Air Quality Project REMOTE SENSING LASER … quality project remote sensing laser measurements of...

US DOT Research and Special Program Administration

Air Quality Project

REMOTE SENSING LASER MEASUREMENTS OF

AIR POLLUTION

YEAR 1 STUDY IN NORTH MISSISSIPPI

YEAR 1 FINAL REPORT UM-CAIT/2001-01

Project

Waheed Udd Associate Professor g

Directo Augu

CENT CED INFRAST NOLOGY

I CARR 7-1848

VOICE: 6http://ww

Director

in, Ph.D., P.E. of Civil Engineerinr, CAIT

st 2001

ER FOR ADVANRUCTURE TECH

THE UNIVERSITY OF MISSISSIPP

IER 203, UNIVERSITY, MS 3867
62-915-5363 FAX: 662-915-5523
w.olemiss.edu/projects/cait/Index.html

Year 1 Final Report/ Uddin UM-CAIT/2001-01 August 2001

1

EXECUTIVE SUMMARY The air quality analysis project, a part of the NCRST-E overall effort, is being conducted by the Center for Advanced Infrastructure Technology (CAIT) at the University of Mississippi, a consortium partner. This national project is focused on measuring air pollution from mobile sources using remote sensing laser technology. The project will model air pollutants and assess air quality impact of transportation systems, which significantly affect quality of life. To enforce 1990 amendments to the Clean Air Act, the Environmental Protection Agency (EPA) has developed the Air Quality Index (AQI) considering standard concentration levels of: ground-level Ozone, Nitrogen Oxides (NOx), particulate matter, CO, and SO2. Ozone, a major air pollutant, is formed by a chemical reaction involving volatile organic compounds (VOC), oxides of nitrogen, and sunlight. Ozone contributes to smog and negatively affects health. Despite considerable regulatory and pollution control efforts over the last three decades, high Ozone concentrations in urban, suburban, and rural areas continue to be a major environmental and health concern. Numerous cities and urban areas are listed as nonattainment areas for Ozone. Vehicular emissions of air pollutants result from fuel combustion, fuel evaporation, and refueling losses. Vehicles are becoming more efficient and cleaner, however, vehicle-miles traveled tripled over the last 40 years. The air quality degradation, affected by diffusion and dispersion of air pollutants, depends upon topographic, climatic, and weather conditions. Ozone and smog created by Nitrogen Dioxide are particularly high during hot days, especially in urban and suburban areas where paved surfaces and constructed roofs cause up to 20 to 30 degree higher temperatures during hot summer days. Traffic gridlocks result in more Ozone pollution. DIfferential Absorption LIDAR (DIAL) has been successfully used to monitor atmospheric pollutants, such as Nitrogen Oxides, Sulfur Dioxide, Hydrocarbons, Ozone, and Mercury vapors. LIDAR is an acronym for LIght Detection And Ranging. It uses laser pulses to transmit and receive electromagnetic radiation. Non-invasive remote sensing DIAL systems operate on the principle that the absorption of light by the atmosphere and air pollutants varies at different wavelengths. The laser is tuned between ultraviolet, visible, near infrared, and thermal infrared spectral regions. The difference in the absorption of light at these different wavelengths can be used to determine the concentration of air pollutants. Extensive literature search and reviews, conducted in North America and Europe, indicate that the tunable DIAL pulse measurement, combined with modern georeferenced location and time stamping, is the most promising remote sensing technology for monitoring concentrations of air pollutants and assessing air quality. Nitrogen Dioxide (NO2), a major precursor of Ozone, has been successfully measured in May 2001 by truck-mounted DIAL equipment. The test site is adjacent to MS Highway 6 in Oxford, Mississippi. Traffic data have been collected on the test site as a part of on-going Intelligent Transportation System (ITS) study for Oxford. Wind and other weather data have been collected from the nearby NOAA weather station in Batesville. The results showed nearly 25 times more NO2 concentration at 10 AM (daytime) compared to the measurement at 11 PM (nighttime) when the traffic was minimal. Further DIAL data collection in Year 2 will be conducted at the NCAT accelerated highway test track site at Auburn University.


2

LIST OF COOPERATING AGENCIES AND ORGANIZATIONS

US DOT Research and Special Program Administration (RSPA) - (Study Sponsor) * Mississippi State University Remote Sensing Technologies Center (NCRST-E Lead University)* NCRST-E => National Consortia for on Remote Sensing in Transportation- Environmental Assessment The University of Mississippi Center for Advanced Infrastructure Technology (NCRST-E Air Quality Project)* Mississippi Department of Environment Quality (DEQ) Office of Pollution Control, Air Division (Observations and DEQ Monitoring Data ) City of Oxford, Mississippi (Traffic volume data by automatic counters; Coordination with Oxford ITS Project) City of Tupelo, Mississippi (Logistic support) Tupelo Regional Airport Authority, Mississippi (Coordination for future air pollution measurements) Mississippi Space Commerce Initiative, NASA Stennis Space Center, Mississippi (LIDAR Evaluation of terrain mapping for highway planning project with CAIT) Space Imaging, Inc, NASA Stennis Space Center, Mississippi (IKONOS Imagery of Oxford area)

* Study partners and their roles

Service Provider for Remote Sensing Laser Measurement of Air Pollution: Skyborne, Inc.

The author appreciates the support of all pertinent individuals from the agencies and organizations listed above, and the comments of NCRST-E advisory committee during the Year1 study meetings. Mr. Phillip Ozdemir, representing Skyborne (New York), worked hard to conduct the field DIAL measurements, process the data, and interpret the results. Thanks are due to Mr. Sergio Garza (doctoral graduate student) who is the primary research assistant on the air quality project, supported by Ms. Yamini Nanagiri, Mr. Xin Chen, Mr. Javier Garcia, and several other graduate students. In addition, the author thanks Ms. Lucy Phillips, Ms. Julia Phillips, Mr. James Caughorn, and other undergraduate senior students who assisted in traffic data collection, field work related to DIAL measurements, weather data collection and analysis, and in the preparation of this report.


3

TABLE OF CONTENTS Page TITLE PAGE EXECUTIVE SUMMARY............................................................................................. 1 LIST OF COOPERATING AGENCIES AND ORGANIZATIONS............................. 2 TABLE OF CONTENTS ............................................................................................... 3 1.0 NCRST-E AIR QUALITY PROJECT................................................................ 4 1.1 BACKGROUND .................................................................................... 4

1.2 OBJECTIVE OF AIR QUALITY PROJECT ........................................ 4 1.3 SCOPE OF THIS REPORT ................................................................... 4

2.0 AIR POLLUTION MEASUREMENTS AND AIR QUALITY ANALYSIS.... 5

2.1 BACKGROUND .................................................................................... 5 2.2 CLEAN AIR ACT REQUIREMENTS .................................................. 6 2.3 AIR POLLUTANT MONITORING IN MISSISSIPPI ....................... 16

3.0 REMOTE SENSING LASER MEASUREMENTS OF AIR POLLUTANTS 20

3.1 PRINCIPLES OF DIAL MASUREMENTS ....................................... 20 3.2 AVAILABLE DIAL TECHNOLOGIES ............................................. 22 3.3 SELECTION OF DIAL TECHNOLOGY AND TEST SITES ........... 31

4.0 YEAR 1 DIAL MEASUREMENTS IN NORTH MISSISSIPPI ................... 37

4.1 DIAL TEST SETUP IN OXFORD, RESULTS, AND DISCUSSIONS . 37 4.2 TRAFFIC AND WEATHER DATA SUMMARY AND ANALYSIS... 49 4.3 CONCLUDING REMARKS ................................................................. 53

5.0 PLANNING FOR YEAR 2 DIAL MEASUREMENTS .................................... 55

5.1 TUPELO AIRPORT ............................................................................ 55 5.2 NCAT TEST TRACK, AUBURN UNIVERSITY, ALABAMA ........ 57

6.0 AIR POLLUTION MODELING ....................................................................... 58

6.1 LITERATURE REVIEW ..................................................................... 58 6.2 MODELING FRAMEWORK AND METHODOLOGIES ................. 58

7.0 REFERENCES………………………………………………………………… 61 ATTACHMENTS ....................................................................................................... 66


4

1.0 NCRST-E AIR QUALITY PROJECT

1.1 BACKGROUND The National Consortia on Remote Sensing in Transportation (NCRST) have been established by the U.S. Department of Transportation Research and Special Program Administration (RSPA), in collaboration with NASA under the funding from the Transportation Equity Act for the 21st Century (TEA-21), Section 5113 to lead in the application of remote sensing and spatial information technologies in the transportation industry [1]. The primary mission of the university consortium for Environmental Assessment (NCRST-E), one of four consortia is to develop and promote the use of remote sensing and geospatial technologies and analysis products by transportation decision-makers and environmental assessment specialists to measure, monitor, and assess environmental conditions in relation to transportation infrastructure [2]. This four-year study started in March 2000. The NCRST-E is expected to incorporate remote sensing technologies for environmental assessment of transportation projects, for streamline of National Environmental Policy Act (1969) guidelines [3], and for environmental monitoring. For acceptance and implementation by transportation agencies and industries, these technologies must be proven to be credible for these applications with proven performance measures and benchmarks. To achieve this objective, NCRST-E has developed demonstration projects in several states including Alabama, Florida, Georgia, Iowa, Mississippi, North Carolina, Tennessee, Virginia, and Washington [2]. The air quality project, a part of the NCRST-E studies, is being conducted at the University of Mississippi by the Center for Advanced Infrastructure Technology (CAIT), a consortium partner. 1.2 OBJECTIVE OF AIR QUALITY PROJECT The NCRST-E air quality project is focused on remote sensing laser measurements of air pollutants and air quality impact of transportation systems. The primary objective is the development of a state-of-the-art protocol for air quality analysis using remote sensing laser measurements of significant air pollutants related to traffic and transportation infrastructure. The project involves: (a) comprehensive literature search and review of air pollutant databases, transportation related air pollution data, air pollutant concentration and dispersion modeling, and available remote sensing technologies, (b) selection of study sites, (c) deployment of selected remote sensing technology for measuring air pollution at selected sites, interpretation and assessment of data, and study of the adverse effects of weather and urban sprawl on air quality, and (d) modeling of air pollution considering traffic, weather, and land use. 1.3 SCOPE OF THIS REPORT This report includes literature reviews on available remote sensing technologies and results of tunable laser measurements on the test site in North Mississippi selected for the Year 1 study.


5

2.0 AIR POLLUTION MEASUREMENTS AND AIR QUALITY ANALYSIS 2.1 BACKGROUND

Quality of the human environment in our society is greatly affected by the quality of air that we breathe and the water that we consume. Increased human activities in the last four decades have disturbed the natural cycles of key elements of life. These concerns led to the creation of a national environmental policy nearly three decades ago. The U.S. Environmental Protection Agency has been created to develop specific regulations and monitoring methods to improve the air and water quality. It ensures that the environment quality is not degraded by urban growth, industrial development, increased transportation facilities, and higher traffic volume. 2.1.1 NEPA Guidelines The 1969 National Environmental Policy Act (NEPA), a five-page long historic document (Public law 91-190, 42 U.S.C. 4321-4347, January 1,1970), was passed by the United States Congress. For achieving environmental protection, NEPA identifies the following three principal elements [3]: (1) declaration of the world’s first policy on the environment (Title 1), (2) establishment of an “action-forcing” mechanism for implementing this policy, and (3) creation of a council on environmental quality (CEQ) for implementing NEPA and elevating environmental concerns directly to the presidential level (Title 2). The CEQ NEPA regulations have the following nine parts [3]: Part 1500- Purpose, Policy, and Mandate, Part 1501- NEPA and Agency Planning, Part 1502- Environment Impact Statement (EIS), Part 1503- Commenting, Part 1504- Predecision Referrals to the Council, Part 1505- NEPA and Agency Decision Making, Part 1506- Other Requirements (while an EIS is in progress), Part 1507- Agency Compliance, and Part 1508- Terminology and Index. Many crucial amendments, statues, regulations, executive orders, and CEQ directives have been issued related to NEPA since that time. The ultimate goal of NEPA is to foster excellent decision making that in the past, mistakenly has been viewed as a lengthy permitting process. Clean Air Act of 1970, Clean Water Act, and Executive Order 11991, Protection of wetlands, issued on May 24, 1977, and their amendments are the most important regulations governing the quality of air and water to enhance the quality of life. More recent guidance includes integration of NEPA with pollution prevention, ISO 14000, and environment justice [3].

