the same. In order for an area to be considered in compliance with the standard, the 99th percentile 24-hour concentration is less than or equal to 150 micrograms per cubic meter [62 FR 38711, July 18, 1997]. For example, if 100 PM10 samples were taken over the course of the year and one sample exceeded 150 micrograms per cubic meter, then the area would meet the standard. If two or more samples were over 150 micrograms per cubic meter, then the area would not be in attainment of the standard. There was also an annual average standard for PM10 until December 17, 2006. EPA revoked the standard because of a lack of evidence of chronic health effects resulting from long-term exposure to moderate levels of PM10. All of Georgia is currently in attainment of the PM10 standard. For additional summary data on this topic, see Appendix A.
36
2006 Georgia Annual Air Quality Report
Average Concentration (g/m3)
35.0 30.0 25.0 20.0 15.0 10.0
5.0 0.0
2002
PM10 Annual Arithmetic Means
2003
2004 Year
2005
2006
Albany Augusta Brunswick Columbus Macon Metro Atlanta Metro Savannah Rome Rossville Sandersville Summerville
Figure 14: PM10 Annual Arithmetic Mean Chart
Figure 14 shows how the metro areas in Georgia stand relative to the annual average PM10 standard. On an annual basis, PM10 levels in Georgia are relatively low. Most inland areas have similar annual average concentrations, and coastal areas typically have less PM10 than inland areas. Longer-term PM10 data generally suggest a slow decrease in PM10 concentrations overall, though 2003 2006 data suggest a minor upturn. This is believed to be a result of natural variation in weather patterns, not an indication of a significant trend toward higher PM10 concentrations.
Figure 15 shows how the same areas compare to the 24-hour standard for PM10, which remains set at 150 g/m3. Because the standard allows one exceedance per year, this chart shows the 2nd highest 24-hour average for each site or metro area. Though there is a great deal of variation from year to year at any given site, the statewide average is relatively stable. Statistical analysis of the statewide average PM10 concentrations from 1998 through 2006 indicates a trend of improvement, with an average decrease of 1.26 g/m3 per year.
37
Statewide (avg)
PM10 24-Hour Design Values
Sandersville
Rossville
Augusta
Columbus
Brunswick Site
Rome
Albany
1998 1999 2000 2001 2002 2003 2004 2005 2006
Metro Atlanta
Summerville
Savannah
Macon 0
10
20
30
40
50
60
70
80
90
Second Highest 24-Hour PM10 Concentration (g/m3)
Figure 15: PM10 24-Hour Averages 38
2006 Georgia Annual Air Quality Report
PM2.5
Particulate matter including PM2.5 consists of the solid particles and liquid droplets found in the air. Individually, these particles and droplets are invisible to the naked eye. Collectively, however, they can appear as clouds or a fog-like haze.
Particulate matter less than or equal to 2.5 microns in diameter is referred to as "fine" particles. In comparison, a human hair is 70-100 microns in diameter. Fine particles are produced by many different sources, including industrial combustion, residential combustion, and vehicle exhaust, so their composition varies widely. Fine particles can also be formed when combustion gases are chemically transformed into particles. Considerable effort is being undertaken by Georgia to analyze the fine PM2.5 particles for the chemical constituents that make up the particles, so pollution control efforts can be focused in areas that create the greatest reductions. We currently monitor fifty-three (53) particle species, which include gold, sulfate, lead, arsenic, and silicon.
Health Impacts Fine particles are of health concern because they can penetrate into the sensitive regions of the respiratory tract and they are linked to the most serious health effects. They can cause persistent coughs, phlegm, wheezing, and physical discomfort.
Several recently published community health studies indicate that significant respiratory and cardiovascular-related problems are associated with exposure to particle levels well below the existing particulate matter standards. These negative effects include premature death, hospital admissions from respiratory causes, and increased respiratory problems. Long-term exposure to particulate matter may increase the rate of respiratory and cardiovascular illness and reduce life span. Some data also suggests that fine particles can pass through lung tissues and into the bloodstream. Children, the elderly, and individuals with cardiovascular disease or lung diseases such as emphysema and asthma are especially vulnerable.
Much new evidence of these health impacts is discussed in EPA's "Air Quality Criteria for Fine Particulate Matter" document. It indicates that fine particles are also thought to enhance the delivery of other pollutants and allergens deep into lung tissue, where their effects are more pronounced. The cancer risk alone of long-term exposure to elevated fine particle concentrations has been estimated to be similar to that of the additional risk to a non-smoker of living with a smoker. Early estimates of the impact of long-term exposure to elevated fine particle concentrations indicate an average life span reduction of one to two years (U.S. EPA, 2004a). Taking this information and that from many other studies into account, EPA strengthened the 24-hour PM2.5 standard effective December 17, 2006.
Fine particles can soil man-made materials, making them look sooty and speeding their deterioration. They also impair visibility and are an important contributor to haze, especially in humid conditions. The visibility effect is roughly doubled at 85% relative humidity as compared to humidities under 60% (U.S. EPA, 2004a).
Measurement Techniques Measurement techniques for PM2.5 are very similar to those for PM10. The official reference method requires that samples are collected on Teflon filters with a PM2.5 sampler for 24
39
hours. A specialized sample-sorting device is used so that the filter collects only particles 2.5 microns in size and smaller. The filters are weighed in a laboratory before and after the sampling period. The change in the filter weight corresponds to the mass of PM2.5 particles collected. That mass, divided by the total volume of air sampled, corresponds to the mass concentration of the particles in the air for that 24-hour period. Only these reference method filters may be used for attainment determinations. But because of the delay in picking up each filter, shipping it to the laboratory, and weighing each filter, weeks pass before the results are known. As such this method is very accurate but is useless when trying to decide if it is safe to go jogging or to let children play outside. As with PM10, continuous samplers for PM2.5 are also used to report "real time" data to support programs like air quality forecasting and public information efforts like the AQI. These instruments produce hourly averaged data that is available almost immediately after the end of each hour. By having information available so quickly, the public can make informed decisions about their levels of physical activity. It is important to note that the EPA does not certify these continuous samplers as being fully equivalent to the reference method when sampling PM2.5. This means that data from these continuous samplers cannot be used for determining if an area is in attainment of the NAAQS; only data from the reference method may be used. Continuous PM2.5 data is reported every hour on the Ambient Air Monitoring web page located at http://www.air.dnr.state.ga.us/amp. Figure 16 and Figure 17 show the location of Georgia's PM2.5 monitors.
40
2006 Georgia Annual Air Quality Report
Sequatchie
Hamilton Marion
Dade Rossville
Catoosa
Walker
Whitfield Murray
Pickens
Dawson
Floyd Coosa HS Bartow
Cherokee
Forsyth
Hall
Gainesville
Madison
Haralson
Paulding
Kennesaw
Cobb
Doraville
Gwinnett TechBarrow
Clarke UGA
Yorkville Macland Ctr Gwinnett E Rivers Sch
Fire Station 8 DeKalb
Walton
College Station Oconee Oglethorpe
Douglas
Fulton
S DeKalbRockdale
Carroll
Forest Park Clayton
Newton
Heard
Coweta
Fayette
Henry
Spalding
Butts
Jasper
Edgefield
McDuffie
Columbia Richmond
Aiken
Bungalow ES GA Med Col
Meriwether
Pike
Lamar
Monroe
Jones
Harris
Columbus Airport
Muscogee
Muscogee Cusseta ES
Chattahoochee Marion Russell
Bibb
GA ForeGstoryrdon
Crawford Allied ChemTwiggs
Warner Robins
Houston
Sandersville
Terrell
Lee
Albany
Dougherty
Worth
Burke
Effingham
Market St
Bryan
Mercer Mid
Chatham
Long
Liberty
McIntosh
Baker
Lanier
Brantley
Glynn
Brunswick
Brooks
Valdosta
Lowndes
Echols
PM 2.5 FRM Urban Areas MSAs Shown as Solid Colors
Figure 16: PM2.5 Federal Reference Method Monitoring Site Map
41
Sequatchie
Hamilton
Rossville
Catoosa
Walker
Whitfield
Murray
Pickens
Dawson
Floyd
Bartow
Coosa HS
Cherokee
Forsyth
Hall
Madison
Yorkville
Paulding Haralson
Douglas
Carroll
Newnan
Heard
Coweta
Cobb
Gwinnett Tech
Gwinnett
DeKalb
Barrow
College Station
UGA
Clarke Oglethorpe
Oconee
Confederate
Fulton
Rockdale
Clayton S DeKalb
Walton
Social Circle
Henry
Newton
Fayette
McDonough
Spalding
Butts
Jasper
Pike Meriwether
Harris
Columbus Airport
Muscogee
Cusseta ESChattahoochee
Russell
Marion
Lamar
Monroe
Jones
GA Forestry
Bibb
Allied Chem
Crawford
Twiggs
Houston
Terrell
Lee
Dougherty Baker
Worth
Lanier
Edgefield
McDuffie
Columbia
Aiken
Richmond Bungalow ES
Burke
General Coffee
Effingham
Lathrop Augusta
Bryan
Chatham
Long
Liberty
Brantley
McIntosh Glynn
Brooks
PM 2.5 Continuous PM 2.5 Speciation
Urban Areas MSAs Shown as Solid Colors
Lowndes
Echols
Figure 17: PM2.5 Monitoring Site Map, Continuous and Speciation Monitors
42
2006 Georgia Annual Air Quality Report Attainment Designation For an area to be in attainment of the national primary and secondary annual ambient air PM2.5 standard, it must have an annual arithmetic mean concentration of less than or equal to 15.0 micrograms per cubic meter. In addition, there was a 24-hour primary and secondary standard that requires that the 98th percentile 24-hour concentration be less than or equal to 65 micrograms per cubic meter [62 FR 38711, July 18, 1997]. This 24-hour limit was reduced to a 35-microgram limit effective December 17, 2006. All sample analyses used for determining compliance with the standards must use a reference method based on information in 40 CFR Appendix L or an equivalent method as designated in accordance with Part 53. Because the PM2.5 standard required three years of monitoring data before attainment or nonattainment could be determined, Georgia's attainment status was not determined until late 2004. As was expected, large portions of the United States were found to be in nonattainment of the standard when EPA made its initial attainment determinations. Based on the three years of data, EPA officially declared several areas of Georgia in nonattainment of the standard. Walker and Catoosa Counties are included in the metro Chattanooga nonattainment area. All of Bibb County and portions of Monroe County have been included in the Macon nonattainment area. Floyd County itself has been declared a nonattainment area. Finally, the metro Atlanta nonattainment area has been also declared. This includes Barrow, Bartow, Carroll, Cherokee, Clayton, Cobb, Coweta, DeKalb, Douglas, Fayette, Forsyth, Fulton, Gwinnett, Hall, Henry, Newton, Paulding, Rockdale, Spalding, and Walton Counties, along with portions of Heard and Putnam Counties. Figure 18 illustrates the boundaries of Georgia's PM2.5 nonattainment areas.
43
CATOOSA WALKER
FLOYD
BARTOW
CHEROKEE
FORSYTH
HALL
PAULDING
COBB
GWINNETT
BARROW
CARROLL
DOUGLAS
Heard partial
COWETA
FULTON
DEKALB ROCKDALE
WALTON
CLAYTON
NEWTON
FAYETTE
HENRY
SPALDING
Putnam partial
Monroe partial
BIBB
Figure 18: Georgia's PM2.5 Nonattainment Area Map 44
2006 Georgia Annual Air Quality Report
PM2.5 Mass Concentration Annual Average (Arithmetic Mean)
Gordon
Sandersville
Rossville
Augusta Bungalow Rd.
Augusta Medical Col.
Yorkville
Columbus Cusseta Rd.
Columbus Airport*
Columbus Health Dept.
Valdosta
Warner Robins
Gainesville
Site
Gwinnett
Brunswick
Atlanta Fire Station # 8
Atlanta E. Rivers School
Rome
Albany
Doraville
South DeKalb
Powder Springs
Kennesaw
Forest Park
Athens UGA**
Athens College Station Rd.*
Savannah Mercer
Savannah Market St.
Macon Forestry
Macon Allied Chem.
0.0
2.5
5.0
7.5
10.0
12.5
15.0
Concentration (g/m3)
* Sites established in 2005; only 2005-2006 data used for graph ** Site was shut down early in 2005; 2003-2005 average does not include three full years
01-03 02-04 03-05 04-06
15 g/m3 limit
17.5
20.0
Figure 19: PM2.5 Design Value, Annual Standard, By Site
As can be seen in Figure 19, many monitoring sites in Georgia do not attain the annual PM2.5 standard. The sites exceeding the standard are generally in north and central Georgia, and the sites attaining the standard are generally in coastal areas.
45
PM2.5 24-Hour Standard
Gordon
Sandersville
Rossville
Augusta Bungalow Rd.
Augusta Med. Col.
Yorkville
Columbus Cussetta Rd.
Columbus Airport*
Columbus H.D.
Valdosta
Warner Robins
Gainesville
Gwinnett
Brunswick
Site
Atl. F.S. # 8**
Atl. E. Rivers Sch.
Rome
Albany
Doraville
South DeKalb
Powder Springs
Kennesaw
Forest Park
Athens
Athens College Station Rd.*
Sav. Mercer
Sav. Market St.
Macon Forestry
Macon Allied Chem.
00-02 01-03 02-04 03-05 04-06
New 35 g/m3 limit
5.0
15.0
25.0
35.0
45.0
55.0
98th Percentile Concentration (g/m3)
* Sites established in 2005; only 2005-2006 data used for graph ** Site was shut down early in 2005; 2003-2005 average does not include three full years
Figure 20: PM2.5 Design Value, Daily Standard, By Site 46
2006 Georgia Annual Air Quality Report
Figure 20 is of particular interest this year, with EPA's release of a stricter 24-hour limit on PM2.5 concentrations. While all Georgia monitors were comfortably achieving the previous 65 g/m3 limit, many are close to the new 35 g/m3 limit, and the Forest Park monitor was over the limit for the 2004-2006 period. It is reassuring to note, though, that while variations in weather patterns have masked any recent progress in controlling annual average PM2.5 concentrations, significant improvement in 24-hour concentrations are visible over the same period. If this trend continues, Georgia may be able to achieve full attainment of the stricter 24-hour standard.
Specific annual summaries for 2006 may be found in Appendix A.
PM2.5 Speciation
As required by the National PM2.5 Speciation program (40 CFR 58), EPD now monitors not only the mass concentration of fine particulate matter (in micrograms per cubic meter of air) but also for the chemical composition of those particles. Doing so serves a number of important purposes.
As controlling fine particulate matter concentrations has now become a national priority through their listing in the National Ambient Air Quality Standards, regulations intended to reduce levels of fine particulate matter are now being implemented on a widespread basis. The desired reduction of fine particulate matter concentrations is expected to produce benefits in human health and to improve visibility by reducing the presence of haze. These particles may have varying health effects depending on their size and chemical composition.
The fine particles that compose fine particulate matter are not uniform. While they are all smaller than 2.5 microns in diameter, their size varies. Some are emitted into the air directly from sources like engine exhaust, fossil fuel combustion, unpaved roads, and the tilling of fields; others are formed in the atmosphere through reactions between gaseous pollutants. Each individual particle, regardless of its source, has a distinct chemical composition. The overall composition of all of the particles that make up the fine particulate matter in a given volume of air may also vary, depending on local sources and a variety of other factors.
Georgia currently monitors fifty-three (53) species, which include gold, sulfate, lead, arsenic, and silicon. However, there are only approximately six (6) chemicals that are detected frequently (the chemical elements typical of the Earth's crust are grouped together as "crustal"). Of these, sulfate and organic carbon are detected in the highest concentrations. Figure 21, Figure 22, Figure 23, and Figure 24 illustrate the average concentrations of these six chemicals from 2003 to 2006. Below the figures is a listing of the most significant chemical constituents of fine particulate matter.
47
Concentration (g/m3)
7
6
5
4
3
2
1
0 Ammonium Ion
Elemental Carbon
2003 PM2.5 Speciation Parameters Arithmetic Mean
Organic Carbon
Sulfate Species
Nitrate
Crustal
Figure 21: 2003 PM2.5 Speciation
Concentration (g/m3)
7
6
5
4
3
2
1
0 Ammonium Ion
Elemental Carbon
2004 PM2.5 Speciation Parameters Arithmetic Mean
Organic Carbon
Sulfate
Species
Nitrate
Crustal
Figure 22: 2004 PM2.5 Speciation
Other Other
Macon Savannah Athens General Coffee Atlanta Rome Columbus Augusta
Macon Savannah Athens General Coffee Atlanta Rome Columbus Augusta
48
Concentration (g/m3)
2006 Georgia Annual Air Quality Report
7
6
5
4
3
2
1
0 Ammonium Ion
Elemental Carbon
2005 PM2.5 Speciation Parameters
Organic Carbon
Sulfate
Species
Nitrate
Crustal
Other
Macon
Athens
General Coffee Atlanta
Rome
Columbus
Augusta
Rossville
Figure 23: 2005 PM2.5 Speciation
Concentration (g/m3)
7
6
5
4
3
2
1
0 Ammonium Ion
Elemental Carbon
2006 PM2.5 Speciation Parameters
Organic Carbon
Sulfate Species
Nitrate
Crustal
Other
Macon
Athens
General Coffee Atlanta
Rome
Columbus
Augusta
Rossville
Figure 24: 2006 PM2.5 Speciation
49
Predominant Species Found in PM2.5 Ammonium Ion is commonly released by fertilizer production, livestock
production, coke production, and some large refrigeration systems. Ironically, it can also be emitted by NOx control systems installed on large fossil fuel combustion systems, which use ammonia or urea as a reactant. Sulfate products form from the oxidation of SO2 in the atmosphere. SO2 is primarily produced by coal burning boilers. Nitrate products are formed through a complex series of reactions that convert NOx to nitrates. Vehicle emissions and fossil fuel burning produce NOx. Crustal products are those components that are the result of weathering of the Earth's crust. They may include ocean salt and volcanic discharges. Crustal products include aluminum, calcium, iron, titanium, and silicon. These components released by metals production, and can also be resuspended in the atmosphere by mechanisms that stir up fine dust, such as mining, agricultural processes, and vehicle traffic. Elemental carbon is carbon in the form of soot. Sources of elemental carbon include diesel engine emissions, wood-burning fireplaces, and forest fires. Organic carbon particles consist of hundreds of organic compounds that contain more than 20 carbon atoms. These particles are released directly and are also formed are formed through a series of chemical reactions in the air, mostly as a result of the burning of fossil fuels and wood.
Data on the composition of the particles that make up fine particulate matter is a useful input to scientific models of air quality, and will help scientists and regulators track the progress and effectiveness of new pollution controls that are implemented. The data will also greatly improve scientific understanding of how particle composition relates to visibility impairment and adverse effects on human health.
Monitoring for chemical speciation for fine particles began late in 2001, so there is limited data available. As the data set becomes more robust, other conclusions may be drawn. Some general observations can already be made. The concentrations of sulfate and organic carbon are less at the General Coffee site than at the remaining seven (7) sites. This is to be expected since the sulfate and organic carbon fractions are mainly caused by human activities. The Douglas-General Coffee site is considered a rural background site and will be used in future comparisons between rural and urban areas.
Now that four years of speciation data is available from a variety of sites, we can begin to evaluate the data for trends. Figure 25, Figure 26, Figure 27, Figure 28, Figure 29, and Figure 30 present a different view of the same data to facilitate visualization of trends. Note that because each chemical component presented is present in widely varying concentrations, each chart uses a different scale for the mass concentrations.
50
2006 Georgia Annual Air Quality Report
Ammonium Arithmetic Mean Mass Concentration
2003-2006
1.8 1.6 1.4 1.2
1 0.8 0.6 0.4 0.2
0 Macon Savannah Athens
General Coffee
Atlanta Site
Rome Columbus Augusta Rossville
2003 2004 2005 2006
g/m3
Figure 25: PM2.5 Speciation, Trends in Ammonium Concentrations
g/m3
Elemental Carbon Arithmetic Mean Mass Concentration
2003-2006
1.4
1.2
1
0.8
0.6
0.4
0.2
0 Macon Savannah Athens
General Coffee
Atlanta Site
Rome Columbus Augusta Rossville
2003 2004 2005 2006
Figure 26: PM2.5 Speciation, Trends in Elemental Carbon Concentrations
51
g/m3
Organic Carbon Arithmetic Mean Mass Concentration
2003-2006
8 7 6 5 4 3 2 1 0
Macon Savannah Athens
General Coffee
Atlanta Site
Rome Columbus Augusta Rossville
2003 2004 2005 2006
Figure 27: PM2.5 Speciation, Trends in Organic Carbon Concentrations
g/m3
Sulfate Arithmetic Mean Mass Concentration
2003-2006
6
5
4
3
2
1
0 Macon Savannah Athens
General Coffee
Atlanta Site
Rome Columbus Augusta Rossville
2003 2004 2005 2006
Figure 28: PM2.5 Speciation, Trends in Sulfate Concentrations
52
2006 Georgia Annual Air Quality Report
Nitrate Arithmetic Mean Mass Concentration
2003-2006
1.2
1
0.8
0.6
0.4
0.2
0 Macon Savannah Athens
General Coffee
Atlanta Site
Rome Columbus Augusta Rossville
2003 2004 2005 2006
g/m3
Figure 29: PM2.5 Speciation, Trends in Nitrate Concentrations
g/m3
Crustal Matter Arithmetic Mean Mass Concentration
2003-2006
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1
0 Macon Savannah Athens
General Coffee
Atlanta Site
Rome Columbus Augusta Rossville
2003 2004 2005 2006
Figure 30: PM2.5 Speciation, Trends in Crustal Matter Concentrations
53
Ammonium Ion concentrations are relatively even statewide, ignoring some unexplained high and low averages observed during 2005.
The Atlanta and Macon areas have the highest elemental carbon concentrations, which is no surprise given their location on major Interstate trucking corridors. Cities with less heavy vehicle traffic have lower concentrations, and the rural site (General Coffee) has the least elemental carbon.
Organic carbon concentrations are also relatively consistent throughout the state, though rural areas like General Coffee are slightly lower. Note how organic carbon concentrations are much higher than typical ammonium ion or elemental carbon concentrations. This means that organic carbon is a relatively large contributor to the total PM2.5 mass concentrations that are a direct regulatory concern.
Sulfate concentrations can be described much the same as organic carbon concentrations. They are relatively consistent statewide, though somewhat lower in rural areas, and their relatively large observed mass means that they are also a major contributor to overall PM2.5mass concentrations.
Nitrate concentrations are relatively small, but are less consistent from site to site. Their concentrations do seem to track somewhat with NOx emission patterns, but the high concentrations at Athens are not readily explained.
Crustal matter concentrations are relatively low and consistent in most areas. Rome and Macon have in some years recorded unexpectedly high crustal matter concentrations. This may be a sign of poor dust control at agricultural, construction, or mining operations in those areas.
For additional PM2.5 speciation data, see Appendix B.
Measurement Techniques Particle speciation measurements require the use of a wide variety of analytical techniques, but all generally use filter media to collect the particles to be analyzed. Laboratory techniques currently in use are gravimetric (microweighing); X-ray fluorescence and particleinduced X-ray emission for trace elements; ion chromatography for anions and selected cations; controlled combustion for carbon; and gas chromatography/mass spectroscopy (GC/MS) for semi-volatile organic particles.
Attainment Designation Particle speciation measurements are performed to support the regulatory, analytical, and public health purposes of the program. There are no ambient air quality standards regarding the speciation of particles.
54
2006 Georgia Annual Air Quality Report
Acid Precipitation
Acid precipitation was monitored in four counties in 2006. The samples were collected weekly and were weighed and analyzed for acidity, conductivity, and selected compounds. The Air Protection Branch operated three of these sites and the Georgia Forestry Commission operated the fourth. There are no national or state standards for acid precipitation, but it is generally desirable for rain to have a relatively neutral pH. This is in contrast to the rain that falls in many regions of industrialized nations, which absorbs ions from air emissions that make it acidic (lower pH numbers). Most of the culprits of this acidification are sulfur and nitrogen compounds; the result is rain that contains excess acidity from sulfuric acid and nitric acid. The excess acidity in the rain causes damage to buildings and vehicles, and can acidify ponds and small lakes to the point of killing off all life in them. Georgia's Acid Rain monitoring network is shown in Figure 31.
55
Sequatchie
Marion
Hamilton
Dade
Catoosa
Walker
Whitfield
Murray
Hiawassee
Summerville
Floyd
Bartow
Pickens Cherokee
Dawson Dawsonville
Hall Forsyth
Haralson Carroll Heard
Paulding
Cobb
Gwinnett
Barrow
Douglas Coweta
Fulton
DeKalb
Clayton
Fayette
Henry
Spalding
Walton
Rockdale
Newton
Butts
Jasper
Madison
Clarke Oconee
Oglethorpe
Meriwether
Pike
Lamar
Monroe
Jones
Harris
Russell
Muscogee
Chattahoochee
Marion
Crawford
Bibb
Twiggs Houston
Columbia McDuffie
McDuffie
Terrell Lee
Dougherty
Worth
Figure 31: Acid Rain Monitoring Site Map
Acid Rain Sites Urban Areas MSAs Shown as Solid Colors
56
2006 Georgia Annual Air Quality Report
As in Figure 32, analysis of statewide average data back to 1982 indicates a small but significant trend toward pH neutrality. The rate of this increase is a pH gain of 0.007 per year over the period shown.
Average pH
Statewide Acid Rain Trends
5.9
5.7
5.5
5.3
5.1
4.9
4.7
4.5
4.3
4.1
1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Monitoring Year
Statewide Avg.
pH of pristine rain
Trendline
Figure 32: Acid Rain Trends, Statewide
57
pH 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Acid Rain, by Location
5.2
5
4.8 Dawsonville Hiawassee
4.6 McDuffie Summerville
4.4
4.2
4
Year
Figure 33: Acid Rain Trends, by Location Analysis of the same data set for individual areas (Figure 33) shows that while through the 1990s Summerville was the most acidic and McDuffie the most neutral, in recent years the difference between these areas has diminished and all of the sites have rainfall with roughly the same pH. Figure 34 depicts a comparison between the Department of Natural Resources (DNR) acid rain monitoring sites and the National Atmospheric Deposition Program/National Trends Network (NADP) annual average pH readings for 2006. As discussed above, Dawsonville, Hiawassee, McDuffie, and Summerville are in the DNR network. The sites that met the completeness criteria for 2005 in the NADP network are in Bellville, Chula, and Okefenokee. It appears that the annual averages for the NADP sites are slightly less acidic. These NADP sites are south of the Fall Line. The less acidic rain in the NADP monitoring areas is interesting. The difference may be related to South Georgia's greater proximity to sea breezes, which carry sodium ions that could affect the pH of rain in those areas.
58
2006 Georgia Annual Air Quality Report
p H
2006 pH for DNR and NADP Sites in Georgia
4.9 4.85
4.8 4.75
4.7 4.65
4.6 4.55
4.5 4.45 Dawsonville (DNR) Hiawassee (DNR)
McDuffie (DNR) Summerville (DNR)
Bellville (NADP)
Site
Chula (NADP)Okefenokee (NADP)
Figure 34: Comparison of DNR and NADP Acid Rain Averages
59
60
2006 Georgia Annual Air Quality Report
Photochemical Assessment Monitoring Stations (PAMS)
General Information Ozone is the most prevalent photochemical oxidant and an important contributor to smog. The understanding of the chemical processes in ozone formation and the specific understanding of the atmospheric mixture in various nonattainment areas nationwide was considered essential by EPA for solving the ozone nonattainment problems and developing a suitable strategy for solving those problems. In February 1993, the EPA revised the ambient air quality surveillance regulations in Title 40, Part 58 of the Code of Federal Regulations (40 CFR Part 58) to include provisions for enhanced monitoring of ozone, oxides of nitrogen, volatile organic compounds (VOCs), selected carbonyl compounds, and monitoring of meteorological parameters. The enhanced monitoring was to be implemented in cities that were classified as serious, severe, or extreme for ozone nonattainment. The classifications were based on the number of exceedances of the ozone standard, and the severity of those exceedances. Nineteen areas nationwide were required to implement a PAMS network. In the Atlanta metropolitan area, a network of four sites (Figure 35) was established beginning in 1993. The monitoring sites were selected based on the purpose of the monitors in light of the prevailing winds in the area. The Yorkville site serves as a rural background site, upwind of the city, which aids in determining the role of transport of pollutants into the Atlanta area. The South DeKalb and Tucker sites are the primary and secondary wind directions for an urban core-type site. These sites are expected to measure the highest precursor concentrations of NOx and VOCs in the Atlanta area. The Conyers site is the downwind site where titration of the precursors has occurred and the ozone concentrations should be at their highest.
61
Floyd
Bartow
Pickens Cherokee
Dawson
Hall Forsyth
Yorkville
Paulding
Haralson Carroll
Douglas
Heard
Coweta
Cobb Fulton Fayette
Gwinnett
Barrow
Tucker
DeKalb
Walton
S DeKalb
Clayton
Rockdale
Conyers
Newton
Henry
Spalding
Butts
Jasper Jasper
Meriwether
Figure 35: PAMS Monitoring Site Map
Pike Lamar
PAMS sites
Monroe
Urban Areas
MSAs Shown as Solid Colors Bibb
62
2006 Georgia Annual Air Quality Report An analysis of the PAMS data finds that the top ten volatile organic compounds (VOCs) for all sites contained the following species consistently across the years: isoprene (which is naturally released by vegetation), m/p xylene, toluene, propane, ethane, isopentane or isopentane/cyclopentane, n-butane and n-pentane. Propane, ethane, isopentane, n-butane, and n-pentane have a limited reactivity for ozone formation and therefore were the most prevalent of the volatile organic compounds measured. However, when the characterization of the top ten species is based upon contributions to ozone formation potential, we find that the list is slightly different. Isoprene, the tracer for VOC emissions from vegetation, is by far the largest contributor to ozone formation at every site. The anthropogenic compounds detected at all sites with the highest ozone formation potential were toluene, m/p xylene, propylene, ethylene, and isopentane. The sources for these five compounds are varied. All five compounds are emitted by mobile sources, with ethylene being an important tracer for vehicle emissions. Toluene, the most abundant species in urban air, m/p xylene, and isopentane also are emitted by solvent use and refinery activities. Isoprene is a 5 carbon organic compound released in large quantities by conifer trees. These trees are very abundant in the Southeastern United States contributing a significant portion to the overall carbon loading of the atmosphere in this region. Isoprene's chemical structure makes it a highly reactive substance with a short atmospheric lifetime and large ozone forming potential. Toluene reaches the air from a variety of sources such as combustion of fossil fuels and evaporative emissions. It has a substituted benzene ring possessing modest atmospheric reactivity. This hydrocarbon is in motor vehicle fuel and is also used a common solvent in many products such as paint.
63
Concentration (ppbC)
Isoprene Yearly Profile, 2003-2006
25 South DeKalb Tucker Yorkville
20 Conyers
15
10
5
0 JanuaryM-0a3rch-0M3 ay-S0J3eupltye-m0N3boevre-m03bJear-n0u3aryM-0a4rch-0M4 ay-S0J4eupltye-m0N4boevre-m04bJear-n0u4aryM-0a5rch-0M5 ay-S0J5eupltye-m0N5boevre-m05bJear-n0u5aryM-0a6rch-0M6 ay-S0J6eupltye-m0N6boevre-m06ber-06
Date
Figure 36: Isoprene Yearly Profile, 2003-2006
Figure 36 and Figure 37 compare the seasonal occurrence of these two compounds throughout the years of 2003 to 2006. These figures combine the 6-day, 24-hour data from the four PAMS sites, and concentrations are given in parts per billion Carbon (ppbC). Evidence of isoprene's natural origin is shown in Figure 36, where the ambient concentration is essentially non-existent from November to May. All four sites exhibit the seasonal cycle, with an occasional spike outside the consistent cycle. The site with the highest concentration of isoprene appears to vary year to year. One would expect to see Yorkville or Conyers with the higher levels, considering the rural area or semi-rural area in which these sites are located. This has been true for most years. As part of the seasonal cycle, in 2003, Conyers had the highest concentration with 16 ppbC, in 2004 Yorkville had the highest with 17 ppbC, in 2005, South DeKalb had the highest with 18 ppbC, and in 2006, Conyers had the highest concentration with 22 ppbC. Tucker had a spike of isoprene concentration (18 ppbC) in November of 2003, which did not fit the seasonal pattern seen for the majority of the four years of data. With the 2006 data, there appears to be a slight upward turn in the highest
64
2006 Georgia Annual Air Quality Report
seasonal levels of isoprene data. However, since there are a limited number of years worth of data at this point, it would be hard to discern a distinguishable trend.
Concentration (ppbC)
Toluene Yearly Profile, 2003-2006
50
45 South DeKalb
40
Tucker
35
Yorkville
Conyers
30
25
20
15
10
5
0 JanuaryM-0a3rch-0M3 ay-S0J3eupltye-m0N3boevre-m03bJear-n0u3aryM-0a4rch-0M4 ay-S0J4eupltye-m0N4boevre-m04bJear-n0u4aryM-0a5rch-0M5 ay-S0J5eupltye-m0N5boevre-m05bJear-n0u5aryM-0a6rch-0M6 ay-S0J6eupltye-m0N6boevre-m06ber-06
Date
Figure 37: Toluene Yearly Profile, 2003-2006
Toluene's atmospheric levels are more or less constant throughout the year, suggesting a constant level of emissions year-round (Figure 37). There is an occasional spike in concentration, but no evident high or low pattern for most of the four years of data. Overall, the PAMS sites that are situated in the urban areas (South DeKalb and Tucker) have slightly higher levels of toluene, while the sites located on the outskirts of the Atlanta metropolitan area (Yorkville and Conyers) show lower levels of toluene. Yorkville appears to have an upward swing throughout the 2006 data. As data is collected in the future, this site can be examined for a possible trend. The jaggedness of these graphs is an artifact of the sampling frequency.
65
The daily profile plots for toluene and isoprene found in Figure 38, produced using data gathered in the summer, show a constant background of toluene emissions with spikes of toluene resulting from morning and evening rush hour traffic. This graphs depicts the typical diurnal, or daily, profile for an urban area. During morning hours, when the nocturnal inversion has not yet broken, emissions become trapped within the boundary layer resulting, in a temporary increase in atmospheric concentration. Nighttime toluene levels are constant from midnight to 5:00 am. From 6:00 am to 7:00 am, enhanced vehicular activity releasing its emissions into an atmosphere with limited dispersing ability produces an increase in the ambient concentration. This behavior is typical of area source anthropogenic emissions with modest to long atmospheric lifetimes. Isoprene, on the other hand, exhibits very different behavior. At night emissions go to zero as photosynthesis ceases. At sunrise (about 6:00 am) concentrations begin to rise and continue to do so throughout the daylight hours. The vertical flux, or mass input per unit area, in the atmosphere of this substance is massive, being only slightly influenced by the enhanced mid-morning mixing. This effect can be seen at 9:00 am when slight drop in concentration occurs followed by a quick resumption in rise.
Typical Urban Diurnal Profile, Toluene & Isoprene
Concentration (ppbC)
12.0 10.0
8.0 6.0 4.0 2.0 0.0
0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time of Day
Figure 38: Toluene & Isoprene, Typical Urban Daily Profile
Toluene Isoprene
66
2006 Georgia Annual Air Quality Report
Carbonyl Compounds
Carbonyl compounds define a large group of substances, which include acetaldehyde, acrolein, and formaldehyde. These compounds can act as precursors to ozone formation. The majority of carbonyl compounds come from vehicle exhaust or the combustion of wood. Depending on the amount inhaled, exposure to these compounds can cause irritation to the eyes, ears, nose, and throat, dizziness, and damage to the lungs. Each of the seven carbonyls compounds that Georgia EPD monitors is discussed further in the following paragraphs. The Tucker site is part of the PAMS network and the South DeKalb site is part of both the PAMS network and the National Air Toxics Trends Stations (NATTS) network. Both of these sites sample every six days. Savannah, Dawsonville, and Brunswick are part of the Air Toxics Network and sample every twelve days. For a map of monitoring locations, see Figure 39.
Acrolein is primarily used as an intermediate in the manufacture of acrylic acid. It can be formed from the breakdown of certain organic pollutants in outdoor air, from forest fires and wildfires, as well as from vehicle exhaust. It is also found in cigarette smoke.
Acetaldehyde is mainly used as an intermediate in the production of other chemicals. Acetaldehyde is formed as a product of incomplete wood combustion (in fireplaces and woodstoves, forest fires, and wildfires), pulp and paper production, stationary internal combustion engines and turbines, vehicle exhaust, and wastewater processing.
Formaldehyde is used mainly to produce resins used in particleboard products and as an intermediate in the production of other chemicals. The major sources of emissions to the air are forest fires and wildfires, marshes, stationary internal combustion engines and turbines, pulp and paper plants, petroleum refineries, power plants, manufacturing facilities, incinerators, cigarette smoke, and vehicle exhaust.
Acetone is used industrially as a reactant with phenol to produce bisphenol A, which is an important component of polymers. It is used in nail polish removers, superglue removers, and as a drying agent. It is also used to dissolve plastic. Acetone is highly volatile and evaporates quickly. Inhalation of acetone can lead to liver damage.
Benzaldehyde is the simplest form of the aromatic aldehydes. It has an almond scent and is used in the food industry. It is also used as an industrial solvent, and is used in making pharmaceuticals, plastic additives, and aniline dyes. Liquid phase oxidation or chlorination of toluene can form benzaldehyde; so can a reaction between benzene and carbon monoxide. The combustion of gasoline, diesel fuel, wood burning, and incinerators emit benzaldehyde into the atmosphere.