2.1.2 Key Air Pollutants Affecting Air Quality Key pollutants affecting air quality include: Carbon Dioxide (CO2), Carbon Monoxide (CO), volatile organic compounds (VOC) or hydrocarbons (HC), Nitrogen Oxides (NOx), particulate matter (PM), and SO2. Ground-level Ozone (O3), a major air pollutant, is formed by a photochemical reaction involving Nitrogen Dioxide (NO2) and VOC in presence of sunlight. Above certain concentration levels, O3, CO, and PM can cause or exacerbate health problems and/or increase mortality rates, making their control an important goal under the Clean Air Act. Carbon Dioxide is not harmful to human health, however, it is a major pollutant from coal burning power plants and from on-road vehicle, aviation, and other transportation related emissions resulting from combustion. It is a “greenhouse gas” that traps in heat within the


6

earth’s atmosphere causing global warming. On the other hand, Carbon Monoxide is a colorless and odorless hazardous gas that once entered into the blood stream through lungs, causes a decrease in the oxygen supply in the blood that may result in death. CO hazards are common in winter when coal or gas fire used in fireplaces, during traffic congestions and gridlocks, at other queuing and stop-and-go locations such as intersections and toll plazas. As the world is becoming more industrialized and dependent on fossil fuel, carbon emission has increased tremendously during the last 100 years. The United States makes up only 4 percent of the world’s population and it produced 24 percent of the world’s total carbon emissions in 1996, as shown in Figure 1 [5]. Primary sources of air pollutant emitters are: �� Point and Area Sources (electric utilities and other fuel combustion industrial processes, such

as: manufacturing, painting and surface coating, metals and chemical processing, petrochemical plants, dry cleaners and others)

�� Road Transportation Vehicles (automobiles, motorcycles, light duty trucks and sport utility vehicles, heavy duty trucks and truck-trailers, and buses)

�� Non- Road Engines (farm equipment, construction equipment, lawn and garden equipment, boats and other marine vessels, railroads, military vehicles, and others)

�� Aviation Sources (air carrier aircraft, general aviation aircraft, military aircraft, spacecraft) �� Miscellaneous Sources (accidental fires, forest wildfires, agricultural fires, health services

cooling towers, windblown dust, even backyard barbecue) �� Natural Emitters (Lightning produces NO2 . Decomposition of fertilizer produces NO2.

Biogenic emission of isoprene from certain trees in mid-eastern US, such as oak trees, is a another source of VOC, as discussed later in this chapter.) Both NO2 and VOC are precursors of O3.

Vehicular emissions of air pollutants result from fuel combustion, fuel evaporation, and refueling losses. About 80 percent of hydrocarbon (or VOC) emission occurs during the first minute of vehicle operation. Vehicles are becoming more efficient and cleaner, as shown in Figure 2. However, vehicle-miles traveled (VMT) tripled over the last 40 years [4]. The air quality degradation is affected by diffusion and dispersion of air pollutants depending upon topographic, climatic, and weather conditions. 2.2 CLEAN AIR ACT REQUIREMENTS 2.2.1 National Ambient Air Quality Standards As a continuation of NEPA legislation, 1970 Clean Air Act, Section 309 Public Law 91-604 § 12(a), 42 U.S.C. § 7609, was passed by the U.S. Congress in 1970 [3]. In response to the Clean Air Act, the U.S. Environmental Protection Agency (EPA) established National Ambient Air Quality Standards for various “criteria” pollutants that adversely affect human health and welfare. In the past, motor vehicles were a source of lead (Pb) emissions but are no longer a major contributor because leaded gasoline is no longer available for transportation usage. Compliance with national (and state standards, if more stringent) ambient air quality standards is an important consideration.

-The 1990 Clean Air Act Amendments direct the Environmental Protection Agency (EPA) to implement strong environmental policies and regulations that will ensure cleaner air quality.


7

Figure 1. Worldwide carbon emission from aviation sources - 1996 data [5]

Figure 2. Reduction in vehicle emissions [4]

Sources of carbon emissions

2 9 1

2 3 8

1 5 3 1 4 5 1 4 01 1 6

1 0 1

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

3 0 0

3 5 0

Japan

German

y

United K

ingdom

Global

Aviatio

n

Canad

aIta

ly

France

Total Carbon Emissions from Global Aviation in Some Industrialized Countries (Source: GAO/RCED-00-57 Aviation’s Effects on the Global Atmosphere, 2000, page 16)

Metric tons of carbon in millions (1996 data)

CO2 – “Greenhouse Gas” Causing Global Warming

CO – Colorless and Odorless Hazardous Gas

In 1996, the United States produced 1,463 million metric tons of carbon, or about 24 percent of the world’s total. The U.S. makes up only 4 % of the world’s population.

CO 11.3 9.5 81.7 3.0

VolatileOrganic

Compounds3.8 2.1 16.8 1.2

NOx 24.5 8.0 3.2 0.6

1967 19671998 1998

Gram

s per mile

CO 11.3 9.5 81.7 3.0

VolatileOrganic

Compounds3.8 2.1 16.8 1.2

NOx 24.5 8.0 3.2 0.6

1967 19671998 19981967 19671998 1998

Gram

s per mile


8 8

These 1990 Clean Air Act Amendments affect all proposed transportation projects. According to Title I, Section 101, Paragraph F, “No federal agency may approve, accept or fund any transportation plan, program or project unless such plan, program or project has been found to conform to any applicable (state) implementation plan in effect under this act.” Title I of the amendments defines conformity as follows: • Conforming to an implementation plan’s purpose of eliminating or reducing the severity and

number of violations of the National Ambient Air Quality Standards (NAAQSs) and achieving expeditious attainment of such standards, and

• Ensuring that such activities will not - cause or contribute to any new violation of any NAAQS in any area, - increase the frequency or severity of any existing violation of any NAAQS in any area, or

delay timely attainment of any NAAQS or any required interim emissions reductions or other milestones in any area. Table 1 shows these standards [6]. An area is in violation of the standard if it exceeds the concentration level for the specified form of the standard and evaluation time frame [4].

Table 1. 1999 Primary Air Quality Standards for Transportation Related Pollutants [4, 6] * New proposed standard 2.2.2 Transportation Related Emissions Figure 3 shows distribution of VOC and NOX emissions from different sources. The NOX emission from the road vehicle decreased about 3 percent during 1970 to 1996 and VOC decreased by 58 percent. These are significant considering several times increase in VMT during this period. Figure 4 shows the 1999 inventory of power plants with VOC and NOX emissions [6]. Figure 5 compares NOX emissions in 1990 and 1999 from utility sources. These emissions have increased in Eastern states, Midwestern states, and California over this period.

Pollutants Maximum Concentration Standard - NAAQS

Ground-level Ozone (O3) 0.12 ppm (1-hour average) Ground-level Ozone (O3) 0.08 ppm (8-hour average) * Particulate matter (PM 2.5) 15 microgram/ m 3 (Annual average) *

Particulate matter (PM 2.5) 65 microgram/ m 3 (24-hour average)* Particulate matter (PM 10) 50 microgram/ m 3 (Annual average)

Particulate matter (PM 10) 150 microgram/ m 3 (24-hour average)

Carbon monoxide (CO) 9 ppm (8-hour average)

Carbon monoxide (CO) 35 ppm (1-hour average) Sulfur dioxide (SO2) 0.03 ppm (Annual average) Sulfur dioxide (SO2) 0.14 ppm (24-hour average) Nitrogen Oxides (NOX) 0.053 ppm (Annual average)

(One of the Primary Precursors of Ozone)


9 9

Figure 3. Distribution of emission sources for VOC and NOX [4]

VOC Emissions, 1996

Point&Area55%

Misc3%

On-road vehicles

29%

Non-road engines

13%

NOX Emissions, 1996

Point&Area55%

Misc1%

On-road vehicles

29%

Non-road engines

13%

16

20

12

8

4

0

NOX Emissions, 1970-96

Mill

ion

shor

t ton

s

On-road vehiclesAll other sources

1970 1974

1978 1982

1986

1990

1994

16

20

12

8

4

0

16

20

12

8

4

0

NOX Emissions, 1970-96

Mill

ion

shor

t ton

s


1970 1974

1978 1982

1986

1990

1994

1970 1974

1978 1982

1986

1990

1994

1970 1974

1978

1982 1986

1990

1994

16

20

12

8

4

0

VOC Emissions, 1970-96


1970 1974

1978

1982 1986

1990

1994

1970 1974

1978

1982 1986

1990

1994

16

20

12

8

4

0

VOC Emissions, 1970-96



10 10

Plants with emissions of NOx above 100 Tons/Year in 1999

Plants with emissions of VOC above 100 Tons/Year in 1999

Figure 4. Distribution of power plant (point) emission sources for NOX and VOC [6]


11 11

Figure 5. Comparison of NOX emissions in 1990 and 1999 from utility sources [6]

Source: http://www.epa.gov/airnow/publications.html Air Quality Index (AQI), EPA-454/R-99-010

Figure 6. The Air Quality Index scale [7, 8]

301+AQI

201-300151-200101-150

51-1000-51

AIR QUALITY

Good

Moderate

Unhealthy(for sensitive groups)

Unhealthy

Hazardous

Very Unhealthy

301+AQI

201-300151-200101-150

51-1000-51

AIR QUALITY

Good

Moderate

Unhealthy(for sensitive groups)

Unhealthy

Hazardous

Very Unhealthy

http://www.epa.gov/airnow/publications.html


12 12

2.2.3 Air Quality Index (AQI) Reporting Scale The Air Quality Index (AQI) has been adopted by the EPA and the States for daily air quality reporting to the general public in accordance with section 319 of the Clean Air Act [7]. For each pollutant, the AQI transforms ambient concentrations to a scale from 0 to 500, keyed to the NAAQS value for each pollutant. In most cases, the value of 100 is associated with the numerical value of the short-term standard (i.e., averaging time of 24-hours or less). Figure 6 shows the current AQI scale and its relationship to public health [7, 8]. The nationwide AQI historical data are accessible on-line through the EPA Air Graphics web page on the Internet [6]. Ozone and NO2 are two major transportation related pollutants that have affected many urban areas in recent years. Over 100 “Ozone” nonattainment areas have been designated by the EPA in the nation. Houston-Galveston, Dallas-Ft. Worth, Beaumont-Port Arthur, and El Paso areas in Texas are in nonattainment status. Under the new standards currently being considered by the EPA, the Austin, San Antonio, Tyler-Longview-Marshall, and Corpus Christi areas could also soon be classified as nonattainment areas [9, 10]. These nonattainment areas in Texas, Los Angeles in California [11], and other cities, as well as in near- nonattainment areas including Atlanta [12, 13], are faced with air quality conditions that may be hazardous to the long term health of residents. Nonattainment areas could lose federal transportation funding if sufficient progress is not made to reduce the objectionably high Ozone levels. Failure to meet standards in areas within the nonattainment states could make such areas, and even the state, less appealing for competitive businesses and potential residents. A comprehensive plan to reduce the high Ozone and NO2 levels and/or prevent any increase in these levels during the summer months requires the state environmental quality control agencies to (a) prepare emission inventories from point and area sources, (b) estimate traffic volumes and patterns to prepare emission inventories mobile sources, (c) operate permanent stations for regular monitoring using point sampling, (d) make regional estimates using regional pollution predictions models, and (e) check regional pollution estimates and permanent station data using random measurements. The remote sensing tunable laser technology being investigated in this study provides the capability of measuring both Ozone and NO2 concentration levels from a sampling distance of one mile or more at several heights to obtain vertical dispersion profiles, as affected by wind. The following sections discuss Ozone and NO2 pollutant formation and dispersion in detail. 2.2.3 Ozone (O3) Ozone (O3) gas forms when three atoms of oxygen are combined in a photochemical process. In the upper atmosphere, Ozone is found naturally and protects the earth from harmful ultraviolet (UV) radiation. But ground level Ozone is a key ingredient in forming urban smog and is considered one of the greenhouse gases. In terms of public health and air quality, we are concerned with both ground level (bad Ozone) and upper atmospheric (good Ozone) [14]. Roughly 6-30 miles above the earth's surface lies the stratosphere where upper-atmospheric Ozone (good Ozone) occurs naturally. In this layer of our atmosphere, Ozone is formed when UV radiation dissociates some O2 molecules into two free oxygen atoms. These free oxygen


13 13

atoms then combine with other O2 molecules to create O3. Since O3 is in turn dissociated by UV rays, a net balance is kept with continual formation and destruction of the oxygen molecules. In the upper atmosphere this "good" Ozone protects life on Earth by absorbing some of the sun's ultraviolet rays [14]. The discovery of a hole in the stratospheric Ozone layer in the last 20 years has been linked to “global warming,” a source of great environment concern. Ground level Ozone (bad Ozone) occurs in the part of the atmosphere directly above the earth's surface, called the troposphere. Unlike the stratospheric Ozone layer, this ground level tropospheric Ozone is toxic to living organisms and a major air pollutant. Ground level Ozone is caused by the release of volatile organic compounds (VOC) and nitrogen dioxides (NO2) into the air. When these precursor pollutants are released into the air, tropospheric Ozone is formed by photochemical reactions with sunlight and heat. Ozone is also considered a greenhouse gas. Ozone absorbs infrared radiation emitted by the Earth's surface and thereby traps heat and warms the troposphere. Consequently, as temperatures rise, Ozone levels rise as well [14]. VOC + NO2 + heat + sunlight = ground level Ozone (O3) Ozone is the most abundant tropospheric oxidant and an important component of photochemical pollution. Elevated concentrations of ground level Ozone that principally occur during the summer months have been shown to be harmful to human health and damaging to vegetation. Regulation of Ozone precursor emissions under the U.S. Clean Air Act of 1970 and its subsequent amendments has been partially successful in reducing human exposure, but many areas of the country are still subject to episodes of high ambient Ozone levels [15]. Early Ozone management strategies emphasized reductions of anthropogenic emissions of VOCs, such as those emitted in automobile exhaust. These strategies have successfully reduced peak Ozone concentrations over time in cities (e.g., Los Angeles) where anthropogenic emissions dominate the ambient VOC mixture. In the eastern United States, however, large emissions of very reactive VOCs from biogenetic sources (e.g., isoprene from oak trees) have been shown to contribute substantially to Ozone formation in both rural and urban areas [16, 17]. Emissions of anthropogenic NOx from transportation, electric utility power plants, and commercial and industrial fuel combustion are all significant in the production of ground level Ozone [18]. Controls on anthropogenic NOx , in addition to controls on anthropogenic VOCs are now thought to be needed to reduce Ozone in regions (e.g., Atlanta) characterized by strong biogenic VOC sources [14, 19]. Mississippi, the first test site for remote sensing laser measurements of air pollution, also falls in this region of high biogenic VOC sources. Ground level Ozone contributes to brown smog formation which is harmful to human health. It is a source of concern in urban and metropolitan areas where paved surfaces, roof tops, and other constructed areas add to higher temperatures. On warm summer days, the air in a city can be 6-8°F hotter than its surrounding areas. Scientists call these cities "Urban Heat-Islands," as shown in Figure 7. The impact of these pollution levels is seen in smog [20].


14 14

2.2.4 Nitrogen Dioxide (NO2) Nitrogen Dioxide (NO2) gas is part of a family of highly reactive gases called nitrogen oxides (NOx), which includes nitric oxide (NO) and nitrous oxide (N2O). Figure 8 shows the nitrogen cycle [21]. NO2 and NO are found in the air as a result of interactions with oxygen.

Figure 7. An illustration of urban heat-island [20]

Figure 8. Nitrogen cycle [21]


15 15

Nitrogen will only react with oxygen in the presence of high temperatures and pressures found near lightning bolts and in combustion reactions in power plants or internal combustion engines. Nitric oxide, NO, and nitrogen dioxide, NO2, are formed under these conditions [21]. For example, when a nitrogen (N2)-oxygen (O2) mixture is heated at very high temperatures, the nitrogen and oxygen will combine to form nitrogen oxide (NO). If the cooling process is slow, the gases will decompose back to their original state. If the cooling process is fast, such as in the internal combustion engines of motor vehicles, the nitrogen oxides will not decompose but instead stay in the NO state. These gases are released into the atmosphere and combine with Ozone (O3) to produce NO2. In the presence of sunlight, NO2 also reacts with other pollutants, such as VOCs, to help form ground-level Ozone. Most NO2 in the atmosphere is formed in this way, although some is released directly from the source [14]. Eventually, nitrogen dioxide may react with rainwater to form nitric acid, HNO3. The nitrates thus formed are utilized by plants as nutrient [21]. NOx can react with SO2 and other chemicals in the air to form acid rain. NOx, when in the presence of water vapor in the atmosphere, partially converts into nitric acid, HNO3. Nitric acid and sulfuric acid combine to create acid rain [14]. NO2 is primarily the result of gases from motor vehicle exhaust and stationary fuel combustion sources like electric utilities and industrial boilers. It can also be produced from gas stoves and heaters. Human activities now account for more than 90 percent of U.S. NO2 emissions, between 20 and 23 million metric tons since 1972 which is double the 1950 value [22]. NO2 absorbs the blue band of light and can lead to the brownish haze, also known as “smog” over metropolitan areas. At a high concentration level, NO2 has a bleach-like odor and may contribute to chronic lung diseases such as asthma. “Silo filler’s disease” is a term used to describe a syndrome of acute pulmonary toxicity experienced by farmers exposed to high levels of NO2 while working in silos that house decomposing fertilizer [23]. NO2 is also a byproduct of the welding process. Patients trapped in closed-space fires may have significant exposure to NO2. Cases have been cited in literature in which hockey players developed pulmonary symptoms consistent with NO2 toxicity after prolonged periods of time spent in ice arenas maintained by the Zamboni type ice-cleaning machines [23]. Concerns with NO2 and O3 problems have prompted researchers and communities to find feasible solutions like reducing the tendency to have high temperatures through planting trees and establishing parks in the city to get more shadow areas, constructing paved surfaces and roofs with highly reflective surfaces, and reducing traffic congestion and gridlocks by adapting flexible hour work days. The Mitsubishi Materials Company in Japan is conducting field trials of sidewalks constructed with NOx absorbing blocks. These titanium dioxide containing blocks create active oxygen when irradiated by ultraviolet rays in the sunlight. Active oxygen oxidizes NOx in the air into nitric acid ions. Most of these ions are washed away. Nitric acid ions remaining on the surface of the block are neutralized by the alkaline nature of the concrete [24, 25].


16 16

2.3 AIR POLLUTANT MONITORING IN MISSISSIPPI Figure 9 shows the locations of Ozone nonattainment areas in the United States [26]. The air pollutant sites established by the Mississippi Department of Environmental Quality (DEQ) for monitoring EPA standards of key air pollutants are shown in Figure 10. Figure 10 also shows the Ozone data collected in the years 1991 and 1999 at the Tupelo Airport station. Figure 11 shows the high concentration Ozone data collected in Desoto County in the year 2000 [26]. Figure 12 shows historical time series data of maximum 1-hour values for Ozone and NOx collected in Desoto County.

Figure 9. Locations of nonattainment areas for Ozone in the United States [26]

17 17

Figure 10. Ozone and PM 10 data at Tu tation [26]

1991

O3 P

Moderate

Good

Unhealthful > 100

Tupelo, Lee County

Map of Mississippi Showing locations of DEQ Monitoring Stations

Southaven, DeSoto County

O xford , Lafayette County

pelo Airport DEQ monitoring s

Year 1 Final Report/ Uddin UM-CAIT/2001-01

August 2001


18 18

Figure 11. Ozone data at DeSoto DEQ monitoring station [26]

DeSoto County, MS

O3

Good

Moderate

Unhealthful


19 19

Figure 12. Ozone and NOx historical annual data collected in DeSoto county

Maximum 1-Hour Value for OzoneDe Soto County, MS (1996 - 2001)

0

20

40

60

80

100

120

140

160

180

200

Year

Ozo

ne c

once

ntra

tion,

ppb

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Unhealthy

EPA Air Quality Standard = 120 ppb120

Data source: EPA Air Data, www.epa.gov/air/data/info.htm

DeSoto

Maximum 1-Hour Value for NOxDe Soto County, MS (1998 - 2001)

0

20

40

60

80

100

120

140

160

180

200

Year

NO

x co

ncen

trat

ion,

ppb

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Data source: EPA Air Data, www.epa.gov/air/data/info.htm

DeSoto


20 20

3.0 REMOTE SENSING LASER MEASUREMENTS AND AIR QUALITY ANALYSIS The national strategy for air pollution abatement relies on the monitoring of air pollutants within the troposhere by the EPA, as required by the Clean Air Act. The implementation of NAAQS values of important pollutants requires each state to establish emission regulations and approval by the EPA. The emission regulations and inventories are based upon the modeled dispersion of pollutants from specific point sources which relate emissions to ambient concentrations. Subsequent compliance monitoring is generally done using a network of fixed air pollution monitoring stations in a geographical area [27]. These stations employ conventional gas analyzer instrumentation and air sampling close to the ground on localized surrounding areas only and are not able to measure the effects of dynamic changes in air shed and mobile sources. Recently, portable remote sensing infrared instruments have been used on the roadside to measure the emission from vehicle exhaust. The instrument consists of a non-dispersive infrared component for detecting CO, CO2, hydrocarbons, and a dispersive ultraviolet spectrometer for measuring nitric oxide [28]. These instruments are not considered remote sensing equipment for the purpose of this study and do not measure air pollution as affected by photochemical reactions, wind, and dispersion characteristics. Because the non-invasive tunable Lidar measurement is real-time and uses a long sampling distance, it is less time consuming and is free from sampling inaccuracies which plague the conventional monitoring stations and direct vehicle exhaust measurements. The effect of transportation and mobile source emissions on air pollution can be assessed, and the uncertainty in many parameters of dispersion models can be reduced if an investment is made on laser based remote sensing monitoring of air pollution. This section provides an overview of principles of tunable pulse laser measurement and available laser technologies for air pollutants. 3.1 PRINICIPLES OF DIAL MEASUREMENTS The majority of gas pollutants present or formed in air exhibit optical absorption bands in the ultraviolet (UV), visible (VIS), or infrared (IR) portions of the spectrum. Figure 13 shows the electromagnetic spectrum in terms of wavelength [29]. The wavelengths of the light range from about 1mm for the far infrared to about 10 nm for the extreme ultraviolet. The concentration of gas pollutants can be monitored by using a pulse laser over a long distance, noting the absorption obtained at one wavelength corresponding to a strong absorption band in the gas, and comparing it with the absorption at an adjacent wavelength where the gas does not absorb [30]. This is the basic principle of tunable pulse LIDAR measurement of concentrations of air pollutants. LIDAR is an acronym for LIght Detection And Ranging. Differential Absorption LIDAR (DIAL) has been used with success to monitor atmospheric pollutants, such as Nitrogen Dioxide, Ozone, Sulfur Dioxide, and Mercury vapors. The laser is tuned between ultraviolet, visible, near infrared, and thermal infrared spectral regions. The difference in the absorption of light at these different wavelengths is used to determine concentration of air pollutants. The general performance specification requirements of the DIAL system as a function of transmitted wavelengths along with the pollutant of interest and the corresponding energy and spectral linewidth requirements are summarized in Table 1 [31].