Butyraldehyde is used in the manufacture of synthetic resins, solvents, and plasticizers. It is emitted into the air by combustion of gasoline, diesel fuel, and wood.
Propionaldehyde is a highly volatile compound that is produced or used in the making of propionic acid, plastics, rubber chemicals, alkyd resins, and is also used as a disinfectant and preservative. It is released into the atmosphere by combustion of gasoline, diesel fuel, wood, and polyethylene.
67
Whitfield Murray
Pickens
Bartow
Cherokee
Dawson
DawsonvilleHall
Forsyth
Paulding Cobb
Douglas
Fulton
Carroll
Barrow
Gwinnett
Tucker
De Kalb
Walton
S DeKalb
Rockdale
Clayton
Newton
Coweta
Fayette
Henry Spalding
Butts
Jasper
Madison
Clarke Oconee
Oglethorpe
Meriwether
Pike
Lamar
Monroe
Jones
Harris
Muscogee
Chattahoochee Russell
Marion
Bibb Crawford
Twiggs
Houston
Terrell
Lee
Dougherty Baker
Worth
Lanier
Brooks Lowndes
Echols
Edgefield
McDuffie
Columbia Richmond
Aiken
Burke
Long
Liberty
E President
McIntosh
Brantley
Glynn
Brunswick
Carbonyls Sites Urban Areas MSAs Shown as Solid Color
Figure 39: Carbonyls Monitoring Site Map
As can be seen in Figure 40 and Figure 41, when the average concentration of all carbonyls is compared with the total number of detections at each of the sampling sites, the carbonyl detections and concentrations tend to track each other directly. It should be noted that South DeKalb's average carbonyl concentration is about half that of Tucker's. This is unexpected 68
2006 Georgia Annual Air Quality Report
due to the relative proximity of the two sites, but may be a result of their positioning relative to local roads. The Tucker site is very close to a secondary road (few feet away), while South DeKalb farther from any roads (hundreds of feet away), is in an area dominated by the intersection of Interstates 20 and 285. It should be noted that the South DeKalb and Tucker sites collect data every six days, while Savannah, Dawsonville, and Brunswick collect data every twelve days. All else being equal, by taking twice as many samples at those locations, one would normally expect to see twice as many detections. To compare the data collected in 2005 and 2006 (Figure 40 and Figure 41), the data is almost exactly the same year to year. The Brunswick site shows a lower average concentration (18.4 g/m3 in 2005 to 10.5 g/m3 in 2006), but overall the number of detections and averages are roughly the same.
Total Number of Detections
Total Average Concentration (g/m3)
2005 Carbonyls, 24-Hour, All Species
35.0
30.0
Total Average
224
Concentration (g/m3)
200
25.0
Total Number of
Detections 20.0
15.0
89
10.0
75
5.0
0.0 Savannah
Dawsonville S. DeKalb Name of Site
Tucker
250
200
150
96
100
50
0 Brunswick
Figure 40: Average 24-Hour Carbonyls Concentration and Number of Detects, 2005
Total Number of Detections
Total Average Concentration (ug/m3)
2006 Carbonyls, 24-Hour, All Species
35.0
250
30.0
Total Average
220
207
200
25.0
Concentration (g/m3) Total Number of
20.0
Detections
150
15.0
10.0
72
81
5.0
90
100
50
0.0 Savannah
Dawsonville
S. DeKalb Name of Site
Tucker
0 Brunswick
Figure 41: Average 24-Hour Carbonyls Concentration and Number of Detects, 2006 69
The average concentration value of all 3-hour samples of carbonyls collected during the summer months (June, July, and August) has been combined for a given hour and is shown in Figure 42 and Figure 43 for the South DeKalb site and Figure 44 and Figure 45 for the Tucker site. The early morning ambient formaldehyde and acetone concentrations at the Tucker site are two times as large as at the South DeKalb site for 2005 and remain higher for 2006, however not twice as large. Acetaldehyde levels differ by a somewhat lesser amount between the two sites. The difference between sites becomes a bit reduced in the afternoon hours. A strong emissions source appears to be present in the vicinity of the Tucker site.
There are a few noticeable changes when comparing the 2005 and 2006 3-hour carbonyl data. The concentrations of acetone are higher in 2006 at both sites, sometimes reaching twice as high for the South DeKalb site. There was a slight increase in the formaldehyde and acetaldehyde concentrations. However, the concentrations of benzaldehyde are more than two times lower at both sites, sometimes reaching seven times lower in 2006. The concentrations of propionaldehyde and butyraldehyde overall show a decrease in 2006. There were no detections of acrolein in 2006, down from about 1 to 4 detections per hour and an average concentration of about 0.57 g/m3 at each hour in 2005. Acetyladehyde, acetone, and formaldehyde continue to be the biggest contributors, with the averages increasing from the 6:00 to 9:00, and again from the 9:00 to 12:00 hours, then decreasing at the 15:00 hour.
2005 South DeK alb Carbonyls, 3-H our
Average Concentration (g/m3)
11 10
9 8 7 6 5 4 3 2
1 0
6:00
9:00 Tim e
12:00
15:00
Form aldehyde A c e to n e A c e ta ld e h yd e B e n za ld e h y d e P ropionaldehyde
B utyraldehyde
A crolein
Figure 42: Average South DeKalb 3-Hour Carbonyls, June-August, 2005
70
2006 Georgia Annual Air Quality Report
2006 South D eK alb C arbonyls, 3-H our
Average Concentration (g/m3)
11 10
9 8 7 6 5 4 3 2 1 0
6:00
9:00 Tim e
12:00
15:00
Form aldehyde A c e to n e A c e ta ld e h y d e B e n za ld e h yd e P ro p io n a ld e h y d e
B u tyra ld e h yd e
A crolein
Figure 43: Average South DeKalb 3-Hour Carbonyls, June-August, 2006
71
2005 Tucker Carbonyls, 3-hour
Average Concentration (g/m3)
11 10
9 8 7 6 5 4 3
2 1 0
6:00 9:00 Tim e
12:00 15:00
Form aldehyde A c e to n e A cetaldehyde B enzaldehyde P ropionaldehyde
B u ty ra ld e h y d e
A crolein
Figure 44: Average Tucker 3-Hour Carbonyls, June-August, 2005
2006 Tucker Carbonyls, 3-hour
Average Concentration (g/m3)
11 10
9 8 7 6 5 4 3 2
1 0
6:00
9:00 Tim e
12:00 15:00
Form aldehyde Acetone A c e ta ld e h yd e B e n z a ld e h y d e P ro p io n a ld e h yd e
B u ty ra ld e h y d e
A c ro le in
Figure 45: Average Tucker 3-Hour Carbonyls, June-August, 2006
Figure 46 and Figure 47 shows the seven (7) species in the analyte group according to their statewide annual abundance, based on number of detections and average concentration. A gradient is evident from this graph, with formaldehyde as the most ubiquitous carbonyl. For 72
2006 Georgia Annual Air Quality Report
the most part, it appears that the number of detections track the average concentration. Acetaldehyde does not follow this pattern, with more detections per concentration. The 2005 and 2006 carbonyl data seem to follow the same pattern. The proportion of each compound remained about the same from 2005 to 2006, with the biggest contributors (formaldeyhde, acetone, and acetaldehyde) remaining the same. One noticeable change is the number of benzaldehyde detections. In 2006, the number of benzaldehyde detections decreased from 50 in 2005 to 17, however the average concentration remained about the same, 4.6 g/m3 in 2005 to 3.6 g/m3 in 2006. This would indicate that although there were fewer detections, each detection would have carried a little bit more weight per detection in 2006.
Total Average Concentration
(g/m3)
2005 Carbonyls, 24-Hour, All Sites
Total Number of Detections
35.0
30.0
182
170
25.0
200 170
150
20.0 15.0
100
10.0
67
41
50 50
5.0 0.0
4
0
FormaldehydeAcetaldehPyrdoepionaldehydBeutyraldehyde
Acrolein
Name of Carbonyls Species
AcetonBeenzaldehyde
Total Average Concentration (g/m3) Total Number of Detections
Figure 46: Average 24-Hour Carbonyls Concentration vs. Number of Detects, by Species, 2005
Total Average Concentration
(g/m3)
Total Number of Detections
2006 Carbonyls, 24-Hour, All Sites
35.0 30.0
190
174
25.0
192
200
150
20.0 15.0
100
10.0 5.0 0.0
61 36 0
50 17 0
FormaldehydeAcetaldehPyrdoepionaldehydBeutyraldehyde
Acrolein
Name of Carbonyls Species
AcetonBeenzaldehyde Total Average Concentration (ug/m3) Total Number of Detections
Figure 47: Average 24-Hour Carbonyls Concentration vs. Number of Detects, by Species, 2006
73
Measurement Techniques A number of methods are used to conduct the PAMS hydrocarbon portion of the analyses. Throughout the year, 24-hour integrated hydrocarbon samples are taken and analyzed in the EPD laboratory for 56 hydrocarbon compounds. A 24-hour integrated carbonyl sample is taken once every sixth day throughout the year and analyzed. During June, July, and August, four integrated three-hour carbonyl samples are taken every third day. All analyses are conducted at the EPD Laboratory. During June, July, and August, hydrocarbon samples are analyzed hourly on-site using a gas chromatography unit with a Flame Ionization Detector (FID). The gas chromatograph produces analyses of the ambient air for the same 56 hydrocarbons. Specific annual summaries for 2006 may be found in Appendix D.
Attainment Designation There are no specific ambient air standards for the hydrocarbon and aldehyde species measured. PAMS measurements are performed to support the regulatory, analytical, and public health purposes of the program. By performing these measurements, we can better serve two major goals. First, by studying local atmospheric chemistry, we improve our ability to control the formation of secondary pollutants like ozone and particulate matter. Second, we are monitoring the concentration of pollutants (aside from the defined criteria air pollutants) expected to be harmful to human health, but are not well enough understood to be regulated. By making such data available, scientists who study human health as it relates to air quality can study how these pollutants may affect our health. When this understanding is further refined, their data can serve to guide policymakers toward making decisions that protect public health.
74
2006 Georgia Annual Air Quality Report
Air Toxics Monitoring
General Information The citizens of Georgia have demonstrated a long-term interest in the quality of Georgia's air. Since the 1970's ambient ozone concentrations have been monitored in several communities throughout the state. As the state's population has grown, more compounds have been monitored in ambient air as required by the Federal Clean Air Act. In 1993 the EPD began to monitor a number of compounds that have no established ambient air standard. The monitoring has been conducted under two efforts, the first being the previously discussed Photochemical Assessment Monitoring Station (PAMS) project, a federally mandated program for areas in serious, severe, or extreme non-attainment of the ozone standard. The second effort is the EPD-sponsored monitoring activities for ambient concentration of hazardous air pollutants (HAPs). That effort was undertaken since monitoring only criteria pollutants would not provide an adequate understanding of the quality of Georgia's air.
In 1994, the EPD conducted an intensive air quality study in Savannah (GADNR, 1996a). In 1996, the EPD conducted an intensive study in Glynn County as part of a multimedia event with EPA (GADNR, 1996b). These studies provided detailed pictures of the air quality in the communities, but the studies were not long-term studies and could not provide information on seasonal variation or trends. A reassessment of the toxic monitoring program occurred, and in 1996 the EPD embarked on an ambitious project of establishing a statewide hazardous air pollutant-monitoring network. The network was not designed to monitor any one particular industry, but to provide information concerning trend, seasonal variation, and rural versus urban ambient concentration of air toxics. To evaluate the rural air quality, two background sites were proposed: one in North Georgia and one in South Georgia. The majority of the other sites were located in areas with documented emissions to the atmosphere of HAPs exceeding one million (1,000,000) pounds per year as indicated by the 1991 Toxic Release Inventory (GADNR, 1993).
By 2002 the Air Toxics Network consisted of fifteen (15) sites statewide, including a collocated site at Utoy Creek and South DeKalb's National Air Toxics Trends site, monitoring for a common set of toxic compounds. The compounds selected for monitoring are from the list of 188 metals and other compounds identified by EPA as being HAPs. These air pollutants can have effects on human health, ranging from causing headaches, nausea, and dizziness to causing cancer, birth defects, problems breathing, and other serious illnesses. These effects can vary depending on how often one is exposed, how long one is exposed, the health of the person that is exposed, and the toxicity of the compound. These air pollutants also affect the environment. Wildlife experiences symptoms similar to those in humans. Pollutants accumulate in the food chain. Many air pollutants can also be absorbed into waterways and have toxic effects on aquatic wildlife. Some of the substances tend to have only one critical effect, while others may have several. Some of the effects may occur after a short exposure and others appear after long-term exposure or many years after being exposed. Exposure is not only through direct inhalation of the pollutant, but also through the consumption of organisms such as fish that have absorbed the pollutant.
Air toxic compounds are released from many different sources, including mobile sources (such as vehicles), stationary industrial sources, small area sources, indoor sources (such as cleaning materials), and other environmental sources (such as wildfires). The lifetime,
75
transportation, and make-up of these pollutants are affected by both weather (rain and wind) and landscape (mountains and valleys). They can be transported far away from the original source, or be caught in rain and brought down to waterways or land. The following section discusses air toxic compounds, possible sources, monitoring techniques, findings for 2006 and a comparison of 2006 data to previous years. In 2004, the Air Toxics Network underwent changes to the detection limits and reporting limits of the chemicals in this network. Each of the new limits is discussed more thoroughly in the following respective sections. Lowering the limit of detection helps the data better represent reality. Instead of only seeing the higher numbers that were detected and using those numbers for average concentrations, one is able to see both sides of the spectrum and have a truer average for each chemical. Also, including the lower concentrations for each chemical allows for a better understanding of what levels can cause chronic health problems. Seeing only the higher levels of concentration, or possibly spikes, only yields data useful for identifying acute health effects. However, with the lower concentration levels included in the data, there can be further assessment of potential chronic health effects. With the lower limits included in the data, one is able to see all possible effects of the chemicals analyzed.
76
2006 Georgia Annual Air Quality Report
Metals
The metals subcategory includes antimony, arsenic, beryllium, cadmium, chromium, cobalt, lead, manganese, nickel, selenium, and zinc.
Antimony is used as a hardener in lead for storage batteries, in matches, as an alloy in internal combustion engines, and in linotype printing machines. Antimony compounds are used in making materials flame-retardant, and in making glass, ceramic enamels, and paints. Forms of the antimony metal are also used in medicines, and can be found in gasoline and diesel exhaust.
Arsenic occurs naturally at trace levels in soil and water. Most people are not exposed to arsenic through air pollution, but it can be found in food. The arsenic found in air comes mainly from the burning of coal or fuel oil, from metal smelters or iron foundries, and from the burning of waste.
Beryllium is a lightweight and rigid metal and used in watch springs, computer equipment, and used in the production of beryllium-copper as an alloying agent. This strong alloy is used to conduct heat and electricity, in spot welding, electrical contacts, and high-speed aircraft. Until 1949, beryllium was used in fluorescent lighting, until it was determined to have caused berylliosis, a disease that primarily affects the respiratory system and skin. Beryllium in ambient air is mainly a result of the burning of coal or fuel oil.
Cadmium emissions, like beryllium and arsenic, are mainly from the burning of fossil fuels such as coal or oil. The incineration of municipal waste and the operation of zinc, lead, or copper smelters also release cadmium to the air. For nonsmokers, food is generally the largest source of cadmium exposure.
Chromium sources include the combustion of coal and oil, electroplating, vehicle exhaust, iron and steel plants, and metal smelters. The emissions from these sources are a combination of elemental chromium and compounds including chromium ions. The most toxic form is hexavalent chromium.
Cobalt is used as a pigment (blue and green coloring agent), as a drying agent for paints, inks and varnishes, and as a catalyst for the petroleum and chemical industries. It is used as an alloy for parts in turbine aircraft engines, corrosion-resistant alloys, magnets, battery electrodes, and steel-belted tires. Cobalt also has a medicinal use as a radioactive metal in radiotherapy. It is also found in gasoline and diesel exhaust. Cobalt is actually necessary to many forms of life, when ingested through the digestive tract, in small amounts, as a micronutrient. It is a central component of vitamin B-12. As with most micronutrients, however, human activity can cause it to accumulate in unnatural locations or in unnatural concentrations. In those cases, it may be harmful and is considered a pollutant.
Lead is used in the manufacturing of batteries. The largest source of lead in the atmosphere used to be from the combustion of leaded gasoline. With the elimination of lead from gasoline, lead levels in the air have decreased considerably. Other sources of lead emissions include combustion of solid waste, coal, oils, emissions from iron and steel production, and lead smelters. Exposure to lead can also occur from food and soil. Children
77
are at particular risk to lead exposure, because they commonly put hands, toys, and other items in their mouths that may come in contact with lead-containing dust and dirt. Leadbased paints were commonly used for many years. Flaking paint, paint chips, and weathered paint powder may be a major source of lead exposure, particularly for children. Manganese is a naturally occurring substance found in many types of rock and soil; it is ubiquitous in the environment and found in low levels in water, air, soil, and food. Manganese can also be released into the air by combustion of coal, oil, wood, the operation of iron and steel production plants. Nickel is found in the air as a result of oil and coal combustion, residential heating, nickel metal refining, lead smelting, sewage sludge incineration, manufacturing facilities, mobile sources, and other sources. Selenium is a byproduct of mining and smelting sulfide ores, such as silver, copper, and pyrite. It is found in soils, and can also be released by burning coal. Selenium has photovoltaic and photoconductive properties and is therefore used in photocells and solar panels. It is used as a pigment (red coloring agent) in enamels and glass. It is also used as a toner in photographs and in photocopying. Selenium is also found in gasoline and diesel exhaust. Selenium is a micronutrient, needed at very low levels for the health of all living creatures. It is normally absorbed through the digestive tract, though, and is not desirable in the air. Zinc is found in gasoline and diesel exhaust. It is used to prevent corrosion of galvanized steel. It is also used in die-casting, and as part of battery containers. Zinc has been used as the primary metal in making the U.S. penny since 1982. Zinc compounds are used in making white pigment, sunscreen, deodorant, calamine lotions, and pigments for glow in the dark items. It is also used in the rubber industry. Like selenium, zinc is also a micronutrient needed for the health of living beings when consumed through the digestive system. When found in the air, though, it may be considered a pollutant. See Figure 48 for a map of monitoring locations for metals.
78
Sequatchie Hamilton
2006 Georgia Annual Air Quality Report
Catoosa Walker
Whitfield Murray
Rome
Floyd Bartow
Pickens Cherokee
Dawson
Gainesville DawsonvilleHall
Forsyth
Madison
Yorkville Paulding Cobb
Gwinnett
Barrow
Haralson
Douglas
Utoy Creek SDDe KeaKlb alb
Walton
Fulton
Rockdale
Carroll
Clayton
Newton
Heard
Coweta
Fayette
Henry
Spalding
Butts
Jasper
Clarke Oconee
Oglethorpe
Milledgeville
Harris
Pike Meriwether
Muscogee
Columbus Univ
Lamar Monroe
Crawford
Bibb
Jones
GA Forestry
Twiggs
Warner Robins
Houston
Russell
Chattahoochee Marion
Edgefield
Columbia McDuffie
Aiken
Augusta
Richmond
Burke
E President
Terrell
Lee
Dougherty Baker
Worth
Brooks
General Coffee
Long
Liberty
McIntosh
Lanier
Lowndes
Valdosta
Echols
Brantley
Glynn Brunswick
Metals Urban Areas MSAs Shown as Solid Colors
Figure 48: Metals Monitoring Site Map
79
Total Detects
Total Detections of Metals Per Site, 2004-2006
550 500 450 400 350 300 250 200 150 100
50 0
AugustaBrunGswenicekral CoffeeColumbDusawsonviGlleainesville* MacMonilledgeville
RomeSavaSnonuathh DeKalbU*toy Creek ValWdoasrtnaer Robins Yorkville 2004
2005
Site
2006
*The South DeKalb sampler runs twice as often as the other sites. Gainesville runs one extra sample a month.
Figure 49: Total Detections of Metals, by Site, 2004-2006
Figure 49 shows the total number of metal species detected at each site for the years 2004 to
2006, including the National Air Toxics Trends Site located at South DeKalb. This site was established January 1st, 2003. Unlike the rest of the network measuring metals, the
equipment at South DeKalb collects only the PM10 fraction of the total suspended solids. The South DeKalb site also has data collected every six days, as opposed to every twelve days at
the other sites. It should also be noted that that Gainesville site has one extra sampling
collection a month. Therefore, it is understandable that these sites have the highest number
of detections. Without these sites included in the graph, the distribution across the sites would be relatively similar. Lower limits of detection (LOD)5 were introduced in September of
2004, resulting in an increase in the number of observations. While this only represented one
5 Limits of detection (LOD) were reduced according to the table below. In addition, all numbers, including those occurring below the designated detection limit, are being reported and included in the data. Values are expressed in g/filter.
Element
Antimony Arsenic Beryllium Cadmium Chromium Cobalt Manganese Nickel Lead Selenium Zinc
Old LOD
1.2 6 1.2 1.2 6 6 1.2 6 1.2 6 6
New LOD
0.2 0.9 0.1 0.05 0.7 0.2 0.3 0.4 0.2 0.2 0.7
80
2006 Georgia Annual Air Quality Report third of a year for 2004, the following graphs include the 2004 data with 2005 and 2006. There has been only two full years of data collected at the lower limits, therefore trends will not be discernible at this time. As more data is collected with consistent lower limits of detection, hopefully trends will be visible. The distribution of metals at the various locations across the state can be examined in Figure 49, as well any changes in the past two years. In 2004 and 2005, Valdosta consistently had fewer detections than most other sites. However in 2006, the number of detections at Valdosta increased to around the same number as the other sites (around 300). Otherwise, there were not dramatic changes from 2005 to 2006. There was a slight decrease in number of detections at Milledgeville and Dawsonville, but most sites had a slight increase or remained the same. The variability across the various sampling locations is modest, considering the vast geographic distribution of the sites, and climatological and anthropogenic influences from nearby urban development.
81
Figure 50 and Figure 51 show the network's frequency of detection and total average concentration by metallic species at all Air Toxics sites during 2004 through 2006. One point of interest when looking at data is to track the number of detections along with the concentration. When examining this aspect, it appears that for most of the metals there are several detections. Therefore, each metal detection contributes little concentration to the overall total concentration. This does not seem to be the case for zinc. While there were about the same number of detections as the other metals, zinc had the highest average concentration for all three years. This would indicate that for each zinc detection, there was a higher concentration of that metal. Some metals including zinc, nickel, antimony, lead, chromium, and cadmium have been associated with emissions from tires and brake linings. The use of vehicles on Georgia's roads could be a reason for higher levels associated with some of these metals. With the concentrations of zinc being much higher than the other metals, zinc is explored further in Figure 52.
82
2006 Georgia Annual Air Quality Report
Total Detections of Metals, 2004-2006
Number of Detects
500
450
400
350
300
250
200
150
100
50
0
Antimony*
Arsenic Berrylium Cadmium Chromium
Cobalt
Lead Manganese
Nickel Selenium
Zinc
2004
Name of Metal
2005
2006
*Antimony was added to the list of metals starting with the 2005 sampling year.
Figure 50: Total Detections of Metals, by Species, 2004-2006
Total Average Concentration (ug/m3)
Total Average Concentration of Metals, 2004-2006
0.4 0.35
0.3 0.25
0.2 0.15
0.1 0.05
0 Antimony*
Arsenic
Berrylium
Cadmium Chromium
Cobalt
Lead Manganese
Name of Metal
Nickel
Selenium
*Antimony was added to the list of metals starting with the 2005 sampling year.
Figure 51: Total Average Concentration of Metals, 2004-2006
Zinc 2004 2005 2006
83
With Figure 52, the total average concentrations of zinc are investigated more closely, divided by site, for 2004 through 2006. Several sites had a consistent level of zinc through the three years, and a few sites even had a decrease in zinc levels with the 2006 data. The General Coffee, Columbus, Gainesville, and Valdosta sites, however, had a noticeable increase in zinc levels from 2005 to 2006. To look at the overall levels of zinc, the Utoy Creek site has consistently had the highest concentration from 2004 to 2006, with levels almost three times as high as the lowest concentrations, which were at the South DeKalb site. It is surprising to see the South DeKalb site, which is situated in the metro Atlanta area, display lower concentrations. The Utoy Creek site is situated at the Utoy Creek Wastewater Treatment Facility. With the particle emissions coming from tires and brake linings, these particles can be washed into waterways. It is a possibility that these emissions are then coming through the wastewater treatment plant, causing the higher numbers to be seen at the Utoy Creek site. Zinc can be used to help keep galvanized steel from corroding, and is possibly used for this reason on the pipes at the wastewater treatment facility. This could be another reason for the higher levels of zinc at the Utoy Creek site.
0.060
Yearly Average Comparison of Zinc, 2004-2006
0.050
Average Concentration (ug/m3)
0.040
0.030
0.020
0.010
0.000 Augusta BrunswGicekneral Coffee ColumbusDawsonville Gainesville
MaconMilledgeville
Rome South
DeKalb
SavannahUtoy Creek
Name of Site
ValdoWstaarner Robins
Yorkville
2004 2005
2006
Figure 52: Yearly Average Comparison of Zinc, by Site, 2004-2006
84
2006 Georgia Annual Air Quality Report
Hexavalent Chromium
Hexavalent chromium (chromium in its +6 oxidation state) in the environment is almost always related to human activity. Hexavalent chromium can be released into the atmosphere through the production of stainless steel, chrome plating, coating processes, and painting. It is also found in vehicle engines. The presence of chromium compounds is common at hazardous waste sites. From locations such as these, exposure of populations residing or working nearby can occur through exposure to air containing particulates or mists of chromium compounds. These particles can also find their way into drinking water if soluble forms of chromium leach into groundwater. Human exposure can also occur through skin contact with soil at hazardous waste sites.
Hexavalent chromium is absorbed most readily thorough the lungs or digestive tract. Other forms of the metal, such as chromium in the +3 oxidation state, occur naturally in the environment and are not as efficient at entering the body. In general, hexavalent chromium compounds are more toxic than other chromium compounds. The toxicity of hexavalent chromium is in part due to the generation of free radicals formed when biological systems reduce hexavalent chromium to the +3 oxidation state.
Effects in humans exposed occupationally to high levels of chromium or its compounds, primarily hexavalent chromium, by inhalation may include nasal septum ulceration and perforation, and other irritating respiratory effects. Cardiovascular effects, gastrointestinal and hematological effects, liver and kidney effects, and increased risks of death from lung cancer may also result from such exposure. In addition to the respiratory effects, exposure to chromium compounds can be associated with allergic responses (e.g., asthma and dermatitis) in sensitized individuals.
Hexavalent chromium dioxide is a tetravalent chromium compound with limited industrial application. It is used to make magnetic tape, as a catalyst in chemical reactions, and in ceramics. Because of its limited industrial uses, the potential for human exposure is less for chromium dioxide than for the more industrially important hexavalent chromium and chromium +3 compounds.
85
This is the second year hexavalent chromium has been monitored at our South DeKalb site. The data for 2005 and 2006 is presented in Figure 53. Observed concentrations range over an order of magnitude, from 0.01 to 0.3 ng/m3 (nanograms per cubic meter). The observed concentrations are represented with the points, while the black line represents a moving average across the data set. At this point, true trends are hard to define, but there may be a suggestion of a seasonal trend with lower concentrations in the winter months. As the data set grows, possible seasonal variation in its concentration, the magnitude of its health risk, and which wind directions are most associated with elevated concentrations will be investigated.
Hexavalent Chromium, 2005-2006
0.35
0.30 0.25
Concentration Moving Average
0.20
Concentration (ng/m3)
0.15
0.10 0.05
0.00 02/05 03/05 04/05 05/05 06/05 07/05 08/05 09/05 10/05 11/05 12/05 01/06 02/06 03/06 04/06 05/06 06/06 07/06 08/06 09/06 10/06 11/06 12/06 Date
Figure 53: Hexavalent Chromium at South DeKalb
86
2006 Georgia Annual Air Quality Report
Volatile Organic Compounds (TO-14/15)
Chlorinated compounds are very stable in the atmosphere, with lifetimes of several years. Dichlorodifluoromethane was the refrigerant of choice for automotive cooling. This material has not been manufactured since the mid-1990s (cars now use R-134a), yet it remains prevalent in the environment. Chloromethane is a volatile industrial solvent. Toluene is major component of paints, solvents and is also present in gasoline. It reaches the atmosphere by way of evaporative emissions as well as incomplete combustion processes. Carbon tetrachloride and the Freons are generally used as refrigerants, industrial solvents, and as fire suppressants (though generally known as Halon in that application). The atmospheric reactivity of aromatic compounds is relatively high, with lifetimes in the weeks to months range. Except for the chlorinated compound, all others are byproducts of burning gasoline.
Figure 54 shows the statewide detection distribution of toxic (TO-14/15) type volatile organic compounds (VOCs) in 2004, 2005, and 2006 across the Air Toxics Network. Again, the South DeKalb site has samples collected every six days, and Gainesville has an extra monthly sampling, compared to the other sites which have samples collected every twelve days. If South DeKalb is excluded, the distribution is relatively even across the state, with the more urban or industrial sites near the upper extreme and the more rural sites near the lower extreme. With lower limits of detection6 beginning in September 2004, the number of detections increased from 2004 to 2005. With two full years of data (2005 and 2006) at the same detection limit, the number of detections are relatively the same. As more data is collected in the future, with a consistent limit of detection, hopefully a trend will be apparent.
Number of Detects
Total Volatile Organic Compunds (TO-14/15) Detected Per Site, 2004-2006
400 350 300 250 200 150 100
50 0 Augusta BrunswGicekneral Coffee ColumbusDawsonville Gainesville
MaconMilledgeville
Rome SavannaShouth DeKalb Utoy Creek
Name of Site
ValdoWstaarner Robins
Yorkville
2004
2005
2006
Figure 54: Total Volatile Organic Compounds Detected per Site, 2004-2006
6 Detection limits for this analyte group, TO-14/15 Toxic VOCs, were halved beginning in September 2004. The old detection limit was 0.5 ppbv for all species except methylene chloride, which was 5 ppbv. The new detection limits are 0.25 ppbv for all species except methylene chloride, which is at 2.5 ppbv.
87
Number of Detections
Volatile Organic Compounds (TO-14/15), 2004-2006 500 450 400 350 300 250 200 150 100
50 0
TDCrihicclhohlrlooorrmoodfeliutfhlouaroonrmeometehtahnaen(eFrmeo/pnMX1e1yth)leyTnl oeclh(umleo/nrpoe-fDBoriemmne(z1teh,n1ye,l1b-eTnrzicehnleo)roeth1a,C2on,y-e4c)X-lToyrhliemenxeeat(hnoye-lDbiemneztehnyelbenzEetnhTey)elbtreanczhelonreoethylenCearbSotynretenterachloridCehlor222of000o000rm456 Name of Compound
Figure 55: Number of Volatile Organic Compounds (TO-14/15) Detected, Select Compounds, 20042006
Total Average Concentration for Select VOCs 40 35 30 25 20 15 10
5 0 TDrCicichhhlloolorrrooomfldueioflturhooamrnoeemtheathnaen(mFerM/epoeXnthy1ylel1nc)eTh(loomlru/oepfn-oDerBmime(n1ez,t1he,yn1leb-Terniczhelnoero) e1t,oh2C-a,4Xny-ceyTl)loreihmneeex(toah-nyDelbimeneztheynlebenEzetThneyetl)braecnhzleonroeethylCeanrebSotnyrteentreachloriCdheloro222f000o000rm456
Name of Compound
Figure 56: Total Average Concentration of Select Volatile Organic Compounds (TO14/15), 20042006 88
Total Average Concentration (g/m3)
2006 Georgia Annual Air Quality Report
When looking at the make up of samples collected, one relationship to consider is that between the concentrations observed compared to the number of detections of that compound. Figure 55 and Figure 56 compare this relationship, with Figure 55 showing the top fifteen compounds of the VOCs group that were detected for 2004-2006 and Figure 56 showing the corresponding compound's total average concentration. Although there are 42 species in this analyte group, only a relatively smaller subset is typically detected with any regularity. The number of detections was derived using any detection that was above half of the limit of detection. To obtain the average concentration for compounds with at least one detection, the half limit of detection for that compound was substituted for any number lower than that compound's half limit of detection. Chloromethane consistently had one of the highest detection rates, but the total average concentration was the third or fourth highest over the three years. This would indicate that the concentrations of chloromethane are relatively low per detection. Conversely, toluene had the third or fourth highest detection rate, but one of the top average concentrations for 2004 through 2006. This would indicate that each detection of toluene has a relatively high concentration compared to the other VOCs.
Concentration (ppb)
Total Volatile Organic Compound (TO-14/15) Loading by Site, 2004-2006
180
2004 160
2005
140
2006
120
100
80
60
40
20
0 Augusta BrunswGicekneral Coffee ColumbusDawsonville Gainesville
MaconMilledgeville
Rome SavannaShouth DeKalbUtoy Creek
ValdoWstaarner Robins
Yorkville
Site
Figure 57: Total Volatile Organic Compound Loading all Species, by Site, 2004-2006
Figure 57 shows the total volatile organic compound concentration, or loading, at each site for 2004 through 2006. This "total loading" measurement is produced by adding up all the detected concentrations of all VOCs, even those below half of the detection limit as discussed earlier. It is intended as a surrogate measure showing general trends in overall VOC concentrations. When considering Figure 57, it is important to note that the South DeKalb and Gainesville sites would appear elevated since these two sites have a larger number of scheduled samplings than the rest of the sites in the network. South DeKalb samples on a 6-day schedule, and Gainesville has an additional sample collected per month
89
over the network's regular sample days. VOC levels at sites located close to or within urban centers (South DeKalb, Utoy Creek) show higher levels of these pollutants, while sites in smaller communities or rural areas (General Coffee, Dawsonville, and Yorkville) show lower levels. There is an obvious increase from 2004 to 2005 since the detection limits were changed only for the last quarter of 2004, in September. When adding the 2006 data to the graph, most of the data did not show dramatic differences, but there are a couple notable changes from the 2005 data. The Milledgeville site had almost a twofold decrease in VOCs concentration from 2005 to 2006. The Augusta site shows an increase of almost double VOCs concentration from 2005 to 2006. Referring to Figure 54, the Augusta site had the highest number of detections after South DeKalb and Gainesville. The Augusta site is located in an urban industrialized area, near an interstate. As more data is collected, these sites will be examined for possible trends. To look at the VOCs loading concentrations in a different way, Figure 58 shows the seasonal concentration of all volatile organic compounds (TO-14/15) throughout the Air Toxics Network for 2004 through 2006. The limits of detection were lowered in September of 2004, causing the last quarter of data to be higher for that year. With a consistent limit of detection for the 2005 data, the higher concentration is found in the third quarter. It is somewhat surprising to see the highest concentration in the third quarter of 2005, since these compounds are usually degraded in chemical activity in the warmer months. For the 2006 data, the pattern of lower concentration is seen in the third quarter. In 2006, the second and fourth quarters are also almost identical, with the fourth quarter showing the highest concentration of the year, by a slight measure. However, overall, with the data represented at this point, each year has varied. It is hard to suggest a trend when there is only two full years worth of data under the newer detection limits. As more data is collected in the coming years, trends may be seen.
90
Concentration (ppb)
2006 Georgia Annual Air Quality Report
Volatile Organic Compounds (TO-14/15) Seasonal Observations Total Loading
Statew ide, 2004-2006
400 350 300 250
200 150 100
50 0 1st
2nd 3rd
Quarter
2006 2005 2004 4th
Figure 58: Volatile Organic Compounds, Seasonal Effects, 2004-2006 For a map of VOC and SVOC monitoring locations, see Figure 59.
91
Sequatchie Hamilton
Catoosa Whitfield Murray
Walker
Floyd Bartow
Rome
Pickens Cherokee
Dawson
DawsonvilleHall Gainesville
Forsyth
Madison
Yorkville Paulding Cobb
Gwinnett
Barrow
Haralson
Douglas
Utoy Creek De Kalb
Walton
Fulton
{
Rockdale
Carroll
S DeKalb
Clayton
Newton
Heard
Coweta
Fayette
Henry
Spalding
Butts
Jasper
Clarke Oconee
Oglethorpe
Milledgeville
Harris
Pike Meriwether
Muscogee
Columbus Univ
Lamar Monroe
Crawford
Bibb
Jones
GA Forestry
Twiggs
Warner Robins
Houston
Chattahoochee Russell
Marion
Edgefield
Columbia McDuffie
Aiken
Augusta
Richmond
Burke
E President
Terrell
Lee
Dougherty
Worth
General Coffee
Long
Liberty
McIntosh
Baker
Brooks
SVOC / VOC Sites { VOC Site
Lanier
Lowndes
Valdosta
Echols
Brantley
Glynn Brunswick
Urban Areas
MSAs Shown as Solid Colors
Figure 59: VOC and SVOC Monitoring Site Map
92
2006 Georgia Annual Air Quality Report
Semi-Volatile Organic Compounds
Polycyclic aromatic hydrocarbons, also called semi-volatile organic compounds (SVOC) are chemical compounds that consist of fused, six-carbon aromatic rings. They are formed by incomplete combustion of carbon-containing fuels such as wood, coal, diesel fuel, fat or tobacco. Over 100 different chemicals are comprised within this designation. Many of them are known or suspected carcinogens. Some environmental facts about this class of compounds are mentioned below.
PAHs enter the air mostly as releases from volcanoes, forest fires, burning coal, and automobile exhaust.
PAHs can occur in air attached to dust particles. Some PAH particles can readily evaporate into the air from soil or surface waters. PAHs can break down by reacting with sunlight and other chemicals in the air over a
period of days to weeks. PAHs can enter water through discharges from industrial and wastewater treatment
plants. Most PAHs do not dissolve easily in water. They stick to solid particles and settle to
the bottoms of lakes or rivers. Microorganisms can break down PAHs in the soil or water after a period of weeks to
months. In soils, PAHs are most likely to stick tightly to particles. Certain PAHs move through
soil to contaminate groundwater. PAH content of plants and animals may be much higher than the PAH content of the
soil or water in which they live. For a map of SVOC monitoring locations, see Figure 59.
93
Number of Detections
Total Semi-Volatile Compounds Detected, Per Site, 2004-2006
6
5
4
3
2
1
0
AugustaBrunGsweincekral CoffeeColumbuDsawsonvillGe ainesville
MacMonilledgeville
RomeSavannaUhtoy Creek Valdoasrtnaer Robins Yorkville
W
2004
Name of Site
2005
2006
Figure 60: Total Semi-Volatile Organic Compound Detections Per Site, 2004-2006
Figure 60 displays the number of detections according to site for 2004 through 2006 for all semi-volatile organic compounds combined in the Air Toxics Network. Detections were counted for any number that was above half of the limit of detection. As can be seen from this graph, the semi-volatile organic compounds are detected much less frequently than the other groups of compounds in the Air Toxics Network. Of the three years represented in this graph, 2005 had the most detections.
Figure 61 and Figure 62 show the total number of times each of the semi-volatile organic compounds were detected and the total average concentration of those compounds across the statewide network of sites from 2004 through 2006. The number of detections was derived using any detection that was above half of the limit of detection. To obtain the average concentration for compounds with at least one detection, the half limit of detection for that compound was substituted for any number lower than that compound's half limit of detection. Even with the few detections and low concentrations, the relationship between these two measures can still be examined. With the 2005 data, the compound with the highest number of detections was benzo(g,h,i)perylene, but this compound had one of the lowest detectable concentrations of the group. This would show that each detection of benzo(g,h,i)perylene had a low concentration. The compound with the highest average concentration was phenanthrene, but this compound had one of the lowest detection rates. In 2006, phenanthrene again had the highest average concentration, though at about half the 2005 concentration level, and had the fewest detections of the compounds detected. This would mean that for the each detection of phenanthrene, there was a higher concentration.
94
2006 Georgia Annual Air Quality Report
Number of Detections
Total Detections of Semi-Volatile Organic Compounds, 2004-2006
18 16 14 2004
2005 12 2006 10
8 6 4 2 0 NaphtAhcaelennaepAhctheennpehthylenFeluPohreennaenthrAenntehraFclueonreanthene PyBreennCzehorB(yaes)naeznnoteh(brBa)efclnuezonorea(kn)tfhlueonDreBaibenenthnzeozno(ee(a)p,Bhy)reaennnzteoh(rga,che,ni)epBIneedrneyzlneoon((ae1),p2y,3re-cnde)pyrene
Name of Compound
Figure 61: Number of Semi-Volatile Organic Compound Detections, by Compound, 2004-2006
Concentration (g/m 3)
0.007 0.006 0.005
Total Average Concentration of Sem i-Volatile Organic Com pounds, 2004-2006
2004 2005 2006
0.004
0.003
0.002
0.001
0 NaphtAhcaelennaepAhtcheennpehthyleneFluPohreenneanthrAennethracFelunoeranthene PyBreennCezhor(yBase)eannnzetoh(rbaB)cfeleunnozeroa(nkt)hfleunoerBaDneibtnhezenonz(eeo)(pay,hrBe)eannnezthor(agc,he,ni)epBeIenrndylzeeonn(oea()1p,y2r,e3n-ced)pyrene
Nam e of Com pound
Figure 62: Total Average Concentration of Semi-Volatile Organic Compounds, 20042006
95
Monitoring Techniques In 2006, samples were collected from a total of fifteen (15) sites, including a collocated site (a site that has two monitors of each type and acts as a quality assurance site for precision and accuracy calculations) and two background (rural) sites.
The compounds sampled at the ATN sites are shown in Appendix E. The list was derived from the 188 compounds EPA has designated as Hazardous Air Pollutants (HAPS). Many of the HAPS do not have standardized ambient air sampling and analytical methods. In order to collect the compounds of interest for the Georgia network, three types of samplers are used at all locations: the HIVOL, PUF, and canister. Also, carbonyls are monitored at three of the air toxics sites (as well as two PAMS sites).
This equipment samples for metals, semi-volatile organic compounds, and volatile organic compounds once every twelve days following a pre-established schedule that corresponds to a nationwide sampling schedule. On the twelfth day the sampler runs midnight to midnight and takes a 24 hour composite sample. The South DeKalb site, which monitors metals and volatile organic compounds, collects samples every six days, as part of the National Air Toxics Trends (NATTS) network.
The HIVOL sampler used for sampling for metals is a timed sampler. The sampler is calibrated to collect 1300 to 2000 liters of air per minute. Particulate material is trapped on an 8" x 10" quartz fiber filter. The particulates include dust, pollen, diesel fuel by-products, particulate metal, etc. The filters are pre-weighed at a remote laboratory prior to use and weighed again after sampling. The filters are subjected to a chemical digestion process and are analyzed on an inductively coupled plasma spectrometer.
The PUF (polyurethane foam) sampler used for sampling for semi-volatile organic compounds is a timed sampler. The sampler is calibrated to collect 198 to 242 liters (L) of air per minute. A multi-layer cartridge is prepared which collects both the particulate fraction and the volatile fraction of this group of compounds. The plug, filter and absorbent are extracted at a remote laboratory and analyzed using a gas chromatograph with an Electron Capture Detector (ECD).
The canister sampler used for sampling volatile organic compounds is also a timed sampler. A SUMMA polished canister is evacuated to a near-perfect vacuum and attached to a sampler with a pump controlled by a timer. The canister is filled to greater than 10 psig. The canister is analyzed using a gas chromatograph with mass spectroscopy detection (GC/MS).
The carbonyl sampler at both PAMS and ATN sites is also a timed sampler. An absorbent cartridge filled with dinitrophenylhydrazine (DNPH) coated silica is attached to a pump to allow approximately 180 L of air to be sampled. The cartridge is analyzed using High Performance Liquid Chromatography.
Some of the chemicals monitored in the Air Toxics Network (ATN) are also monitored at sites in the PAMS network. While the monitoring schedule and some analysis methods are different at the PAMS and ATN sites, the PAMS sites were also evaluated and compared to concentrations measured at nearby ATN sites for this analysis.
96
2006 Georgia Annual Air Quality Report Attainment Designation Currently, there are no attainment standards for the air toxics compounds, with the exception of lead, which has its designation as a criteria pollutant. Air toxics measurements are performed to support the regulatory, analytical, and public health purposes of the program. While it is understood that these compounds are toxic, it is not well understood what airborne concentrations of each compound may be harmful. By collecting data about their current concentrations, researchers can later compare our data with health data to determine what levels of each compound may be safe. Eventually, this information can be used in making policy decisions about regulating limits like the NAAQS for these other pollutants.
97
98
2006 Georgia Annual Air Quality Report
Meteorological Report
Variance in climate across north and central Georgia is largely a function of terrain. Most of the northern half of the state is made up of rolling hills with elevation ranging from 400 feet to between 800 and 1100 feet. The northeast region of the state occupied by the Appalachian Mountains is termed the "Northeast Georgia Mountains" and has a climate similar to the rest of North Georgia. The average annual rainfall is between 55 and 60 inches of rain, with the driest months on average being September and October, and the wettest month being March.
The climate of central Georgia, which includes the metropolitan areas of Atlanta and Macon, involves summers of warm, humid weather and variable temperatures during the winter months. The average date of first freeze is in mid-November. The average date of the last freeze in the spring is mid to late March. Average amounts of rainfall reach between 45 and 50 inches, with September and October averaging as the driest month and the wettest being March.
According to the National Weather Service, the weather across north and central Georgia during 2006 was predominantly characterized by warm and dry conditions (see Table 4 and Table 5), primarily due to persistence of high pressure aloft and near the surface, and the Gulf being cutoff. January set the trend with unseasonably mild temperatures and rainfall deficits recorded for Athens, Columbus, and Macon. Average monthly temperatures in February were colder than normal, with near or above normal rainfall recorded at most of the metro sites.
Atlanta
19712000
J
F
M
A
M
J
J
A
S
O
N
D
Yearly +/-
2006 5.1 5.50 2.93 2.48 2.86 5.80 1.31 8.66 3.31 3.04 4.39 3.08 -1.74
30 yr avg
5.0
4.68
5.38
3.62
3.95
3.63
5.12
3.67
4.09
3.11
4.10
3.82
Athens
19712000
2006 4.2 4.71 2.53 2.35 2.17 1.93 3.66 5.76 2.22 3.52 3.18 3.91
30 yr avg
4.6
4.39
4.99
3.35
3.86
3.94
4.41
3.78
3.53
3.47
3.71
3.71
-7.63
Macon
19712000
2006 2.0 3.81 1.24 2.48 1.52 5.12 3.42 2.28 1.46 2.23 3.05 5.99 -10.38
30 yr avg
5.0
4.55
4.90
3.14
2.98
3.54
4.32
3.79
3.26
2.37
3.22
3.93
Columbus 2006 2.9 4.46 2.77 2.91 3.18 1.77 2.86 3.77 3.49 3.14 4.63 2.89
19712000
30 yr avg
4.7
4.48
5.75
3.84
3.62
3.51
5.04
3.78
3.07
2.33
3.97
4.40
(Data compiled by National Weather Service at Peachtree City)
-9.73
Table 4: Monthly Rainfall For 2006 and 30-Year Average, Selected Cities
99
City
Mean Temperature
for 2006
Normal Mean Temperature
Mean Temperature Departure from
Normal
Atlanta
63.5
62.1
+1.4
Athens
63.2
61.5
+1.7
Macon
65.2
63.7
+1.5
Columbus
67.1
65.1
+2.0
(Data compiled by National Weather Service at Peachtree City)
Total Rainfall for 2006
48.46"
40.20"
34.62"
38.84"
Normal Total
Rainfall
Total Rainfall Departure
from Normal
50.38" 47.83" 45.00" 48.57"
-1.74 -7.63 -10.38" -9.73"
Table 5: Temperature and Rainfall Statistics for 2006, Selected Cities
In March, temperatures moderated again to near normal or to slightly above normal monthly readings. However, abnormally dry conditions returned resulting in hefty rainfall deficits for the month. The drier than average conditions continued in April as Atlanta, Athens, Columbus and Macon all recorded deficits for the 2nd consecutive month.
May temperatures were the closest to average for any month during the year. An unseasonably cool 10-day period during the middle of the month offset an early heat wave, which occurred toward the end of the month. This heat was also accompanied by a prolonged dry spell, which contributed to rainfall deficits of 1.69 inches at Athens and 1.46 inches at Macon. The early summer heat continued through June, as 90 degrees was reached 17 times in Atlanta. The heat wave accompanied by localized rainfall initiated a moderate drought for most of central and northern Georgia by mid June. July will long be remembered for its persistent 100+ degree heat over a broad area of the Great Plains and Midwest. In Atlanta, the mercury topped 90 degrees or more on 23 days, the most July occurrences since 1996. In August, more frequent thunderstorm activity brought some wellneeded relief from the persistent drought. However, not all areas benefited from the generally localized rainfall. Whereas Atlanta had its' wettest August since 1967 with 8.66 inches of rainfall, Macon received only 2.28 inches, which was 1.51 inches below normal. One big difference was a major rain event that occurred on the last day of the month. A slow moving upper level wave over Mississippi/Alabama approached Georgia on the heels of Tropical Storm Ernesto. A persistent training effect of thunderstorms set up over Atlanta, but not across central Georgia. The result, a record amount of rainfall (3.72") for the day in Atlanta, but just 0.15 inches fell in Macon.
September finally brought relief from the heat with all four metro cities recording a below normal monthly temperature. The cooler than average temperatures continued in October. A welcomed change was a more active lower Mississippi valley storm track, which brought near normal or above normal precipitation to the region. November provided near normal to above normal rainfall, while experiencing very close to average monthly temperatures. December began where November left off with 70 degree or more high temperatures on the 1st in Atlanta, Columbus, and Macon. This warmth soon ended however, as some of the coldest temperatures in 2 years were ushered in for the 7th through 11th. Mild temperatures continued though late December and led to all four cities recording monthly averages well above normal.
100
2006 Georgia Annual Air Quality Report
Summary of Meteorological Measurements (2006)
A complete suite of meteorological instrumentation is used to characterize meteorological conditions around metropolitan Atlanta. The basic surface meteorological parameters measured at the Photochemical Assessment Monitoring Sites (PAMS) are shown in Table 6. The PAMS sites are Conyers, South DeKalb, Tucker, and Yorkville. All PAMS sensors measure hourly-averaged scalar wind speed and vector-averaged wind direction at the 10meter level, and hourly-averaged surface temperature, relative humidity and barometric pressure at the 2-meter level. Several sites include instruments to record hourly-averaged precipitation, global solar radiation, and total ultraviolet radiation. The standard deviation of the wind direction is also computed at one of the PAMS sites (South DeKalb). Other surface meteorological measurements were made across the state in 2006 and are also shown in Table 6.
Statewide Monitoring
Sites
Wind Speed (m/s)
Wind Direction (deg)
Sigth (deg)
Relative Humidity
(%)
Solar Radiation
(W/m2)
Total Ultraviolet Radiation
(W/m2)
Barometric Pressure (mb)
Precip. (in)
Temp (C)
Conyers
a
a
a
a
a
a
a
South DeKalb
a
a
a
a
a
a
a
Tucker
a
a
a
a
a
a
a
a
Yorkville
a
a
a
a
a
a
a
Fort Mountain
a
a
a
a
Brunswick
a
a
Confederate Avenue
a
a
Dawsonville
a
a
Savannah E. President
a
a
Macon
a
a
Douglasville
a
a
Fayetteville
a
a
Newnan
a
a
Savannah L&A
a
a
Table 6: Meteorological Parameters Measured, 2006
101
Metro Atlanta had 30 ozone violations during (May through September) 2006, three of which were consecutive code reds on June 21-23rd. This particular case study in June will be described below. Augusta and Macon had 4 ozone violations each in 2006, with north Georgia having one ozone violation at Dawsonville on July 11th. This was considered to be a fairly typical ozone season for Metro Atlanta, with a fairly wide distribution of violations throughout the season as a whole. Monthly time series plots of ozone predictions and observations for Metro Atlanta during 2006 ozone season are shown in Figure 63 and Figure 64. Overall team forecasting performance for the 2006 ozone season was 83% on an event to non-event basis and 60% on an AQI basis. Overall performance for PM2.5 forecasting between May and September for 2006 was 83.7% on an Air Quality Index (AQI) basis. Of particular interest was the lack of tropical activity during 2006. Only 10 named tropical cyclones occurred with only 3 tropical storms making landfall in the United States. Alberto and Ernesto were the only two tropical storms to affect the Southeast United States. Both systems seemed to play a minimal role in surface ozone concentrations around Metro Atlanta in 2006.
102
2006 Georgia Annual Air Quality Report
Maximun O3 concentrations (ppbv)
2006
120
O3 observed O3 predicted
100
80
60
40
MAY
20 140
120
100
80
60
40
JUN
20
120
110
JUL
100
90
80
70
60
50
40
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31
Day of the month
(Data compiled by Georgia Tech Dr. Carlos Cardelino)
Figure 63: Forecasted and Observed 8-hr Ozone for Metro Atlanta, May-July 2006
2006
O3 observed O3 predicted
Maximun 8-hrs O3 concentrations (ppbv)
100
80
60
AUG
40
100
80
60
40
SEP
20
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 Day of the month
(Data compiled by Georgia Tech Dr. Carlos Cardelino)
Figure 64: Forecasted and Observed 8-hr Ozone for Metro Atlanta, August-September 2006
103
Observations of PM2.5 for Metro Atlanta are presented in Figure 65. As the figure shows, most PM2.5 violations occur during summer, however, some have been observed during winter months as well. Code orange violations for PM2.5 during summer 2006 were observed during June, July, and August. Overall team performance, as presented in Figure 66, has shown to improve relative to previous years, since the team began forecasting in October 2003, with an averaged forecasting accuracy shown to be near 80% for 2006. Of particular interest is the lower accuracy shown during the transitional months, relative to winter and summer months, possibly due to less air stagnation events. The shift in storm track and upper level jet position, as well as fire/smoke activity, play an important role in the frequency of air stagnation events and particle pollution episodes across the Southeast. PM2.5 forecasting by EPD will expand in the future to include the Columbus/Phenix City area, as well as continued O3 forecasting in Macon and Augusta.
PM 2.5 O bservations
30
G reen Y e llo w O range
25
Number of days
20
15
10
5
0
OND J FMAM J J ASOND J FMAM J J ASOND J FMAM J J ASOND J FM
Oct 2003 - Mar 2007
(Data compiled by Georgia Tech Dr. Carlos Cardelino)
Figure 65: PM2.5 Observations For Atlanta Area By AQI Category, October 2003 May 2007
104
Accuracy (%)
2006 Georgia Annual Air Quality Report
TEAM PERFORMANCE
100 95 90 85 80 75 70 65 60 OND J FMAMJ JASOND J FMAMJ JASONDJ FMAMJ JASOND J FM
October 2003 - March 2007
(Data compiled by Georgia Tech Dr. Carlos Cardelino)
Figure 66: Forecast Team Performance, Atlanta PM2.5, October 2003 May 2007
Select Meteorological and Air Quality Case Studies for 2006
(see Appendix C for relevant imagery)
April 14, 2006 - Several ambient monitoring sites in Metropolitan Atlanta experienced highly elevated 1-hour PM2.5 readings during the late afternoon/evening hours of April 14, 2006. The three sites, Confederate Avenue, South DeKalb, and Gwinnett, experienced peak hourly values of 54.7 g/m3, 47.8 g/m3, and 55.2 g/m3 respectively, during that time period.
Synoptic conditions involved a Gulf surface ridge holding firm over the region, with north Georgia positioned on the eastern flank of a large upper level ridge. A weak impulse over the Great Lakes region with associated frontal boundary remained to the north during the period. Sounding data from Peachtree City shows a dry, stable air mass with light westerly flow at the surface and northwesterly flow aloft over the top of the ridge. Stable conditions and limited vertical mixing allowed PM2.5 24-hour averaged levels to reach moderate values.
The fire and smoke product from the Hazard Mapping System (HMS), depicts numerous hotspots across the state on April 14th. Several fires were in close proximity and southwest
105
of Atlanta, as well as central Alabama. Backward trajectory analysis at the 1500m level shows that these fires likely had an impact on our region through particulate matter transport, as low level trajectories wrapped around from the southwest.
June 20-24, 2006 - This was an interesting ozone and PM2.5 episode, because the forecasting team correctly forecasted the middle of the episode but missed the beginning and end. The synoptic situation for this period was dominated by strong mid-Atlantic ridging which promoted enhanced subsidence over north Georgia. The RUC/MAPS analysis showed a dry, stable pocket over north Georgia on the 21 June at 1600Z. This was accompanied by hot downslope flow down the eastern slopes of the Appalachians. The dry pocket over north Georgia was enhanced by subsidence from outflow from an MCS that had developed along a lee-side trough over South Carolina. This dry pocket of air and MCS outflow boundary is well depicted in the visible satellite imagery shown in Appendix C. This particular exceedance was interesting and difficult to forecast because metro Atlanta reached code red for ozone on June 23rd, and this was a fairly isolated pocket of dry air situated between two convective outflow boundaries.
June 26, 2006 - Elevated ozone readings were observed on June 26, 2006 for the Evans monitoring site, located at latitude: 33.582 and longitude: -82.13134. The Evans site is located just west of the Augusta, Georgia metropolitan area. One-hour ozone readings peaked above 0.110 parts per million (ppm) at four different hours during the day, with a maximum reading of 0.309 ppm at 1800.
Synoptic conditions for this time period involved a tropical wave just off the Florida coast with outer rain bands reaching into the northern and central parts of Georgia. Moist and unstable conditions, coupled with the outer rain bands and leftover outflow boundaries, allowed for heavy scattered afternoon thunderstorms across the area. The multiple elevated readings at Evans were observed in the presence of such heavy storms, which suggests several possibilities as to their occurrence:
1) Lightning-induced ozone - Lightning is known to produce nitrogen oxides within thunderstorms. These chemicals can react with others in the presence of sunlight to produce ozone. Above the Earth's surface in the free troposphere (3-8 miles high), during the summer months, lightning activity increases NOx by as much as 90 percent and ozone by more than 30 percent. Since most lightning occurs inside a storm, the added ozone tends to show up several miles high rather than near the earth's surface.
While lightning was prevalent in and around the Evans area, it is difficult to prove that the sudden increases in surface concentrations of ozone were directly related to lightning or a cloud-to-ground lightning strike.
2) Tropopause Fold - Tropopause folding in the upper-levels is consistent with the formation of upper fronts and is one of the contributors to lower-level cyclogenesis. A tropopause fold is literally a break in the tropopause between the troposphere and the stratosphere. Large horizontal and vertical wind shears such as the jet streaks create this folding process as lower stratospheric air intrudes into upper-troposphere air. For a tropopause fold to occur there needs to be a deep upper low and strong jet aloft.
3) Electrical Noise Another possible cause of the elevated surface ozone readings at Evans is that of electrical noise or power surges from nearby lightning
106
2006 Georgia Annual Air Quality Report strikes. Thunderstorms and lightning are the most dramatic and destructive causes of power line problems and can create power surges. These surges are literally surges of overexcited electrons, which contain a lot of energy in short duration bursts. Indirect effects of lightning strikes or discharge include over voltages in electric and telecommunication lines. Although each of these factors has the slight possibility of being the cause of the elevated ozone readings at the Evans site on June 26, electrical noise from nearby lightning seemed more probable.
107
108
2006 Georgia Annual Air Quality Report
Quality Assurance
The purpose of this report is to provide ambient air quality users and the general public with a summary of the quality of the 2006 ambient air monitoring data in quantifiable terms. This is the second edition of the report and presents an overview of various quality assurance and quality control activities. The tables included in this report provide summary data for ambient air monitoring stations in the statewide network.
The Georgia Air Protection Branch mission is to promote and protect public health, welfare, and ecological resources through effective and efficient reduction of air pollutants while recognizing and considering the effects on the economy of the state. The Ambient Air Monitoring Program provides a key element of that mission through collecting and reporting on quality information on a large number of pollutants and for a vast air-monitoring network. The Ambient Air Monitoring Programs, directed by state law, conducts various monitoring projects in support of the Department of Natural Resources (DNR), Environmental Protections Division (EPD), and the United States Environmental Protection Agency (U.S. EPA). The monitoring projects include gaseous criteria and non-criteria pollutants, particulate matter, air toxics, non-methane hydrocarbons, and meteorological parameters. Data from these monitoring sources provide the means to determine the nature of the pollution problem and assess the effectiveness of the control measures and programs.
It is the goal of the Ambient Monitoring Program to provide accurate, relevant, and timely measurements of air pollutants and their precursors associated with the corresponding meteorological data to support Georgia's Air Protection Branch for the protection of environment and public health. The Quality Assurance Program conducts various quality assurance activities to ensure that data collected comply with procedures and regulations set forth by the U.S. EPA and can be considered good quality data and data for record.
What is quality assurance? Quality assurance is an integrated system of management activities that involves planning, implementing, assessing, and assuring data quality through a process, item, or service that meets users needs for quality, completeness, representativeness and usefulness. Known data quality enables users to make judgment about compliance with quality standards, air quality trends and health effects based on sound data with a known level of confidence. The objective of quality assurance is to provide accurate and precise data, minimize data loss due to malfunctions, and to assess the validity of the air monitoring data to provide representative and comparable data of known precision and accuracy.
Quality assurance (QA) is composed of two activities: quality control and quality assessment. Quality control (QC) is composed of a set of internal tasks performed routinely at the instrument level that ensures accurate and precise measured ambient air quality data. Quality control tasks address sample collection, handling, analysis, and reporting. Examples include calibrations, routine service checks, chain-of-custody documentation, duplicate analysis, development and maintenance of standard operating procedures, and routine preparation of quality control reports.
109
Quality assessment is a set of external, quantitative tasks that provide certainty that the quality control system is satisfactory and that the stated quantitative programmatic objectives for air quality data are indeed met. Staff independent of those generating data perform these external tasks. Tasks include conducting regular performance audits, on site system audits; inter-laboratory comparisons, and periodic evaluations of internal quality control data. Table 7 illustrates the types of performance audits currently performed by the QA program in 2006. Field and laboratory performance audits are the most common. System audits are performed on an as needed basis or by request. Whole air sample comparisons are conducted for the toxic air contaminants and non-methane hydrocarbons.
Air Monitoring
Program
Gaseous Pollutants Particulate
Matter
Air Toxic Contaminants
Non-Methane Hydrocarbons
Field Performance
Audit X
X
X
X
Laboratory Performance
Audit X
X
X
X
System Audit
X X
X
Whole Air
Audit
X X
Meteorology
X
X
Table 7: Audits Performed for Each Air Monitoring Program in 2006
Quality Control and Quality Assessment
The Quality Assurance Program supports all ambient monitoring programs undertaken by Georgia EPD, which in 2006 includes gaseous pollutants, particulate pollutants, air toxics contaminants, non-methane hydrocarbons and meteorological sensors run by the Ambient Monitoring Program. There are approximately 65 air monitoring sites operating in Georgia. Appendix F provides information about the air-monitoring network (i.e., sampling schedules, number of instruments, collection/analysis method, etc.).
The air quality monitors collect data in both real-time and on a time integrated basis. The data is used to define the nature, extent, and trends of air quality in the state; to support programs required by state and federal laws; and to track progress in attaining air quality standards. The precision and accuracy necessary depends on how the data will be used. Data that must meet specific requirements (i.e., criteria pollutants) are referred to as controlled data sets. Criteria for the accuracy, precision, completeness, and sensitivity of the measurement in controlled data sets must be met and documented.
Air Quality Data Actions (AQDA) are key tools used by the Quality Assurance Program to confirm the data set meets the established control limits. They are initiated generally by auditors upon a failed audit and resolved after a review of calibrations, precision checks, and
110
2006 Georgia Annual Air Quality Report
audit results. The AQDA must confirm that an analyzer/sampler has operated within Georgia Air Sampling Network's control limits of 15% (10% for PM10 and 4% for PM2.5), or for sitting temperature conditions. Otherwise, further action is taken.
Data without formal data quality objectives (i.e., toxics) are called descriptive data sets. The data quality measurements are made as accurately as possible in consideration of how the data are being used. Quantified quality assessment results describe the measurement variability in standard terminology, but no effort is made to confine the data set to values within a predetermined quality limit.
The Georgia Air Sampling Network's (GASN) Quality Assurance Program is outlined in a fivevolume Quality Assurance Manual. The volumes, listed below, guide the operation of the quality assurance programs used by the GASN.
Volume I: Quality Assurance Plan Volume II: Standard Operating Procedures for Air Quality Monitoring Volume III: Laboratory Standard Operating Procedures Volume IV: Monitoring Methods for the State Ambient Air Quality standards Volume V: Audit Procedures for Air Quality Monitoring
Volume I lists the data quality objectives and describes quality control and quality assessment activities used to ensure that the data quality objectives are met.
Gaseous Pollutants
Ambient concentrations of carbon monoxide (CO), nitrogen dioxide (NO2), ozone (O3), sulfur dioxide (SO2) and hydrogen sulfide (H2S) are continuously monitored by an automated network of stations run by the Georgia Ambient Air Monitoring Program. Exposure to these pollutants may cause adverse health effects such as: respiratory impairment, fatigue, permanent lung damage, and increased susceptibility to infection in the general population. Gaseous criteria and non-criteria pollutant data are a controlled data set and are subject to meeting mandatory regulations.
Sampling Cone
Accuracy (field): Annually, EPA conducts field through-the probe (TTP) performance audits for gaseous pollutants to verify the system accuracy of the automated methods and to ensure the integrity of the sampling system.
111
Accuracy is represented as an average percent difference. The average percent difference is the combined differences from the certified value of all the individual audit points. The upper and lower probability limits represent the expected accuracy of 95 percent of all the single analyzer's individual percent differences for all audit test levels at a single site. Overall, the responses of the individual analyzers indicate that as a whole, the network is providing accurate data. Ninety-five percent of the gaseous pollutant instruments audited in 2006 were found to be operating within the Georgia Ambient Air Monitoring control limits (15%). The most common causes for audit failure are malfunctions within the instrument and leaks in the sampling system. Table 8 and Figure 67 summarize the 2006 performance audit results for the criteria pollutants. UL stands for upper limit and LL stands for lower limit.
Number Average Probability Limits
Pollutant of
%
Analyzers Difference 95%UL 95%LL
CO
3
8.33
14.68
1.98
NO2
5
-0.84
3.44
-5.12
O3
24
2.14
5.38
-1.11
SO2
8
3.54
13.54 -6.47
Source: Quality Assurance Unit, Accuracy Estimates
Table 8: Results for Criteria Pollutants Performance Audits
2006 Georgia Gaseous Criteria Air Pollutants Overall Accuracy
20.00%
Overall Average % Difference
15.00%
10.00%
5.00%
0.00%
-5.00%
-10.00%
Lower 95% Probability Limit Overall Average % Difference Upper 95% Probability Limit
O3 -1.11% 2.14% 5.38%
SO2
NO2
-7.16%
-5.63%
3.54%
-0.84%
14.23%
3.95%
Gaseous Criteria Air Pollutants
CO 0.56% 8.33% 16.11%
Figure 67: Gaseous Criteria Pollutants Accuracy Analysis
In general, the Georgia Ambient Monitoring Program satisfied the requirements of the United States Environmental Protection Agency's (U.S. EPA) 40 CFR Part 58 and U.S. EPA's Quality Assurance Handbook for Air Pollution Measurement Systems, Volume II, August
112
2006 Georgia Annual Air Quality Report
1998. Compliance with these regulations is necessary if the data are to be considered datafor-record.
Precision (field): On a weekly basis, site operators confirm the linear response of the instrument by performing zero, precision and span checks. The zero precision check confirms the instrument's ability to maintain a stable reading. The span precision check confirms the instrument's ability to respond to a known concentration of gas. The degree of variability in each of these weekly measurements is computed as the precision of that instrument's measurements.
Annually, the Quality Assurance Unit conducts a precision data analysis as an overall indicator of data quality. The analysis addresses three parameters: precision data submission, precision data validity, and a combination of the two referred to as data usability rates. The precision performance goal for all three parameters is 85%. The submission rate is the number of precision points submitted for a pollutant divided by the expected number of bi-weekly submissions. Data validity is the percent difference of the actual and indicated values of each precision check. These differences should not exceed 15% for gaseous analyzers. Usable data rates are determined by multiplying the data submission and data validity rates. They indicate the completeness of verifiable air quality data on the official database.
Particulate Matter
Particulate matter is a mixture of substances that include elements such as carbon, metals, nitrates, organic compounds and sulfates; complex mixtures such as diesel exhaust and soil. Particles with an aerodynamic diameter of 10 microns or smaller pose an increased health risk because they can deposit deep in the lung and contain substances that are particularly harmful to human health. Respirable particulate matter (PM10) and fine particulate matter (PM2.5) increase the chance of respiratory disease, lung damage, cancer, and premature death.
Particulate matter monitoring is conducted using both manual and continuous type samplers. Manual samplers are operated on a six-day sampling schedule for PM10, and a similar, or more frequent schedule, for PM2.5. The Georgia Ambient Monitoring particulate program also includes total suspended particulates (TSP) sulfate, mass and lead monitoring.
Particulate matter is a controlled data set, and as such is subject to formal data quality objectives and federal and state regulations.
Accuracy (field): The accuracy of particulate samplers is determined by comparing the instrument's flow rate to a certified variable orifice (PM10 and TSP), or a calibrated mass flow meter (TEOM, BAM, and PM2.5 samplers) that is certified against a National Institute of Standards and Technology (NIST) traceable flow device or calibrator. Since an accurate measurement of particulate matter is dependent upon flow rate, the Ambient Monitoring Program conducts annual flow rate audits at each site. The average percent difference between the sampler flow rates and the audit flow rates represents the combined differences
113
from the certified value of all the individual audit points for each sampler. The upper and lower probability limits represent the expected flow rate accuracy for 95 percent of all the single analyzer's individual percent differences for all audit test levels at a single site.
Overall, the 2006 flow audit results indicate that the flow rates of samplers in the network are almost all within bounds. Approximately ninety-eight percent of the instruments audited in 2006 operated within the Georgia Ambient Monitoring Program's control limits. The 2006 performance audit results are listed below in Table 9 and Figure 68.
Pollutant
Number Average
of
%
Samplers Difference
Audited
Probability Limits 95%UL 95%LL
PM10
3
1.8
2.84
0.76
PM2.5
35
0.74
2.94
-1.47
PM10 Partisol 13
1.36
3.87
-1.16
Source: Quality Assurance Program, Accuracy Estimates
Table 9: Results for Particulate Sampler Performance Audits
2006 Particulate Air Pollutants
Overall Average % Difference
5.00%
4.00%
3.00%
2.00%
1.00%
0.00%
-1.00%
-2.00%
Upper 95% Probability Limit Overall Average Percent Difference Lower 95% Probability Limit
PM10 2.84% 1.80% 0.76%
PM2.5 2.94% 0.74% -1.47%
Particulate Air Pollutants
PM10 Partisol 3.87% 1.36% -1.16%
Figure 68: Particulate Air Pollutants Accuracy Analysis
Precision (field): Precision data for non-continuous particulate samplers is obtained through collocated sampling whereby two identical samplers are operated side-by-side and the same laboratory conducts filter analyses. Collocated samplers are located at selected sites and are intended to represent overall network precision. Validity of the data is based on the percent difference of the mass concentrations of the two samplers. In 2006, collocated PM2.5 samplers were operated at Augusta Medical College, Atlanta E. Rivers, Columbus Health
114
2006 Georgia Annual Air Quality Report
Department, Atlanta Doraville Health Department, Savannah Scott School and Macon Allied. Collocated PM10 samplers were operated at Atlanta Fire Station #8 and Macon Allied. Collocated TSP samplers were operated at Atlanta Utoy Creek and Atlanta DMRC.
Particulate samplers (collocated PM10 and TSP) must have mass concentrations greater than or equal to 20 g/m3 to be used in data validity calculations. The difference between the mass concentrations must be no greater than 5 g/m3. If the mass concentrations are greater than 80 g/m3, the difference must be within 7% of each other. TSP (lead) samplers must have both mass concentrations greater than or equal to 0.15 g/m3 to be used in data validity calculations. For collocated PM2.5 samplers, data probability limits validity is based on the sampler's coefficient of variation, which cannot exceed 10%. Both sample masses must also be greater than 6 g/m3.
Precision for continuous PM2.5 monitors is based on the comparison of the sampler's/analyzer's indicated and actual flow rates. The differences between the flow rates must be within 15.
Accuracy (lab): Annual performance audits for PM10 and PM2.5 mass analysis programs include an on-site check and assessment of the filter weighing balance, relative humidity and temperature sensors, and their documentation. The performance audits conducted in 2006 found that the Ambient Monitoring Program was operating in accordance with U.S. EPA guidelines and that the data were of good quality and should be considered data-for-record.
Precision (lab): Laboratories perform various quality control tasks to ensure that quality data are produced. Tasks include duplicate weighing on exposed and unexposed filters, replicate analysis on every tenth filter, and a calibration of the balance before each weighing session. Upon receipt of particulate matter filters from the field, laboratory staff has up to 30 days to analyze the PM10 and PM2.5 samples. Filters are visually inspected for pinholes, loose material, poor workmanship, discoloration, non-uniformity, and irregularities, and are equilibrated in a controlled environment for a minimum of 24 hours prior to the filters being weighed. If room conditions are not within the established U.S. EPA control limits, weighing is done only after the proper environment is re-established and maintained for 24 hours.
115
In 2006, there were no occurrences in which the Georgia's Ambient Monitoring laboratory balance room was outside of control limits. The analytical precision results indicate that the Ambient Monitoring Program is providing precise particulate matter data. Table 10 and Table 11 show the unexposed and exposed filter replicate results for the Air Protection Branch's (APB) laboratory in 2006.
QC Checks for Pre-weighed Filters
PM10
Total # of sample analyzed Total # of replicates Total % replicated Total # out-of-range
1189 59 5 0
Source: Laboratory Section, Quality Control Report
PM2.5
5419 361
7 0
Table 10: Summary of Unexposed Filter Mass Replicates
QC Checks for Pre-weighed Filters
PM10
Total # of samples analyzed Total # of replicates Total % replicated Total # out-of-range
1026 51 5 0
Source: Laboratory Section, Quality Control Report
Table 11: Summary of Exposed Filter Mass Replicates
PM2.5
4494 300
7 0
116
2006 Georgia Annual Air Quality Report
Air Toxics
In 1996, the Air Protection Branch established an Air Toxics Network in major urban areas of the state to determine the average annual concentrations of air toxics. The program was established to assess the effectiveness of control measures in reducing air toxics exposures. Compounds identified as air toxics vaporize at ambient temperatures, play a critical role in the formation of ozone, and have adverse chronic and acute health effects. Sources of air toxics include motor vehicle exhaust, waste burning, gasoline marketing, industrial and consumer products, pesticides, industrial processes, degreasing operations, pharmaceutical manufacturing, and dry cleaning operations. Under the current air toxic sampling schedule, ambient air is collected in a stainless steel canister (or cartridge) every 12 days over a 24hour sampling period at each of the network stations. Toxic particulate samples are also collected and analyzed for air toxic contaminants to support the Georgia Air Toxic and Control Unit. By using a low-flow multi-channel sampler capable of sampling onto filters or cartridges, ambient air is collected and analyzed for carbonyl and polycyclic aromatic hydrocarbons (PAH) compounds (also called semi-volatile organic compounds) and toxic metals. The quality of the air toxic data set is governed by a series of quality assurance activities, including audits. However, because this is a descriptive data set, no mandatory corrections are made to the data based on audit results. The laboratory and monitoring staff are made aware of any exceedance found during an audit, and every effort is made to ensure that the data collected is as accurate as possible.