21 21

Figure 13. The UV and blue visible bands for DIAL measurements of Ozone and NO2 [29]

Table 1. General System specifications for DIAL technology to measure gas pollutants [31]

Wavelength Range Required

Pollutant of Interest

Detection Requirement

Energy @ 30Hz

Spectral Linewidth Required

250-260 nm Benzene, Toluene, Xylene

low ppb 25 mJ 3 GHz

280-300 nm Ozone (O3) 80-100 ppb 25 mJ 3 GHz 298-300 nm Sulfur Dioxide

(SO2) 20 ppb 25 mJ 3 GHz

250 nm Formaldehyde (CH2O)

low ppb 25 mJ 3 GHz

435 nm Nitrogen Dioxide (NO2)

30 ppb 25 mJ 3 GHz

350 nm Molecular Chlorine (Cl2)

low ppb 25 mJ 3 GHz

368 nm Nitrous Acid (HNO2)

low ppb 25 mJ 3 GHz

2.47 µm Hydrofluoric Acid (HF)

low ppb 25 mJ 600 MHz

1.65 µm Methane (CH4) low ppb 25 mJ 600 MHz 1.78 µm Hydrochloric

Acid (HCl) low ppb 25 mJ 600 MHz

The electromagnetic spectrum

10-310-410-510-6 10-2 10-1 109108107106105104103102101

Cosmic rays

γγγγ rays

X rays

Ultraviolet (UV)

VisibleNear-IRM

id-IRThermal IR

Microwave

Television

andradioWavelength (µµµµm) Wavelength (µµµµm)

0.4 0.5 0.6 0.7 (µµµµm)

UV Near-infrared (IR)

Visible(1 mm) (1 m)

BLU

E

GR

EEN

RE

DThe electromagnetic spectrum

10-310-410-510-6 10-2 10-1 109108107106105104103102101

Cosmic rays

γγγγ rays

X rays

Ultraviolet (UV)

VisibleNear-IRM

id-IRThermal IR

Microwave

Television


0.4 0.5 0.6 0.7 (µµµµm)


Visible(1 mm) (1 m)

BLU

E

GR

EEN

RE

D

10-310-410-510-6 10-2 10-1 109108107106105104103102101

Cosmic rays

γγγγ rays

X rays

Ultraviolet (UV)

VisibleNear-IRM

id-IRThermal IR

Microwave

Television


0.4 0.5 0.6 0.7 (µµµµm)


Visible(1 mm) (1 m)

BLU

E

GR

EEN

RE

D


22 22

3.2 AVAILABLE DIAL TECHNOLOGIES Progress in remote sensing of the atmosphere closely follows progress in laser technology [27, 31, 32, 33]. A mobile DIAL system has been developed in Europe, for mapping concentrations of air pollutants [34, 35]. A comparative study was done between Lyon, France; Stuttgart, Germany; Geneva, Switzerland; and Berlin, Germany. Its continuous monitoring from selected building tops is being used for long-term evolution of pollution distributions. Truck mounted and/or airborne DIAL technology uses active tunable laser sensors that are capable of measuring real-time air pollutants in the atmosphere with high temporal and spatial resolutions. Several remote sensing air pollution monitoring systems using DIAL technology have been developed during the last two decades in the United States [36, 37, 38, 39, 40], Germany [34, 35, 41], Japan [42], and Russia [43]. Most of these pulse LIDAR systems are costly, require water-cooling, have a dedicated electrical supply system, and generate electrical interference. Efforts are being made for the development of alternative, cheaper LIDAR technology [44] and use of Computed Tomography [45]. The use of LIDAR in the infrared range for hydrocarbons (VOCs) has been demonstrated in recent studies [37, 41]. Passive Fourier Transform Infrared (FTIR) remote sensing systems have been developed for measuring air pollution [46]. The laser systems operating in the infrared (IR) band offer a higher relative degree of eye safety as compared to the visible and UV regions [36]. However, the data interpretation and discrimination of several air pollutants in the IR range become complex. The reliable pulse laser DIAL technology has been selected in this study for further investigation and use on selected highway test sites to get independent Ozone or NO2 concentrations related to transportation and its effects on air quality. The tunable DIAL remote sensing technology on mobile (truck or van) and airborne platforms can provide real-time measurement and data analysis, wide-area and vertical profile monitoring from a distance of up to 2 km or more, and multi-pollutant concentration measurements. Cost and complexity of operation so far have prevented wide application. An overview of key DIAL technologies is presented in the following sections. 3.2.1 Elight Laser System The following description is extracted from Reference 35. Ozone concentrations at ground level are determined by photochemical processes and horizontal and vertical transports. These processes take place not only at ground level, but mainly in different altitudes of the troposphere. For explaining the origins of average Ozone concentrations (critical for the vegetation) as well as the causes of short episodes with high Ozone levels (critical for the human health), it is necessary to determine the vertical distribution and the long-range transport of Ozone and its precursors. The best possibility for monitoring air pollution in three dimensions over large time and space scales is provided by Lidar systems. Elight Lasers Systems GmbH developed new Lidar system with pulsed solid state laser sources for stand-alone operations. It uses a flash-lamp pumped Nd:YAG laser with a repetition rate of 20 Hz for the simulation of an optical parametric oscillator (OPO) with internal sum frequency generator (SFG). The pumping energy of the OPO with SFG is approximately 2 microjoule (mJ)


23 23

at a pulse duration of 5 ns. The OPO and SFG crystals are turned shot-by-shot. The beam is expanded to a diameter of 50 mm. The alternating laser pulses are transmitted via an emitter telescope vertically into the atmosphere. The light returning from the atmosphere is collected by the detection optics which have a diameter of 300 mm. The Ozone Profiler is designed for measuring the ground level Ozone concentration. This compact system allows the monitoring of the vertical Ozone distribution from about 100 m up to 2000 m with a reported accuracy of a few ppb and the spatial resolution down to several meters. Furthermore, the system delivers information about the vertical distribution of aerosols (as extinction), which is very important regarding the formation of secondary aerosols by photochemical processes [35]. The Ozone Profiler DIAL system sends out short laser pulses into the atmosphere. All along its path the light is scattered by the molecules of the air (Rayleigh scattering) and by small particles (Mie scattering). A small fraction of the light is backscattered to the Lidar system and is received by a telescope and a sensitive detector that is integrated into the system. The received signal is acquired as a function of time. With the constant value of the velocity of light and the time that the emitted and backscattered light needs until its detection, the distance from the backscattering location can be determined. In this way it is possible to get information about the spatial distribution of molecules and aerosols along the beam path, as shown in Figure 14. To determine the spatial distribution of Ozone, the DIAL technique is applied. It is based on the specific light absorption by molecules. According to the spectroscopic properties of the specific molecule (Ozone), the absorption depends on the wavelength of the incident light. Therefore, a DIAL System sends out laser pulses that alternate in two different wavelengths. One of the wavelengths (λon) is chosen for high selective absorption by Ozone (λoff). In the absence of this pollutant, the atmospheric return signals of both wavelengths are nearly equal. If the laser beam meets higher concentrations of Ozone (as sketched in Figure 15), a significant intensity difference in both signals is observed. The difference of the acquired signals allows the calculation of the Ozone concentration as a function of the distance along the beam [35]. 3.2.2 NOAA ETL Lidar and OPAL Systems Located in Boulder, Colorado, the Environmental Technology Laboratory (ETL) supports the environmental monitoring and stewardship charter of the National Oceanic and Atmospheric Administration (NOAA) by performing oceanic and atmospheric research and developing new remote sensing systems. ETL operates two Ozone DIAL systems - a ground-based Ozone DIAL and an airborne Ozone DIAL - which have been deployed to many field projects in the United States to study Ozone pollution [38, 39]. The following description is based on Reference 39. Lidar transmits short pulses of laser light into the atmosphere. The laser beam loses light to scattering and attenuation as it travels. At each range, some of the light is backscattered into a detector, as shown in Figure 16 (a). Because the light takes longer to return from more distant ranges, the time delay of the return pulses can be converted to the corresponding distance between the atmospheric scatter and the Lidar. The end result is a profile of atmospheric scattering versus distance, as shown in Figure 16 (b). Analysis of this signal can yield information about the distribution of aerosols in the atmosphere.


24 24

Figure 14. Measurement principle of laser system [35]

Figure 15. Measurement principle of Elight DIAL System [35]

Figure 16. Measurement principle of ETL Lidar [39]


25 25

The amount of backscatter indicates the density of the scatters. This can be used to measure cloud base height or track plumes of pollution. Properties of the atmosphere can also be deduced from the Lidar return signals. A frequency shift in the light due to the Doppler effect permits the measurement of wind speeds. By detecting the amount of depolarization, one can discriminate between liquid droplets and nonspherical ice particles. A Raman Lidar detects particular atmospheric components (such as water vapor) by measuring the wavelength-shifted return from selected molecules [39]. Differential Absorption Lidar or DIAL uses absorption, as evidenced by reduced backscatter from greater distances, to measure the concentration of atmospheric gases. Profiles of the atmospheric component constituents (such as aerosol particles, ice crystals, water vapor, or trace gases and Ozone), as a function of altitude or location are necessary for weather forecasting, climate modeling, and environmental monitoring. DIAL measurements of Ozone require less precise laser frequencies because Ozone has a broad (~200 nm) absorption band, as shown in Figure 18, instead of narrow lines like water vapor. This requires that the on- and off- wavelengths be chosen with sufficient wavelength separation to ensure a significant difference in their absorption. However, uncertainty in the changes of aerosol scattering and patterning between the wavelengths can introduce error. The difference in aerosol scattering at λon and λoff can be extrapolated from a third measurement taken at a longer wavelength, λa. Therefore, DIAL measurements with three wavelengths can be used to determine Ozone concentration profiles to good accuracy [39]. The ETL’s Ozone Profiling Atmospheric Lidar (OPAL) equipment is housed in a mobile laboratory constructed from a modified sea-going cargo container, shown at the 1999 Atlanta SuperSite (Figure 18). It is transported via an air-ride semi-tractor trailer. The OPAL system has been used for Ozone formation/evolution monitoring, Ozone transport studies, and Aerosol study in polluted areas. Some results are shown in Figure 19. Field projects include: • Aircraft Plume Study, May 2001 • TexAQS (Texas Air Quality Study), August/September 2000, Houston, Texas • 1999 SOS Atlanta SuperSite (Southern Oxidants Study), August 1999, Atlanta, Georgia • SOS '99 (Southern Oxidants Study), June/July 1999, Nashville, Tennsessee • SCOS (1997 Southern California Ozone Study), June-October 1997, El Monte, California • LIFT (Lidars In Flat Terrain), August 1996, Champain/Urbana, Illinois • 1995 Victorville Ozone Transport Experiment, August 1995, Victorville, California Typical equipment specifications are shown in Table 2 [39].


26 26

Figure 17. Measurement principle of ETL OPAL [39]

Figure 18. ETL OPAL equipment [39]

Figure 19. Comparison of ETL OPAL measurements [39]


27 27

Table 2. Typical specifications of ETL OPAL DIAL system, 1999 [39]

Wavelength 266 nm 289 nm 299 nm 355 nm Pulse energy 6 mJ 2 mJ 2 mJ 6 mJ Pulse rate 1 - 10 Hz Transmitter Configuration Multibeam Single beam Scanning Directions Vertical, horizontal, and in a single vertical plane Minimum range 60 m Aerosol: Maximum range 1.5 - 3 km 6 - 7 km 6 - 7 km 9 km Range Resolution 15 m Ozone: Maximum range 3 km

Range Resolution below 1 km: 60 - 90 m 1 - 2 km: 100 - 150 m 2 - 3 km: 150 - 300 m

Accuracy: Better than +/- 10 ppb

3.2.3 Skyborne DIAL System An airborne optical remote sensing system designed and built by Skyborne is reported as the world's most sensitive airborne Lidar system, based on its ability to detect less than 1 part per trillion of mercury vapor in the atmosphere [40]. High-rep-rate, narrow linewidth excimer-pumped dye laser systems are preferred for airborne or truck mounted mobile applications to measure several types of pollutants. In comparison, all-solid-state tunable YAG-pumped OPO systems have been used for some airborne applications and ground-based monitoring systems for reasons of simplicity, reliability and convenience. However, these are used for limited numbers of pollutants. The Skyborne airborne or truck mounted tunable Lidar system is able to detect a wide variety of air pollutants. Remote concentration measurements can be performed on NO2, Ozone, benzene, NO, SO2, mercury vapor (Hg), and other target atoms and molecules with suitable spectral absorption characteristics. Typical detection sensitivities are 1 to 100 ppb at ranges of several kilometers, depending on signal averaging and other considerations (absorption cross section, linewidth, atmospheric conditions, etc.). All Lidar systems designed by Skyborne meet or exceed ANSI Z.136 standards for eye safety when used in accordance with the manufacturer's operating directions [40]. The following description is based on References 40 and 47. The Skyborne airbone DIAL system is composed of two high–rep rate narrow linewidth excimer–pumped dye laser systems which can be easily tuned across the entire UV-VIS-NIR spectral range from about 2,500 Angstroms to about 11,000 Angstroms using a range of laser dyes solvents and non-linear frequency-doubling crystals. Figure 20 shows the airborne system in DC-3 aircraft and Figure 21 shows a schematic of the system setup.