Flow audits of the toxic metal, VOCs, semi-VOCs and carbonyl samplers are typically conducted annually at each site to ensure the accuracy of measuring toxic metals and carbonyl compounds. Flow rates are a determining factor in calculating concentration and are included as part of the Quality Assurance Program. Overall, the 2006 results indicate that the samplers maintained stable flows. Although toxics data are a descriptive data set, completeness is issued based on the operating parameters of the sampler. Corrections are made to the data if an audit is found to be outside the Air Toxic Program control limits.
Precision (field and lab): As part of the Air Toxic Program laboratory analyses, internal QC techniques such as blanks, control samples, and duplicate samples are applied to ensure the precision of the analytical methods and that the toxics data are within statistical control. Precision data for non-continuous toxics particulate samplers are obtained through collocated sampling whereby two identical samplers operate side-by side simultaneously and the same laboratory conducts filter analyses. The collocated toxic sampler located at Utoy Creek is intended to represent overall network precision.
In 2006, all compounds analyzed were within their respective control limits and results for blanks, spikes, and duplicate samples as established in the Laboratory QC Manual. Duplicate analyses were performed on 10% of the toxic samples. In 2006, all duplicate results (concentrations must be greater than five times the published limit of detection) were within the established limits for all target analytes. Data exceeding duplicate criteria of three times the assigned percent relative standard deviation (from control samples collected during the control limit evaluation) are deleted from the toxics database and samples reanalyzed. Table 12 shows the total precision concentrations for the Georgia Air Toxics Network.
117
Parameter
TSP Metals IO-2.1 Antimony (TSP) Arsenic (TSP) Berrylium (TSP) Cadmium (TSP) Chromium (TSP) Cobalt (TSP) Lead (TSP) Manganese (TSP) Nickel (TSP) Selenium (TSP) Zinc (TSP)
AQS Parameter
Code
Completeness
Overall Avg.
Std. Dev.
Lower 95% Probability Limit
Upper 95% Probability Limit
12102 12103 12105 12110 12112 12113 12128 12132 12136 12154 12167
100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%
-1.59% 43.46% 0.11% 0.06% 0.00% 0.00% -19.76% 44.09% 10.68% 65.98% -5.01% 65.50% -0.90% 39.71% 2.69% 45.87% -5.16% 84.81% -2.05% 45.27% 0.30% 30.34%
-61.8% 0.0% 0.0% -80.9% -80.8% -95.8% -55.9% -60.9% -122.7% -64.8% -41.8%
58.6% 0.2% 0.0% 41.4% 102.1% 85.8% 54.1% 66.3% 112.4% 60.7% 42.4%
VOCs TO-14/15 Freon 113 Freon 114 1-3-Butadiene Cyclohexane Chloromethane DiChloroMethane (Methylene Chloride) Chloroform Carbon Tetrachloride TriChloroFluoroMethane (Freon 11) Chloroethane 1-1-DiChloroEthane MethylChloroform (1,1,1-Trichloroethane) Ethylene DiChloride (1,2-Dichlororethane) TetraChloroEthylene 1-1-2-2-Tetrachloroethane Bromomethane 1-1-2-Trichloroethane DiChloroDiFluoroMethane TriChloroEthylene 1-1-DiChloroEthylene 1-2-DiChloroPropane Trans-1-3-DiChloroPropylene cis-1-3-DiChloroPropylene cis-1-2-DiChloroEthene Ethylene DiBromide (1,2-Dibromoethane) HexaChloroButadiene
43207 43208 43218 43248 43801 43802 43803 43804 43811 43812 43813 43814 43815 43817 43818 43819 43820 43823 43824 43826 43829 43830 43831 43839 43843 43844
100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%
7.80% 41.07% 0.00% 0.00% 0.00% 0.00% -0.72% 52.01% 0.44% 19.97% 0.00% 0.00% 8.26% 64.73% 2.78% 17.71% 2.43% 19.24% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 13.76% 72.73% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 2.98% 18.72% 0.72% 3.99% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
-49.1% 0.0% 0.0% -72.8% -27.2% 0.0% -81.5% -21.8% -24.2% 0.0% 0.0% 0.0% 0.0% -87.0% 0.0% 0.0% 0.0% -23.0% -4.8% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
64.7% 0.0% 0.0% 71.4% 28.1% 0.0% 98.0% 27.3% 29.1% 0.0% 0.0% 0.0% 0.0% 114.6% 0.0% 0.0% 0.0% 28.9% 6.2% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
Table 12: Total Precision Concentrations for the Georgia Air Toxics Network
118
2006 Georgia Annual Air Quality Report
Parameter
AQS Parameter
Code
Completeness
VOCs TO-14/15 (continued) Vinyl Chloride (Chloroethene) M/P Xylene (m- & p-Dimethylbenzene) Benzene Toluene Ethylbenzene O-Xylene (o-Dimethylbenzene) 1-3-5-TriMethylBenzene 1-2-4-TriMethylBenzene Styrene 4-EthylToluene Chlorobenzene 1-2-DiChloroBenzene 1-3-DichloroBenzene 1-4-DichloroBenzene BenzylChloride 1-2-4-TriChloroBenzene
43860
45109
45201 45202 45203 45204 45207 45208 45220 45228 45801 45805 45806 45807 45809 45810
96.7%
96.7%
96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7% 96.7%
Overall Avg.
Std. Dev.
Lower Upper
95%
95%
Probability Probability
Limit
Limit
0.0%
-1.7%
-3.8% -1.9% -2.0% -3.3% 0.0% -2.8% -0.1% 0.0% 0.0% 2.3% 0.0% 6.9% 0.0% 2.3%
0.0%
8.5%
10.1% 7.3% 6.7% 7.5% 0.0% 8.9% 6.9% 0.0% 0.0% 12.4% 0.0% 37.1% 0.0% 12.4%
0.0%
-13.5%
-17.8% -12.0% -11.2% -13.8% 0.0% -15.1% -9.6% 0.0% 0.0% -14.9% 0.0% -44.6% 0.0% -14.9%
0.0%
10.0%
10.2% 8.2% 7.3% 7.1% 0.0% 9.6% 9.4% 0.0% 0.0% 19.5% 0.0% 58.4% 0.0% 19.5%
SVOCs TO-13 Naphthalene Acenaphthene Acenaphthylene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Chrysene Benzo(a)anthracene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(e)pyrene Dibenzo(a-h)anthracene Benzo(g,h,I)perylene Benzo(a)pyrene Indeno(1-2-3-cd)pyrene
17141 17147 17148 17149 17150 17151 17201 17204 17208 17215 17220 17223 17224 17231 17237 17242 17243
100.0% 100.0% 96.7% 96.7% 100.0% 96.7% 100.0% 100.0% 100.0% 100.0% 96.7% 100.0% 100.0% 100.0% 96.7% 100.0% 100.0%
6.0% 6.7% 6.9% 0.0% -22.0% 0.0% -15.6% 6.7% 6.7% 0.0% 0.0% 0.0% 0.0% 6.7% 6.3% -6.7% 0.0%
36.8% 36.5% 37.1% 0.0% 61.3% 0.0% 105.9% 36.5% 36.5% 0.0% 0.0% 0.0% 0.0% 36.5% 36.6% 36.5% 0.0%
-45.0% -43.9% -44.5% 0.0% -106.9% 0.0% -162.4% -43.9% -43.9% 0.0% 0.0% 0.0% 0.0% -43.9% -44.4% -57.3% 0.0%
57.0% 57.3% 58.4% 0.0% 62.9% 0.0% 131.1% 57.3% 57.3% 0.0% 0.0% 0.0% 0.0% 57.3% 57.1% 43.9% 0.0%
Table 12: Total Precision Concentrations for the Georgia Air Toxics Network (continued)
119
Figure 69, Figure 70, and Figure 71 show the yearly precision numbers for the Utoy Creek collocated air toxics site.
120
Overall Average % Difference
2006 Georgia Air Toxics TSP Metals Monitoring Concentration Collocated Precision (Utoy Creek Site)
200.0%
150.0%
100.0%
50.0%
0.0%
-50.0%
-100.0%
-150.0%
-200.0%
Upper 95% Probability Limit Average % Difference Low er 95% Probability Limit
Antimony (TSP)
58.6% -1.59% -61.8%
Arsenic (TSP)
0.2% 0.11% 0.0%
Berrylium (TSP)
0.0% 0.00% 0.0%
Cadmium (TSP)
41.4% -19.76% -80.9%
Chromium Cobalt (TSP) Lead (TSP) Manganese Nickel (TSP)
(TSP)
(TSP)
102.1%
85.8%
54.1%
66.3%
112.4%
10.68%
-5.01%
-0.90%
2.69%
-5.16%
-80.8%
-95.8%
-55.9%
-60.9%
-122.7%
TSP Metals HAPs
Selenium (TSP)
60.7% -2.05% -64.8%
Zinc (TSP)
42.4% 0.30% -41.8%
Figure 69: Metals Monitoring, Collocated Precision
121
Figure 70: VOC Monitoring, Collocated Precision
VOCs HAPs
Freon 113
Freon 114
1-3-Butadiene
Cyclohexane
Chloromethane DiChloroMethane(Methylene
Chloride) Chloroform
Carbon Tetrachloride TriChloroFluoroMethane(Freon
11) Chloroethane
1-1-DiChloroEthane MethylChloroform(1,1,1EthyTlernicehlDoriCoehtlhoarindee)(1,2-
Dichlororethane) TetraChloroEthylene
1-1-2-2-Tetrachloroethane
Bromomethane
1-1-2-Trichloroethane
DiChloroDiFluoroMethane
TriChloroEthylene
1-1-DiChloroEthylene
1-2-DiChloroPropane
Trans-1-3-DiChloroPropylene
cis-1-3-DiChloroPropylene
cis-1-2-DiChloroEthene Ethylene DiBromide(1,2-
Dibromoethane) HexaChloroButadiene
Vinyl Chloride(Chloroethene) M/P Xylene (m & pDimethylbenzene) Benzene
Toluene
Ethylbenzene
O-Xylene (o-Dimethylbenzene)
1-3-5-TriMethylBenzene
1-2-4-TriMethylBenzene
Styrene
4-EthylToluene
Chlorobenzene
1-2-DiChloroBenzene
1-3-DichloroBenzene
1-4-DichloroBenzene
BenzylChloride
1-2-4-TriChloroBenzene
200% 150% 100%
50% 0%
-50% -100% -150% -200%
Overall Average % Difference
Lower 95% Probability Limit
Overall Average % Difference
Upper 95% Probability Limit
2006 Georgia Air Toxics VOCs Monitoring Concentrations Collocated Precision (Utoy Creek)
2006 Georgia Air Toxics Semi-VOCs Monitoring Concentration Collocated Precision (Utoy Creek)
Overall Average % Difference
200.0%
150.0%
100.0%
50.0%
0.0%
-50.0%
-100.0%
-150.0%
-200.0% NaphthaleneAcenaphthenAecenaphthylene
FluorenePhenanthrene Anthracene Fluoranthene
Pyrene
ChrByesneznoe(a)anthBreanczeon(eb)fluoraBnethnezon(ek)fluoranthenBeenzoD(ieb)epnyzreon(ae-h)anthBraecneznoe(g,h,I)perylenBeenzoI(nad)peynroe(n1e-2-3-cd)pyrene
Semi-VOCs HAPs
Upper 95% Probability Limit
Overall Average % Difference
Lower 95% Probability Limit
Figure 71: SVOC Monitoring, Collocated Precision 123
Stainless steel canisters used to collect ambient air samples are also checked for contamination. Canisters are analyzed for aromatic and halogenated hydrocarbons. One canister per batch of eight is assayed to ensure individual compound measurements fall below the limit of detection. In the event a compound exceeds canister cleanliness criteria, the canister and all other canisters represented in the batch are re-cleaned until compounds meet the cleanliness criteria. In addition, Xontech 910A air samplers are checked for cleanliness. Failed air collection media are re-cleaned and re-tested until they pass Xontech 910A cleanliness criteria. Overall, the network is providing precise air toxic contaminants data.
Accuracy (field): The accuracy of air toxic samples is determined by comparing the instrument's flow rate to a certified variable orifice (PM10 and TSP), or a calibrated mass flow meter (PM2.5 samplers) that is certified against a National Institute of Standards and Technology (NIST) traceable flow device or calibrator. Since an accurate measurement of particulate matter is dependent upon flow rate, the Ambient Monitoring Program conducts annual flow rate audits at each site. The average percent difference between the sampler flow rates and the audit flow rates represents the combined differences from the certified value of all the individual audit points for each sampler. The upper and lower probability limits represent the expected flow rate accuracy for 95 percent of all the single analyzer's individual percent differences for all audit test levels at a single site.
Overall, the 2006 flow audit results indicate that the flow rates of samplers in the network are almost all within bounds. Approximately ninety-eight percent of the instruments audited in 2006 operated within the Georgia Air Toxics Network control limits. The total 2006 performance audit results for the Air Toxics Network are listed below in Table 13 and Figure 72, showing the network's sampling flow rate accuracy.
Air Toxics Network
Average
Upper Limit
Lower Limit
Std. Dev
11 TSP Metals (15 Samplers)
2.0% 6.0% -2.0% 2.0
17 PUF Semi-VOCs (15 samplers) -0.1% 5.9% -6.1% 3.1
42 Canister VOCs (15 samplers) 2.3% 9.8% -5.1% 3.8
7 Carbonyls (3 Samplers)
1.4% 17.1% -14.3% 7.9
Table 13: Yearly Summary of Flow Rate Accuracy Performance Audit, Air Toxics Network
124
2006 Georgia Annual Air Quality Report
Overall Average % Difference
50.0% 40.0% 30.0% 20.0% 10.0%
0.0% -10.0% -20.0% -30.0% -40.0% -50.0%
Upper 95% Probability Limit Overall Average % Difference Lower 95% Probability Limit
2006 Georgia Air Toxics Network Overall Average Sampling Flow Rate Accuracy
11 TSP Metals (15 Samplers) 6.0% 2.0% -2.0%
17 PUF Semi-VOCs 42 Canister VOCs (15
(15 samplers)
samplers)
5.9%
9.8%
-0.1%
2.3%
-6.1%
-5.1%
Air Toxic Speciated HAPs
7 Carbonyls (3 Samplers) 17.1% 1.4% -14.3%
Figure 72: Yearly Summary of Sampling Flow Rate Accuracy, Air Toxics Network
125
NATTS
There are currently 188 hazardous air pollutants (HAPs), or air toxics, regulated under the Clean Air Act (CAA). These compounds have been associated with a wide variety of adverse human health and ecological effects, including cancer, neurological effects, reproductive effects, and developmental effects. According to the Government Performance Results Act (GPRA), the U.S. Environmental Protection Agency (U.S. EPA) is committed to reducing air toxics emissions by 75 percent from 1993 levels in order to significantly reduce Americans' risk of cancer and of other serious health effects caused by airborne toxic chemicals. Early efforts toward this end have focused on emissions reductions through the assessment of technical feasibility. However, as new assessment tools are developed, more attention is being placed on the goal of risk reduction associated with exposure to air toxics. To meet the GPRA goals, the National Air Toxics Trends Station (NATTS) network has been established, consisting of 23 stations in the contiguous 48 states, with one in Georgia. Having data of sufficient quality is paramount for a network such as the NATTS. As such, Georgia has closely followed the Quality System (QS) for the NATTS, established by U.S. EPA, two aspects of which are Technical Systems Audits (TSAs) and Instrument Performance Audits (IPAs) of each network station and its affiliated sample analysis laboratory. Another integral part of the QS is the quarterly analysis of performance evaluation (PE) samples. Furthermore, the sampling and analytical techniques selected to collect and quantify the air toxics of concern must demonstrate acceptable analytical and overall sampling precision as well as suitable overall method detection limits that are compatible with expected ambient air toxics concentrations. There are 23 sites in the NATTS network. Georgia joined the network with one site established in Decatur at the South DeKalb Monitoring Station. The EPA Region in which the sites are located, the location of the sites (site identifier), whether the site is located in an urban or rural area, and the unique AQS identification code (site code) for all the sites are given in Table 14.
126
2006 Georgia Annual Air Quality Report
Region I I I II II III IV IV IV
IV
IV V V V VI VI VII VIII
Site Identifier Boston-Roxbury, MA Chittenden City, VT Providence, RI Bronx, NY Rochester, NY Washington, DC Chesterfield, SC Decatur, GA Hazard, KY Hillsborough City, Tampa, FL Pinellas City, Tampa, FL Dearborn, MI Mayville, WI Northbrook, IL Deer Park, TX Harrison County, TX St. Louis, MO Bountiful, UT
VIII Grand Junction, CO
IX Phoenix, AZ IX San Jose, CA X La Grande, OR X Seattle, WA
Type Urban Rural Urban Urban Urban Urban Rural Urban Rural
Urban
Urban Urban Rural Urban Urban Rural Urban Urban
Rural
Urban Urban Rural Urban
AQS Site Code 25-025-0042 50-007-0007 44-007-0022 36-005-0110 36-055-1001 11-001-0043 45-025-0001 13-089-0002 21-193-0003
12-057-3002
12-103-0026 26-163-0033 55-027-0007 17-031-4201 48-201-1039 48-203-0002 29-510-0085 49-011-0004 08-077-0017,
-0018 04-013-9997 06-085-0005 41-061-0119 53-033-0080
Table 14: NATTS Sites with EPA Region Numbers and AQS Site Codes
127
Several Measurement Quality Objectives (MQOs) have been established for the NATTS
network in order to ensure that only data of the highest quality are collected by the NATTS
network, and to meet the NATTS Data Quality Objective (DQO): "to be able to detect a 15
percent difference (trend) between two consecutive 3-year annual mean concentrations within acceptable levels of decision error"7. The four compounds of primary importance to the
NATTS program are benzene, 1,3-butadiene, formaldehyde, and PM10 arsenic. The MQOs for these four compounds are summarized in Table 15 below.
Compound
Precision Completeness (Coefficient
of Variation)
Benzene
> 85 %
< 15 %
1,3-Butadiene
> 85 %
< 15 %
Formaldehyde
> 85 %
< 15 %
Arsenic
> 85 %
< 15 %
Laboratory Bias
< 25 % < 25 % < 25 % < 25 %
Method
Detection
Limit
(MDL)
0.044 g/m3
0.020 g/m3
0.014 g/m3
0.046 ng/m3
Table 15: Measurement Quality Objectives for the NATTS Program
The MQOs require that (1) sampling occurs every sixth day and is successful 85 percent of the time; (2) precision as measured by the coefficient of variation (CV) be controlled to less than 15 percent; and (3) that laboratory (measurement) bias be less than 25 percent. Data acquired to assess compliance with the above stated MQOs are derived from a variety of sources. These sources are given in Table 16.
Criteria Completeness Precision Bias Laboratory Bias - Field
MDL
Data Source Air Quality System (AQS)
AQS and Proficiency Testing
Proficiency Testing
Audits of sampler flowrates
Laboratories
MQO Limit < 15 % < 15 %
< 25 %
< 10 % 0.046 ng/m3 to
0.044 g/m3
Table 16: MQO Data Sources for the Georgia NAATS Program
The Air Quality System (AQS) database contains raw data that is used to assess data completeness, and to estimate precision from results of replicate analyses and collocated
7 Quality Assurance Handbook for Air Pollution Measurement System. Volume 1. Principles. EPA-600/R94/038A, January 1994.
128
2006 Georgia Annual Air Quality Report
sampling. In addition, results from the analysis of proficiency testing samples allow one to calculate laboratory precision and bias. Field bias is evaluated by measurement of sampler flow rates during on-site Instrument Performance Audits.
Completeness (of NATTS Data): The AQS database was accessed and the raw data records analyzed 23 compounds having the AQS codes given in Table 17. The completeness of the 2006 AQS dataset was assessed for four compounds: benzene, 1,3-butadiene, formaldehyde, and arsenic. The results are shown in Table 18. The presence of 61 concentration values in the database indicates 100 percent completeness, since sampling is to occur every sixth day. Furthermore, the completeness data presented here are composite values for both the primary and collocated sampler at the South DeKalb site. Primary and collocated data are differentiated in AQS by use of parameter occurrence codes (POCs).
Compound Name AQS Code
Benzene
45201
1,3-Butadiene
43218
Carbon Tetrachloride
43804
Chloroform
43803
1,2-Dibromoethane
43843
1,2-Dichloropropane
43829
1,2-Dichloroethane
43815
Dichloromethane
43802
1,1,2,2-Tetrachloroethane 43818
Tetrachloroethylene
43817
Trichloroethylene
43824
Vinyl Chloride
43860
Cis-1,3-Dichloropropene 43831
Trans-1,3-Dichloropropene 43830
Formaldehyde
43502
Acetaldehyde
43503
Arsenic
82103
Beryllium
82105
Cadmium
82110
Lead
82128
Manganese
82132
Mercury
82142
Nickel
82136
Table 17: 23 Selected HAPs and Their AQS Parameter Codes
129
Completeness of Compound by AQS Number and by Name
45201
43218
43502
82103
Site
benzene 1,3-butadiene formaldehyde arsenic
Decatur, GA 90%
89%
95%
90%
Table 18: Percent Completeness of Georgia's 2006 AQS Data, Selected Compounds
Precision (of NATTS Data): The precision of the data in AQS for Georgia was assessed for all 23 compounds identified in Table 17. Replicate data transactions were retrieved from AQS. For a given sampling moment, site, and compound, AQS replicate data can have three separate transactions with Precision IDs of 1, 2, or 3. Concentration information is given in the transaction labeled with Precision ID 1; the value obtained from the replicate analysis of the primary sample is recorded with Precision ID 2; and the replicate analysis of the collocated sample is given in the transaction with Precision ID 3. Thus, both the analytical and overall sampling and analysis precision can be determined. Table 19 reveals that Georgia had large CV values for arsenic, beryllium, cadmium, lead, and manganese.
Compound Name AQS Code Precision Estimate
Benzene
45201
-6.6%
1,3-Butadiene
43218
3.3%
Carbon Tetrachloride
43804
-5.4%
Chloroform
43803
-5.4%
1,2-Dibromoethane
43843
-6.3%
1,2-Dichloropropane
43829
4.6%
1,2-Dichloroethane
43815
1.1%
Dichloromethane
43802
-0.3%
1,1,2,2-Tetrachloroethane 43818
-1.4%
Tetrachloroethylene
43817
-11.1%
Trichloroethylene
43824
0.8%
Vinyl Chloride
43860
-5.3%
Cis-1,3-Dichloropropene 43831
2.4%
Trans-1,3-Dichloropropene 43830
4.5%
Table 19: Laboratory Analytical Precision Estimate
130
2006 Georgia Annual Air Quality Report
Photochemical Assessment Monitoring
In 1996, the Air Protection Branch began a routine seasonal sampling program to gather information about non-methane hydrocarbon (NMHC) species that were precursors to ozone formation in high ozone areas. In 1994, Federal regulations required states to establish photochemical assessment monitoring stations (PAMS) as part of their State Implementation Plan monitoring networks in areas designated as serious or higher for ozone. Monitoring is to continue until the ozone standard is reached. The PAMS program is intended to supplement ozone monitoring and add detailed sampling for its precursors. PAMS sites collect data on real-time total NMHC, PAMS speciated VOCs, carbonyls, and various meteorological parameters at ground level and aloft. As this is a descriptive data set, there are currently no mandatory data quality objectives or regulations for the data. However, efforts are made to ensure that accurate data are collected and that the analyzers are operating within PAMS audit standards.
Accuracy (field and lab): Laboratory performance audits are conducted annually to assess the laboratory's ability to measure ambient levels of hydrocarbons. Through the probe sampler performance audits are typically conducted annually at each monitoring site to assess the integrity of the sampling, analysis, and transport system. The 2006 laboratory performance audit results are shown in Table 20 and Figure 73. The 2006 audit results were within the PAM's control limits of 20%, with the exception of isoprene and n-decane.
2006 PAMS Speciated VOCs
Audits
Number of
GC/FIDs Audited
Ethane
5
Propane
5
Isobutane
5
Trans-2-Butene
5
n-Pentane
5
2-Methyl-Pentane
5
Isoprene
5
n-Hexane
5
Benzene
5
n-Heptane
5
Toluene
5
Ethylbenzene
5
n-Decane
5
1,2,3-TrimethylBenzene
5
Average % Difference
-3.1 -10.2 -13.1 -21.5 -37.9 -20.1 -41.5 -4.0 -9.4 -9.2 -12.2 -14.3 -20.1
-22.1
Probability Limits
95% LL
-9.3 -14.7 -17.5 -55.9 -99.1 -52.0 -59.5 -19.8 -20.0 -22.0 -23.7 -26.6 -45.9
-35.4
95% UL
3.2 -5.8 -8.7 13.0 23.4 11.9 -23.4 11.9 1.1 3.7 -0.7 -2.0 5.7
-8.7
Table 20: Laboratory Speciated VOC Audit Results for PAMS Network
131
% Difference
2006 Georgia PAMS NetworkGC/FID Continuous PAMS Speciated VOCs Monitoring Overall Average Accuracies
100.0% 80.0% 60.0% 40.0% 20.0% 0.0% -20.0% -40.0% -60.0% -80.0%
-100.0% Ethane Propane IsobuTtraannes-2-Butenen-P2e-nMtaenthePylA-PMenStanSepeIscopiaretneednV-HOexCasneinBeAnuzednietnS-HteapntadnaerdTolueEntehylbenze1n,e2,n3--Dtriemceatnheyl-Benzene
Upper 95% Probability Limit Average % Difference Lower 95% Probability Limit
Figure 73: VOC Monitoring Accuracy Analysis
Meteorology
The Ambient Monitoring Program monitors meteorological parameters such as wind speed, wind direction, ambient temperature, relative humidity, barometric pressure, total ultra violet radiation, precipitation and total solar radiation. Real-time meteorological data are generated to characterize meteorological processes such as transport and diffusion, and to make air quality forecasts and burnday decisions. The data are also used for control strategy modeling case study analysis and urban airshed modeling. A state/local meteorology subcommittee of the Air Monitoring Technical Advisory Commission (AMTAC) agreed to define the level of acceptability for meteorological data as those used by the U.S. EPA for both the Prevention of Significant Deterioration (PSD) and Photochemical Assessment Monitoring Stations (PAMS) programs. The Quality Assurance Unit audits to those levels.
The data variability collected by this element of the monitoring program is generally described as meeting or not meeting the PSD requirements. No mandatory corrections are made to the data. However, station operators are notified whether they passed the audit or not. Most operators make an effort to meet the audit standards. The wind speed, wind direction, 132
2006 Georgia Annual Air Quality Report
ambient temperature and relative humidity data sets are controlled data sets, and subject to meeting PAMS objectives. Since the inception of the meteorological audit program, the data quality has improved significantly.
Accuracy (field): The accuracy of meteorological sensors is checked by annual performance audits. Figure 74 summarizes the 2006 audit results. The average difference (average degree difference with respect to ambient temperature) represents the combined differences from the certified value of all the individual audit points for each sensor. The upper and lower probability limits represent the expected accuracy of 95 percent of all the single sensor's individual percent differences for all audit test levels at a single site.
% Difference
2006 Georgia EPD Meteorological Measurements Network Performance Overall Average Audit Results
10.0%
8.0%
6.0%
4.0%
2.0%
0.0%
-2.0%
-4.0%
-6.0%
-8.0%
-10.0%
Wind speed
Wind Direction
Ambient
Relative humidity
temperature
Barometric pressure
Meteorological Parameters
Precipitation
Upper 95% Probability Limit Average % Difference Lower 95% Probablity Limit
Figure 74: Meteorological Measurements Accuracy Results
133
Quality Control Reports
Quality Control (QC) reports are summaries of the quality control activities conducted by the laboratory to support accurate and precise measurements. These activities include: blanks, duplicates, controls, spiked samples, limits of detection, calibrations, and audit results.
Standards Laboratory
The U.S. EPA Region IV Standards Laboratory yearly performs technical support and certification services for Georgia's ozone primary standard. Flow rate transfer standards and certification of compressed gas cylinders are sent to the manufactures for re-certification to ensure that all are traceable to standards of the NIST. A calibration establishes a correction factor to adjust or correct the output of an instrument; a certification establishes traceability of a transfer standard to a NIST-traceable standard; and verification establishes comparability of a standard to a NIST-traceable standard of equal rank.
Laboratory and Field Standard Operating Procedures
Standard Operating Procedures (SOPs) are guidance documents for the operation of quality assurance programs used by the Georgia Ambient Monitoring Program. The SOPs are intended for field operators and supervisors; laboratory, data processing and engineering personnel; and program managers responsible for implementing, designing, and coordinating air quality monitoring projects. Each SOP has a specific method that must be followed to produce data-for-record. The SOPs are developed and published to ensure that, regardless of the person performing the operation, the results will be consistent.
Siting Evaluations
To generate accurate and representative data, ambient monitoring stations should meet specific siting requirements and conditions. It is assumed that the stations meet the siting criteria in place at the time initial operation began. The siting requirements of the AMP Quality Assurance Manual Volume II; 40 CFR 58, Appendix E; U.S. EPA's Quality Assurance Handbook Volume IV: U.S. EPA's Prevention of Significant Deterioration (PSD); and U.S. EPA's PAMS guidelines, present siting criteria to ensure the collection of accurate and representative data. The siting criterion for each pollutant varies depending on the pollutant's properties, monitoring objective and intended spatial scale. The U.S. EPA's siting criteria are stated as either "must meet" or "should meet". According to 40 CFR 58, Appendix E, the "must meet" requirements are necessary for high quality data. Any exception from the "must meet" requirements must be formally approved through the Appendix E waiver provision. The "should meet" criteria establish a goal for data consistency. Siting criteria are requirements for locating and establishing stations and samplers to meet selected monitoring objectives, and to help ensure that the data from each site are collected uniformly. There are four main monitoring objectives: to determine highest concentrations expected to occur in the area covered by the network; to determine representative concentrations in areas of high population density; to determine the impact on ambient pollution levels of significant sources
134
2006 Georgia Annual Air Quality Report or source categories; and to determine general background concentration levels. Typical siting designations are: micro, middle, neighborhood, and regional. These designations represent the size of the area surrounding the monitoring site which experiences relatively uniform pollutant concentrations. Typical considerations for each of these site designations are, for example, the terrain, climate, population, existing emission sources, and distances from trees and roadways. Siting evaluations are conducted annually by the Quality Assurance Unit. Physical measurements and observations include probe/sensor height above ground level, distance from trees, type of ground cover, residence time, obstructions to air flow, and distance to local sources. These measurements and observations are taken to determine compliance with 40 CFR Part 58, Appendix E requirements.
135
136
2006 Georgia Annual Air Quality Report
2005-2006 Risk Assessment Discussion
Introduction
In 2005 and 2006, ATN samples were collected from a total of fifteen sites, including a collocated site (a site that has two monitors of each type and acts as a quality assurance site for precision and accuracy calculations), and two background (rural) sites. The compounds sampled at the ATN sites are shown in Table 21. The list was derived from the 188 compounds EPA has designated as Hazardous Air Pollutants (HAPS). Many of the HAPS do not have standardized ambient air sampling and analytical methods. In order to collect the compounds of interest for the Georgia network, at least three types of samplers are used at all locations (except South DeKalb): HIVOL, PUF, and canister. A carbonyls sampler is located at the Brunswick, Dawsonville, Savannah, South DeKalb (NATTS and PAMS), and Tucker (PAMS) sites. This equipment samples for metals, semi-volatile organic compounds, volatile organic compounds, and carbonyls once every twelve days following a preestablished schedule that corresponds to a nationwide sampling schedule. On the twelfth day the sampler runs midnight to midnight and takes a 24-hour composite sample. Exceptions to this sampling schedule are the South DeKalb and Tucker sites. The South DeKalb site samples for metals, volatile organic compounds, and carbonyls but not semivolatile organic compounds, every 6 days as part of the National Air Toxics Trends Station (NATTS) and PAMS network. The Tucker site samples only carbonyls and volatile organic compounds, every 6 days as part of the PAMS network.
Some of the chemicals monitored in the Air Toxics Network (ATN) are also monitored at sites in the Photochemical Assessment Monitoring Stations (PAMS) network. While the monitoring schedule and some analysis methods are different at the PAMS sites and ATN sites, several of the compounds from the PAMS sites were also evaluated and compared to values measured at nearby ATN sites for this report.
Results and Interpretation
The air toxic data ((volatile organic compounds (VOC), semi-volatile organic compounds, and metals)) collected during 2005 and 2006 from the Air Toxics Network was evaluated to assess the potential for health concerns. The data collected for the group of chemicals known as carbonyls were assessed separately from the other air toxics because those chemicals were only monitored at three of the ATN sites and two of the PAMS locations.
The initial evaluation consisted of a comparison of the monitored results to "health based" screening values. These values were calculated using procedures recommended in EPA's latest guidance on risk assessment for air toxics (U.S EPA, 2006). Briefly, EPA's prioritized chronic dose-response values for both noncancer (reference concentrations, RfC) and cancer (inhalation unit risks, IUR) were used to generate screening air concentrations. To screen for noncancer effects, the reference concentration was used as a starting point. However, to account for possible exposure to multiple contaminants, the screening air concentration was obtained by dividing the RfC by 10. Screening values for the cancer endpoint were determined by calculating air concentrations equivalent to a risk level of one in one million. Most screening values utilized in this assessment are listed in Appendix A of the previously
137
mentioned guidance document (U.S. EPA, 2006). These screening values and the chemicals monitored are displayed in Table 21. For a limited number of chemicals, other resources such as toxicity values from EPA's Region IX risk program were used to calculate conservative screening values. These compounds are indicated with an asterisk. When available, both the name derived from the International Union of Chemistry (IUC) and the common names are given. It is important to emphasize that the screening values were calculated in a very conservative manner. Assumptions were made that accounted for the potential for continuous exposure to air toxics for 24 hours per day for 70 years. The conservative screening process was utilized so that the chance of underestimating the potential for health impacts would be minimized, as chemicals were excluded from further quantitative analysis. Because results for many of the chemicals assessed were routinely below detection limits of the analytical methods available, the initial review of the data was based on an assessment of the number of chemicals detected and the frequency with which they were detected. The process included determining how often (if at all) a chemical was detected (present), if it was present above detection limits, and if those concentrations were above screening values of concern.
138
2006 Georgia Annual Air Quality Report
Chemical
Screen Value Chemical (g/m3)
Screen Value (g/m3)
Metals
Antimony
0.02 Cobalt
0.01
Arsenic
0.00023 Lead
0.15
Beryllium
0.00042 Manganese
0.005
Cadmium
0.00056 Nickel
0.0021
Chromium
0.000083 Selenium
2
Chromium VI
0.000083 Zinc
N/A
Semi-Volatiles
Acenaphthene
0.3 Chrysene
0.091
Acenaphthylene
0.3 Dibenzo(a,h)anthracene
0.00083
Anthracene
0.3 Fluoranthene
0.3
Benzo(a)anthracene
0.0091 Fluorene
0.3
Benzo(b)fluoranthene
0.0091 Ideno(1,2,3-c,d)pyrene
0.0091
Benzo(k)fluoranthene
0.0091 Naphthalene
0.029
Benzo(g,h,i)perylene
0.3 Phenanthrene
0.3
Benzo(a)pyrene
0.00091 Pyrene
0.3
Benzo(e)pyrene
0.3
Volatile Organic Compounds
Benzene
0.13 1,2-Dimethylbenzene (o-Xylene)
10
Benzenecarbonal (Benzaldehyde)
370* 1,3 and 1,4-Dimethylbenzene (m/p-Xylene)
10
Benzyl chloride
0.02 Ethanal (Acetaldehyde)
0.45
Bromomethane (Methyl bromide)
0.5 Ethylbenzene
100
1,3-Butadiene Butanal (Butyraldehyde)
0.03 Ethenylbenzene (Styrene)
100
N/A 1-Ethyl,4-methyl benzene (4-Ethyltoluene)
N/A
Chlorobenzene (Phenyl chloride)
100 Freon 113
31000*
Chloroethane (Ethyl chloride)
1000 Hexachloro-1,3-Butadiene(Hexachlorobutadiene) 0.045
Chloroethene (Vinyl chloride)
0.11 Methanal (Formaldehyde)
0.98
Chloromethane (Methyl chloride)
9.0 Methybenzene/Phenylmethane (Toluene)
40
Cyclohexane 1,2-Dibromoethane (Ethylene bromide) 1,2-Dichlorobenzene
6200* 0.002 0.091
Propanal (Propionaldehyde) 2-Propanone (Acetone) Propenal (Acrolein)
N/A 3300* 0.002
1,3-Dichlorobenzene
110* 1,1,2,2-Tetrachloroethane
0.017
1,4-Dichlorobenzene
0.091 Tetrachloroethene (Perchloroethylene)
0.17
Dichlorodifluoromethane (Freon 12)
210* Tetrachlormethane (Carbon tetrachloride)
0.067
1,1-Dichloroethane (Ethylidene chloride)
0.63 1,2,4-Trichlorobenzene
20
cis-1,2-Dichloroethene
370 1,2,4-Trimethylbenzene
6.2*
1,1-Dichloroethene (1,1-Dichloroethylene)
210* 1,3,5-Trimethylbenzene
6.2*
Dichloromethane (Methylene chloride)
2.10 1,1,1-Trichloroethane (Methylchloroform)
100
1,2-Dichloropropane (Propylene chloride)
0.3 1,1,2-Trichloroethane
0.063
cis-1,3-Dichloropropene
N/A Trichloroethene (Trichloroethylene)
0.5
trans-1,3-Dichloropropene
N/A Trichlorofluoromethane (Freon 11)
1,1-Dichloro-1,2,2,2-tetrafluoroethane(Freon114) N/A Trichloromethane (Chloroform)
730* 9.8
*From EPA's Region IX Risk Program
Table 21: Compounds Monitored and Screening Values Used in Initial Assessment
139
Table 22 and Table 23 summarize the total number of chemicals monitored at each site (excluding carbonyls), the number of chemicals detected, and the number of chemicals detected above the health based screening values for 2005 and 2006. Seventy chemicals were monitored at all the ATN sites, except the South DeKalb site, where 54 air toxic chemicals were monitored. In 2005, twenty-four of the 70 sampled compounds were not detected at the sites, and an additional 13 compounds had 3 or fewer sites with detections. The number of chemicals that were detected at concentrations above the screening levels was even less, with a mean value of 6.2. Of the three categories of chemicals measured at all sites (VOC, semi-VOC, metals), most of the chemicals that were detected above screening values belonged to the metals group. In 2006, thirty-three of the 70 compounds were not detected and thirteen compounds had 3 or fewer sites with detections. The number of chemicals above the screening value was an average of 5.5. The group most detected above the screening value was metals in 2006 as well.
Location
County
Number of Compounds Monitored
Number of Compounds
Detected
Number Greater than Screening
Value
Augusta
Richmond
70
36
8
Brunswick
Glynn
70*
32
6
Columbus
Muscogee
70
30
5
Dawsonville
Dawson
70*
23
6
Douglas
Coffee
70
34
5
Gainesville
Hall
70
29
7
Macon
Bibb
70
28
6
Milledgeville
Baldwin
70
30
5
Rome
Floyd
70
33
7
Savannah
Chatham
70*
30
6
South DeKalb
DeKalb
54*
30
6
Utoy Creek
Fulton
70
35
9
Valdosta
Lowndes
70
30
5
Warner Robins
Houston
70
32
7
Yorkville
Paulding
70
27
5
* 7 additional chemicals were monitored at these locations, but that information is summarized in Table 32.
Table 22: Summary of Chemicals Analyzed in 2005
140
2006 Georgia Annual Air Quality Report
Location
County
Number of Compounds
Monitored
Number of Compounds
Detected
Number Greater than Screening
Value
Augusta
Richmond
70
27
5
Brunswick
Glynn
70*
18
3
Columbus
Muscogee
70
25
6
Dawsonville
Dawson
70*
16
5
Douglas
Coffee
70
21
6
Gainesville
Hall
70
22
7
Macon
Bibb
70
19
5
Milledgeville
Baldwin
70
20
4
Rome
Floyd
70
24
6
Savannah
Chatham
70*
21
6
South DeKalb
DeKalb
54*
22
5
Utoy Creek
Fulton
70
23
7
Valdosta
Lowndes
70
23
6
Warner Robins
Houston
70
24
6
Yorkville
Paulding
70
16
5
* 7 additional chemicals were monitored at these locations, but that information is summarized in Table 33.
Table 23: Summary of Chemicals Analyzed in 2006
Table 24 and Table 25, on the following pages, show only the chemicals that were detected above screening values at each site in 2005 and 2006. They also provide detailed information on how often they were detected (frequency), and the overall average (mean) in micrograms per cubic meter. The number of detects were counted as any number that was above half the limit of detection. The average was computed using the sample value when it was above half the limit of detection and substituting half the limit of detection if the sample value was below this limit.
141
Location Augusta
Brunswick Columbus Dawsonville Douglas Gainesville
Macon Milledgeville
Chemical Arsenic Chromium Manganese Nickel Naphthalene Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride
Mean (g/m3)
9.9 x 10-4 2.06 x 10-3 1.38 x 10-2 3.54 x 10-3 3.88 x 10-2 9.16 x 10-1 4.41 x 10-1 2.38 x 10-1
5.7 x 10-4 2.06 x 10-3 2.95 x 10-3 5.97 x 10-1 5.04 x 10-1 2.38 x 10-1
8.3 x 10-4 1.48 x 10-3 8.62 x 10-3 8.90 x 10-1 4.41 x 10-1
8.4 x 10-4 5.29 x 10-3 9.07 x 10-3 4.79 x 10-3 4.21 x 10-1 5.04 x 10-1
9.2 x 10-4 1.58 x 10-3 4.26 x 10-1 5.04 x 10-1 2.04 x 10-1
8.0 x 10-4 7.05 x 10-3 1.24 x 10-2 5.9 x 10-3 5.15 x 10-1 5.04 x 10-1 3.39 x 10-1
7.0 x 10-4 1.82 x 10-3 7.67 x 10-3 4.80 x 10-1 5.04 x 10-1 2.38 x 10-1
6.4 x 10-4 1.52 x 10-3 1.34 x 10-2 4.58 x 10-1 4.41 x 10-1
Detection Frequency 23/27 26/27 26/27 26/27 1/28 30/30 12/30 1/30 12/25 25/25 25/25 26/27 16/27 1/27 21/30 28/30 28/30 30/30 14/30 24/29 29/29 29/29 29/29 28/29 18/29 25/29 28/29 24/27 16/27 1/27 41/41 41/41 41/41 41/41 41/41 25/41 9/41 22/29 29/29 29/29 27/28 15/28 1/28 21/30 30/30 30/30 30/30 14/30
Table 24: Site-Specific Detection Frequency and Mean Chemical Concentration, 2005
142
2006 Georgia Annual Air Quality Report
Rome Savannah South DeKalb Utoy Creek
Valdosta Warner Robins Yorkville
Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Nickel Benzene Carbon tetrachloride Tetrachloroethene 1,4-Dichlorobenzene Arsenic Cadmium Chromium Manganese Nickel Naphthalene Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Naphthalene Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride Tetrachloroethene
1.16 x 10-3 1.75 x 10-3 1.46 x 10-2 2.32 x 10-3 9.27 x 10-1 4.41 x 10-1 6.45 x 10-1
8.8 x 10-4 3.92 x 10-3 8.21 x 10-3 4.47 x 10-3 5.65 x 10-1 5.04 x 10-1
7.0 x 10-4 1.5 x 10-3 2.26 x 10-3
1.30 5.04 x 10-1 3.05 x 10-1 2.31 x 10-1
1.06 x 10-3 5.7 x 10-4 2.63 x 10-3 1.20 x 10-2 2.42 x 10-3 3.85 x 10-2 9.59 x 10-1 5.04 x 10-1 3.05 x 10-1
8.2 x 10-4 1.44 x 10-3 5.65 x 10-1 4.41 x 10-1 2.04 x 10-1
7.0 x 10-4 1.91 x 10-3 6.36 x 10-3 4.03 x 10-2 7.73 x 10-1 5.04 x 10-1 4.07 x 10-1
1.27 x 10-3 1.69 x 10-3 5.51 x 10-3 4.0 x 10-1 4.41 x 10-1 2.38 x 10-1
28/30 30/30 30/30 30/30 28/28 12/28 17/28 23/28 27/28 27/28 27/28 27/28 16/28 50/55 55/55 55/55 53/53 28/53 10/53 5/53 29/30 30/30 30/30 30/30 30/30 1/30 29/29 17/29 8/29 8/27 26/27 28/28 14/28 1/28 23/30 30/30 30/30 1/30 29/29 17/29 3/29 26/29 27/29 29/29 25/27 14/27 1/27
Table 24: Site-Specific Detection Frequency and Mean Chemical Concentration, 2005 (continued)
143
Location Augusta Brunswick Columbus
Dawsonville Douglas
Gainesville
Macon Milledgeville
Chemical Arsenic Chromium Manganese Benzene Tetrachloroethene Arsenic Chromium Benzene Arsenic Chromium Manganese Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Arsenic Chromium Manganese Benzene
Mean (g/m3) 1.39 x 10-3 1.42 x 10-3 1.18 x 10-2
1.1 8.62 x 10-1 7.4 x 10-4 1.73 x 10-3 5.15 x 10-1 9.8 x 10-4 3.18 x 10-3 1.17 x 10-2 5.57 x 10-3 8.84 x 10-1 8.62 x 10-1 7.1 x 10-4 1.73 x 10-3 5.06 x 10-3 3.22 x 10-3 4.45 x 10-1 9.0 x 10-4 2.67 x 10-3 5.15 x 10-3 3.55 x 10-3 4.26 x 10-1 7.87 x 10-1 8.6 x 10-4 2.33 x 10-3 8.29 x 10-3 4.67 x 10-3 6.45 x 10-1 7.87 x 10-1 8.53 x 10-1 2.08 x 10-3 3.67 x 10-3 9.18 x 10-3 4.62 x 10-3 5.02 x 10-1 6.1 x 10-4 1.4 x 10-3 6.68 x 10-3 5.35 x 10-1
Detection Frequency 27/31 31/31 31/31 20/30 1/30 20/28 26/28 12/29 23/31 31/31 31/31 31/31 25/31 1/31 23/29 28/29 29/29 29/29 12/30 26/29 28/29 29/29 29/29 7/30 1/30 36/42 42/42 42/42 42/42 31/43 1/43 1/43 22/29 28/29 29/29 29/29 20/29 18/28 26/28 28/28 18/29
Table 25: Site-Specific Detection Frequency and Mean Chemical Concentration, 2006
144
2006 Georgia Annual Air Quality Report
Rome Savannah South DeKalb Utoy Creek Valdosta Warner Robins Yorkville
Arsenic Chromium Manganese Benzene Dichloromethane Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene
1.17 x 10-3 1.69 x 10-3 7.2 x 10-3 7.48 x 10-1
3.67 9.59 x 10-1 8.8 x 10-4 1.72 x 10-3 5.26 x 10-3 2.45 x 10-3 6.17 x 10-1 7.87 x 10-1 8.6 x 10-4 2.79 x 10-3 3.25 x 10-3 9.45 x 10-1 8.6 x 10-1 1.09 x 10-3 4.14 x 10-3 9.63 x 10-3 4.07 x 10-3 8.25 x 10-1 7.87 x 10-1 8.71x 10-1 9.4 x 10-4 4.41 x 10-3 8.34 x 10-3 5.41 x 10-3 5.96 x 10-1 8.56 x 10-1 7.8 x 10-4 4.5 x 10-3 8.35 x 10-3 6.33 x 10-3 6.04 x 10-1 9.75 x 10-1 1.06 x 10-3 3.12 x 10-3 5.55 x 10-3 3.72 x 10-3 4.34 x 10-1
27/30 30/30 30/30 12/25 1/25 4/25 25/30 28/30 29/30 29/30 18/31 1/31 49/59 58/59 58/59 51/61 2/61 27/31 31/31 31/31 31/31 26/31 1/31 1/31 25/31 29/31 31/31 30/31 20/29 1/29 26/31 31/31 31/31 31/31 16/27 3/27 26/29 29/29 29/29 29/29 8/24
Table 25: Site-Specific Detection Frequency and Mean Chemical Concentration, 2006 (continued)
145
Formula For Calculating Risk Using IUR For Carcinogens
Risk = IUR*Conc
Formula For Calculating Hazard Quotient Using RfC For Noncarcinogens
HQ = Conc RfC
Equation Parameters Risk Theoretical lifetime cancer risk (unitless probability) HQ Hazard quotient (unitless ratio) Conc Measured ambient air concentration in g/m3 IUR Inhalation unit risk (1/(g/m3)) RfC Reference concentration (g/m3)
Figure 75: Formulas For Calculating Risk and Hazard Quotient Figure 75 shows the formulas used to calculate cancer risk and non-cancer hazard for chemicals that were carried beyond the screening process into the quantitative assessment.
146
2006 Georgia Annual Air Quality Report Table 26 and Table 27, on the following pages, show the theoretical cancer risk and noncancer hazard that would result from an individual breathing air containing the detected chemicals at the estimated concentrations daily for seventy years, or a full lifetime. These cancer risk and hazard quotient estimates are likely conservative because they were calculated assuming continuous exposure to outdoor air at breathing rates typical of moderate exertion. Real risk cannot be calculated, but may be substantially lower. Lifetime cancer risks for the limited number of chemicals exceeding screening values (and excluding that from carbonyls) exceeded 1 x 10-6 or one in one million, a value generally deemed as insignificant. However, lifetime cancer risks for these chemicals did not exceed 1 x 10-4 or one in ten thousand. This value is generally taken as a crude upper limit for "allowable" risk in many regulatory contexts. Individual hazard quotients (HQs) are ratios that relate daily exposure concentrations, or dose, to a concentration or an amount thought to be without appreciable risks of causing deleterious non-cancer effects in sensitive individuals as well as the general population. HQ values less than 1.0 indicate the air "dose" is less than the amount required to cause toxic effects other than cancer. HQs for all individual ATN chemicals (excluding carbonyls) were well below 1.0 at all of the sites monitored in 2005 and 2006.
147
Location Augusta
Brunswick Columbus Dawsonville General Coffee Gainesville
Macon Milledgeville
Chemical Arsenic Chromium Manganese Nickel Naphthalene Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride
Cancer Risk 4 x 10-6 2 x 10-5
2 x 10-6 1 x 10-6 7 x 10-6 7 x 10-6 1 x 10-6 2 x 10-6 2 x 10-5 1 x 10-6 5 x 10-6 8 x 10-6 1 x 10-6 4 x 10-6 2 x 10-5
7 x 10-6 7 x 10-6 4 x 10-6 6 x 10-5
2 x 10-6 3 x 10-6 8 x 10-6 4 x 10-6 2 x 10-5 3 x 10-6 8 x 10-6 1 x 10-6 3 x 10-6 8 x 10-5
3 x 10-6 4 x 10-6 8 x 10-6 2 x 10-6 3 x 10-6 2 x 10-5
4 x 10-6 8 x 10-6 1 x 10-6 3 x 10-6 2 x 10-5
4 x 10-6 7 x 10-6
Hazard Quotient 0.03 0.02 0.3 0.04 0.1 0.03 0.002
0.0009 0.02 0.02 0.03 0.02 0.003
0.0009 0.03 0.01 0.2 0.03 0.002 0.03 0.05 0.2 0.05 0.01 0.003 0.03 0.02 0.01 0.003
0.0008 0.03 0.07 0.2 0.07 0.02 0.003 0.001 0.02 0.02 0.2 0.02 0.003
0.0009 0.02 0.02 0.3 0.02 0.002
Table 26: Cancer Risk and Hazard Quotient by Location and Chemical, 2005
148
2006 Georgia Annual Air Quality Report
Rome Savannah South DeKalb Utoy Creek
Valdosta Warner Robins Yorkville
Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Nickel Benzene Carbon tetrachloride Tetrachloroethene 1,4-Dichlorobenzene Arsenic Cadmium Chromium Manganese Nickel Naphthalene Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Naphthalene Benzene Carbon Tetrachloride Tetrachloroethene Arsenic Chromium Manganese Benzene Carbon tetrachloride Tetrachloroethene
7 x 10-6 2 x 10-5
1 x 10-6 7 x 10-6 7 x 10-6 4 x 10-6 4 x 10-6 5 x 10-5
2 x 10-6 4 x 10-6 8 x 10-6 3 x 10-6 2 x 10-5 1 x 10-6 1 x 10-5 8 x 10-6 2 x 10-6 3 x 10-6 5 x 10-6 1 x 10-6 3 x 10-5
1 x 10-6 1 x 10-6 7 x 10-6 8 x 10-6 2 x 10-6 4 x 10-6 2 x 10-5 4 x 10-6 7 x 10-6 1 x 10-6 3 x 10-6 2 x 10-5
1 x 10-6 6 x 10-6 8 x 10-6 2 x 10-6 5 x 10-6 2 x 10-5
3 x 10-6 7 x 10-6 1 x 10-6
0.04 0.02 0.3 0.03 0.03 0.002 0.002 0.03 0.04 0.2 0.05 0.02 0.003 0.02 0.02 0.03 0.04 0.003 0.001 0.0003 0.04 0.03 0.03 0.2 0.03 0.1 0.03 0.003 0.001 0.03 0.01 0.02 0.002 0.0008 0.02 0.02 0.1 0.1 0.03 0.003 0.002 0.04 0.02 0.1 0.01 0.002 0.0009
Table 26: Cancer Risk and Hazard Quotient by Location and Chemical, 2005 (continued)
149
Location Augusta Brunswick Columbus
Dawsonville General Coffee
Gainesville
Macon Milledgeville
Chemical Arsenic Chromium Manganese Benzene Tetrachloroethene Arsenic Chromium Benzene Arsenic Chromium Manganese Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Arsenic Chromium Manganese Benzene
Cancer Risk 6 x 10-6 2 x 10-5
9 x 10-6 5 x 10-6 3 x 10-6 2 x 10-5 4 x 10-6 4 x 10-6 4 x 10-5
3 x 10-6 7 x 10-6 5 x 10-6 4 x 10-6 2 x 10-5
2 x 10-6 3 x 10-6 4 x 10-6 3 x 10-5
2 x 10-6 3 x 10-6 1 x 10-5 4 x 10-6 3 x 10-5
2 x 10-6 5 x 10-6 1 x 10-5 5 x 10-6 9 x 10-6 4 x 10-5
2 x 10-6 4 x 10-6 3 x 10-6 2 x 10-5
4 x 10-6
Hazard Quotient 0.05 0.01 0.2 0.04 0.003 0.02 0.02 0.02 0.03 0.03 0.2 0.06 0.03 0.003 0.02 0.02 0.1 0.04 0.01 0.03 0.03 0.1 0.04 0.01 0.004 0.03 0.02 0.2 0.05 0.02 0.004 0.003 0.07 0.04 0.2 0.05 0.02 0.02 0.01 0.1 0.02
Table 27: Cancer Risk and Hazard Quotient by Location and Chemical, 2006
150
2006 Georgia Annual Air Quality Report
Rome Savannah South DeKalb Utoy Creek Valdosta Warner Robins Yorkville
Arsenic Chromium Manganese Benzene Dichloromethane Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Arsenic Chromium Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Carbon tetrachloride Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene Tetrachloroethene Arsenic Chromium Manganese Nickel Benzene
5 x 10-6 2 x 10-5
6 x 10-6 2 x 10-6 6 x 10-6 4 x 10-6 2 x 10-5
1 x 10-6 5 x 10-6 1 x 10-5 4 x 10-6 3 x 10-5 2 x 10-6 7 x 10-6 5 x 10-6 5 x 10-6 5 x 10-5
2 x 10-6 6 x 10-6 1 x 10-5 5 x 10-6 4 x 10-6 5 x 10-5
3 x 10-6 5 x 10-6 5 x 10-6 3 x 10-6 5 x 10-5
3 x 10-6 5 x 10-6 6 x 10-6 5 x 10-6 4 x 10-5
2 x 10-6 3 x 10-6
0.04 0.02 0.1 0.02 0.004 0.004 0.03 0.02 0.1 0.03 0.02 0.004 0.03 0.03 0.04 0.03 0.003 0.04 0.04 0.2 0.05 0.03 0.004 0.003 0.03 0.04 0.2 0.06 0.02 0.003 0.03 0.05 0.2 0.07 0.02 0.004 0.04 0.03 0.1 0.04 0.01
Table 27: Cancer Risk and Hazard Quotient by Location and Chemical, 2006 (continued)
151
Table 28 and Table 29 show total or aggregate theoretical cancer risk and hazard indices
(added hazard quotients) for the chemicals (VOCs, semi-VOCS, and metals) carried through
the quantitative assessment. It is considered appropriate to treat the potential for effects in
an additive manner and to sum cancer risk and hazard quotients. For example, if cancer risk for two separate chemicals were 1 x 10-4 and 2 x 10-4, then the sum or aggregate cancer risk would equal 3 x 10-4. Likewise, if cancer risk for two separate chemicals were 1 x 10-4 and 1 x 10-5, then total cancer risk for the two would equal 1.1 x 10-4, or rounded to 1 x 10-4.
Similarly, if hazard quotients were 0.6 and 0.5 for two different chemicals it would indicate
that each chemical alone is not likely to result in detrimental effects. However, summing the
two would yield a Hazard Index (HI) of 1.1 or rounded to 1. Comparing this value to the
threshold value of 1.0, this HI suggests at least the potential for detrimental effects from the
combination of the two chemicals.
Location Augusta Brunswick Columbus Dawsonville Douglas Gainesville Macon Milledgeville Rome Savannah South DeKalb Utoy Creek Valdosta Warner Robins Yorkville
Cancer Risk 4 x 10-5 4 x 10-5 4 x 10-5 8 x 10-5 4 x 10-5 1 x 10-4 4 x 10-5 3 x 10-5 5 x 10-5 7 x 10-5 5 x 10-5 6 x 10-5 4 x 10-5 4 x 10-5 4 x 10-5
Hazard Index 0.5 0.09 0.3 0.3 0.06 0.4 0.3 0.4 0.4 0.3 0.1 0.5 0.06 0.3 0.2
Table 28: Aggregate Cancer Risk and Hazard Indices for Each Site, Excluding Carbonyls, 2005
152
2006 Georgia Annual Air Quality Report
Location Augusta Brunswick Columbus Dawsonville Douglas Gainesville Macon Milledgeville Rome Savannah South DeKalb Utoy Creek Valdosta Warner Robins Yorkville
Cancer Risk 4 x 10-5 3 x 10-5 6 x 10-5 3 x 10-5 5 x 10-5 6 x 10-5 6 x 10-5 3 x 10-5 4 x 10-5 4 x 10-5 5 x 10-5 8 x 10-5 7 x 10-5 7 x 10-5 5 x 10-5
Hazard Index 0.3 0.06 0.4 0.2 0.2 0.3 0.4 0.2 0.2 0.2 0.1 0.4 0.4 0.4 0.2
Table 29: Aggregate Cancer Risk and Hazard Indices for Each Site, Excluding Carbonyls, 2006
In 2005, the aggregate theoretical cancer risk (excluding carbonyls) for all ATN sites exceeded 1 x 10-6, with risks ranging from 3 x 10-5 at the Milledgeville site and 1 x 10-4 at the
Gainesville site. Hazard indices (HIs) were well below 1, with no value exceeding 0.5. In 2006, the aggregate theoretical cancer risk for the ATN sites ranged from 3 x 10-5 for Brunswick, Dawsonville, and Milledgeville to 8 x 10-5 at Utoy Creek. The hazard indices (HIs)
were also below 0.5 for 2006. These findings suggest little potential for non-cancer effects
from the chemicals assessed in this study.
Some data collected from the PAMS network was evaluated in conjunction with the ATN data. The PAMS network is a federally mandated network required to monitor for ozone precursors in those areas classified as serious, severe, or extreme for ozone non-attainment. Fifty-four (54) chemicals are monitored on six-day intervals at these sites. In Georgia, PAMS sites are located in Conyers, South DeKalb, Tucker, and Yorkville. Of the 54 chemicals monitored at these sites, many are ozone precursors, and are not truly comparable to the chemicals monitored at the ATN sites, or appropriate to evaluate as air toxics. Therefore, for this study, only twelve chemicals were assessed for their potential to have detrimental effects on human health if present in ambient air. Those twelve chemicals were benzene, cyclohexane, ethyl benzene, p-ethyltoluene, n-hexane, 1,2,3-trimethylbenzene, 1,2,4trimethylbenzene, 1,3,5-trimethylbenzene, styrene, toluene, m,p-xylene, and o-xylene.
153
Of those twelve chemicals evaluated from the PAMS network, only benzene, m/p xylenes, and 1,2,4-trimethylbenzene were found in concentrations above the screening values in 2005 and 2006. The 1,2,4-trimethylbenzene data was not reportable for some of the sites. The 1,2,4-trimethylbenzene data for the Conyers site was not reportable for 2005, and not reportable for the Conyers and South DeKalb sites in 2006, therefore it could not be evaluated. Table 30 and Table 31 show the number of samples collected, first and second highest sample values (1st and 2nd Max), averages (means) in micrograms per cubic meter, hazard quotients (HQ) and cancer risk (CR) for chemicals evaluated in the quantitative assessment at each of the four PAMS sites for 2005 and 2006. In 2005, benzene was detected consistently and when evaluated as a potential carcinogen, produced theoretical cancer risks as great as 7 x 10-5 at the South DeKalb site. Although it could not be evaluated for Conyers, 1,2,4-trimethylbenzene was detected above the screening value at all the other sites, and when evaluated as a non-cancer hazard, produced HQs ranging from 1 to 3. In 2005, the m/p-xylenes were found at the South DeKalb and Tucker sites, and produced HQs of only 0.1. In 2006, the same three PAMS compounds (benzene, m/p-xylenes, and1,2,4trimethylbenzene) were found above the screening value. Benzene was consistently found at all four PAMS sites, producing a theoretical cancer risk as great as 6 x 10-5. In 2006, the highest theoretical cancer risk values for benzene were at both the South DeKalb and Yorkville PAMS sites. For the Yorkville site, this theoretical cancer risk value for benzene increased from 2 x 10-5 in 2005 to 6 x 10-5 in 2006. For South DeKalb this theoretical cancer risk value for benzene decreased from 7 x 10-5 in 2005 to 6 x 10-5 in 2006. In 2006, the m/pxylene compounds were found only at the South DeKalb and Tucker sites, and produced a non-cancer hazard quotient of only 0.1, as was the case for 2005. Although 1,2,4trimethylbenzene could not be evaluated for the South DeKalb and Conyers sites in 2006, it was evaluated for the Tucker and Yorkville sites. It was found above the screening value only for the Tucker site in 2006. The non-cancer HQs produced for the Tucker site for 1,2,4trimethylbenzene decreased from 2 in 2005 to 1 in 2006.
Location
Chemical
Mean
Samples 1st Max 2nd Max (g/m3)
HQ
CR
Conyers
Benzene
58
15.0
8.6
4.76
0.2
4 x 10-5
South DeKalb
Benzene
54
27.2
21.1
8.78
0.3
7 x 10-5
m/p Xylenes
54
43.4
37.4
13.86
0.1
1,2,4-Trimethybenzene 54
68.8
54.1
15.83
3
Tucker
Benzene
48
51.1
13.1
8.24
0.3
6 x 10-5
m/p Xylenes
48
91.2
33.0
14.46
0.1
1,2,4-Trimethybenzene 48
78.7
63.9
15.39
2
Yorkville
Benzene
55
6.7
4.8
2.30
0.08 2 x 10-5
1,2,4-Trimethybenzene 55
20.2
19.7
6.29
1
Table 30: Summary Data for Select VOCs at PAMS Sites, 2005
154
2006 Georgia Annual Air Quality Report
Location
Chemical
#
Mean
Samples 1st Max 2nd Max (g/m3)
HQ
CR
Conyers
Benzene
59
14.7
12.1
3.74
0.1
3 x 10-5
South DeKalb Benzene
62
25.9
23.3
7.09
0.2
6 x 10-5
m/p Xylenes
62
52.1
47.8
12.51
0.1
Tucker
Benzene
59
17.6
16.6
6.87
0.2
5 x 10-5
m/p Xylenes
59
39.1
36.1
12.38
0.1
1,2,4-Trimethybenzene 59
30.0
20.7
6.93
1
Yorkville
Benzene
55
156.5
111.8
7.06
0.2
6 x 10-5
Table 31: Summary Data for Select VOCs at PAMS Sites, 2006
The carbonyls (acetaldehyde, acetone, acrolein, benzylaldehyde, butyraldehyde, formaldehyde, and propionaldehyde) were measured at only three of the ATN sites (Brunswick, Savannah, and Dawsonville) and two of the PAMS sites (South DeKalb and Tucker). For that reason, their results are displayed separately from the rest of the data. Detection frequency, average (mean) concentration in micrograms per cubic meter, cancer risk, and non-cancer HQs for the carbonyls are shown in Table 32 and Table 33. These tables also show the sum of the cancer risk and hazard quotients, which are the aggregate cancer risk and hazard index (HI), per site. Of the seven carbonyls sampled, acetaldehyde, formaldehyde, and acrolein were detected above the screening value for 2005, and only acetaldehyde and formaldehyde were detected above the screening value for 2006. Acetaldehyde and formaldehyde were evaluated as carcinogens, and acrolein as a noncarcinogen. All the sites monitoring for acetaldehyde and formaldehyde detected these compounds with a relatively high detection frequency. In 2005, acetaldehyde was detected 64% to 98% of the time, and formaldehyde was detected 75% to 100% of the time, with the Dawsonville site having the lowest detection rates and the Tucker site having the highest. However, acetaldehyde and formaldehyde had relatively low theoretical cancer risks, ranging from 1 x 10-8 to 8 x 10-6, and relatively low hazard quotients, ranging from 0.1 to 2. Acrolein was detected only at the South DeKalb and Tucker sites in 2005. Detection frequencies of acrolein were relatively low, ranging from approximately 3 to 4%. However, when acrolein was detected, the hazard quotients were quite high, at 28. In 2006, acetaldehyde was detected 61% to 100% of the time while formaldehyde was detected 77% to 100% of the time. With both compounds, the Dawsonville site had the lowest detection frequency, and the South DeKalb and Tucker sites had the highest detection frequency. The higher detection frequency at the South DeKalb and Tucker sites possibly reflects the urban setting in which these sites are located. The theoretical cancer risk and hazard quotients for these carbonyls were relatively low for 2006 as well, with theoretical cancer risks ranging from 1 x 10-8 to 8 x 10-6, and hazard quotients at 1 or below.
155
Location Brunswick Dawsonville Savannah South DeKalb
Tucker
Chemical
Acetaldehyde Formaldehyde SUM
Acetaldehyde Formaldehyde SUM
Acetaldehyde Formaldehyde SUM
Acetaldehyde Acrolein Formaldehyde SUM
Acetaldehyde Acrolein Formaldehyde SUM
Detection Frequency
Mean (g/m3)
Cancer Risk
Hazard Quotient
25/28
2.94
6 x 10-6
0.3
25/28
4.37
2 x 10-8
0.4
6 x 10-6
0.7
18/28
1.13
2 x 10-6
0.1
21/28
2.14
1 x 10-8
0.2
2 x 10-6
0.3
18/26
1.52
3 x 10-6
0.2
24/26
2.96
2 x 10-8
0.3
3 x 10-6
0.5
56/58
3.06
7 x 10-6
0.3
2/58
5.68 x 10-1
28
58/58
7.54
4 x 10-8
0.8
7 x 10-6
29
53/54
3.80
8 x 10-6
0.4
2/54
5.67 x 10-1
28
54/54
16.57
9 x 10-8
2
8 x 10-6
30
Table 32: Summary Observations, Cancer Risk, and Hazard Quotient from Carbonyls, 2005
156
2006 Georgia Annual Air Quality Report
Location Brunswick Dawsonville Savannah South DeKalb Tucker
Chemical
Acetaldehyde Formaldehyde SUM
Acetaldehyde Formaldehyde SUM
Acetaldehyde Formaldehyde SUM
Acetaldehyde Formaldehyde SUM
Acetaldehyde Formaldehyde SUM
Detection Frequency
23/29 28/29
19/31 24/31
19/28 26/28
56/56 55/55
57/57 57/57
Mean (g/m3)
1.48 3.58
1.10 1.68
1.32 2.88
2.93 8.41
4.62 13.59
Cancer Risk
6 x 10-6 2 x 10-8 6 x 10-6
2 x 10-6 1 x 10-8 2 x 10-6
3 x 10-6 2 x 10-8 3 x 10-6
7 x 10-6 4 x 10-8 7 x 10-6
8 x 10-6 9 x 10-8 8 x 10-6
Hazard Quotient
0.2 0.4 0.6
0.1 0.2 0.3
0.1 0.3 0.4
0.3 0.9 1
0.5 1 2
Table 33: Summary Observations, Cancer Risk, and Hazard Quotient from Carbonyls, 2006
157
Summary and Discussion
In both 2005 and 2006, there were 70 air toxics compounds monitored at the 15 sites across the state, with the exception of the South DeKalb site that monitored 54 air toxic compounds. In 2005, of these compounds monitored, 24 were not detected and 13 compounds were detected at three sites or less. 52.6% of the compounds detected above the screening value were in the metals category, 44.2% were in the volatile organic compounds category, and 3.2% were in the semi-volatile organic compounds category. The average number of compounds that were above the screening value for 2005 was 6.2. Even though more compounds were monitored and more compounds were detected in 2005 (with the lower detection limits that started in September of 2004), this number is not significantly higher than the average number of compounds above the screening value for the 2004 data. In 2004, the average number of compounds detected was 5.5. In 2006, 33 compounds were not detected, and 13 compounds were detected at three sites or less. In 2006, 65.9% of the compounds detected were in the metals category and 34.1% were in the volatile organic compounds category. There were no semi-volatile organic compounds found above the screening value in 2006. The average number of compounds detected above the screening value for 2006 was 5.5, the same number of compounds above the screening value for 2004.
Benzene is one of the volatile organic compounds found at all sites. In 2005, average benzene concentrations at the ATN sites ranged from 0.4 to 1.3 g/m3. These concentrations correspond to the predicted theoretical lifetime cancer risk in the range of 3 x 10-6 to 1 x 10-5. Average concentrations of benzene measured in the PAMS network ranged from 2.30 to 8.78 g/m3. These concentrations correspond to predicted theoretical lifetime cancer risks in the range of 2 x 10-5 to 7 x 10-5 for the PAMS sites. In 2006 benzene concentrations ranged from 0.43 to 1.1 g/m3 at the ATN sites, with corresponding predicted theoretical lifetime cancer risks ranging from 3 x 10-6 to 9 x10-6. At the PAMS sites for 2006, benzene concentrations ranged from 3.74 to 7.09 g/m3, with corresponding predicted lifetime cancer risks ranging from 3 x 10-5 to 6 x 10-5. Major sources of benzene to the environment include automobile service stations, exhaust from motor vehicles, and industrial emissions (ATSDR, 1997). Most data relating effects of long-term exposure to benzene are from studies of workers employed in industries that make or use benzene, where people were exposed to amounts hundreds or thousands of times greater than those reported herein. Under these circumstances of high exposure, benzene can cause problems in the blood, including anemia, excessive bleeding, and harm to the immune system. Exposure to large amounts of benzene for long periods of time may also cause cancer of the bloodforming organs, or leukemia (ATSDR, 1997). The potential for these types of health effects from exposure to low levels of benzene, as reported in this study, are not well understood. Benzene has been determined to be a known carcinogen (U.S. EPA, 2000) and was evaluated as such in this study.
In 2005, carbon tetrachloride (CCl4) was detected at all fifteen sites with a detection frequency of approximately 53%. Lifetime theoretical cancer risks calculated from the mean concentrations of CCl4 were in the range of 8 x 10-6. In 2006 carbon tetrachloride was found at only 4 sites, with a detection frequency of 2.3% to 4.2%, and lifetime theoretical cancer risks were around 1 x 10-5. CCl4 was used to produce refrigeration fluids, as propellants for aerosol cans, as a pesticide, in fire extinguishers, as a spot cleaner, and as a degreasing agent (ATSDR, 2005a). Because of concerns regarding CCl4's toxicity, these uses have been stopped or severely restricted. When exposures to CCl4 are relatively large, it can
158
2006 Georgia Annual Air Quality Report
damage the liver, the kidneys, and the nervous system. U.S. EPA has classified CCl4 as a probable human carcinogen (U.S. EPA, 1991a).
In 2005, tetrachloroethene was detected at eleven of the fifteen sites, but with a relatively low frequency at most sites. At five sites there was only one detection, and at the other six sites detection frequencies ranged from 10% to 61%. In 2006, tetrachloroethene was detected at eight of the fifteen sites, again with a low detection frequency. Five of the sites each had one detection and the other three sites had a detection frequency that ranged from 3% to 16%. Tetrachloroethene is used for dry cleaning fabrics, where it is often called perchloroethylene, referred to as "perc". It is also used as a metal degreaser. In the body, it acts as a central nervous system depressant. For this study, the chemical was evaluated as a carcinogen. Theoretical cancer risk calculated from the mean ambient air concentrations (accounting for non-detected samples) was approximately 1 x 10-6 to 4 x 10-6 for 2005 and 5 x 10-6 to 6 x 10-6 for 2006. It should be noted that almost half of these contributions arise from measurements made on one day of sampling. With that in mind, the estimate may not be a reasonable estimate of risk considering the low frequency of detection.
In 2005, two compounds were found with very low detection frequencies: naphthalene and 1,4-dichlorobenzene. Napthalene was the only semi-volatile organic compound found above the screening value in 2005. It was detected at only three of the fifteen sites, and detected only one time at each site. The theoretical lifetime cancer risk was approximately 1 x 10-6, which includes adding the half detection limit for the non-detected samples. 1,4Diclorobenzene was found at one ATN site, with approximately a 9.4% detection rate. The estimated lifetime cancer risk was 3 x 10-6. However, with such low detection numbers for both of these compounds, these may not be reasonable estimates of risk. In 2006, neither of these compounds were detected above the screening value.