28 28

Figure 20. The Skyborne airborne DIAL system [Coutesy of Skyborne, Inc]

Figure 21. A schematic of Skyborne DIAL system [Coutesy of Skyborne, Inc]


29 29

The lasers are tuned to the absorption region of interest. One is locked on to the absorption line of the target molecule (viz., 2,536.5 Angstrom Hg resonance line), and the other is slightly detuned, i.e. displaced a certain distance away from that absorption line in k space, creating a discrete differential absorption pulse pair (signal and reference wavelengths) that can be used to probe the airspace beneath the airplane. The signal and reference pulses are sequentially transmitted into the sampling space beneath the plane and are received back aboard the plane after reflecting from aerosols and other natural constituents of the atmosphere via elastic backscattering. The time delay between the two pulses is approximately 150 microseconds. Free space optics are used to project the beams from the plane, and a nadir-looking, large-aperture receiver is used to collect the backscattered light. The laser firing circuitry, detection electronics, and sophisticated program control allow continuous longpath DIAL concentration measurements every 1-3 feet beneath the plane’s belly along the flightpath at normal sorty speeds. A range gate can be used to vary the “sounding depth” of the measurement. Because of the large-aperture, high-efficiency receiver design, extremely low levels of transmitted light are needed to perform a remote spectroscopic measurement. In one recent instance, Skyborne was able to perform background-level concentration measurements of atomic Hg in the atmosphere using about 0.5 microjoules of transmitted energy at both the on-resonance and off-resonance wavelengths. The optical transmitter consists of a power supply, two high-rep rate excimer lasers, a cooling system for the excimer lasers, two high-rep rate narrow line dye lasers, two temperature-controlled, vibration-damped frequency doubling crystals, and projection optics. The optical transmitter is under complete computer control. Either or both excimer pump lasers may be fired, the dye lasers may be scanned, the crystal phase matching angles may be adjusted, and the rep-rate of either or both laser systems may be adjusted using a microcomputer program. The power for the optical transmitter comes from two 28 volt, 200 amp generators which are attached to the plane’s engines. These are the plane’s main generators, which also feed the avionics and the electromechanical features on the plane. DC power from the main generator bus is allowed to pass back into the cabin after passing through a master cutoff switch/circuit breaker in the DC-3 cockpit. A bank of Avionics Instruments Model 1000 inventors are used to invert the DC current into 3 phase 60 cycle, 208 volt AC power which is used to feed the excimers’ West German power supplies. The intrinsic firing control circuitry of the lasers allows the excimers to be fired at a rate of 200 Hertz. However, the power available on the DC bus currently limits operation to about 70 Hertz. Current consumption is monitored by the co-pilot. The excimer lasers (Lambda Physik EMG 53S MSC) are operated using xenon chloride which lases at 308 nanometers. The laser gas is usually obtained in a pre-mixed form. Both EMG 53S MSC excimer lasers are specified to provide 100 millijoule pulses of t = 18 nanoseconds at 308 nm using fresh xenon chloride mix. When the excimer lasers were first delivered, each laser delivered considerably more than specifications, one delivering 125 millijoule pulses and the other delivering 110 millijoules consistently. The output energies of the excimers can be continuously adjusted using the high voltage controls on the power supply modules which vary the high voltage on the discharge circuitry from 18 to 21 KV. In general, the high voltage is set at an intermediate value to take advantage of the greater pulse intensity regularity at lower


30 30

voltage settings. Approximately 90% of the energy taken by the lasers is emitted as heat which must be dissipated for reliable operation. Operating at 70 Hertz, the excimers dissipate approximately 990 watts of waste heat. The cooling system employs a 2’ x 2’ x 2’ aluminum chiller with an integral pump which can hold 200 pounds of readily-available ice. The large amount of cooling capacity (42,000 btus) of this thermal mass allows operation of the Lidar system for a 3-4 hour measurement sorty. The excimers pump the two dye lasers (a pair of Lambda Physik FL3002’s) which are mounted in bunk-bed fashion, one atop the other. Two narrow-bandwidth photon sources are necessary in order to keep the time interval between the two sequentially transmitted pulses as short as possible. The dye lasers employ an oscillator and amplifier and are connected to the main control computer via an IEEE 488 link. The dye lasers have a high peak power (20 MW) and a high average power (10 Watts) capability. The linewidth of the FL3002 is 0.2 cm-1. The output from the dye lasers can be used to drive nonlinear optical crystals to obtain the second harmonic output in the ultraviolet, if so desired. In the case of mercury measurements, BBO crystals were used with green light to provide outputs in the 2537 Angstrom mercury absorption region. The total generated UV was about 1.0 microjoules. The outputs from the two laser systems are combined in a 2” diameter quartz beam, which consists of a CVI Laser #26-50 quartz plate mounted in an ORIEL 14501 two axis adjustable two inch mount. A fraction of the energy of each beam is diverted into a calibration section to provide shot-to-shot measurements of differential absorption cross sections and initial intensities, while another fraction of the energy of each beam is transmitted into the probe space. A beam steering mirror placed after the beam combines, allows the axis of the transmitted optical beams to be made parallel with the receiver axis. The geometry of the beam projection and calibration optical sections has been carefully designed to minimize initial optical alignment times and to maximize transmitted pulse and intensity. The optical receiver consists of a quartz condensing lens which has a focal length of 27 inches and a diameter of 18 inches. The lens was fabricated from a single piece of fused silica and precision ground to the desired surface flatness. The lens is mounted in a precision two-axis aluminum mount, which allows for alignment with the transmitted beam and the flight axis of the aircraft. The aluminum mount also consists of a fixed photodetector mount located at the focal point of the lens and integral light-absorbing optics cover to eliminate stray light. The thickness of the lens at the midpoint is approximately 3.75 inches. The internal efficiency of transmission at 2537 Angstroms is 94%. The use of a single condensing lens instead of a multi-element receiver (viz., Newtonian reflector) considerably reduces the difficulty of optical alignment of the system. A DIAL concentration measurement requires knowledge of the transmitted and received optical intensities at both the on and off resonance wavelengths as well as a knowledge of the differential absorption cross section and pathlength. To determine these parameters on a shot-to-shot basis, the airborne Lidar system contains an optical calibration section near where the two light pulses are directed out of the plane. The calibration section consists of two photodetectors and a temperature-monitored sealed optical cell, which contains a known concentration of the


31 31

target gas. In the case of mercury, the cell consists of a 1 cm pathlength quartz cube which contains a saturated vapor pressure of HG issuing from a small liquid droplet in the base. A fraction of the on and off resonance beams are directed through this cell and fall upon a photodetector, while another fraction impact on a photodetector before passing through the cell. The digitized intensities of these four values in conjunction with the two received intensities and a knowledge of the remote sounding depth are used to determine the relative concentration of the target gas. The temperature of the calibration cell is monitored with a digital thermometer accurate to about 0.2 degrees Celsius. Detailed procedures for the measurements of NO2 and O3 are described later. 3.3 SELECTION OF DIAL TECHNOLOGY AND TEST SITES The remote sensing tunable pulse Lidar technology implemented by Skyborne, Inc. presented the most promising option for measuring NO2 and O3 air pollutants using the same basic truck mounted DIAL equipment. The cost was reasonable within the Year 1 budget resources of the NCRST-E Air Quality Project. The following information of measurement methodology for measuring NO2 and O3 were provided by Skyborne [47]. 3.3.1 Tropospheric Ozone The Ozone molecule can be measured in the Huggins-Hartley absorption band using two wavelengths, which are chosen for maximum differential absorption species in SO2. Therefore, care must be taken to minimize SO2 interference in the measurement. The two wavelengths chosen to measure Ozone are 2,923 and 2,940 Angstroms (A) because at these two wavelengths, the absorption due to SO2 is the same and therefore cancel out. Table 3 shows the absorption coefficients for Ozone and SO2 in this wavelength interval [48]:

Table 3. The absorption coefficients for Ozone and SO2

Wavelength (A) Absorption Coefficient Sigma, SO2 (cm-1 atm-1)

Absorption Coefficient Sigma, O3 (cm-1 atm-1)

2,923 26.0 28.0 2,933 14.0 24.4 2,940 26.0 22.0 2,952 12.5 19.5

Note: 1 micrometer (µm) = 1x 104 Angstroms (A) Thus, for Ozone, σ(1) - σ(2) = 6.0 (cm-1 atm-1), while for SO2, σ(1) - σ(2) = approximately zero. Figure 22 shows the absorption spectra of these two molecules around 3,000 Angstroms [48]. Another off-resonance wavelength, which can be used is 3010 Angstroms because at this wavelength the SO2 absorption is also approximately 26 cm-1 atm-1. However, this wavelength is outside the bandpass of inexpensive, commercially available, multi-cavity interference filters, so it must be used at night.


32 32

Figure 22. Absorption spectra of O3 and SO2 [48]

The following two optoelectronic models show two different scenarios in order to give the reader a flavor for what is happening in a truck mounted Skyborne EarthStar DIAL measurement. In Scenario One, the 3,010 Angstroms off-resonance wavelength is used at night at a range of 3.5 kilometers (with one hour signal averaging, i.e., 18,000 pulses) where the differential absorption of 0.359% is clearly resolvable against a minimum detectable fractional change of 0.346%. The concentration detectable in the target zone at 3.5 kilometers is 10 ppb and the target zone is 100 meters length. This shows the system operating at the limit of its practical detection capability. In Scenario Two, the model shows an attempted measurement made during the day using the 2,940 Angstroms off-resonance wavelength at a range of 4 kilometers using a 5 minute signal averaging time. This measurement is doomed to fail (even though the hypothetical target gas concentration is 1 ppm) because the operator is trying to operate the system outside its range. The differential absorption from the 1 ppm * 100 meter concentration “anomaly” at 4 kilometers is 8.78%. However, the system is only able to discriminate a 29.43% fractional signal change at the received power level of 3.24 e-11 watts. This type of analysis may be extended to other hypothetical scenarios in order to systematically build up a table of detection sensitivities versus range, length of target zone (i.e., vertical resolution), signal averaging time, and diurnal measurement epoch. In each case, the question of whether the differential absorption from a given concentration-pathlength “anomaly” is resolvable against the minimum detectable fractional change in the lowest received power level is determined on a case by case basis. Up to 1 or 2 kilometers for reasonable signal averaging times (i.e., 5 to 60 minutes), the system has sub-part per billion detection sensitivity. Table 4 shows Skyborne system specifications for measuring O3 [47].