In 2005 and 2006, along with benzene, there were two other VOCs detected above the screening value at the PAMS sites. These two VOCs were 1,2,4-trimethylbenzene and m/pxylenes. As stated earlier, in 2005, the Conyers 1,2,4-trimethylbenzene data was not reportable, and in 2006 the South DeKalb and Conyers sites had 1,2,4-trimethylbenzene data that was not reportable. Therefore, 1,2,4-Trimethylbenzene was assessed at three sites in 2005 and two sites in 2006. It was detected above the screening value at the three other PAMS sites in 2005, and only at the Tucker site in 2006. M/p-xylenes were detected above the screening value at only the South DeKalb and Tucker sites in both 2005 and 2006. 1,2,4Trimethylbenzene occurs naturally in coal tar and petroleum crude oil. It is a component of gasoline, and has other uses in industry as an intermediate in the production of dyes, drugs, and coatings. Exposure to very large amounts of 1,2,4-trimethylbenzene may cause skin and respiratory irritancy and nervous system depression, fatigue, headache, and drowsiness. However, risks resulting from exposure to low ambient concentrations of 1,2,4trimethylbenzene have not been studied extensively (U.S. EPA, 1994a). For this study, 1,2,4-trimethylbenzene was evaluated as a non-carcinogen with potential to cause central nervous system and irritant effects (U.S. EPA, 2004b). 1,2,4-Trimethylbenzene HQs ranged from approximately 1 to 3 for the South DeKalb, Tucker, and Yorkville sites for 2005, and was 1 for the Tucker site in 2006. M/p-xylenes were also evaluated as a non-carcinogen, and the HQs were 0.1 for each site (South DeKalb and Tucker) for both 2005 and 2006. M/p-xylenes are released into the air from auto exhaust, industrial emissions, and used as solvents. It also has the potential to cause harm to the central nervous system with long-term exposure (ATSDR, 2006b).
159
In 2005 and 2006, four metals, arsenic, chromium, manganese, and nickel, were evaluated in the quantitative assessment. Manganese was detected at eleven sites in 2005 and thirteen sites in 2006. Manganese is a trace element, and small amounts are needed to support good health. However, exposure to very large amounts through inhalation can result in neurological effects (ATSDR, 2000a). Manganese was evaluated as a neurotoxin, but did not contribute significantly in the quantitative assessment with HQs less than 0.3 for 2005 and less than 0.2 for 2006. These HQs suggest that there is little potential for neurological effects from ambient air concentrations of manganese.
In 2005 and 2006, arsenic was found at all fifteen ATN sites. Arsenic occurs naturally in soil and rocks, and was used extensively in the past as a pesticide on cotton fields and in orchards (ATSDR, 2005b). However, the majority of arsenic found in the atmosphere comes from the burning of coal and oil, incineration, and smelting operations. Arsenic has been recognized as a human poison since ancient times. Inhalation of large quantities of some forms of arsenic may cause irritation of the throat and upper respiratory tract. Long-term exposure either by inhalation or ingestion may result in a unique pattern of skin changes, and circulatory and peripheral nervous disorders (ATSDR, 2005b). Inhalation of some forms of arsenic may also cause cancer, so arsenic was evaluated as a carcinogen in this assessment. Theoretical lifetime cancer risks estimated from the data collected in 2005 ranged from 2 x 10-6 to 7 x 10-6 and from the 2006 data ranged from 3 x 10-6 to 9 x 10-6.
In both 2005 and 2006, chromium was detected at all fifteen ATN sites, with a detection frequency ranging from 93% to 100%. In 2005, the theoretical cancer risk ranged from 2 x 10-5 to 8 x 10-5, and in 2006 the theoretical cancer risk ranged from 2 x 10-5 to 5 x 10-5. In 2005, the site with the highest theoretical cancer risk was Gainesville, with 8 x 10-5, however this theoretical cancer risk dropped to 3 x 10-5 in 2006. Chromium is a naturally occurring element and is common in low amounts in foodstuffs (ATSDR, 2000b). Natural processes such as wind generating dust and even volcanoes may release chromium into the atmosphere. However, many human activities such as coal and oil combustion, electroplating, smelting, and iron and steel production also release it into the atmosphere.
The chemistry of chromium is complex. It may occur in different forms or oxidation states in the environment, having very different degrees of toxicity. Chromium+3 is the form that often predominates in the natural environment, and is also an essential element required for good nutrition. Hexavalent chromium (chromium+6) is the most toxic form of chromium and is often related to releases from industrial activities (ATSDR, 2000b). Inhaling large amounts of chromium+6 may cause upper respiratory track irritation, and chromium+6 has also been shown to be a carcinogen, causing increases in the risk of lung cancer (ATSDR, 2000b).
Studies have shown that in ambient air, even near industrial sites, chromium+6 is usually only a small portion of total chromium, with measured values for chromium+6 accounting for a range of values from 1 to 25% of total chromium (ATSDR, 2000b). At the South DeKalb site, sampling has begun for chromium+6, however the concentrations detected were below the screening value and were not evaluated further. The South DeKalb site is located within and representative of an urban area. Since the chromium+6 concentrations were below the screening value for the South DeKalb site, this could indicate that chromium+6 levels are low throughout the network. The other sites that measure for chromium, measure for the total form. Therefore, the measurements used in this study were for the total form, and distinctions cannot be made as to how much of the different states are present at the other
160
2006 Georgia Annual Air Quality Report
ATN sites. In the interest of conservativeness, chromium was evaluated with the most stringent toxicity index as chromium+6, even though the chromium metal measured was not in this most toxic form. Data collected on the ratio of chromium+6 to total chromium (ATSDR, 2000b) indicates that this process may appreciably overestimate risk. Further work is needed to better understand chemical forms of chromium in Georgia's air, and determine if chromium is an important contributor to risk.
In 2005, nickel was detected at eight of the fifteen ATN sites, with theoretical lifetime cancer risk ranging from 1 x 10-6 to 3 x 10-6. When detected, nickel had a high detection frequency, occurring in 96% to 100% of the collected samples. In 2006, nickel was detected above the screening value for eleven of the fifteen sites. The theoretical cancer risk ranged from 1 x 10-6 to 3 x 10-6, as it did for 2005. The detection frequency was also high in 2006, ranging from 97% to 100% detection, with eight of the eleven sites having 100% detection frequency. Nickel is a naturally occurring element used in many consumer and industrial products such as stainless steel, alloys, and coins, and is also released in the burning of oil and coal. If large amounts are breathed, nickel can cause damage to the lungs and nasal cavities, and can be carcinogenic (ATSDR, 2005d).
Carbonyls were monitored at five sites in Georgia in 2005 and 2006. Three sites, Brunswick, Dawsonville and Savannah are ATN sites, while the other two sites, South DeKalb and Tucker, are in the PAMS network. Three carbonyls- formaldehyde, acetaldehyde, and acrolein- were detected with sufficient frequency, and have sufficient potential for toxicity to be included in the quantitative assessment.
Formaldehyde, the simplest of the aldehydes, is produced in small quantities by natural processes, and from the fertilizer, paper, and manufactured wood products industries (ATSDR, 1999). It is also found in vehicle emissions. Formaldehyde is a health concern because of its respiratory irritancy and potential as a carcinogen. It may cause irritation of the eye, nose, throat, and skin, and has the potential under certain exposure scenarios to cause cancers of the nose and throat (ATSDR, 1999). Acetaldehyde, as an intermediate product of plant respiration and a product of incomplete combustion, is ubiquitous in the environment. Acetaldehyde, like formaldehyde, is also a concern as an upper respiratory irritant, and because of its potential to cause nasal tumors in animal studies. However, research has shown it to be significantly less potent than formaldehyde (U.S. EPA, 1987; U.S. EPA 1991b).
In 2005 and 2006, formaldehyde and acetaldehyde were detected at all five locations where carbonyls were assessed. As noted in past studies, concentrations of these aldehydes are much higher at the PAMS sites (South DeKalb and Tucker) compared to the ATN sites. The greatest difference is noted in formaldehyde concentrations, which ranged from approximately 1.68 to 4.37 g/m3 at ATN sites over the two years, compared to a range of 7.54 to 16.57 g/m3 at the PAMS sites. The reason for these differences is not clear at this time. However, it may be related to differences in siting criteria between the two networks. Type II PAMS sites are intentionally located in "urban core" locations to monitor precursors of ozone, but ATN sites are not. Because of this difference, vehicle emissions may play a greater role in measurements made at PAMS sites compared to ATN sites. When the theoretical cancer risk for formaldehyde was evaluated, the risk ranged from 1 x 10-8 to 9 x 10-8 for both 2005 and 2006, with no change in risk for each site. When acetaldehyde was
161
evaluated for theoretical cancer risk, the risk ranged from 2 x 10-6 to 7 x 10-6 for both years, also with no change from 2005 to 2006.
It should be noted that the current guidance used by EPD for the air toxics evaluation recommends using a cancer toxicity value for formaldehyde developed by U.S. EPA's Office of Air Quality Planning and Standards (OAQPS). This value, which takes into account physiologically based pharmacokinetic modeling for formaldehyde, is less restrictive by a factor of 2400 compared to the cancer toxicity value available in U.S. EPA's Integrated Risk Information System (IRIS). If the more conservative toxicity factor were used, the original theoretical risk found (1 x 10-8 to 9 x 10-8) would result in theoretical formaldehyde risk values in the range of 2 x 10-5 to 2 x 10-4 for 2005 and 2006, with formaldehyde contributing much more significantly to overall risk.
In 2005, acrolein was detected at two sites, with the detection frequency of 2% to 4%. The concentrations detected yielded values for the annual average (using half the detection limit for non-detected samples) around 0.57 g/m3. These concentrations were sufficient to yield HQs at approximately 28. In 2006, acrolein was not detected at the sites monitoring for carbonyls. Acrolein may enter the environment as a result of combustion of trees and other plants, tobacco, gasoline, and oil. Additionally, it has a number of industrial uses as a chemical intermediate (ATSDR, 2005c). The potential for acrolein to cause health effects is not well understood. At very low concentrations, it is an upper respiratory irritant. At very high concentrations it may produce more serious damage to the lining of the upper respiratory tract and lungs (ATSDR, 2005c; U.S. EPA, 2003).
As stated previously, the estimates of risk presented herein are likely overestimates due to conservative assumptions used in this exercise. Conservative assumptions were used to estimate the potential for possible exposures (high inhalation rates and long term exposure) and toxicity values. In the absence of good exposure information, this practice is warranted to decrease the potential for underestimating risk.
The results presented herein suggest that the majority of calculated risk is due to a small number of chemicals. The risk values presented in this report should not be interpreted as indicators of true or "real" risk, but for relative comparisons of a chemicals contribution to aggregate risk, or for comparisons of risk between locations within the monitoring network or in other areas of the country.
162
2006 Georgia Annual Air Quality Report
Outreach and Education
One of the most important tasks of the Ambient Monitoring Program is maintaining effective public outreach and education. The program seeks to address the air quality issues that are most vital to the citizens of Georgia by identifying the pollutants that represent the greatest risks, continually monitoring them, and communicating the monitoring results directly with the public. The goal is to provide an understanding of the presence of air pollution throughout the state and to educate the public on the steps they can take to improve air quality. This is done by issuing smog alerts and information provided in the Air Quality Index (AQI), maintaining a partnership with the Clean Air Campaign in the metro Atlanta nonattainment area, and other outreach strategies aimed at keeping the public up to date on air quality issues.
What is the Clean Air Campaign? The Clean Air Campaign (CAC) is a not-for-profit organization that works to reduce traffic congestion and improve air quality in the metro Atlanta nonattainment area through a variety of voluntary programs and services, including free employer assistance, incentive programs, public information and children's education. EPD is a proud funding sponsor of the CAC.
The CAC works with more than 300 public and private sector employers, representing several hundred thousand employees, to reduce the number of single-occupancy vehicle commuters in metro Atlanta year-round. The program has helped reduce emissions and vehicle miles traveled by encouraging people to alter their commuting habits and to reconsider behaviors-driving in particular.
In addition to addressing commuters' driving habits, CAC utilizes the Air Quality Index (AQI) to relay air quality information to metro Atlanta residents.
The Air Quality Index The Air Quality Index (AQI) is a national air standard rating system developed by the U.S. Environmental Protection Agency. The AQI is used state wide to provide the public, on a daily basis, with an analysis of air pollution levels and possible related health risks. Generally, an index scale of 0 to 500 is used to assess the quality of air, and these numbers are synchronized with a corresponding descriptor word such as: Good, Moderate, Unhealthy for Sensitive Groups, Unhealthy and Very Unhealthy. To protect public health the EPA has set an AQI value of 100 to correspond to the NAAQS for the following pollutants: Ozone (O3), Sulfur Dioxide (SO2), Carbon Monoxide (CO), Particulate Matter 10 (PM10), Particulate Matter 2.5 (PM2.5), and Nitrogen Dioxide (NO2). The AQI for a reporting region equates to the highest rating recoded for any pollutant within that region. Therefore, the larger the AQI value, the greater level of air pollution present, and the greater expectation of potential health concerns. However, this system only addresses air pollution in terms of acute health effects over time periods of 24 hours or less and does not provide an indication of chronic pollution exposure over months or years. Figure 76 shows how the recorded concentrations correspond to the AQI index values, descriptors and health advisories.
163
Maximum Pollutant Concentration
PM2.5 PM10 SO2 O3
O3 CO NO2
(24hr) (24hr) (24hr) (8hr) g/m3 g/m3 ppm ppm
(1hr) (8hr) (1hr) ppm ppm ppm AQI
Value
Descriptor
EPA Health Advisory
0 15.4
0 54
0 0.034
0 0.064
None
0 4.4
None 0 to 50
Good (green)
Air quality is considered satisfactory, and air pollution poses little or no risk.
Moderate Air quality is acceptable; however, for some pollutants there
15.5 40.4
55 154
0.035 0.144
0.065
0.084
None
4.5 9.4
None
51 to 100
(yellow)
may be a moderate health concern for a very small number of people. For example, people who are unusually sensitive to the condition of the air may experience respiratory
symptoms.
Unhealthy for Members of sensitive groups (people with lung or heart
40.5 65.4
155 254
0.145 0.224
0.085
0.104
0.125
0.164
9.5 12.4
None
101 to 150
Sensitive Groups
(orange)
disease) are at greater risk from exposure to particle pollution. Those with lung disease are at risk from exposure to ozone. The general public is not likely to be affected in this range.
65.5 150.4
255 354
0.225 0.304
0.105
0.124
0.165
0.204
12.5
15.4
None
151 to 200
Unhealthy (red)
Everyone may begin to experience health effects in this range. Members of sensitive groups may experience more serious health effects.
150.5 250.4
355 424
0.305 0.604
0.125
0.374
0.205
0.404
15.5
30.4
0.65 201 to 1.24 300
Very Unhealthy
(purple)
AQI values in this range trigger a health alert. Everyone may experience more serious health effects. When the AQI is in this range because of ozone, most people should restrict their outdoor exertion to morning or late evening hours to avoid high ozone exposures.
250.5 500.4
425 604
0.605 1.004
None
0.405
0.604
30.5
50.4
1.25 301 to 2.04 500
Hazardous (maroon)
AQI values over 300 trigger health warnings of emergency conditions. The entire population is more likely to be affected.
Figure 76: The AQI 164
2006 Georgia Annual Air Quality Report
Each day the AQI value for the metropolitan areas in Georgia are available to the public through the Environmental Protection Division's website http://www.air.dnr.state.ga.us/amp/. An analysis of AQI values for the 2006 monitoring year is listed in Table 34.
AQI Summary by Region
Number of Days
AQI Category
Good
Unhealthy Moderate for Sensitive Unhealthy
Groups
Very Unhealthy
Pollutants Monitored
(0-50) (51-100) (101-150) (151-200) (201-300) in 2006
Athens
2006 211
149
5
0
0
O3, PM10, PM2.5
Atlanta
2006 119
210
28
8
O3, SO2, CO,
0
NO2, PM10,
PM2.5
Augusta
2006 194
158
5
0
0
O3, SO2, PM10, PM2.5
Brunswick*
2006 256
128
0
0
0
O3, SO2, PM10, PM2.5
Columbus
2006 200
160
5
0
0
O3, SO2, PM10, PM2.5
Macon
2006 195
166
4
0
0
O3, PM10, PM2.5
Savannah
2006 210
155
0
0
0
O3, SO2, PM10, PM2.5
* Values do not add up to 365 because of limited monitoring in cool-weather months.
Air quality in this area is likely cleaner during these months than during the warmer months.
Table 34: AQI Summary Data, 2006
How does Georgia's Ambient Monitoring Program (AMP) Cooperate with The Clean Air Campaign (CAC)?
The Ambient Monitoring Program is responsible for measuring air pollutant levels in metro Atlanta and throughout the state. Equipment at fourteen continuous monitoring stations across metro Atlanta is used for these measurements of particulate matter (PM), sulfur dioxide (SO2), carbon monoxide (CO), nitrogen dioxide (NO2), and ozone (O3). This data is reported hourly on a website which is maintained and updated by the Ambient Monitoring Program. Based on these levels, AMP calculates the Air Quality Index (AQI), which represents overall air quality in a way that is quick and easy for the general public to understand. The Ambient Monitoring Program's website is linked to a website maintained by CAC. The AQI is then
165
displayed on The Clean Air Campaign's website. The CAC also distributes AQI information to people who have signed up to receive daily air quality forecasts via e-mail. When a smog alert is forecasted, an automated fax blast informs all local media as well. Through these connections, thousands of metro Atlanta citizens and businesses keep abreast of current air quality conditions. The Ambient Monitoring Program also encourages the public to access the CAC's website and become aware of what voluntary measures they can take to improve local air quality.
Media Outreach
The Ambient Monitoring Program is in constant touch with citizens as well as the news media through phone calls, the AMP web site and media interviews. At many times throughout the year, the demand for a story puts AMP in the limelight. The program manager and staff of the Ambient Monitoring Program make themselves available to television and newspaper reporters, thus educating the public about the AQI, the statewide air monitors, and the Clean Air Campaign.
Other Outreach Opportunities
Meteorologists In cooperation with The Clean Air Campaign, forecasters from the ambient monitoring program visit the weather centers of Atlanta's top four commercial television stations. During these visits, the group is briefed on how each station's weather team receives and uses ambient monitoring information in their daily smog forecasts. The EPD/Clean Air Campaign team provides input and direction to the weathercasters as to how they can best use the data to maximize the usefulness of this information for their viewers.
Elementary and Middle Schools Educating school children and incorporating air quality information into the classroomlearning environment is also an outreach strategy for the Ambient Monitoring Program. AMP staff visits Georgia classrooms to discuss air quality, forecasting, and monitoring. Each program presented by the AMP is designed to supplement grade-specific curricula. Learning opportunities include meteorological lessons, such as weather patterns and conditions, as well as forecasting techniques.
In many situations, these lessons involve hands-on activities and mini-field trips to the monitoring sites. High School students simulate forecasting conditions and use scientific methods to create their own forecasts. AMP Staff also participate in Career Days at both elementary and high schools to draw excitement into environmental and meteorological careers.
Colleges and Universities The Ambient Monitoring Program works with colleges and universities in several capacities. Utilizing a more technical, advanced approach, AMP has participated in several college-level seminars, providing scientific expertise on the subject of meteorology and forecasting. Through this close contact with university staff, AMP staff have co-authored scientific papers
166
2006 Georgia Annual Air Quality Report in peer-reviewed scientific journals. AMP Staff provide technical data to professors as well as students, thus incorporating real-time data into college courses and projects. Additionally, AMP contracts with Georgia Institute of Technology in a joint forecasting effort.
Monitoring Data Requests AMP also regularly receives requests for specific, detailed monitoring data from members of the research community and the broader public. Completely fulfilling the needs of these data users often also requires not just providing such data, but also providing guidance on how the data can be interpreted and what the limitations of the data set may be. We welcome these opportunities to serve the public and the research community and to ensure that the data we collect is put to its fullest and most advantageous use in protecting the health and welfare of Georgia's citizens and the state's natural environment.
EPA AIRNOW Website Georgia supplies ozone and particulate matter data to the US EPA every hour for pollution mapping activities. AIRNOW is a cooperative effort between EPA, states, and local air pollution control agencies to provide near real-time information on ground level ozone and PM2.5 concentrations. EPA uses the data to produce maps that display ozone and PM2.5 contours covering the Midwest, New England, Mid-Atlantic, southeastern, south central and Pacific coastal regions of the country. Color-coded, animated concentration gradient ozone maps are created that show daily ozone formation and transport at various spatial scales. The information is available on the EPA's AIRNOW website at: http://airnow.gov. See Figure 77 for a sample map.
Figure 77: Sample AIRNOW Ozone Concentration Map
167
The AIRNOW Data Management Center (DMC) regularly evaluates the performance of monitoring agencies that participate in the AIRNOW project based on three criteria:
1. Percent of hourly data files received 2. Average arrival time (earlier in the hour is better) 3. Percent completeness of the data within the submission files Since the last reporting period, the AIRNOW DMC has changed their evaluation system. Instead of ranking the participating agencies, they are placed into a three-tier system based on their performance for these criteria. Georgia's evaluation results are as shown in Table 35.
Evaluation Criteria Percent of hourly files received Timeliness of file submissions Data completeness within files Source: AIRNOW DMC, 2007
Ozone Season (May 1-September 30, 2006)
Top Tier Top Tier Middle Tier
PM2.5 Season (whole year) Top Tier Middle Tier Middle Tier
Table 35: AIRNOW Participation Evaluation Results
While these results are already a relatively strong showing, we believe that the larger monitoring networks are at a relative disadvantage given how these evaluation criteria were implemented. These results are among the best evaluations given to monitoring networks of Georgia's size.
EPD Website and Call-In System The Ambient Monitoring Program also provides a public-access web site with Georgiaspecific current and historical air quality data more promptly and with more detail than what is available at the AIRNOW web site. AMP's web site provides hourly information about current pollutant concentrations from Georgia's continuous and semi-continuous monitoring equipment, and is updated with each hour's data only 15 minutes after the hour ends. The site also offers downloads of bulk data, and electronic copies of Annual Reports such as this one, on a self-serve basis to facilitate research projects and satisfy public interest on these topics. Finally, the Ambient Monitoring Program also maintains an automated dial-in system that provides current air quality information for those who may not have ready access to the Internet. These resources are listed below.
Ambient Monitoring Program Web Site: http://www.air.dnr.state.ga.us/amp
Call-In System: (800)427-9605 (statewide) (404)362-4909 (metro Atlanta free calling zone)
168
2006 Georgia Annual Air Quality Report
Appendix A: Additional Criteria Pollutant Data
Carbon Monoxide (CO)
1-Hour and 8-Hour Maximum Observations
Units: parts per million
Site ID
City
County
Site Name
Hours Measured
130890002 Decatur DeKalb 131210099 Atlanta Fulton 132230003 Yorkville Paulding
South DeKalb Roswell Road
Yorkville
8651 8629 8527
Max
1 - Hour
1st
2nd
10.800 2.956
Obs. 35
0
3.5
3.2
0
1.000 0.942 0
Max 8 -
Hour
1st
2nd
2.4 2.2
Obs. 9
0
1.8 1.8 0
0.8 0.8 0
Nitrogen Dioxide (NO2)
Units: parts per million
Site ID
City
County
Site Name
130890002 130893001 131210048 132230003 132470001
Decatur Tucker Atlanta Yorkville Conyers
DeKalb DeKalb Fulton Paulding Rockdale
South DeKalb Idlewood Road Georgia Tech
Yorkville Monastery
Hours Measured
8587 8222 8049 8044 8150
Annual Arithmetic
Mean
0.0155 0.0126 0.0179 0.0034 0.0057
Obs. > Std. (0.053)
0 0 0 0 0
Nitric Oxide (NO)
Units: parts per million
Site ID
City
County Site Name
130890002 130893001 131210048 132230003 132470001
Decatur Tucker Atlanta Yorkville Conyers
DeKalb DeKalb Fulton
South DeKalb
Idlewood Road
Georgia Tech
Paulding Yorkville
Rockdale Monastery
Hours Measured
8587
8222
8049 8044 8150
1st Max
0.561 0.252 0.584 0.216 0.083
Annual Arithmetic Mean 0.0316
0.0098
0.0149 0.0052 0.0059
169
Oxides of Nitrogen (NOx)
Units: parts per million
Site ID
City
County
Site Name
130890002 130893001 131210048 132230003 132470001
Decatur Tucker Atlanta Yorkville Conyers
DeKalb DeKalb Fulton Paulding Rockdale
South DeKalb Idlewood Road Georgia Tech
Yorkville Monastery
Hours Measured
8587 8222 8049 8044 8150
1st Max 0.614 0.297 0.648 0.219 0.109
Annual Arithmetic Mean
0.0444 0.0200 0.0301 0.0057 0.0080
Reactive Oxides of Nitrogen (NOy)
Units: parts per million
Site ID
City
County
Site Name
Hours Measured 1st Max
130210013 130730001 130890002 130893001
Macon Evans Decatur Tucker
Bibb Lake Tobesofkee Columbia Riverside Park DeKalb South DeKalb DeKalb Idlewood Road
7407 8364 8049 7693
0.0958 0.0897 0.200* 0.200*
Annual Arithmetic Mean
0.00566
0.00667
0.04045 0.02413
* The NOy instrument is specialized for measurement of trace concentrations, so its range is only 00.200 ppm. Actual 1st Max appears to have exceeded the instrument's measurement range. Since all ambient concentrations exceeding the instrument's range are recorded as 0.200 instead of the actual (higher) value, the reported annual arithmetic mean may be biased slightly downward from the true concentration.
170
Sulfur Dioxide (SO2)
3-Hour and 24-Hour Maximum Observations
Units: parts per million
Site ID
City
County
130090001 130210012 130510021 130511002 131110091 131150003 131210048 131210055
Milledgeville
Baldwin
Macon
Bibb
Savannah Savannah McCaysville
Chatham Chatham Fannin
Rome Atlanta Atlanta
Floyd Fulton Fulton
Site Name
Baldwin Co. Airport Georgia Forestry Comm. East
President St. W. Lathrop & Augusta Ave.
Elementary School
Coosa Elem. School
Georgia Tech
Confederate Ave.
Hours Measured
8344 7767 8539 8478 8643 8406 8453 8383
Max 24 - Hour
1st
2nd
0.027 0.018
0.008 0.008
0.025 0.023 0.031 0.018 0.010 0.007
0.020 0.018 0.019 0.018 0.021 0.018
Obs. 0.14 0 0 0 0 0 0 0 0
Max 3 - Hour
1st
2nd
0.095 0.076
0.024 0.022
0.078 0.065 0.081 0.074 0.027 0.025
0.087 0.083 0.074 0.073 0.057 0.049
Obs. 0.5
0 0 0 0 0 0 0 0
Annual Arithmetic
Mean 0.0028
0.0020
0.0027 0.0027 0.0017
0.0030 0.0034 0.0033
Ozone (O3)
1-Hour Averages
Units: parts per million
Site ID
City
County
130210012 130210013 130510021 130550001 130590002 130670003 130730001 130770002 130850001 130890002 130893001 130970004 131130001 131210055 131270006 131350002 131510002 132130003 132150008 132151003 132230003
Macon Macon Savannah Summerville Athens Kennesaw Evans Newnan Dawsonville Decatur Tucker Douglasville Fayetteville Atlanta Brunswick Lawrenceville McDonough Chatsworth Columbus Columbus Yorkville
Bibb Bibb Chatham Chattooga Clarke Cobb Columbia Coweta Dawson DeKalb DeKalb Douglas Fayette Fulton Glynn Gwinnett Henry Murray Muscogee Muscogee Paulding
132450091 Augusta Richmond
132470001 132611001
Conyers Leslie
Rockdale Sumter
Site Name
GA Forestry Comm. Lake Tobesofkee E. President Street DNR Fish Hatchery College Station Rd.
Georgia National Guard Riverside Park
Univ. of West Georgia GA Forestry Comm.
South DeKalb Idlewood Road W. Strickland St. Georgia DOT Confederate Ave. Risley Middle School Gwinnett Tech. County Extension Office Fort Mountain Columbus Airport Columbus Crime Lab
Yorkville Bungalow Road Elementary School Conyers Monastery Leslie Community Center
Days Measured
245 243 242 245 245 245 238 245 237 243 227 245 245 245 241 244 245 228 245 245 245
239
240 242
1st Max
0.102 0.101 0.085 0.089 0.106 0.133 0.138 0.109 0.106 0.155 0.145 0.137 0.130 0.134 0.085 0.143 0.114 0.096 0.096 0.112 0.123
0.102
0.134 0.091
2nd Max 0.097 0.101 0.080 0.087 0.102 0.110 0.110 0.103 0.087 0.147 0.125 0.121 0.126 0.129 0.078 0.131 0.113 0.084 0.095 0.100 0.110
0.101
0.128 0.085
Number of Days 0.125 0 0 0 0 0 1 1 0 0 2 2 1 2 2 0 1 0 0 0 0 0
0
2 0
Ozone (O3)
8-Hour Averages
Units: parts per million
Site ID
City
County
Site Name
130210012 130210013 130510021 130550001 130590002 130670003 130730001 130770002 130850001 130890002 130893001 130970004 131130001 131210055 131270006 131350002 131510002 132130003 132150008 132151003 132230003
Macon Macon Savannah Summerville Athens Kennesaw Evans Newnan Dawsonville Decatur Tucker Douglasville Fayetteville Atlanta Brunswick Lawrenceville McDonough Chatsworth Columbus Columbus Yorkville
Bibb Bibb Chatham Chattooga Clarke Cobb Columbia Coweta Dawson DeKalb DeKalb Douglas Fayette Fulton Glynn Gwinnett Henry Murray Muscogee Muscogee Paulding
132450091 Augusta Richmond
132470001 132611001
Conyers Leslie
Rockdale Sumter
GA Forestry Comm. Lake Tobesofkee E. President Street DNR Fish Hatchery
College Station Road Georgia National Guard
Riverside Park Univ. of West Georgia GA Forestry Comm.
South DeKalb Idlewood Road W. Strickland St. Georgia DOT Confederate Ave. Risley Middle School Gwinnett Tech. County Extension Office Fort Mountain Columbus Airport Columbus Crime Lab.
Yorkville Bungalow Road Elementary School Conyers Monastery Community Center
Days Measured
244 242 240 245 244 245 236 244 236 242 226 245 244 245 239 243 245 228 245 245 242
239
237 240
1st Max 0.091 0.090 0.076 0.083 0.090 0.108 0.090 0.088 0.086 0.121 0.120 0.108 0.103 0.118 0.077 0.119 0.103 0.085 0.087 0.078 0.096
0.093
0.108 0.088
2nd Max 0.088 0.086 0.070 0.076 0.088 0.093 0.085 0.087 0.080 0.103 0.105 0.105 0.103 0.106 0.070 0.111 0.099 0.077 0.087 0.076 0.094
0.091
0.104 0.082
3rd Max 0.083 0.084 0.069 0.075 0.087 0.093 0.076 0.086 0.080 0.098 0.103 0.096 0.102 0.100 0.069 0.108 0.098 0.077 0.086 0.075 0.086
0.083
0.100 0.078
4th Max 0.077 0.083 0.069 0.073 0.086 0.093 0.074 0.086 0.079 0.096 0.094 0.095 0.090 0.092 0.069 0.096 0.095 0.075 0.080 0.073 0.080
0.083
0.099 0.077
Number of Days 0.085
2 2 0 0 5 11 2 4 1 10 8 8 6 11 0 8 9 1 3 0 3
2
10 0
Lead (Pb)
Quarterly Composite Averages
Units: micrograms per cubic meter
Site ID
City
County
Site Name
130890003 Atlanta DeKalb DMRC
132150011
Columbus
Muscogee
Cusseta School
Number of Observations
(months)
12
12
1st Quarter Composite
Avg.
0.10
0.10
2nd Quarter Composite
Avg.
0.10
0.10
3rd Quarter Composite
Avg.
0.10
0.10
4th Quarter Composite
Avg.
0.10
0.10
Number of
Values 1.5
0
0
Note: The analysis method used for lead cannot reliably distinguish values smaller than 0.20. This is known as the Method Detection Limit (MDL). In cases where the analysis results in a raw data value less than the MDL for that method, EPA requires us to report a concentration of one half the MDL. For many purposes, however, these values could alternatively be interpreted as "Not Detected".
2006 Georgia Annual Air Quality Report
Fine Particulate Matter (PM2.5)
Annual Arithmetic Mean Integrated Sampling (midnight to midnight) Using Federal Reference Method
Units: micrograms per cubic meter
Site ID
City
County
Site Name
130210007 Macon
130210012 Macon
130510017 Savannah 130510091 Savannah
130590002 Athens
130630091
Forest Park
130670003 Kennesaw
130670004
Powder Springs
130890002 Decatur
130892001 Doraville
130950007 Albany
131150005 Rome 131210032 Atlanta 131210039 Atlanta
Bibb Bibb Chatham Chatham Clarke Clayton Cobb
Cobb DeKalb DeKalb Dougherty Floyd Fulton Fulton
Allied Chemical
GA Forestry Comm. Market
St. Mercer School College Station
Rd. Georgia
DOT GA National Guard Macland Aquatic Center South DeKalb Police Dept. Turner Elem. School Coosa High E. Rivers School Fire Station 8
Days Measured
119 119 111 112 107 119 118
112 333 324 111 114 345 83
98th Percen-
tile 29.6 29.4 27.6 25.6 29.5* 31.2 31.8
33.6 32.1 28.8 34.6 33.5 32.0 32.8*
Values Exceeding Applicable
Daily Standard**
0
0
0 0
0
0
0
0
0 0
0
0 0 0
Arithmetic mean in bold indicates values above the 15.0 g/m3 annual standard.
Annual Arithmetic Mean 16.92
14.08
13.89 13.80
15.75*
16.67
16.24
15.82
15.41 14.45
14.90
15.95 15.30 18.38*
* Insufficient data to compute per EPA rules. These values are partial-year estimates. ** The daily standard was reduced from 65 to 35 g/m3 effective December 17, 2006.
175
Site ID
City
131270006 Brunswick
131350002 131390003 131530001
Lawrenceville
Gainesville
Warner Robins
131850003 Valdosta
132150001 132150008
Columbus Columbus
132150011 Columbus
132230003 132450005
Yorkville Augusta
132450091 Augusta
132950002 Rossville 133030001 Sandersville 133190001 Gordon
County
Glynn Gwinnett
Hall Houston Lowndes Muscogee Muscogee Muscogee Paulding Richmond Richmond Walker Washing-
ton Wilkinson
Site Name
Days 98th
Meas- Percen-
ured
tile
Risley
Middle
113
School
Gwinnett Tech
54
Fair St. Elem.
110
Robins AFB
60
S.L.
Mason
61
Elem.
Health Dept.
118
Columbus Airport
71
Cusseta
Rd.
111
School
Yorkville 116
Medical College
116
Bungalow
Rd.
114
School
Health Dept.
57
Health Dept.
59
Police Dept.
120
26.4 30.7 28.9 26.0 25.7 27.8 29.5 28.4* 32.4 29.9 31.1 27.3 29.8 29.5
Values Exceeding Applicable
Daily Standard**
0
0 0 0
0
0 0
0 0 0
0
0 0 0
Arithmetic mean in bold indicates values above the 15.0 g/m3 standard. * Insufficient data to compute per EPA rules. These values are partial-year estimates. ** The daily standard was reduced from 65 to 35 g/m3 effective December 17, 2006.
Annual Arithmetic Mean
11.63
16.86 13.84 14.54
12.77
15.28 16.09
14.93* 13.95 16.11
16.34
14.45 16.09 15.40
176
Fine Particulate Matter (PM2.5)
2006 Georgia Annual Air Quality Report
Annual Arithmetic Mean Semi-Continuous Measurements
Units: micrograms per cubic meter
Site ID
City
County
130210012 130511002 130590002
Macon Savannah
Athens
Bibb Chatham
Clarke
130770002 Newnan
Coweta
130890002
Decatur
DeKalb
131210055
Atlanta
Fulton
131350002 Lawrenceville Gwinnett
131510002 McDonough Henry
132150008 Columbus Muscogee
Site Name
GA Forestry Comm. Market Street College
Station Rd. Univ. of West Georgia South DeKalb
Confederate Avenue Gwinnett Tech County
Extension Office
Columbus Airport
Hours Measured
8611 8487 8498
8566
8639 8670 8458
8629
8647
98th Percen-
tile Daily Value 30.5 29.4 30.7
33.7
34.3 39.8 36.4
34.1
28.5
132230003
Yorkville
Paulding Yorkville 8514 36.7
132450091
Augusta
Richmond
Bungalow Rd. School
8272
31.9
132970001 Social Circle Walton
DNR Fish Hatchery
8616
31.3
Values Exceeding
Daily Standard*
0 0 0
0
0 0 0
0
0 0 0 0
Annual Arithmetic Mean 14.00 14.69 14.07
14.96
14.88 19.66 15.34
15.37
14.13 13.82 14.75 13.80
Arithmetic mean in bold indicates value above the 15.0 g/m3 standard. These semi-continuous methods for measuring PM2.5 are not approved for use in making attainment determinations. * The daily standard was reduced from 65 to 35 g/m3 effective December 17, 2006.