200

100

50

20

10

5

22600 2700 2800 2900 3000 3100

O3

SO 2

Wavelength (A)

Absorption coefficients of SO2 and O3


33 33

3.3.2 Tropospheric Nitrogen Dioxide Nitrogen dioxide is relatively easy to measure because its absorption band lies in the visible part of the spectrum, which can be directly accessed by the output of the optical parametric oscillator without the need for frequency doubling. Thus, rather high pulse energies are available. Also, around 4,500 Angstroms, the nitrogen dioxide absorption is largely interference free. (The strong visible absorption band of this molecule, making it easy to measure, also makes it lose the sky’s lovely blue color; thus, the brown color of smog is not just due to the presence of soot and particulate matter, it is also due to a paucity of blue light.) Nitrogen dioxide can be measured using 4,478.5 Angstroms as the on-resonance wavelength and 4,500.0 Angstroms as the off-resonance wavelength. The on-resonance absorption cross section is 6.38 e-19 cm2/molecule; the off-resonance absorption cross section is 3.5 e-19 cm2/molecule. The NO2 spectrum, shown in Figure 23, follows from Ahmed [48]. Results from the use of the optoelectronic model to assess system sensitivity versus range indicate that the Skyborne EarthStar I system has part per trillion (ppt) level detection limits over much of its operating range. The optoelectronic model shows a typical measurement scenario in which the Lidar system is used to detect an average concentration of 2 ppb at 5 kilometers in a range increment of 100 meters (vertical resolution) using a 5 minute signal averaging tie under night sky conditions. The minimum fractional change in the lowest received power level (on-resonance) which is detectable is 0.020%. The concentration-pathlength product “anomaly” under consideration gives a 0.031% differential absorption signal. Therefore, it is resolvable under the assumed conditions. Table 4 shows Skyborne system specifications for measuring NO2 [47].

Figure 23. Absorption spectra of NO2 [48]

4,300 4,400 4,500 4,600 4,700 4,800

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

NO2 : 2.6 X 1017 cm-3

L = 10 cmLINE BROADENINGEQUIVALENT TOAMBIENT CONDITIONS

WAVELENGTH (A)

NO2 ABSORPTION CURVE

NO

2 ABS

OR

PTIO

N C

OEF

FIC

IEN

T (c

m-1

) x L

(cm

)


34 34

Table 4. Skyborne truck mounted DIAL - Optoelectronic model system specifications for measuring NO2 and O3 on MS Highway 6 Test Site in Oxford, Mississippi [47].

March 18, 2001

On-resonance Off-resonance On-resonance Off-resonanceWavelength (Angstroms): 4,478 4,500 2,923 2,940 Laser Pulse Energy (mj): 9 9 1 1Laser Pulsewidth (ns): 18 18 18 18

Beamsteering mirror, M1, efficiency: 85.00% 85.00% 85.00% 85.00%Beamsplitter, BS1, efficiency: 50.00% 50.00% 50.00% 50.00%Total transmitting optics efficiency: 42.50% 42.50% 42.50% 42.50%

Transmitted Signal (watts): 212,500 212,500 23,611 23,611

Absorption Cross-section (cm2/mol): na na na naAbsorption Cross-section (cm-1*atm-1): 1.72E+01 9.41E+00 2.80E+01 2.20E+01Visibility (km): 20 20 20 20Mie Scattering Coefficient (km-1): 0.33474754 0.330482914 1.02183825 1.006452348Rayleigh Scattering Coefficient (km-1): 0.44 0.44 0.14 0.14Other atmospheric attenuation (km-1): 0.05 0.05 0.05 0.05Total Volume Extinction Coefficient (km-1): 0.82474754 0.820482914 1.21183825 1.196452348

Ambient Conc. of Target Molecule (atm): 1E-08 1E-08 3E-08 3E-08

Length of Target Zone (m): 1E+02 1E+02 1E+02 1E+02Conc. of Target Mol. in Target Zone (atm): 5E-08 5E-08 4E-08 4E-08

Atmospheric Reflectivity (%): 0.66% 0.65% 0.99% 0.97%

Receiver Area (cm2): 1,241 1,241 1,241 1,241 Receiver Efficiency (%): 80.00% 80.00% 80.00% 80.00%Narrow Bandpass Filter Trans. @ lamba 0.36 0.36 0.15 0.15

Range (km): 0.5 0.5 0.5 0.5

Total Atmospheric Transmission: 42.718097% 43.404897% 27.062038% 28.048014%Difference -0.687% -0.986%

Loss from Scattering: 56.165436% 55.978098% 70.235038% 69.773536%Difference 0.187% 0.462%

Absorption from Background: 1.705292% 0.936586% 8.056874% 6.386914%Difference 0.769% 1.670%

Absorption from Target Zone: 0.856313% 0.469395% 1.113751% 0.876139%Difference 0.387% 0.238%

Received Optical Signal (watts): 2.71468E-05 2.74314E-05 1.19399E-06 1.2211E-06Photocathode Responsivity (amps/watts): 1.00E-03 1.00E-03 2.00E-03 2.00E-03Photocathode Output (milliamps): 2.71468E-08 2.74314E-08 2.38797E-09 2.4422E-09Amplification: 1E+08 1E+08 2E+08 2E+08Input resistance (ohms): 50 50 50 50Voltage (millivolts): 135.7341308 137.1572305 23.87972792 24.4220059

Detector: Thorn 9798 Thorn 9798 EMR RbTe EMR RbTeBandwidth (Hz): 1.00E+07 1.00E+07 1.00E+07 1.00E+07Dark Current (amps): 6.00E-10 6.00E-10 1.50E-11 1.50E-11Number of pulses: 15,000 15,000 15,000 15,000Electronic Charge, q (coulombs): 1.60E-19 1.60E-19 1.60E-19 1.60E-19Signal to Noise Ratio: 10 10 10 10Min. detectable change, dark i limit: 0.0401% 0.0399% 0.1342% 0.1327%

Average 0.040% 0.134%

Minimum Detectable Power (watts): 1.13208E-09 1.13208E-09 8.94986E-11 8.94986E-112012 Counts expected: 271.5 274.3 47.8 48.8

Target gas: OzoneTarget gas: Nitrogen Dioxide

Optoelectronic ModelOxford, MS Highway 6 Project: Range = 500 m ( 1,640 ft )


35 35

3.3.3 Test Site for Year 1 DIAL Measurements Pursuant to an extensive literature review on current remote sensing methods used and data collected in North America and Europe to measure air pollution described in earlier sections, potential test sites were investigated. It is important to study and quantify the effects of traffic type and volume, traffic gridlock, and urban growth on air quality by studying rural to heavily populated urban areas. The research plan for the air pollution and transportation study related pollution modeling over four years is primarily focused on traffic characteristics, vehicle emissions, climatic parameters, and other significant urban/rural/socio-economic issues that may influence the air pollutant measurements and air quality models. An experiment design has been developed to select sampling sites ranging from rural to heavily urban areas. Other factors associated with congestion, commuter traffic, and driving distances are considered. The following primary study sites have been identified for DIAL measurements of air pollution and air quality analysis.

1. Rural area and low traffic site, Northern Mississippi - Oxford, a rural small university town in Northern Mississippi, is surrounded by three major state highways (MS Highway 6, MS Highway 7, and SR 30), and Interstate I-55 about 24 miles to the west. MS Highway 6 is the first test site.

2. Rural area, heavy traffic site, Auburn, Alabama – The test site is the Asphalt Highway

test Track, located about 15 miles from Auburn University and Opelka, Alabama. This 1.7 mile, heavy-duty asphalt pavement test track has been constructed by the National Center for Asphalt Technology (NCAT) at Auburn University. The construction was completed in September 2000, and since then, it has been subjected to accelerated heavy test truck traffic at about 5 million standard Equivalent Single Axle Load (ESAL) applications per year. This is a candidate site for Year 2 measurements.

3. Urban/ suburban area and heavy to medium traffic site, Northern Mississippi –

Southaven, a growing commercial center in Northern Mississippi, is near the Mississippi state line and the Memphis International Airport. This area, between I-55 and US Highway 78, has numerous commercial distribution centers and heavy truck traffic. This site will be used for Year 3 and/or 4 measurements.

4. Urban/ suburban area and heavy traffic site, Atlanta area, Georgia - A candidate site.

Mississippi Highway 6 in Oxford near the University of Mississippi is the first pilot test site. The University of Mississippi is assisting the City of Oxford for the deployment of an Intelligent Transportation System (ITS) using a 1.18 million dollar US DOT federal grant. An important part of this project is an airborne laser mapping and color aerial photogrammetry of the entire city and the development of a digital terrain model. It also involves a comprehensive geographical information system (GIS) for the entire city of Oxford, an asset management software, a traffic surveillance study, and a real-time traffic tracking study for improving emergency response and incident management. Figure 24 shows a map of Oxford area and the location of MS Highway 6 test site.


36 36

Figure 24. MS Highway 6 test site for Year 1 DIAL measurements of NO2 and O3

Oxford ITS Project

LIDAR Mapping

Oxford

Tupelo

City of Oxford

Lafayette County Mississippi

Air Quality Test Site MS Highway 6


37 37

4.0 YEAR 1 DIAL MEASUREMENTS IN NORTHERN MISSISSIPPI Airborne LIDAR and color aerial photography survey missions were carried out for the Oxford area in March 2001 and April 2001 as a part of the Intelligent Transportation System project (cooperative feature). This will be used to develop a GIS map for Oxford. The air quality project will have access to the GIS and a digital terrain model being developed for the Oxford ITS project. This will provide a GIS layer of orthorectified color raster image of Oxford, a GIS vector layer of roadway planimetrics, a GIS layer of 2-ft contours, and a GIS layer of traffic volume data throughout the city including the MS Highway test site, West Oxford Exit Intersection on MS Highway 6, and nearby Jackson Avenue. An image by IKONOS satellite has been acquired through the Mississippi Space Commerce Initiative (MSCI) project. Figure 25 shows the image and photo of traffic data collection by CAIT project staff on MS Highway 6. 4.1 DIAL TEST SETUP IN OXFORD, RESULTS, AND DISCUSSIONS 4.1.1 Skyborne Truck Mounted DIAL Setup The Skyborne Skylar I airborne UV DIAL system was re-configured and modified to fit on a truck in New York for this series of measurements. The skyborne truck was driven to Oxford during the second week of May 2001 for measuring air pollutants over the highway section adjacent to the University of Mississippi campus. The DIAL truck was setup at an athletic field on the University of Mississippi campus (a typical rural area with low to medium traffic volume). The measurements were performed on this selected test site, about 750 ft north of MS Highway 6 West. Figure 26 shows the Skyborne DIAL truck and tunable laser equipment. The Skyborne high-rep-rate, widely-tunable UV DIAL system has been described previously in Section 3.2.3. This Lidar system was designed to be operated from an aircraft for downward-looking measurements of trace gas species in the troposphere. The system consists of two narrow-linewidth LAMBDA PHYSIK dye lasers pumped by two Lambda Physik XeCl excimer lasers. The rep rate of the system is continuously adjustable from 0 to 200 Hertz. Narrow linewidth light is generated in the dye lasers and transmitted to the measurement region by free space projection optics which also tap-off a small percentage of the light for calibration purposes. The light is backscattered from the atmosphere and collected by an 18" diameter quartz lens. The intensity of the backscattered light is then determined by sensitive photomultiplying tubes (viz., EMI 9798QB). In simple terms, the DIAL long-path optical measurement, which the Skyborne system makes, is merely a normal absorption spectroscopy measurement which is made in situ through the atmosphere over a given pathlength rather than through a conventional sample gas cell. The intensity of the backscattered light is determined by the spectroscopic absorption cross-section of the target molecule at the on- and off- resonance wavelengths chosen for the measurement, the pathlength the light travels over is measured, and the concentration of the molecule is thereby determined. By comparing the intensity of the backscattered light to the transmitted light at on- and off-resonance wavelengths, a reading of the concentration of the target air pollutant can be made [40, 47].


38 38

Figure 25. IKONOS image of Oxford area near MS Highway 6 West and DIAL test site (courtesy of UM Geoinformatics Center and Space Imaging, Inc.)