177
Particulate Matter (PM10)
24-Hour Integrated Measurements
Units: micrograms per cubic meter
Site ID
City
County
130210007 Macon 130510014 Savannah
Bibb Chatham
130550001 Summerville Chattooga
Site Name
Allied Chemical Shuman
School
DNR Fish Hatchery
Days Measured
1st Max
Number Values
150
59
72
0
58
44
0
58
67
0
Annual Arithmetic Mean 31.9
23.2
19.7
130892001 Doraville
DeKalb Police Dept.
58
50
0
22.9
130950007 Albany
Dougherty
Turner Elementary
56
57
0
29.0
130970003 Douglasville
Douglas
Beulah Pump Station
59
47
0
20.7
131150005 Rome
Floyd
Coosa High School
57
50
0
26.1
131210001 Atlanta
Fulton
Fulton Co. Health Dept.
59
48
0
23.3
131210032 Atlanta
Fulton
E. Rivers School
54
53
0
26.3
131270004 Brunswick
Glynn
Arco Pump Station
50
48
0
18.9
132150011 Columbus
Muscogee
Cusseta Rd. Elem. School
58
48
0
25.3
132450091 Augusta
Richmond
Bungalow Rd. Elem. School
51
54
0
24.5
Univ. of GA
132550002 Griffin
Spalding Experiment
58
43
0
23.3
Station
133030001 Sandersville Washington Health Dept.
59
54
0
26.4
178
2006 Georgia Annual Air Quality Report
Particulate Matter (PM10)
Hourly Averages of Semi-Continuous Measurements
Units: micrograms per cubic meter
Site ID City County Site Name
131210048 Atlanta Fulton
Georgia Tech
Hours Measured
8495
1st Max 57
Annual Arithmetic
Mean
23.1
179
180
2006 Georgia Annual Air Quality Report
Appendix B: Additional PM2.5 Particle Speciation Data
Particle Speciation- 2006 Statewide Average
Crustal 2%
Nitrate 4%
Other 13%
Ammonium Ion 10%
Elemental Carbon 5%
Sulfate 28%
Organic Carbon 38%
181
Particle Speciation- Macon 2006
Other 12%
Crustal 3%
Nitrate 4%
Ammonium Ion 8% Elemental Carbon 5%
Sulfate 26%
Organic Carbon 42%
Particle Speciation- Athens 2006
Other 13%
Crustal 1%
Nitrate 7%
Ammonium Ion 11%
Elemental Carbon 4%
Sulfate 26%
Organic Carbon 38%
182
2006 Georgia Annual Air Quality Report
Particle Speciation- General Coffee 2006
Other 15%
Crustal 3%
Nitrate 3%
Ammonium Ion 8% Elemental Carbon 4%
Sulfate 27%
Organic Carbon 40%
Particle Speciation- Atlanta 2006
Other 13% Crustal 2%
Nitrate 5%
Ammonium Ion 10%
Elemental Carbon 8%
Sulfate 28%
Organic Carbon 34%
183
Particle Speciation- Rome 2006
Other 14%
Crustal 2%
Nitrate 4%
Ammonium Ion 10%
Elemental Carbon 4%
Sulfate 30%
Organic Carbon 36%
Particle Speciation- Columbus 2006
Other 13%
Crustal 2%
Nitrate 3%
Ammonium Ion 9%
Elemental Carbon 4%
Sulfate 28%
Organic Carbon 41%
184
2006 Georgia Annual Air Quality Report
Particle Speciation- Augusta 2006
Other 14%
Crustal 2%
Nitrate 4%
Ammonium Ion 10%
Elemental Carbon 4%
Sulfate 28%
Organic Carbon 38%
Particle Speciation- Rossville 2006
Other 14%
Crustal 2%
Nitrate 6%
Ammonium Ion 11% Elemental Carbon 4%
Sulfate 30%
Organic Carbon 33%
185
186
2006 Georgia Annual Air Quality Report
Appendix C: Additional Meteorological Data
Case Study April 14, 2006
187
188
2006 Georgia Annual Air Quality Report
Case Study June 20-24, 2006
Near 100 degF
189
Ridge Axis
190
2006 Georgia Annual Air Quality Report 191
Case Study June 26, 2006
01 Hour Averages 0.5
0.36875
0.2375
0.10625
-0.025 06/26/06 00:00
06/26/06 06:00
06/26/06 12:00
06/26/06 18:00
EVANS O3 PPM
0.5 0.36875 0.2375 0.10625 -0.025
192
2006 Georgia Annual Air Quality Report 193
194
2006 Georgia Annual Air Quality Report
Appendix D: Additional PAMS Data
PAMS Continuous Hydrocarbon Data (June- August 2006)
(concentrations in parts per billion Carbon (ppbC))
Name
Site # Samples Avg.
1st Max
PAMSHC
S. DeKalb 1696
57.41
Tucker*
Conyers
2028
33.83
Yorkville
1831
20.89
TNMOC
S. DeKalb 1696
76.19
Tucker*
Conyers
2028
175.29
Yorkville
1831
23.84
Ethane
S. DeKalb 1552
3.189
Tucker*
Conyers
2028
1.893
Yorkville
1829
1.104
Ethylene
S. DeKalb 1552
2.232
Tucker*
Conyers
2028
0.455
Yorkville
1829
0.370
Propane
S. DeKalb 1552
4.394
Tucker*
Conyers
2028
2.798
Yorkville
1829
2.564
Propylene
S. DeKalb 1552
1.574
Tucker*
Conyers
2028
0.531
Yorkville
1829
0.344
Acetylene
S. DeKalb 1552
0.93
Tucker*
Conyers
2028
0.16
Yorkville
1829
0.27
n-Butane
S. DeKalb 1552
2.365
Tucker*
Conyers
2028
1.146
Yorkville
1829
0.769
Isobutane
S. DeKalb 1552
1.397
Tucker*
Conyers
2028
0.522
Yorkville
1829
0.351
*Data not reportable at this time, addendum to follow
327.2
149.4 105.5 406.0
350.0 108.4 11.06
7.77 13.97 14.84
3.09 2.77 43.31
11.43 26.89 7.49
2.36 1.80 11.7
2.9 3.2 20.18
6.01 4.87 9319
3.56 3.42
2nd Max 322.7
135.0 88.0 403.8
350.0 91.8 10.96
7.41 7.73 14.75
2.98 2.76 34.54
11.19 16.26 7.08
2.23 1.70 7.2
2.1 1.6 12.27
5.96 4.72 7.81
3.46 1.94
195
PAMS Continuous Hydrocarbon Data (June-August 2006)(continued)
(concentrations in ppbC)
Name
Site # Samples Avg.
1st Max 2nd Max
trans-2-Butene
S. DeKalb 1552
0.088
8.52
3.65
Tucker*
Conyers
2028
0.008
1.09
0.55
Yorkville
1829
0.099
2.01
1.89
cis-2-Butene
S. DeKalb 1552
0.05
2.09
1.65
Tucker*
Conyers
2028
0.006
0.49
0.48
Yorkville
1829
0.081
1.85
1.84
n-Pentane
S. DeKalb 1552
2.04
16.83
13.72
Tucker*
Conyers
2028
0.989
18.87
15.88
Yorkville
1829
0.610
3.90
3.36
Isopentane
S. DeKalb 1552
3.499
42.94
28.11
Tucker*
Conyers
2028
2.023
17.95
13.34
Yorkville
1829
1.022
8.97
7.92
1-Pentene
S. DeKalb 1552
0.053
1.13
1.09
Tucker*
Conyers
2028
0.007
0.30
0.29
Yorkville
1829
0.005
0.01
0.01
trans-2-Pentene
S. DeKalb 1552
0.097
4.3
4.29
Tucker*
Conyers
2028
0.008
0.46
0.41
Yorkville
1829
0.006
0.39
0.26
cis-2-Pentene
S. DeKalb 1552
0.044
4.83
4.81
Tucker*
Conyers
2028
0.005
0.22
0.20
Yorkville
1829
0.005
0.24
0.22
3-Methylpentane
S. DeKalb 1552
0.6
16.39
5.53
Tucker*
Conyers
2028
0.219
2.25
2.24
Yorkville
1829
0.090
1.11
1.04
n-Hexane
S. DeKalb 1625
1.264
8.28
8.2
Tucker*
Conyers
2017
0.368
3.04
2.61
Yorkville
1828
0.234
1.14
1.06
n-Heptane
S. DeKalb 1669
0.54
7.04
3.96
Tucker*
Conyers
2017
0.188
1.72
1.68
Yorkville
1828
0.018
0.52
0.50
*Data not reportable at this time, addendum to follow
196
2006 Georgia Annual Air Quality Report
PAMS Continuous Hydrocarbon Data (June-August 2006)(continued)
(concentrations in ppbC)
Name
Site # Samples Avg.
1st Max 2nd Max
n-Octane
S. DeKalb 1669
0.151
2.65
1.81
Tucker*
Conyers
2017
0.041
1.77
1.70
Yorkville
1828
0.007
0.76
0.40
n-Nonane
S. DeKalb 1669
0.173
1.68
1.55
Tucker*
Conyers
2017
0.061
0.90
0.84
Yorkville
1828
0.007
1.17
0.33
n-Decane
S. DeKalb 1669
1.122
5.33
4.71
Tucker*
Conyers
2017
0.072
2.50
2.39
Yorkville
1828
0.031
3.04
2.63
Cyclopentane
S. DeKalb 1552
0.191
14.78
4.34
Tucker*
Conyers
2028
0.052
0.60
0.60
Yorkville
1829
0.011
0.35
0.29
Isoprene
S. DeKalb 1552
6.2
46.64
37.4
Tucker*
Conyers
2028
7.847
110.93 73.50
Yorkville
1829
7.619
93.81
73.39
2,2-Dimethylbutane
S. DeKalb 1552
0.24
3.96
3.05
Tucker*
Conyers
2028
0.020
0.54
0.51
Yorkville
1829
0.010
0.36
0.33
2,4-Dimethylpentane S. DeKalb 1669
0.254
2.89
2.12
Tucker*
Conyers
2017
0.043
0.86
0.80
Yorkville
1828
0.009
2.45
2.34
Cyclohexane
S. DeKalb 1669
0.146
1.51
1.39
Tucker*
Conyers
2017
0.034
2.05
1.69
Yorkville
1828
0.008
1.42
0.71
3-Methylhexane
S. DeKalb 1669
1.222
7.49
5.35
Tucker*
Conyers
2017
0.250
2.96
2.54
Yorkville
1828
0.030
2.00
1.55
2,2,4-Trimethylpentane S. DeKalb 1669
1.866
17.27
12.21
Tucker*
Conyers
2017
0.564
4.14
4.07
Yorkville
1828
0.655
2.07
1.93
*Data not reportable at this time, addendum to follow
197
PAMS Continuous Hydrocarbon Data (June-August 2006)(continued)
(concentrations in ppbC)
Name
Site
# Samples Avg.
1st Max 2nd Max
2,3,4-Trimethylpentane S. DeKalb 1669
0.472
6.0
3.36
Tucker*
Conyers
2017
0.098
1.18
1.17
Yorkville
1828
0.023
5.11
0.54
3-Methylheptane
S. DeKalb 1669
0.184
3.7
2.05
Tucker*
Conyers
2017
0.030
0.99
0.98
Yorkville
1828
0.006
0.48
0.48
Methylcyclohexane
S. DeKalb 1669
0.245
3.13
2.48
Tucker*
Conyers
2017
0.100
1.33
1.27
Yorkville
1828
0.013
0.69
0.62
Methylcyclopentane S. DeKalb 1669
0.574
5.36
4.03
Tucker*
Conyers
2017
0.106
1.38
1.29
Yorkville
1828
0.016
3.49
0.45
2-Methylhexane
S. DeKalb 1669
0.521
5.67
4.02
Tucker*
Conyers
2017
0.133
1.82
1.58
Yorkville
1828
0.017
3.81
3.58
1-Butene
S. DeKalb 1552
0.171
2.37
1.84
Tucker*
Conyers
2028
0.148
0.53
0.52
Yorkville
1829
0.006
0.29
0.29
2,3-Dimetylbutane
S. DeKalb 1552
0.121
4.71
3.31
Tucker*
Conyers
2028
0.049
2.60
2.46
Yorkville
1829
0.020
0.52
0.45
2-Methylpentane
S. DeKalb 1552
0.673
8.71
6.06
Tucker*
Conyers
2028
0.262
2.38
2.24
Yorkville
1829
0.218
1.81
1.64
2,3-Dimethylpentane S. DeKalb 1669
0.432
3.59
2.98
Tucker*
Conyers
2017
0.096
1.53
1.23
Yorkville
1828
0.024
3.21
2.84
n-Undecane
S. DeKalb 1669
0.265
3.4
3.08
Tucker*
Conyers
2017
0.127
2.06
0.73
Yorkville
1828
0.156
3.35
3.31
*Data not reportable at this time, addendum to follow
198
2006 Georgia Annual Air Quality Report
PAMS Continuous Hydrocarbon Data (June-August 2006)(continued)
(concentrations in ppbC)
Name
Site # Samples Avg.
1st Max 2nd Max
2-Methylheptane
S. DeKalb 1669
0.112
2.84
1.64
Tucker*
Conyers
2017
0.033
1.12
1.10
Yorkville
1828
0.005
0.22
0.21
m & p Xylenes
S. DeKalb 1669
2.365
38.53
22.97
Tucker*
Conyers
2017
0.670
5.97
5.78
Yorkville
1828
0.197
1.93
1.82
Benzene
S. DeKalb 1669
1.71
13.46
10.26
Tucker*
Conyers
2017
0.560
3.34
3.19
Yorkville
1828
0.103
3.71
2.27
Toluene
S. DeKalb 1669
6.167
59.4
53.04
Tucker*
Conyers
2017
2.031
15.21
14.32
Yorkville
1828
0.819
5.14
4.81
Ethylbenzene
S. DeKalb 1669
0.892
12.03
7.07
Tucker*
Conyers
2017
0.215
2.39
2.25
Yorkville
1828
0.056
1.09
0.86
o-Xylene
S. DeKalb 1669
1.043
24.21
12.32
Tucker*
Conyers
2017
0.251
2.18
2.17
Yorkville
1828
0.063
1.29
0.70
1,3,5-Trimethylbenzene S. DeKalb 1669
0.394
3.97
2.62
Tucker*
Conyers
2017
0.188
4.01
2.95
Yorkville
1828
0.007
0.51
0.50
1,2,4-Trimethylbenzene S. DeKalb 1669
1.192
10.94
7.06
Tucker*
Conyers
2017
0.289
2.54
2.43
Yorkville
1828
0.046
0.96
0.60
n-Propylbenzene
S. DeKalb 1669
0.208
1.88
1.41
Tucker*
Conyers
2017
0.073
0.76
0.76
Yorkville
1828
0.007
0.87
0.66
Isopropylbenzene
S. DeKalb 1669
0.021
1.77
0.6
Tucker*
Conyers
2017
0.009
2.80
0.38
Yorkville
1828
0.007
2.96
0.01
*Data not reportable at this time, addendum to follow
199
PAMS Continuous Hydrocarbon Data (June-August 2006)(continued)
(concentrations in ppbC)
Name
Site # Samples Avg.
1st Max 2nd Max
o-Ethyltoluene
S. DeKalb 1669
0.351
3.06
2.31
Tucker*
Conyers
2017
0.083
0.94
0.85
Yorkville
1828
0.012
0.59
0.40
m-Ethyltoluene
S. DeKalb 1669
2.479
13.97
13.69
Tucker*
Conyers
2017
3.138
17.66
17.62
Yorkville
1828
0.616
4.34
3.47
Pinene and p-
S.DeKalb 1669
0.627
5.29
4.43
Ethyltoluene
Tucker*
Conyers
2017
0.074
7.24
6.95
Yorkville
1828
0.012
0.85
0.49
m-Diethylbenzene
S. DeKalb 1669
0.031
1.22
0.84
Tucker*
Conyers
2017
0.271
1.30
1.19
Yorkville
1828
0.008
0.53
0.50
p-Diethylbenzene
S. DeKalb 1669
0.216
3.0
1.9
Tucker*
Conyers
2017
0.056
0.79
0.78
Yorkville
1828
0.008
0.48
0.41
Styrene
S. DeKalb 1669
0.242
7.85
2.07
Tucker*
Conyers
2017
0.125
1.07
1.05
Yorkville
1828
0.025
1.16
0.77
Beta Pinene and 1,2,3- S. DeKalb 1669
2.950
24.76
21.74
Trimethylbenzene
Tucker*
Conyers
1860
4.944
49.03
30.97
Yorkville
1828
1.104
13.97
7.73
*Data not reportable at this time, addendum to follow
200
2006 Georgia Annual Air Quality Report
PAMS 2006 24-hour Canister Hydrocarbons
(concentrations in parts per billion Carbon (ppbC))
Name PAMSHC TNMOC Ethane Ethylene Propane Propylene Acetylene n-Butane Isobutane
Site
S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville S. DeKalb Tucker Conyers Yorkville
#Samples #Detects Avg. 1st Max 2nd Max
62
61 137.34 470.0 408.0
59
59 69.34 200.0 170.0
59
59 144.47 350.0 350.0
55
55 29.84 120.0 66.0
62
61 403.23 1400.0 1100.0
59
59 179.36 370.0 330.0
59
59 376.27 720.0 710.0
55
55 131.33 300.0 290.0
62
57
5.35 14.0 14.0
59
58
5.28 13.0 10.0
59
57
4.17 8.8 8.5
55
54
3.67 8.2 6.3
62
11
0.54 5.2 4.0
59
6
0.39 7.9 3.8
59
8
0.30 2.8 2.1
55
7
0.21 1.5 1.3
62
60
5.96 15.0 14.0
59
59
5.34 13.0 12.0
59
59
4.14 9.4 9.1
55
55
3.50 8.2 8.2
62
55
1.32 5.2 5.2
59
59
1.09 4.0 3.2
59
48
0.50 2.4 1.9
55
7
0.12 0.4 0.3
62
59
2.61 10.0 10.0
59
59
2.14 8.3 7.2
59
59
1.24 4.6 3.7
55
54
0.74 3.9 1.6
62
61
6.87 55.0 32.0
59
59
6.37 22.0 21.0
59
59
2.79 17.0 8.7
55
55
1.60 4.3 4.1
62
61
2.60 16.0 12.0
59
59
2.28 8.1 7.0
59
53
0.91 5.1 3.0
55
39
0.47 1.4 1.4
201
PAMS 2006 24-hour Canister Hydrocarbons (continued)
Name
(concentrations in ppbC)
Site #Samples #Detects Avg.
1st Max 2nd Max
trans-2-Butene
S. DeKalb
62
9
0.15 1.0 0.9
Tucker
59
8
0.13 0.5 0.5
Conyers
59
1
0.12 1.5
Yorkville
55
ND
cis-2-Butene
S. DeKalb
62
15
0.23 1.5 1.2
Tucker
59
7
0.12 0.3 0.3
Conyers
59
8
0.14 0.8 0.5
Yorkville
55
ND
n-Pentane
S. DeKalb
62
61 3.48 16.0 14.0
Tucker
59
59 4.10 17.0 15.0
Conyers
59
56 1.79 9.6 5.9
Yorkville
55
51 0.61 1.3 1.1
Isopentane
S. DeKalb
62
61 7.23 31.0 25.0
Tucker
59
59 6.90 18.0 18.0
Conyers
59
59 3.10 12.0 9.8
Yorkville
55
53 1.63 3.8 3.6
1-Pentene
S. DeKalb
62
21 0.20 1.3 0.8
Tucker
59
14 0.16 0.7 0.5
Conyers
59
4
0.12 0.7 0.4
Yorkville
55
ND
trans-2-Pentene
S. DeKalb
62
28 0.32 2.5 2.0
Tucker
59
29 0.29 1.2 0.9
Conyers
59
4
0.13 1.4 0.2
Yorkville
55
ND
cis-2-Pentene
S. DeKalb
62
3
0.13 1.0 0.8
Tucker
59
3
0.11 0.4 0.3
Conyers
59
ND
Yorkville
55
ND
3-Methylpentane
S. DeKalb
62
56 1.08 5.1 4.6
Tucker
59
56 0.98 3.0 2.6
Conyers
59
43 0.41 2.0 1.8
Yorkville
55
10 0.13 0.3 0.3
n-Hexane
S. DeKalb
62
56 1.18 5.2 4.9
Tucker
59
57 1.09 3.2 2.6
Conyers
59
46 0.52 3.2 2.1
Yorkville
55
21 0.16 0.4 0.4
n-Heptane
S. DeKalb
62
48 0.58 3.0 2.6
Tucker
59
53 0.62 1.9 1.6
Conyers
59
22 0.25 1.5 1.4
Yorkville
55
3
0.11 0.3 0.2
202
2006 Georgia Annual Air Quality Report
PAMS 2006 24-hour Canister Hydrocarbons (continued)
Name
(concentrations in ppbC)
Site #Samples #Detects Avg. 1st Max 2nd Max
n-Octane
S. DeKalb
62
12
0.17 1.1 0.9
Tucker
59
19
0.18 0.7 0.6
Conyers
59
4
0.13 1.1 0.7
Yorkville
55
ND
n-Nonane
S. DeKalb
62
16
0.16 1.1 0.7
Tucker
59
22
0.22 1.1 0.6
Conyers
59
3
0.12 0.9 0.5
Yorkville
55
ND
n-Decane
S. DeKalb
62
28
0.26 1.8 0.9
Tucker
59
32
0.30 1.5 1.0
Conyers
59
6
0.13 0.7 0.4
Yorkville
55
ND
Cyclopentane
S. DeKalb
62
15
0.17 0.9 0.8
Tucker
59
8
0.13 0.5 0.4
Conyers
59
1
0.1 0.2
Yorkville
55
ND
Isoprene
S. DeKalb
62
39
3.48 16.0 14.0
Tucker
59
36
2.46 12.0 12.0
Conyers
59
34
4.05 22.0 21.0
Yorkville
55
27
3.31 19.0 18.0
2,2-Dimethylbutane
S. DeKalb
62
27
0.32 2.1 2.1
Tucker
59
21
0.23 1.2 0.9
Conyers
59
19
0.47 5.4 3.8
Yorkville
55
ND
2,4-Dimethylpentane S. DeKalb
62
18
0.19 1.1 1.1
Tucker
59
8
0.13 0.6 0.5
Conyers
59
2
0.11 0.3 0.2
Yorkville
55
1
0.10 0.2
Cyclohexane
S. DeKalb
62
6
0.13 0.7 0.6
Tucker
59
5
0.12 0.5 0.4
Conyers
59
5
0.12 0.5 0.4
Yorkville
55
32
1.76 11.0 6.7
3-Methylhexane
S. DeKalb
62
61
1.29 4.1 3.7
Tucker
59
59
1.60 3.4 3.0
Conyers
59
56
1.36 15.0 12.0
Yorkville
55
51
0.65 1.6 1.4
2,2,4-Trimethylpentane S. DeKalb
62
59
1.76 7.1 6.4
Tucker
59
58
1.49 4.5 4.1
Conyers
59
49
0.61 2.4 2.2
Yorkville
55
13
0.16 0.5 0.5
203
PAMS 2006 24-hour Canister Hydrocarbons (continued)
(concentrations in ppbC)
Name
Site #Samples #Detects Avg. 1st Max 2ndMax
2,3,4-Trimethylpentane S. DeKalb
62
41
0.48 2.4 2.2
Tucker
59
40
0.37 1.5 1.4
Conyers
59
12
0.19 1.1 0.9
Yorkville
55
ND
3-Methylheptane
S. DeKalb
62
5
0.14 0.9 0.7
Tucker
59
5
0.13 0.5 0.5
Conyers
59
ND
Yorkville
55
ND
Methylcyclohexane
S. DeKalb
62
18
0.19 1.1 1.0
Tucker
59
12
0.17 0.8 0.7
Conyers
59
3
0.11 0.5 0.3
Yorkville
55
ND
Methylcyclopentane S. DeKalb
62
46
0.55 2.5 2.5
Tucker
59
42
0.45 1.5 1.3
Conyers
59
21
0.29 3.4 1.7
Yorkville
55
ND
2-Methylhexane
S. DeKalb
62
43
0.57 3.2 2.8
Tucker
59
47
0.56 2.1 1.6
Conyers
59
16
0.44 5.5 5.2
Yorkville
55
2
0.11 0.6 0.3
1-Butene
S. DeKalb
62
24
0.29 2.3 1.7
Tucker
59
19
0.22 1.1 0.9
Conyers
59
7
0.66 12.0 7.5
Yorkville
55
1
0.10 0.2
2,3-Dimenthylbutane S. DeKalb
62
38
0.42 2.6 2.3
Tucker
59
36
0.34 1.4 1.1
Conyers
59
6
0.14 0.9 0.7
Yorkville
55
ND
2-Methylpentane
S. DeKalb
62
58
2.30 8.5 7.4
Tucker
59
59
2.37 5.8 5.5
Conyers
59
51
1.12 4.0 3.4
Yorkville
55
33
0.82 4.0 3.5
2,3-Dimethylpentane S. DeKalb
62
45
0.41 1.6 1.5
Tucker
59
37
0.30 1.1 1.0
Conyers
59
21
0.19 0.7 0.5
Yorkville
55
1
0.10 0.2
n-Undecane
S. DeKalb
62
19
0.18 0.8 0.7
Tucker
59
24
0.22 0.9 0.8
Conyers
59
3
0.11 0.5 0.3
Yorkville
55
2
0.11 0.3 0.2
204
2006 Georgia Annual Air Quality Report
PAMS 2006 24-hour Canister Hydrocarbons (continued)
(concentrations in ppbC)
Name
Site #Samples #Detects Avg. 1st Max 2ndMax
2-Methylheptane
S. DeKalb
62
7
0.14 0.8 0.7
Tucker
59
6
0.13 0.5 0.5
Conyers
59
3
0.12 0.6 0.5
Yorkville
55
ND
m & p Xylenes
S. DeKalb
62
60
2.88 12.0 11.0
Tucker
59
59
2.85 9.0 8.3
Conyers
59
53
1.24 14.0 5.7
Yorkville
55
37
0.32 0.8 0.8
Benzene
S. DeKalb
62
61
2.22 8.1 7.3
Tucker
59
59
2.15 5.5 5.2
Conyers
59
59
1.17 4.6 3.8
Yorkville
55
53
2.21 49.0 35.0
Toluene
S. DeKalb
62
61
7.41 26.0 25.0
Tucker
59
59
7.42 19.0 16.0
Conyers
59
59
3.20 10.0 10.0
Yorkville
55
55
5.24 15.0 14.0
Ethylbenzene
S. DeKalb
62
50
0.78 3.5 3.4
Tucker
59
49
0.77 2.6 2.3
Conyers
59
24
0.26 2.1 1.6
Yorkville
55
ND
o-Xylene
S. DeKalb
62
54
0.96 4.2 4.1
Tucker
59
55
0.97 3.2 3.0
Conyers
59
36
0.53 11.0 3.6
Yorkville
55
2
0.10 0.2 0.2
1,3,5-Trimethylbenzene S. DeKalb
62
36
0.33 1.9 1.6
Tucker
59
45
0.46 1.8 1.5
Conyers
59
17
0.20 0.7 0.7
Yorkville
55
ND
1,2,4-Trimethylbenzene S. DeKalb
62
++
Tucker
59
56
1.41 6.1 4.2
Conyers
59
++
Yorkville
55
9
0.13 0.4 0.3
n-Propylbenzene
S. DeKalb
62
9
0.15 0.9 0.7
Tucker
59
12
0.16 0.9 0.6
Conyers
59
4
0.12 0.5 0.5
Yorkville
55
ND
Isopropylbenzene
S. DeKalb
62
4
0.11 0.6 0.3
Tucker
59
2
0.11 0.6 0.2
Conyers
59
1
0.11 0.5
Yorkville
55
2
0.11 0.6 0.3
++ Data not reportable
205
PAMS 2006 24-hour Canister Hydrocarbons (continued)
(concentrations in ppbC)
Name
Site #Samples #Detects Avg. 1st Max 2ndMax
o-Ethyltoluene
S. DeKalb
62
30
0.32 1.9 1.2
Tucker
59
34
0.47 1.9 1.8
Conyers
59
17
0.21 1.9 1.4
Yorkville
55
9
0.13 0.7 0.7
m-Ethyltoluene
S. DeKalb
62
48
0.74 3.7 3.2
Tucker
59
52
0.84 3.7 2.8
Conyers
59
25
0.23 1.1 1.1
Yorkville
55
ND
p-Ethyltoluene
S. DeKalb
62
38
0.50 2.0 1.8
Tucker
59
41
0.40 2.0 1.4
Conyers
59
38
0.44 1.8 1.8
Yorkville
55
5
0.11 0.3 0.2
m-Diethylbenzene
S. DeKalb
62
21
0.20 1.0 1.0
Tucker
59
24
0.23 0.7 0.6
Conyers
59
25
0.25 1.3 1.1
Yorkville
55
28
0.39 1.5 1.1
p-Diethylbenzene
S. DeKalb
62
11
0.16 1.1 0.7
Tucker
59
15
0.18 1.0 0.8
Conyers
59
3
0.12 0.6 0.5
Yorkville
55
ND
Styrene
S. DeKalb
62
22
0.28 1.5 1.2
Tucker
59
40
0.74 3.1 2.8
Conyers
59
27
0.60 6.4 6.2
Yorkville
55
34
0.61 2.2 1.9
1,2,3-Trimethylbenzene S. DeKalb
62
33
0.33 1.6 1.4
Tucker
59
28
0.30 1.5 1.4
Conyers
59
13
0.20 1.3 1.1
Yorkville
55
ND
ND indicates no detection
206
2006 Georgia Annual Air Quality Report
Appendix E: Additional Toxics Data
(concentrations in micrograms per cubic meter (g/m3))
Name
Site
Antimony
Milledgeville
Macon
Savannah
General Coffee
Dawsonville
South DeKalb**
Rome
Utoy Creek
Brunswick
Gainesville
Warner Robins
Valdosta
Columbus
Yorkville
Augusta
Arsenic
Milledgeville
Macon
Savannah
General Coffee
Dawsonville
South DeKalb**
Rome
Utoy Creek
Brunswick
Gainesville
Warner Robins
Valdosta
Columbus
Yorkville
Augusta
Beryllium
Milledgeville
Macon
Savannah
General Coffee
Dawsonville
South DeKalb**
2006 Metals
Total Samples
28 29 30 29 29 59 30 31 28 42 31 31 31 29 31 28 29 30 29 29 59 30 31 28 42 31 31 31 29 31 28 29 30 29 29 59
# of Detects
27 28 29 27 26 57 30 31 28 42 29 29 29 29 31 18 22 25 26 23 49 27 27 20 36 26 25 23 26 27 ND ND ND ND ND ND
Avg. 0.00068 0.00090 0.00081 0.00050 0.00064 0.00128 0.00146 0.00167 0.00047 0.00117 0.00074 0.00071 0.00118 0.00069 0.00271 0.00061 0.00208 0.00088 0.00090 0.00071 0.00086 0.00117 0.00109 0.00074 0.00086 0.00078 0.00094 0.00098 0.00106 0.00139
1st Max 0.00220 0.00434 0.00266 0.00418 0.00225 0.00515 0.00389 0.00367 0.00206 0.00462 0.00192 0.00164 0.00300 0.00244 0.01189 0.00125 0.04310 0.00271 0.00206 0.00138 0.00874 0.00263 0.00226 0.00516 0.00182 0.00323 0.00208 0.00349 0.00210 0.01048
2nd Max 0.00142 0.00231 0.00241 0.00126 0.00168 0.00472 0.00301 0.00351 0.00132 0.00229 0.00179 0.00144 0.00250 0.00186 0.00854 0.00104 0.00114 0.00209 0.00178 0.00137 0.00209 0.00254 0.00223 0.00118 0.00157 0.00203 0.00198 0.00305 0.00196 0.00274
207
2006 Metals (continued)
(concentrations in g/m3)
Total
# of
Name
Site
Samples Detects Avg.
Beryllium
Rome
30
ND
(continued) Utoy Creek
31
ND
Brunswick
28
ND
Gainesville
42
ND
Warner Robins
31
1
0.00003
Valdosta
31
ND
Columbus
31
ND
Yorkville
29
ND
Augusta
31
ND
Cadmium
Milledgeville
28
28 0.00029
Macon
29
29 0.00029
Savannah
30
29 0.00029
General Coffee
29
29 0.00019
Dawsonville
29
28 0.00016
South DeKalb**
59
58 0.00017
Rome
30
30 0.00024
Utoy Creek
31
31 0.00025
Brunswick
28
28 0.00033
Gainesville
42
42 0.00025
Warner Robins
31
31 0.00030
Valdosta
31
31 0.00015
Columbus
31
31 0.00019
Yorkville
29
29 0.00021
Augusta
31
31 0.00026
Chromium
Milledgeville
28
26 0.00140
Macon
29
28 0.00367
Savannah
30
28 0.00172
General Coffee
29
28 0.00267
Dawsonville
29
28 0.00173
South DeKalb**
59
58 0.00279
Rome
30
30 0.00169
Utoy Creek
31
31 0.00414
Brunswick
28
26 0.00173
Gainesville
42
42 0.00233
Warner Robins
31
31 0.00450
Valdosta
31
29 0.00441
Columbus
31
31 0.00318
Yorkville
29
29 0.00312
Augusta
31
31 0.00142
Chromium+6* South DeKalb
57
52 0.00005
1st Max
0.00006
0.00120 0.00177 0.00130 0.00072 0.00125 0.00191 0.00140 0.00102 0.00216 0.00123 0.00147 0.00060 0.00059 0.00109 0.00134 0.00284 0.03921 0.00803 0.01868 0.01340 0.04708 0.00994 0.02881 0.01692 0.02071 0.06513 0.04743 0.04350 0.05167 0.00213 0.00030
2nd Max
0.00113 0.00122 0.00067 0.00046 0.00030 0.00126 0.00124 0.00082 0.00107 0.00093 0.00141 0.00036 0.00050 0.00109 0.00133 0.00240 0.02516 0.00314 0.01196 0.00304 001268 0.00299 0.01660 0.00289 0.01447 0.01723 0.02578 0.00913 0.00508 0.00212 0.00020
208
2006 Georgia Annual Air Quality Report
2006 Metals (continued)
(concentrations in g/m3)
Total
# of
Name
Site
Samples Detects Avg.
Cobalt
Milledgeville
28
11 0.00008
Macon
29
16 0.00019
Savannah
30
17 0.00015
General Coffee
29
15 0.00013
Dawsonville
29
10 0.00010
South DeKalb**
59
20 0.00009
Rome
30
17 0.00012
Utoy Creek
31
27 0.00018
Brunswick
28
7 0.00007
Gainesville
42
29 0.00016
Warner Robins
31
12 0.00014
Valdosta
31
14 0.00014
Columbus
31
23 0.00019
Yorkville
29
12 0.00011
Augusta
31
24 0.00014
Lead
Milledgeville
28
28 0.00296
Macon
29
29 0.00278
Savannah
30
29 0.00362
General Coffee
29
29 0.00347
Dawsonville
29
29 0.00230
South DeKalb**
59
58 0.00265
Rome
30
30 0.00414
Utoy Creek
31
31 0.00336
Brunswick
28
28 0.00221
Gainesville
42
42 0.00312
Warner Robins
31
31 0.00270
Valdosta
31
31 0.00281
Columbus
31
31 0.00421
Yorkville
29
29 0.00252
Augusta
31
31 0.00450
Manganese Milledgeville
28
28 0.00668
Macon
29
29 0.00918
Savannah
30
29 0.00526
General Coffee
29
29 0.00515
Dawsonville
29
29 0.00506
South DeKalb**
59
58 0.00367
Rome
30
30 0.00720
Utoy Creek
31
31 0.00963
Brunswick
28
28 0.00354
1st Max 0.00023 0.00177 0.00145 0.00044 0.00074 0.00059 0.00034 0.00046 0.00024 0.00118 0.00094 0.00101 0.00100 0.00076 0.00025 0.00791 0.00638 0.01455 0.03041 0.00986 0.03974 0.01242 0.00637 0.00703 0.00667 0.00547 0.00743 0.00931 0.00508 0.01480 0.01359 0.02903 0.00959 0.01242 0.01737 0.01922 0.01696 0.01714 0.00703
2nd Max 0.00018 0.00061 0.00023 0.00043 0.00026 0.00029 0.00025 0.00041 0.00013 0.00037 0.00081 0.00042 0.00075 0.00018 0.00025 0.00477 0.00411 0.00608 0.00760 0.00414 0.00437 0.00698 0.00579 0.00466 0.00541 0.00442 0.00580 0.00886 0.00466 0.00953 0.01326 0.01711 0.00958 0.01216 0.01287 0.01082 0.01690 0.01646 0.00700
209
2006 Metals (continued)
(concentrations in g/m3)
Total
# of
Name
Site
Samples Detects Avg.