Air Quality Test Site MS Highway 6 West

Traffic Data Collection

MS Highway 6 East to Tupelo

Jackson Avenue

West Oxford Exit

To I-55, Batesvill

Traffic Counts

MS Highway 6 West Oxford


39 39

Survey of test site

DIAL truck Newtonian telescope inside DIAL truck (pointed towards target)

Figure 26. Skyborne DIAL truck setup at Oxford test site near MS Highway 6 West

MS Highway 6 WestMS Highway 6 WestMS Highway 6 West


40 40

The equation used to derive the concentration is a variant of the well known Lambert-Beer Law of optical absorption. The real-time in situ character of the long-path spectroscopic measurement allows the concentrations of trace gases to be measured as they naturally exist in the atmosphere, over pathlengths which are more representative of actual volume-averaged concentrations than point monitors. There is no competing species to NO2 in the 4,500 Angstrom band. There is a rather large differential volume scattering coefficient between 308 nanometers and 4,500 Angstroms, due to differential Rayleigh and Mie scattering; however, because of the convenient availability of the XeCl 308 nanometer output and because of the possibility that simultaneous (or near-simultaneous) measurements of both NO2 and O3 could be made with the proper combination of wavelengths, 3080 Angstroms (308 nanometers) was chosen as the on-resonance wavelength for Ozone for this work, 4478.5 Angstroms (447.85 nanometers) was chosen as the on-resonance wavelength for nitrogen dioxide, and 4500 Angstroms (450 nanometers) was chosen as the off-resonance wavelength for both NO2 and O3. Previously, Uchino [50] used a single frequency output of a XeCl laser to measure stratospheric O3; and others have used the frequency-doubled Nd:YAG line at 532 nanometers as an off-resonance line to correct for aerosol scattering effects, but to our knowledge, this was the first use of the 308 XeCl line and 4,500 Angstrom line in combination for O3 measurements. When making aircraft measurements, the Lidar system configuration consists of two separately pumped FL3002e dye lasers which are mounted one on top of the other in “bunkbed” fashion. One is tuned to the on-resonance wavelength of the atom or molecule of interest and the other is tuned to the off-resonance wavelength. Rapid switching back and forth between the two measurement wavelengths is performed by cybernetic control. Switching times of approximately 1 millisecond are typical by first triggering one laser and then the other. The rapid switching back and forth between the on-and off-resonance wavelengths significantly reduces the effect that non-stationary aspects of the atmosphere have on differential scattering and also reduces the effect that changing background light levels have on the differential signal. But while using two laser trains significantly increases the accuracy of the system, there are increased expenses associated with the care, complexity, and maintenance of the second laser system. Due to the monetary constraints of this project, only a single laser system was used. Measurements at different wavelengths were made by sequentially tuning from one wavelength to another. Other modifications to the airborne DIAL system which were made to allow ground-based horizontal measurements of NO2 and O3 included replacing the large 18" diameter quartz lens as the receiving element with a 72" focal length, 17.5" diameter Newtonian telescope (Coulter Optical); mounting the system in a truck for movement of the system to/from the field sites and for housing and protection of the system while at field measurement site; and using a rotary phase convertor for electric power. Furthermore, because of problems experienced with a CAMAC crate (Bi Ra Systems Model 4450), the measurements of the electrical outputs of the photomultiplying tubes which measure the optical signal strengths of the transmitted and received Lidar returns (as well as the calibration returns) were made with either a Beckman Industrial Circuitmate 9020 20 Megahertz oscilloscope, or with a Tektronix Model 11401 Digitizing Oscilloscope with the Model 11A32 two channel amplifier plug-in. The Beckman


41 41

9020 oscilloscope has a 1 Mega-ohm input impedance (35 picofarad capacitance) with a voltage resolution of 500 microvolts. The Tektronix 11401 has a selectable 1 Megaohm (15 picofarad capacitance) or 50 ohm input impedance with a voltage resolution of 50 microvolts and a time resolution of 500 picoseconds. Furthermore, this scope is capable of displaying a digital readout of pulsed waveform parameters such as integrated pulse energy and peak to peak voltage. Using its two input channels, displayed on a single display screen, it is easy to compare the signal strengths of the transmitted and received Lidar returns [47]. 4.1.2 DIAL Measurement Principle and Test Setup The sensitivity of a DIAL measurement is proportional to the standard deviation of signal returns, the optical pathlength measured over, the absolute magnitude of the differential absorption cross-section, and the square root of the number of pulses integrated over [40, 47]. The equation for sensitivity is: Nmin = ½ * rho *1/(sigma1-sigma2) * 1/L *1/n^0.5 Eq. 1 Where Nmin = Minimum Detectable Concentration (atm) rho = Standard Deviation of Signal Returns (%) sigma1 = On-resonance Absorption Cross-section (atm-1 cm-1) sigma2 = Off-resonance Absorption Cross-section (atm-1 cm-1) L = Optical Pathlength (cm) n = Number of pulses integrated over The standard deviation term, rho, is the minimum change in signal that can be differentiated from the noise (for a signal-to-noise ratio of one). It incorporates all of the noise terms of the measurement, including: optical noise associated with the non-stationary character of the atmosphere and the electrical noise associated with the photodetectors and measuring electronics. One can see from this equation that in order to increase the sensitivity of the measurement, as well as maximizing the difference in absorption cross sections between the on and off-resonance wavelengths, it is also necessary to increase the pathlength as much as possible. Increasing the number of pulses integrated also increases the accuracy. Constraints on the pathlength are set by the measurement site and constraints on the physical space posed by the measurement site. Constraints on the number of pulses integrated over are set by the rep rate of the apparatus (e.g., 50 Hertz) and by the signal averaging time (i.e., the patience of the system operator). Certainly, a convenient signal averaging time can be found within a 1 to 10 minute period without straining the patience of the system operator. The magnitude of the Lidar optical signal received by the photodetector is proportional to the magnitude of the transmitted signal, the receiver area, and other factors. The equation for the received signal strength (no retroreflector) at one wavelength is: Pr(f,R)=Pt * k * A * exp { -2 [Sigma(f)*N +B] dR } / pi R^2 Eq. 2


42 42

where Pr(f,R) = Received Optical Power (watts) Pt(f,R) = Transmitted Optical Power (watts) k = Optical Efficiency (%) A = Receiver Diameter (cm^2) Sigma(f) = Optical Absorption Coefficient (cm-1 atm-1) N = Concentration of Target Molecule (atm) B = Volume Scattering Coefficient (cm-1) f = Frequency of Measurement index R = Optical Pathlength (cm) For horizontal measurements during these field tests, a retroreflector has been used. When a retroreflector is used instead of the naturally-occurring constituents of the atmosphere to reflect the transmitted light, the denominator of the right hand side of Equation 2 is replaced by a factor equal to the reflectivity of the retroreflector at that particular wavelength multiplied by an overlap function which expresses the percentage of the outgoing beam cross-section which is captured by the retroreflector and returned to the receiver multiplied by a factor which expresses the overlap of the returned beam with the receiver field of view. This factor may vary from 1 to 25% depending on the size of the retroreflector distance it is placed from the Lidar system and optical alignment. In any event, Equation 2 may be divided by an identical equation expressing the power which is received at the other wavelength. The ratio of the returned signal intensities at the two wavelengths (on and off-resonance) is then obtained, viz.: ln [( P’r(f,R)/P’t )/ (Pr(f,R)/Pt) ] = { (Sigma’(f)-Sigma(f) )* N +(B’-B)}* R Eq. 3 In general, for wavelengths close together, it is safe to assume that B’-B is equal to zero; but for some wavelength pair separations, this assumption is incorrect and it is necessary to include a correction to adjust for the effect this term has in modifying the logarithm of the ratio of ratios. The way to adjust for it is to calculate the Rayleigh and Mie scattering at each wavelength, add them together to obtain B’ or B, and then substitute them into the above equation. Then the concentration, N, is obtained by the use of this equation. For the measurements of nitrogen dioxide, because the on and off-resonance wavelengths are so close together, it will be assumed that B’ = B; therefore, B’-B vanishes. But for the differential absorption measurements on Ozone at 3080 Angstroms (on-resonance) and 4500 Angstroms (off-resonance), it would be necessary to include a correction factor. An absorption cross section of 3.5 cm-1atm-1 has been calculated for Ozone at 308 nanometers (nm) for 0.7 nm linewidth light from a XeCl laser following the procedure of Uchino [50], compared to 4.0 cm-1 atm-1 at 308 nm and 0.007 cm-1 atm-1 at 450 nm reported by Inn and Tanaka [51]. Thus, the differential absorption coefficient between Ozone at 308 nm (an average

43 43

of Uchino and Inn and Tanaka, 3.75 cm-1 atm-1) and Ozone at 450 nm is 3.75 - 0.007 = 3.74 cm-1 atm-1. And the differential absorption of NO2 between 4,478.5 Angstroms (on-resonance wavelength ) and 4,500.0 Angstroms (off-resonance wavelength), based on the work of Ahmed [48] is 7.8 cm-1 atm-1. Using these values of the differential absorption cross sections and an expected standard deviation of 2% (after signal averaging) in the normalized signal ratio, detection sensitivities for different ranges have been computed using Equation 1, as shown in Table 5.

Table 5. Detection sensitivities for different ranges [47] Calibration of a DIAL system can be performed nown concentration pathlength product into the beam n the strength of the signal returns. For cells with twice the pathl ule concentration, the effect is twice as great on the logarithm of I/ used to determine the atmospheric number density of the target mo g the line, connecting the added optical densities until zero ed. At this point, the logarithm of received versus transmitted int nd the value is used in conjunction with the path length and ab ine the atmospheric concentration of the target molecul d addition method [47]. Figure 27 shows some calibration steps. dar signal in the blue band of NO2.

Nitrogen Dioxide (NO2) Range (meters) Detection Sensitivity (atm)

50 2.6e-7 100 1.3e-7 150 8.5e-8 200 6.4e-8 250 5.1e-8 500 2.6e-8 1,000 1.3e-8

Ozon

Range (meters) (atm) 50 100 150 200 250 500 1,000

by inserting optical cells of a kpath and observing the effect oength or twice the target molecI(0). This method may also be lecule by graphically continuin added optical density is reach

ensities (Eq. 3) is determined asorption cross section to determe. This is known as the standarFigure 28 shows a real-time Li

e (O3) Detection Sensitivity

5.3e-7 2.7e-7 1.8e-7 1.3e-7 1.1e-7 5.3e-8 2.7e-8


August 2001


44 44

NO2 Calibration cell showing brownish color against blue sky

Figure 27. Skyborne DIAL calibrations at Oxford test site

Laser beam in NO2 band Skyborne Excimer Laser System

Figure 28. Real-time Lidar signal of Skyborne DIAL system at Oxford test site


45 45

4.1.3 DIAL Measurement Protocol The DIAL measurements were performed following the termination of the Spring semester from May 17 to May 25, 2001 at the University of Mississippi athletic field which is adjacent to Highway 6 West in Oxford, MS. The highway is elevated relative to the athletic fields and the measurement site by approximately 50 ft (15.2 m). The Lidar truck was parked approximately 750 ft (228.6 m) from the edge of the highway, and was orientated approximately parallel to the highway. Measurements were first performed to assess the potential usage of an upward-looking optical configuration to obtain a vertical profile of pollutant concentration. 1Once this is done, the optical axis of the outgoing beam and the receiving telescope will be aligned for all other altitudes. However, difficulties in acquiring the signal from a hard-target reflector (viz., a low-lying cloud with sufficient reflectivity) made this type of measurement impossible. Instead, horizontal measurements were made using the retroreflector. At this site, a 212 ft (64.6 m) ranging distance was available over a horizontal path at a height of about 10 ft (3.05 m) approximately parallel to the highway. A 2" corner cube front-surface-coated hollow retroreflector was set up at this distance to reflect the transmitted laser light back to the Lidar system. The 17.5" diameter Newtonian telescope was pointed at the target, and the on and off-resonance beams were directed out of the back of the Lidar truck toward the retroreflector. The use of the 10.0 cm calibration cell to obtain absorption cross sections was limited by two problems in the field. Therefore, published absorption cross sections which appear in the literature [48] were instead used to derive the concentration of the target molecule. The first problem involved NO2. The measured absorption cross section of 1.08 cm-1 atm-1 appeared too small. This could be due to the dimerization of NO2 to N2O4. This is temperature dependent. With regard to Ozone, problems occurred with the beamsplitter. Due to fluorescence of the anti-reflection coating on the beamsplitter when excited by the 308 nm excimer light, it was impossible to reliably determine the fraction of light reflected by this beamsplitter at this wavelength. The laser beam could be seen as blue light at nighttime in the NO2 spectral band, as shown in Figure 28. The laser beam in UV range for Ozone measurement was not visible. During the measurements at this site, moderate traffic was observed on the highway during day, and very low traffic was observed during the night. Traffic data are presented in Section 5.