Manganese Gainesville
42
42 0.00829
continued
Warner Robins
31
31 0.00835
Valdosta
31
31 0.00834
Columbus
31
31 0.01169
Yorkville
29
29 0.00555
Augusta
31
31 0.01177
Nickel
Milledgeville
28
27 0.00117
Macon
29
29 0.00462
Savannah
30
29 0.00245
General Coffee
29
29 0.00355
Dawsonville
29
29 0.00322
South DeKalb**
59
58 0.00325
Rome
30
30 0.00173
Utoy Creek
31
31 0.00407
Brunswick
28
28 0.00208
Gainesville
42
42 0.00467
Warner Robins
31
31 0.00633
Valdosta
31
30 0.00541
Columbus
31
31 0.00557
Yorkville
29
29 0.00372
Augusta
31
31 0.00172
Selenium
Milledgeville
28
27 0.00111
Macon
29
28 0.00094
Savannah
30
28 0.00070
General Coffee
29
27 0.00059
Dawsonville
29
27 0.00075
South DeKalb**
59
56 0.00080
Rome
30
30 0.00106
Utoy Creek
31
31 0.00175
Brunswick
28
27 0.00054
Gainesville
42
42 0.00097
Warner Robins
31
30 0.00086
Valdosta
31
29 0.00060
Columbus
31
28 0.00090
Yorkville
29
28 0.00102
Augusta
31
31 0.00123
1st Max 0.02860 0.02768 0.03935 0.03335 0.02611 0.02627 0.00250 0.03972 0.00954 0.02129 0.05955 0.04708 0.00994 0.03643 0.01598 0.09369 0.07324 0.08073 0.06982 0.06527 0.00554 0.00378 0.00284 0.00186 0.00192 0.00189 0.00262 0.00259 0.00391 0.00128 0.00297 0.00273 0.00183 0.00380 0.00207 0.00349
2nd Max 0.01788 0.02302 0.02009 0.02999 0.01304 0.02601 0.00201 0.03429 0.00568 0.01518 0.00358 0.02684 0.00396 0.01390 0.00336 0.02556 0.07056 0.02925 0.02771 0.00965 0.00397 0.00243 0.00152 0.00170 0.00145 0.00150 0.00243 0.00234 0.00380 0.00124 0.00211 0.00162 0.00118 0.00229 0.00202 0.00329
210
2006 Georgia Annual Air Quality Report
(concentrations in g/m3)
Name
Site
Zinc
Milledgeville
Macon
Savannah
General Coffee
Dawsonville
South DeKalb**
Rome
Utoy Creek
Brunswick
Gainesville
Warner Robins
Valdosta
Columbus
Yorkville
Augusta
2006 Metals (continued)
Total
# of
Samples Detects Avg.
28
28
0.01531
29
29
0.02288
30
29
0.01649
29
29
0.02526
29
29
0.02060
59
58
0.01364
30
30
0.02434
31
31
0.05110
28
28
0.01602
42
42
0.03357
31
31
0.01758
31
31
0.02265
31
31
0.03475
29
29
0.01794
31
31
0.02394
1st Max 0.02564 0.05070 0.02723 0.08946 0.18794 0.08346 0.04500 0.28105 0.03414 0.36650 0.04028 0.04743 0.08063 0.05580 0.08166
2nd Max 0.02441 0.04895 0.02674 0.06051 0.05955 0.02814 0.03216 0.18954 0.02817 0.10153 0.03749 0.03998 0.06929 0.03318 0.03760
*Hexavalent Chromium, sample collected every 6 days ** Hi-Vol PM10 selected Total Suspended Particulates, sample collected every 6 days
211
2006 Semi-Volatile Compounds
(concentrations in g/m3)
Total # of
Name
Site Samples Detects Avg.
Acenaphthene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner Robins
30
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Acenaphthylene
Milledgeville 29
ND
Macon
30
ND
Savannah
28
ND
General
Coffee
29
ND
Dawsonville 29
ND
Rome
27
ND
Utoy Creek
30
ND
Brunswick
19
ND
Gainesville
40
ND
Warner
Robins
29
ND
Valdosta
30
ND
Columbus
30
ND
Yorkville
27
ND
Augusta
30
ND
Anthracene
Milledgeville 28
ND
Macon
28
ND
Savannah
26
ND
General Coffee
28
ND
Dawsonville 28
ND
Rome
27
ND
Utoy Creek
29
ND
Brunswick
17
ND
1st Max 2nd Max
212
2006 Georgia Annual Air Quality Report
2006 Semi-Volatile Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Anthracene (continued) Gainesville
38
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
29
ND
Yorkville
26
ND
Augusta
29
ND
Benzo(a)anthracene
Milledgeville 30
1 0.000146 0.00208
Macon
31
1 0.000153 0.00236
Savannah
29
ND
General
Coffee
30
1 0.000140 0.00192
Dawsonville 30
ND
Rome
28
2 0.000159 0.00222
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner Robins
30
1 0.000081 0.00013
Valdosta
31
ND
Columbus
31
1 0.000148 0.00223
Yorkville
28
ND
Augusta
31
1 0.000135 0.00181
Benzo(b)fluoranthene
Milledgeville 29
1 0.000107 0.00091
Macon
31
1 0.000114 0.00117
Savannah
29
ND
General Coffee
30
1 0.000104 0.00083
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner Robins
30
ND
Valdosta
31
ND
Columbus
31
1 0.000112 0.0011
Yorkville
28
ND
Augusta
31
1 0.000103 0.00081
2nd Max 0.00019
213
2006 Semi-Volatile Compounds (continued)
(concentrations in g/m3) Name
Total # of
Site
Samples Detects Avg. 1st Max
Benzo(k)fluoranthene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
1 0.000081 0.00015
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
1 0.000083 0.00016
Gainesville
41
ND
Warner
Robins
30
1 0.000081 0.00015
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Benzo(a)pyrene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
1 0.000317 0.00035
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
1 0.000325 0.00059
Augusta
31
ND
Benzo(e)pyrene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
1 0.000079 0.00008
Rome
31
ND
Utoy Creek
31
ND
Brunswick
20
ND
2nd Max
214
2006 Georgia Annual Air Quality Report
2006 Semi-Volatile Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Benzo(e)pyrene (continued) Gainesville
41
1 0.000079 0.00008
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Benzo(g,h,i)perylene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Chrysene
Milledgeville 31
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
2nd Max
215
2006 Semi-Volatile Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Dibenzo(a,h)anthracene Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Fluoranthene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
1 0.000325 0.00058
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
1 0.000317 0.00034
Yorkville
28
ND
Augusta
31
ND
Fluorene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
2nd Max
216
2006 Georgia Annual Air Quality Report
2006 Semi-Volatile Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Fluorene (continued)
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Indeno(1,2,3-cd)pyrene Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
Naphthalene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
2nd Max
217
2006 Semi-Volatile Compounds (continued)
(concentrations in g/m3) Name
Total # of
Site
Samples Detects Avg. 1st Max
Phenanthrene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
1 0.003201 0.00446
Yorkville
28
ND
Augusta
31
ND
Pyrene
Milledgeville 30
ND
Macon
31
ND
Savannah
29
ND
General
Coffee
30
ND
Dawsonville 30
ND
Rome
28
ND
Utoy Creek
31
ND
Brunswick
20
ND
Gainesville
41
ND
Warner
Robins
30
ND
Valdosta
31
ND
Columbus
31
ND
Yorkville
28
ND
Augusta
31
ND
2nd Max
ND indicates no detection
218
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg.
Freon 113
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Freon 114
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
1,3-Butadiene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
1st Max
2nd Max
219
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg. 1st Max
1,3-Butadiene
Rome
25
ND
(continued)
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Cyclohexane
Milledgeville
29
16 0.8936 6.1988
Macon
29
2
0.6181 5.8544
Savannah
31
4
0.4805 1.1709
General
Coffee
30
1
0.4471 0.9298
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
3
0.9656 13.7751
Utoy Creek
31
ND
Brunswick
29
1
0.5166 2.9272
Gainesville
43
ND
Warner
Robins
27
1
0.4413 0.7232
Valdosta
29
1
0.4631 1.3775
Columbus
31
3
0.4799 1.3431
Yorkville
24
ND
Augusta
30
7
0.5527 2.3418
Chloromethane
Milledgeville
29
29 1.0263 1.3012
Macon
29
29 1.0947 1.5904
Savannah
31
31 1.1033 1.5697
General
Coffee
30
30 1.1601 1.6317
Dawsonville
30
30 0.9831 1.2599
South DeKalb* 61
61 1.0053 1.5491
Rome
25
25 0.9749 1.5697
Utoy Creek
31
31 1.0061 1.6110
Brunswick
29
29 1.0876 1.4458
Gainesville
43
43 0.9732 1.2599
Warner
Robins
27
27 1.0434 1.4045
Valdosta
29
29 1.0847 1.4871
Columbus
31
31 1.0167 1.6524
2nd Max
1.7219 0.4477 0.9987
0.4477
0.8265 1.2398 1.2806 1.4871 1.5284 1.6317 1.2393 1.5284 1.2393 1.3425 1.4252 1.2599 1.2806 1.3425 1.2393
220
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Chloromethane
(continued)
Yorkville
24
24 0.9269 1.4665
Augusta
30
30 1.0740 1.5284
Dichloromethane
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
1
3.6668 8.3337
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Chloroform
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
1
0.6108 0.6348
Rome
25
ND
Utoy Creek
31
2
0.6246 0.9767
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
1
0.6180 0.8302
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Carbon tetrachloride Milledgeville
29
ND
Macon
29
ND
Savannah
31
1
0.7873 0.8178
2nd Max 1.2599 1.4871
0.6837
221
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg. 1st Max
Carbon tetrachloride General Coffee 30
1
0.7873 0.8178
(continued)
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
1
0.7873 0.8178
Brunswick
29
ND
Gainesville
43
1
0.7870 0.8178
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Trichlorofluoromethane Milledgeville
29
28 1.1869 1.7983
Macon
29
29 1.3255 1.8545
Savannah
31
30 1.2282 2.2479
General
Coffee
30
29 1.2429 1.8545
Dawsonville
30
30 1.2588 1.6927
South DeKalb* 61
60 1.3307 2.2479
Rome
25
25 1.2835 2.0231
Utoy Creek
31
31 1.2907 2.3603
Brunswick
29
28 1.1947 1.7421
Gainesville
43
43 1.5970 2.4164
Warner
Robins
27
27 1.2301 1.6297
Valdosta
29
29 1.2693 2.0231
Columbus
31
30 1.2572 2.0231
Yorkville
24
23 1.1110 1.4611
Augusta
30
30 1.2719 1.9669
Chloroethane
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
2nd Max
1.5735 1.7983 1.7983 1.7983 1.5735 1.9669 1.8545 2.0231 1.5735 2.2479 1.5735 1.8545 1.7983 1.4611 1.7983
222
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Chloroethane
Warner Robins 27
ND
(continued)
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
1,1-Dichloroethane
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Methyl chloroform
Milledgeville
29
17 2.3593 6.0016
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
30 13.0035 41.4658
2nd Max 5.4560 30.0082
223
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg.
1st Max
Ethylene dichloride
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Tetrachloroethylene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
2
0.8599 1.4240
Rome
25
4
0.9589 1.9665
Utoy Creek
31
1
0.8706 1.5597
Brunswick
29
ND
Gainesville
43
1
0.8532 1.0850
Warner
Robins
27
3
0.9745 2.5769
Valdosta
29
1
0.8558 1.0850
Columbus
31
1
0.8619 1.2884
Yorkville
24
ND
Augusta
30
1
0.8623 1.2884
1,1,2,2-
Tetrachloroethane
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
2nd Max
1.0172 1.8309 2.5090
224
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
1,1,2,2-
Tetrachloroethane
South DeKalb* 61
ND
(continued)
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Bromomethane
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
1,1,2-Trichloroethane Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
2nd Max
225
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg. 1st Max
1,1,2-Trichloroethane Warner
(continued)
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Dichlorodifluoromethane Milledgeville
29
29 2.0598 3.0658
Macon
29
29 2.3053 3.4119
Savannah
31
31 2.0704 3.3130
General
Coffee
30
30 2.1807 3.1152
Dawsonville
30
30 2.1526 2.7691
South DeKalb* 61
61 2.8615 47.9644
Rome
25
25 2.0847 3.3130
Utoy Creek
31
31 2.1390 3.5603
Brunswick
29
29 2.1297 3.4119
Gainesville
43
43 2.1458 3.0163
Warner
Robins
27
27 2.1556 3.1152
Valdosta
29
29 2.1263 3.1152
Columbus
31
31 2.1948 3.4613
Yorkville
24
24 1.9346 2.6702
Augusta
30
30 2.1938 2.9174
Trichloroethylene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
2nd Max
2.7196 2.9669 3.2141
2.9174 2.7691 3.1152 2.9669 3.0163 2.8185 2.9669
2.8680 3.0163 3.4119 2.4724 2.8185
226
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
1,1-Dichloroethylene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
1,2-Dichloropropane Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
2nd Max
227
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
trans-1,3-
Dichloropropylene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
cis-1,3-
Dichloropropylene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
2nd Max
228
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
cis-1,2-Dichloroethene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Ethylene dibromide
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Hexachlorobutadiene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
2nd Max
229
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg. 1st Max
Hexachlorobutadiene Rome
25
ND
(continued)
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Vinyl chloride
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
m/p Xylene
Milledgeville
29
3
0.5677 0.8687
Macon
29
1
0.5467 0.6515
Savannah
31
9
0.6964 1.7809
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
32 0.9716 4.3436
Rome
25
10 0.8835 2.3890
Utoy Creek
31
12 0.8708 3.6052
Brunswick
29
4
0.5804 1.1293
Gainesville
43
15 0.7364 2.3890
Warner
Robins
27
3
0.6161 2.2152
Valdosta
29
7
0.6500 2.0415
Columbus
31
13 0.8827 2.3890
2nd Max
0.8687 1.5637 3.1274 2.3890 2.4324 0.9121 2.3021 0.7818 1.1728 2.3455
230
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
m/p Xylene (continued) Yorkville
24
ND
Augusta
30
14 1.3610 6.0810
Benzene
Milledgeville
29
18 0.5348 0.8944
Macon
29
20 0.5017 0.8305
Savannah
31
18 0.6167 1.7888
General
Coffee
30
7
0.4264 0.6708
Dawsonville
30
12 0.4451 0.8305
South DeKalb* 61
51 0.9447 3.5137
Rome
25
12 0.7481 2.7790
Utoy Creek
31
26 0.8248 2.5554
Brunswick
29
12 0.5149 1.3097
Gainesville
43
31 0.6448 1.7888
Warner
Robins
27
16 0.6040 2.2360
Valdosta
29
20 0.5964 1.4374
Columbus
31
25 0.8841 2.0763
Yorkville
24
8
0.4339 0.6069
Augusta
30
20 1.1010 3.8331
Toluene
Milledgeville
29
22 1.0359 2.5991
Macon
29
17 0.7469 1.9211
Savannah
31
21 1.4922 9.4172
General
Coffee
30
3
0.4778 0.6027
Dawsonville
30
9
0.4966 0.6780
South DeKalb* 61
56 2.6124 10.1706
Rome
25
23 2.6850 10.9239
Utoy Creek
31
29 15.8111 41.4359
Brunswick
29
17 0.8755 4.1436
Gainesville
43
36 1.2952 5.6503
Warner
Robins
27
18 0.9982 5.6503
Valdosta
29
19 1.0339 5.2736
Columbus
31
27 1.9442 7.1571
Yorkville
24
17 1.6284 6.7804
Augusta
30
20 2.6356 12.0540
2nd Max
5.6466 0.8305 0.7347 1.2458
0.6389 0.6708 3.1943 1.6291 2.4277 0.8305 1.3416
1.1499 1.3735 2.0763 0.5750 3.8331 2.2978 1.4314 4.8969
0.5274 0.6027 9.4172 6.7804 37.6687 2.1848 4.8969
2.6368 3.7669 4.5202 5.6503 10.1706
231
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg. 1st Max
Ethylbenzene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
1
0.5450 0.6081
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
10 0.5800 1.4334
Rome
25
2
0.5530 0.6950
Utoy Creek
31
2
0.5640 1.1293
Brunswick
29
ND
Gainesville
43
1
0.5485 0.7818
Warner
Robins
27
1
0.5437 0.5647
Valdosta
29
1
0.5437 0.5647
Columbus
31
3
0.5563 0.7818
Yorkville
24
ND
Augusta
30
6
0.6457 1.7374
o- Xylene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
2
0.5457 0.6081
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
9
0.5846 1.5202
Rome
25
4
0.5655 0.8253
Utoy Creek
31
2
0.5766 1.4334
Brunswick
29
ND
Gainesville
43
1
0.5505 0.8687
Warner
Robins
27
1
0.5502 0.7384
Valdosta
29
1
0.5512 0.7818
Columbus
31
5
0.5773 0.9121
Yorkville
24
ND
Augusta
30
7
0.7087 2.2152
1,3,5-Trimethylbenzene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
2nd Max
0.9121 0.6515 0.6081
0.6515 1.5637 0.5647
0.9556 0.7818 0.6950
0.8253 1.9980
232
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
1,3,5-Trimethylbenzene Rome
25
ND
(continued)
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
1
0.6170 0.6883
1,2,4-Trimethylbenzene Milledgeville
29
ND
Macon
29
ND
Savannah
31
1
0.6217 0.8357
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
12 0.6717 1.3765
Rome
25
3
0.6352 0.9832
Utoy Creek
31
5
0.6486 1.2782
Brunswick
29
ND
Gainesville
43
1
0.6197 0.8357
Warner
Robins
27
1
0.6209 0.7866
Valdosta
29
1
0.6171 0.6883
Columbus
31
5
0.6597 1.0324
Yorkville
24
ND
Augusta
30
7
0.8054 2.4089
Styrene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
2
0.5685 1.3211
Brunswick
29
ND
Gainesville
43
ND
2nd Max
1.1799 0.7374 0.8849 0.9832 2.2123
0.8524
233
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Styrene
Warner
(continued)
Robins
27
ND
Valdosta
29
6
0.8722 5.1141
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Benzene, 1-ethenyl-4-
methyl
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
1
0.6153 0.6391
Chlorobenzene
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
2nd Max 2.3440
234
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
Chlorobenzene
Valdosta
29
ND
(continued)
Columbus
31
ND
Yorkville
24
ND
Augusta
30
1
0.5764 0.5987
1,2-Dichlorobenzene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
1,3-Dichlorobenzene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
2nd Max
235
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Name
Site
Total # of Samples Detects Avg. 1st Max
1,4-Dichlorobenzene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
Benzyl chloride
Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
Rome
25
ND
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
1,2,4-Trichlorobenzene Milledgeville
29
ND
Macon
29
ND
Savannah
31
ND
General
Coffee
30
ND
Dawsonville
30
ND
South DeKalb* 61
ND
2nd Max
236
2006 Georgia Annual Air Quality Report
2006 Volatile Organic Compounds (continued)
(concentrations in g/m3)
Total # of
Name
Site
Samples Detects Avg. 1st Max
1,2,4-Trichlorobenzene Rome
25
ND
(continued)
Utoy Creek
31
ND
Brunswick
29
ND
Gainesville
43
ND
Warner
Robins
27
ND
Valdosta
29
ND
Columbus
31
ND
Yorkville
24
ND
Augusta
30
ND
2nd Max
ND indicates no detection *sample collected every 6 days
237
2006 Carbonyl Compounds, 24-hour
(concentrations in micrograms per cubic meter)
Name
Total
# of
Site
Samples Detects Avg. 1st Max 2nd Max
Formaldehyde
Savannah
28
Dawsonville
31
S. DeKalb*
55
Tucker*
57
Brunswick
29
Acetaldehyde
Savannah
28
Dawsonville
31
S. DeKalb*
56
Tucker*
57
Brunswick
29
Propionaldehyde Savannah
28
Dawsonville
31
S. DeKalb*
56
Tucker*
57
Brunswick
29
Acrolein
Savannah
28
Dawsonville
31
S. DeKalb*
56
Tucker*
57
Brunswick
29
Butyraldehyde
Savannah
28
Dawsonville
31
S. DeKalb*
56
Tucker*
57
Brunswick
29
Acetone
Savannah
28
Dawsonville
31
S. DeKalb*
56
Tucker*
57
Brunswick
29
Benzaldehyde
Savannah
28
Dawsonville
31
S. DeKalb*
56
Tucker*
57
Brunswick
29
ND indicates no detection
*sample collected every 6 days
26 2.877 6.882 5.406
24 1.678 5.222 3.844
55 8.405 14.353 14.235
57 13.594 49.059 26.412
28 3.575 6.611 6.611
19 1.318 3.681 2.612
19 1.100 3.072 2.517
56 2.929 5.135 5.018
57 4.618 45.118 8.118
23 1.480 3.400 2.665
2
0.604 1.212 1.094
7
0.662 1.372 1.239
19 0.733 2.494 2.147
26 0.850 5.776 2.314
7
0.623 1.522 0.856
ND
ND
ND
ND
ND
ND
1
0.563 0.600
14 0.660 2.100 1.700
17 0.707 2.594 1.665
4
0.638 1.900 1.144
24 2.998 6.882 6.250
29 2.989 7.111 6.389
56 6.116 10.529 10.235
57 7.029 18.118 16.143
26 3.625 7.222 6.556
1
0.586 1.222
1
0.719 5.433
7
0.797 4.276 3.835
6
0.867 6.000 5.941
2
0.600 1.222 1.011
238
2006 Georgia Annual Air Quality Report
2006 Carbonyl Compounds, 3-hour (June-August)
Name
(concentrations in micrograms per cubic meter) Site Time # Samples # Detects Avg. 1st Max 2nd Max
Formaldehyde
S. DeKalb 0600
27
27
5.07 9.00 8.39
Tucker
0600
29
29
7.84 14.67 13.72
S. DeKalb 0900
25
24
9.36 20.61 17.94
Tucker
0900
27
27
10.61 21.83 15.83
S. DeKalb 1200
27
27
10.77 21.89 16.56
Tucker
1200
28
28
10.98 18.11 17.50
S. DeKalb 1500
27
25
9.95 20.11 16.06
Tucker
1500
27
27
10.68 15.89 15.39
Acetaldehyde
S. DeKalb 0600
27
27
2.00 3.89 3.46
Tucker
0600
29
29
2.68 4.72 4.44
S. DeKalb 0900
25
24
3.13 5.28 5.23
Tucker
0900
27
27
3.55 8.94 7.06
S. DeKalb 1200
27
27
3.70 5.83 5.26
Tucker
1200
28
28
3.97 8.89 6.22
S. DeKalb 1500
27
25
3.19 4.83 4.49
Tucker
1500
27
27
3.38 6.17 4.60
Propionaldehyde
S. DeKalb 0600
27
8
0.16 0.67 0.63
Tucker
0600
29
15
0.33 0.95 0.88
S. DeKalb 0900
25
14
0.33 0.87 0.79
Tucker
0900
27
17
0.41 1.24 0.77
S. DeKalb 1200
27
17
0.39 0.81 0.78
Tucker
1200
28
23
0.55 1.01 0.83
S. DeKalb 1500
27
16
0.37 0.88 0.88
Tucker
1500
27
18
0.43 0.82 0.81
Acrolein
S. DeKalb 0600
27
ND
Tucker
0600
29
ND
S. DeKalb 0900
25
ND
Tucker
0900
27
ND
S. DeKalb 1200
27
ND
Tucker
1200
28
ND
S. DeKalb 1500
27
ND
Tucker
1500
27
ND
Butyraldehyde
S. DeKalb 0600
27
9
0.15 0.58 0.56
Tucker
0600
29
20
0.50 2.27 1.53
S. DeKalb 0900
25
20
0.62 2.59 1.54
Tucker
0900
27
21
0.67 2.29 1.92
S. DeKalb 1200
27
21
0.64 2.12 1.95
Tucker
1200
28
26
0.63 2.10 1.52
S. DeKalb 1500
27
19
0.65 2.47 2.04
Tucker
1500
27
24
0.73 1.71 1.64
239
2006 Carbonyl Compounds, 3-hour (June-August) (continued)
Name
(concentrations in micrograms per cubic meter) Site Time # Samples # Detects Avg. 1st Max 2nd Max
Acetone
S. DeKalb 0600
27
27
5.30 8.89 8.17
Tucker
0600
29
29
6.69 17.22 12.17
S. DeKalb 0900
25
24
6.78 12.44 11.56
Tucker
0900
27
27
6.90 14.61 11.67
S. DeKalb 1200
27
27
8.36 13.83 13.11
Tucker
1200
28
28
7.51 14.78 13.06
S. DeKalb 1500
27
25
7.49 11.89 11.78
Tucker
1500
27
27
6.89 12.17 10.61
Benzaldehyde
S. DeKalb 0600
27
2
0.10 1.82 0.93
Tucker
0600
29
4
0.52 7.50 2.89
S. DeKalb 0900
25
2
0.43 5.78 4.86
Tucker
0900
27
3
0.61 7.17 5.25
S. DeKalb 1200
27
2
0.32 5.20 3.57
Tucker
1200
28
3
0.49 5.28 4.49
S. DeKalb 1500
27
3
0.47 5.29 5.12
Tucker
1500
27
3
0.58 6.22 5.41
240
2006 Georgia Annual Air Quality Report
Appendix F: Monitoring Network Survey
Georgia Gaseous Criteria Pollutant Monitoring as of January 2006
Parameter Measured
Sampling Schedule Number of GASN Sites
Method Used
EPA Reference
Method
Data Availability
Ozone
Nitrogen Dioxide
Carbon Monoxide
Continuous hourly average
Sulfur Dioxide
24
5
3
8
Ultraviolet photometry
Ultraviolet photometry
Ultraviolet photometry
Ultraviolet photometry
Nondispersive
Infrared photometry
Nondispersive
Infrared photometry
Ultraviolet fluorescence detector
Spectrophotometry (pararosaniline method)
Planning and Technical Support Division, Air Quality Data Branch, U.S. EPA Air Quality System (AQS)
241
Georgia Ambient Air Particulate Matter Monitoring as of January 2006
Parameter Measured
Sampling Schedule
Collection Method Sampling Media
PM10
PM2.5
Mass (integrated)
Every 6 days
Mass sequential, single channel Teflon filter 46.2mm,
Mass (semicontinuous)
Mass (integrated)
Mass (semicontinuous)
Speciated
Continuous hourly
averages
TEOM; BAM
Proprietary filter; filter tape
Varies; every day, every third day, or every sixth day
FRM sampler
Teflon filter 46.2mm
Continuous hourly
averages
TEOM
Proprietary filter
1 in 6 days 1 in 3 days for South DeKalb
Speciation air sampling
system (SASS) Teflon, nylon & quartz filter
46.2mm
Number of
Sites
15
Analyzed
1
28
12
8
Number of Collocated
Sites
Analysis Method
3
Method 016 Electronic analytical balance
0
5
0
0
Method 079; TEOM
gravimetric at 50 degrees C; Method 122
Beta Attenuation
Monitor
Method 055 Electronic analytical balance
Method 703 R&P
TEOM with SCC at 30 degrees C
Method 055 Electronic analytical balance Method 014
x-ray fluorescence Method 062 filter preparation Method 064 Ion chromatography Method 065 Thermal/optical
carbon
Data Availability
Planning and Technical Support Division, Air Quality Data Branch, U.S.EPA Air Quality System (AQS)
242
2006 Georgia Annual Air Quality Report
Georgia Organic Air Toxic Contaminant Monitoring as of January 2006
Parameter Volatile Organic Measured Compounds (VOCs)
Carbonyls
Semi - VOCs
Metals
Method
TO-14A/15
TO-11A
TO 13A
10-2.I
Sampling Schedule
Collection Equipment
Sampling Media
Number of Sites
Analyzed Number of Collocated
Sites
Data Availability
Every 12 days, 24-hour;
1 in 6 day schedule for South DeKalb
AVOCS or ATEC2200
Polished stainless steel canister
Every 12 days, 24-hour
ATEC100
DNPH-coated silica cartridges
15**
3
Every 12 days, 24-hour
PUF sampler Polyurethane
Foam filter
15**
Every 12 days, 24-hour; 1 in 6 day
schedule for South DeKalb*
High volume TSP
Quartz microfiber filter 8 x 10
inch
15**
1
0
1
1
Planning and Technical Support Division, Air Quality Data Branch, U.S.EPA Air Quality System (AQS)
* Sampler at this site is a PM10 Hi-Vol ** 14 GA ATN sites, 1 NATTS (South DeKalb)
243
PAMS Monitoring as of January 2006
Parameter
Sampling Schedule
Collection Equipment Sampling
Media Number of
Sites
Analysis Method
Data Availability
54 PAMS-Speciated VOCs & Total NMHC
Continuous 54-PAMS Speciated
VOCs & Total NMHC
Carbonyl Compounds
24-hour 1 in 6 day schedule (all year)
ATEC 2200
Polished stainless steel canister
Continuous hourly average
(June-August) Perkin-Elmer
HC GC
Direct injection
3-hour sample (June-August); 24-hour, 1 in 6 day
(all year)
ATEC 8000; PUF Sampler
DNPH coated silica gel Cartridge;
Polyurethane Foam
4
4
2
PAMS GC/FID
GC/FID
High performance liquid chromatograph/ultraviolet
detector
Planning and Technical Support Division, Air Quality Data Branch, U.S.EPA Air Quality System (AQS)
244
Georgia Meteorological Monitoring as of January 2006
Parameter Wind Speed
Measured
(m/s)
Sampling Schedule
Wind Direction (degrees)
Ambient Relative Atmosphere
Temperature Humidity Pressure
(C)
(%)
(mb)
Solar
Radiation (w/m2)
Continuous hourly average
Precip (in)
Sig. Theta (deg)
Total Ultraviolet Radiation
Number of Sites
14
14
5
5
4
3
2
1
3
Method Used
Data Availability
Propeller or cup
anemometer
Wind vane potentiometer
Aspirated Thermocouple or thermistor
Thin film capacitor
Pressure transducer
Thermopile or Tipping pyranometer bucket
Planning and Technical Support Division, Air Quality Data Branch, U.S.EPA Air Quality System (AQS)
Wind direction
UV radiometer
245
246
2006 Georgia Annual Air Quality Report
Appendix G: Siting Criteria
Instrument PM10, AISI Nephelo-
meter Dichot, TEOM, PM2.5
Lead, TSP
O3
CO
NO2
Height Above Ground
Micro Other
2-7m 2-15m
2-7m 2-15m
207m 2-15m
3-15m 3-15m
2.5 3.5m
3-5m
3-15m 3-15m
Space Between Samplers
2m 1m
2m
1m
Height Above Obstructions
1m
2 times height of obstacle
above inlet
1m
Distance From
Obstacles
2 times height or obstacle
above inlet
2 times height or obstacles
above inlet
2 times height of obstacles
above inlet
2 times height of obstacles
above inlet
Micro: must be no trees between sampler and road Others: must be 10m if trees, 5m above sampler
2 times height of obstacle
above inlet
Distance From Tree
Dripline
Should be 20m, must be
10m if considered
an obstruction Should be 20m, must be
10m if considered
an obstruction Micro and middle: no
trees between sampler and source Neighborhoo d: should be 20m, must be
10m if considered
an obstruction Should be 20m, must be
10m if considered
an obstruction
Micro: must be no trees
between sampler and
road Others: must
be 10m if trees, 5m
above sampler
Should be 20m, if
individual tree 5m above probe, must be 10m from dripline
Distance from Walls,
Parapets, etc. 2m 2m
2m
1m
1m
1m
Airflow Arc
270
270
270
270, or on side
of building
180
270, or on side
of building
180
270, or on side
of building
180
247
Instrument
Height Above Ground
Micro Other
Space Between Samplers
SO2
3-15m 3-15m
H2S
3-15m 3-15m
CH4, THC, NMHC, PAMS
3-15m
3-15m
Toxics: Gaseous 910, 910A, 929, 920
3-15m
3-15m
Temperature and Relative 1.25-2m
Humidity
2.252m
Wind Speed and
Direction
Solar Radiation
Height Above Obstructions
1m
1m
1m
2m
Distance From
Obstacles
2 times height of obstacle
above inlet
2 times height of obstacle
above inlet
2 times height of obstacle
above inlet 2 times height of obstacle above inlet 4 times height of obstacle above sensor
1.5 times height of obstacle
above sensor
Distance From Tree
Dripline
Should be 20m, must be
10m if considered
an obstruction Should be 20m, must be
10m if considered
an obstruction Should be 20m, must be
10m in direction of urban core
1 tower width from tower side
2 tower widths from tower side, 1 tower width from tower
top
Distance from Walls,
Parapets, etc. 1m
1m
1m
4.5m
Airflow Arc
270, or on side
of building
180
270, or on side
of building
180
270, or on side
of building
180
248
2006 Georgia Annual Air Quality Report
Appendix H: Instrument and Sensor Control Limits
ARB'S CONTROL AND WARNING LIMITS
Control 15%
15%
10%
LIMITS
Warning 10%
10%
7%
4% (Flow) 5% (Design) 20%
None None None
INSTRUMENT
All gaseous criteria and non-criteria analyzers Total suspended particulate (TSP) samplers PM10 Dichotomous (Dichot), Lead (Pb), Tapered Element Oscillating Microbalalance (TEOM), Toxic Air Contaminant (XonTech920) Samplers, Beta Attenuation Monitors (BAM), and Carbonyl (XonTech9250) Samplers
PM2.5
Laboratory audits (Toxics, PAMS, Motor Vehicle Exhaust and Total Metals)
ACCEPTANCE CRITERIA FOR METEOROLOGICAL (MET) SENSORS
LIMITS
1.0 Celsius (0.5C PAMS only) 2.25 mm of Mercury (Hg) 3% RH for 10-90% RH 5% RH for <10% or >90% RH 5% Watts/m2 Less than or equal to 5 combined accuracy and orientation error 0.25 m/s between 0.5 and 5m/s and less that 5% difference above 5 m/s Less than or equal to 0.5m/s
0.25 m/s between 0.5 and 5 m/s and less than 5% difference above 5 m/s Less than or equal to 0.5 m/s
SENSOR
Ambient Temperature Barometric Pressure Relative Humidity Solar Radiation Wind Direction
Horizontal Wind Speed Horizontal Wind Speed Starting Threshold Vertical Wind Speed Vertical wind Speed Starting Threshold
249
250
2006 Georgia Annual Air Quality Report
References
http://www.airnow.gov/index.cfm?action=static.aqi. "Air Quality Index (AQI) - A Guide to Air Quality and Your Health."
AIRNOW DMC. 2006 Data Polling Summary. Poster presented at the National Air Quality Conference. Sonoma Technology, Petaluma, California.
ATSDR, 1992. Toxicological Profile for 1,3-Butadiene. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
ATSDR, 1997. Toxicological Profile for Benzene. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
ATSDR, 1999. Toxicological Profile for Formaldehyde. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Registry, Atlanta, Georgia.
ATSDR, 2000a. Toxicological Profile for Manganese. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
ATSDR, 2000b. Toxicological Profile for Chromium. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Registry, Atlanta, Georgia.
ATSDR, 2001. Toxicological Profile for 1,2-Dichloroethane. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
ATSDR, 2005a. Toxicological Profile for Carbon Tetrachloride. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
ATSDR, 2005b. Toxicological Profile for Acrolein, Draft for Public Comment. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
ATSDR, 2005c. Toxicological Profile for Arsenic, Draft For Public Comment. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Registry, Atlanta, Georgia.
ATSDR, 2005d. ToxFAQS for Nickel. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Registry, Atlanta, Georgia.
ATSDR, 2006a. Toxicological Profile for Dichlorobenzenes. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia.
251
ATSDR, 2006b. Medical Mangement Guidelines for Xylenes. U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Registry, Atlanta, Georgia.
Code of Federal Regulations, Title 40, Volume 2, Parts 50 to 51, Revised as of July 1, 1998.
Code of Federal Regulations, Title 40, Protection of the Environment, Part 58, Ambient Air Quality Surveillance, September 2006.
Georgia Department of Natural Resources. Georgia Ambient Monitoring Network Plan.
Georgia Department of Natural Resources. Georgia Ambient Air Monitoring Quality Assurance Manual, Quality Assurance Plan.
GADNR, 1993. Toxic Release Inventory Report, 1991. Georgia Department of Natural Resources, Environmental Protection Division. Atlanta, Georgia.
GADNR, 1996a. The 1994 Chatham County Air Toxics Study, Georgia Department of Natural Resources, Environmental Protection Division. Atlanta, Georgia.
GADNR, 1996b. 1996 Glynn county Initiative: air Toxics Dataset Ground Level Measurements, Georgia Department of Natural Resources, Environmental Protection Division. Atlanta, Georgia.
GADNR, 2006. 2005 Ambient Air Surveillance Report. Georgia Department of Natural Resources, Environmental Protection Division. Atlanta, Georgia.
U.S. EPA, 1987. Health Assessment Document for Acetaldehyde. U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA, 1991a. Integrated Risk Information System, Carbon Tetrachloride. U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA, 1991b. Integrated Risk Information System, Acetaldehyde. U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA, 1994a. OPPT Chemical Fact Sheet, Chemicals in the environment: 1,2,4trimethylbenzene (CAS No. 95-63-6). U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA, 1994b. Quality Assurance Handbook for Air Pollution Measurement System. Volume 1: Principles. EPA-600/R-94/038A, January 1994.
U.S. EPA, 1998. Quality Assurance Handbook for Air Pollution Measurement System. Volume 1: Principles. EPA-600/R-94/038B, April 1998.
U.S. EPA, 2000. Integrated Risk Information System, Benzene. U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA, 2002. Integrated Risk Information System, 1,3-Butadiene. U.S. Environmental Protection Agency, Washington, D.C. 252
2006 Georgia Annual Air Quality Report U.S. EPA, 2003. Integrated Risk Information System, Acrolein. U.S. Environmental Protection Agency, Washington, D.C. U.S. EPA, 2004a. Air Quality Criteria for Particulate Matter. U.S. Environmental Protection Agency, Washington, D.C. U.S. EPA, 2004b. Provisional Peer Reviewed Toxicity Value Database. U.S. Environmental Protection Agency, Region IV, Atlanta, Georgia. U.S. EPA, 2006. A Preliminary Risk-Based Screening Approach for Air Toxics Monitoring Data Sets. U.S. Environmental Protection Agency, Washington, D.C.
253