1In this sense, downward-looking measurements from an airplane are actually easier than upward-looking ones. In order to align the system, all that is necessary is to fly the plane low enough (e.g., 1,000 feet) so that the projected laser light reflecting from the ground can be seen on the photodetector face. Once this is done, the optical axis of the outgoing beam and the receiving telescope will be aligned for all other altitudes. The aircraft may then fly to higher elevations, and measurements may be performed not using the ground as a hard-target retroreflector but by instead using the atmosphere. The advantage of this is that it produces lower noise. The standard deviation of signal returns from the atmosphere is always much lower than the standard deviation of ground returns; thus, a higher sensitivity is possible.


46 46

4.1.4 Results of Air Pollution Measurements and Discussions Near-simultaneous measurements of Ozone and NO2 were made using the differential wavelength pairs by making only minor modifications of the tunable DIAL system. Three different measurements were made on NO2 concentration. The first was made on the evening of May 19 at 10 PM. The second was made on the evening of May 24 at 11 PM. The last measurement was made during the morning of May 25 at 10 AM. On May 24, the concentration of O3 was measured within 30 minutes after the NO2 measurement. The raw Lidar data and detailed results of air pollution measurements are included in the Attachment. Table 6 shows the summary results of NO2 and O3 measurements.

Table 6. Summary of Oxford measurements of air pollution The 17.5" diameter Newtonian telescope was pointed at the target, and the on and off-resonance beams were directed out from the back of the Lidar truck toward the retroreflector. The Newtonian telescope was set up pointing toward the target retroreflector at nighttime because it was easy to see the blue laser beam hitting the target and the back reflection, as shown in Figures 26 and 28. Full daylight measurements of NO2 using sequential tuning between the on and off resonance wavelengths were made. Also in Oxford, quantitative spectroscopic measurements were performed in-house on Ozone and nitrogen dioxide using a 10 cm sample cell. Figure 27 shows the sample cell filled with a concentration of approximately 20,000 ppm nitrogen dioxide. The brownish yellow color of the gas is evident against the blue sky. To obtain a calibrated mix of Ozone, an Ozone generator was

Pollutant Test Date Time Concentration Sensitivity Comments (ppb) (ppb) _ NO2 May 19 2200 hrs 238 218 Fertilizer, Recent rain NO2 May 24 2300 hrs 37.4 174 Nighttime NO2 May 25 1000 hrs 943 3,430 Full daylight Partly cloudy O3 May 24 2330 hrs 9,670 18,700 450 nm off-resonance Nighttime


47 47

used. It produces a known concentration of Ozone by high voltage electric discharge in pure oxygen fed gas. As shown in Table 6, the daytime measured concentration of NO2 on May 25 was nearly 25 times higher than the concentration measured during the night on May 24. This significant increase in the NO2 concentration can be contributed to higher highway traffic volume that was 6.4 times more at that hour of the daytime on MS Highway 6. Note that the relationship between hourly traffic volume and NO2 concentration is highly nonlinear. An exceptionally high concentration of atmospheric nitrogen dioxide was obtained at the Oxford test site on the evening of May 19 at 10 PM which persisted for about 15 minutes. One hypothesis is that lightning activity that evening in the Oxford area produced copious amounts of NO2 in the troposphere which settled low to the ground in the athletic field. A stronger alternative hypothesis to explain this high nighttime reading is that nitrogen dioxide can be produced and released into the atmosphere from air/water/fertilizer reactions on open fields. Since reactions involving nitrogen dioxide can produce O3 under certain conditions [49], and because large areas of Mississippi have fertilizer applied to them on a regular basis, it is possible that this ground-level source represents a large area source of NO2, which in turn directly affects the concentrations of atmospheric Ozone. This hypothesis deserves to be investigated further because fertilizer was spread on the athletic field two to three days before these measurements were taken. It is already known to scientists that a significant of amount of nitrates and nitrogen from fertilizers and urban runoff, major sources of pollution, enter the Mississippi River and ultimately reach the Gulf [52]. Threat to water quality due to fertilizer runoff is also recognized by the EPA [53]. The available horizontal pathlength limited the sensitivity of the nitrogen dioxide measurements. The exact sensitivity depended on the magnitude of the signal strengths received at on- and off- resonance wavelengths (see Table 5). The instrument sensitivity in the first measurement was 218 ppb. The instrument sensitivity in the second measurement was 174 ppb. For nitrogen dioxide measurements, the photomultiplying tubes were adjusted to a linear response region, their outputs measured on the Tektronix 11401, and the transmitted wavelength was switched back and forth between on- and off-resonance using the diffraction grating controls on the FL3002 dye laser (which is accurate to 0.005 Angstroms). Signal averaging times of approximately 3 to 10 minutes were used at pulse rates of 20 to 35 Hertz allowing signal integrations between 100 and 1,000 pulses. For Ozone measurements, the 450 nm off-resonance wavelength was first transmitted and a measure of its return intensity was made. The amplifier cuvette and focusing lens was then removed from the FL3002, and a small 1" diameter front surface coated mirror was used to direct the 308 nm amplifier pump light along the optical axis of the dye laser, and to block-off the blue light from the oscillator/preamplifier. Careful adjustment was made to direct the 308 nm on-resonance light along the same light path followed by the 450 nm off-resonance blue light. Good alignment was possible using a paper target mounted on the truck back so the UV light would hit the same spot marked previously by the blue light. For the Ozone measurements, the


48 48

photomultiplying tubes were adjusted to a linear response region and their outputs were measured on the Tektronix 11401. Signal averaging of 200 pulses on both the on- and off- resonance wavelength signal intensities was performed. One night-time measurement was made on Ozone in Oxford using the wavelengths described above, but the available horizontal pathlength limited the system sensitivity. The instrument sensitivity in the first measurement was only 18,700 ppb. It was not possible to make a daytime measurement of ambient and/or elevated Ozone concentrations due to the lack of a suitable narrow bandpass filter for the 308 nm light. With the cooperation of the City of Tupelo, a visit was made to Tupelo during May 30 - June 1. The DIAL tests site was on McCollugh Boulevard, off US Highway 78 East, about three to four miles northwest of Tupelo Regional Airport. No measurement was made on May 30 because of lightning, thunderstorm, heavy rainfall, and disruption of electricity in the area west of the airport. Laser ranging was performed across a large open field, and a 2" retroreflector was placed in a tree on the other side of the railroad tracks at a height of approximately 20 feet (6.1m). The total horizontal ranging distance available at this site was 579 ft (176.6 m). This distance was measured using a Bushnell Range Pro 500 laser rangefinder, which was accurate to +/- 3 ft. In Tupelo, two different DIAL readings were made of nitrogen dioxide, and one was made of Ozone. All of the measurements were made during the evening of May 31st following a break in the afternoon rain. The measured NO2 concentration at 10:00 PM was 13.8 ppb, and almost no detection was made at 10:40 PM. The available horizontal ranging distance at Tupelo was greater than at Oxford, but still not as far as desired. As at Oxford, the available horizontal pathlength limited the sensitivity of the DIAL measurements. The exact sensitivity of each measurement depended on the magnitude of the signal strengths received at the on- and off- resonance wavelengths. The instrument sensitivity in the first nitrogen dioxide measurement was 64.1 ppb. The instrument sensitivity in the second nitrogen dioxide measurement was 64.3 ppb. The instrument sensitivity for the O3 measurement was 33.7 ppm (without correcting for differential scattering effects). The concentration of O3 measured at 12:30 AM was 11.3 ppm. The Spike tests of O3 were also made, and the ambient atmospheric concentration was computed by the standard addition method. For the spike tests, two spectroscopic cells (Hellma Cells, Inc.), one 1 cm in length and the other 2 cm in length, were passed in front of the beam. Ozone generated by the Ozone generator was allowed to pass through both cells via a system of small plastic piping. The same O3 concentration was thus present in each cell, and each presented a distinct concentration*pathlength (“CL”) “anomaly” when introduced into the beam, differing by a factor of 2. In this way, the linearity of the system was checked. The insignificant amounts of O3 and NO2 concentration measured in Tupelo is reasonable considering the effects of heavy rain on both days, and very small traffic on surrounding roads during the night. The traffic data measured on the test site are shown in the next section.


49 49

4.2 TRAFFIC AND WEATHER DATA SUMMARY AND ANALYSIS Extensive traffic data were collected on MS Highway 6 adjacent to the test site and at the West Oxford Intersection during April – May 2001. Both the manual traffic counts by the CAIT student staff and the automated counters (in cooperation with the City of Oxford staff for Oxford ITS project) were used in this study. Figure 29 shows the traffic data collection locations on MS Highway 6. 4.2.1 Analysis and Results of Traffic Data Manual traffic data were collected over several days for each of the four lanes of MS Highway 6 (two lanes in each direction) during peak and non-peak hours. Manual traffic counts included both vehicle counts and car/truck classifications. Automated counters were used to collect continuous vehicle counts for several seven-day periods during the last weeks of the spring semester and at the end of the semester. Figure 30 shows typical hourly volume distributions. Table 7 shows daily traffic volume distributions estimated for this segment of MS Highway 6. These data provide the latest information on traffic volume and daily/weekly/seasonal patterns. Figure 31 shows the total hourly traffic volume for May 19 (about 16,000 vehicles/day). Figure 32 shows the total hourly traffic volume for May 24 and 25 (about 22,000 vehicles/day). The NO2 concentrations measured by the remote sensing DIAL method are also shown on these figures, indicating a strong positive correlation between traffic volume and time of air pollution measurements. About 25 times higher concentration level and 6.4 times higher hourly traffic volume were observed in the daytime, compared to the nighttime data.

Table 7. Typical daily traffic volume distributions, MS Highway 6, Oxford test site

4.2.2 Results of Weather Data Figure 33 shows hourly ambient air temperature and wind speed data collected from a nearby NOAA SURFRAD weather station in Batesville [54]. The weather data were downloaded for 2001 including test days in May. Figure 33 also indicates that a higher concentration of NO2 in the daytime is associated with a higher air temperature, compared to nighttime data. Additional weather data are being collected from weather stations in Tupelo and Southaven.

Mon 19,658 14.755%Tue 19,778 14.846%Wed 20,082 15.074%Thu 21,106 15.842%Fri 22,718 17.052%Sat 16,263 12.207%Sun 13,620 10.223%

Total 133,225 100.000%

Total Traffic Volume

Percentage from TotalDay

Weekly Distribution By Day


50 50

Installation of automated traffic counters West Oxford Exit Intersection on MS Highway 6 (Traffic control by MS Highway Patrol) (Manual traffic counts)

Figure 29. Traffic data collection, MS Highway 6, Oxford test site

Figure 30. Typical hourly traffic volume distribution, MS Highway 6, Oxford test site

Automatic Traffic Counts April 25, 2001

Air Quality Test Site MS Highway 6, Oxford

Manual Traffic Counts April 24, 2001

Average Hourly Distribution, Outside LaneHighway 6, Oxford, MSFrom Automatic Counts

0%

2%

4%

6%

8%

10%

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

23:0

0

0:00

1:00

2:00

3:00

4:00

5:00

6:00

Time (Hrs)

Perc

enta

ge D

istri

butio

n

Hwy 6 East, Outside Lane, WeeklyAverage Distribution

Hwy 6 West, Outside Lane, WeeklyAverage DistributionHwy 6, 2 Lanes Combined, WeeklyAverage Distribution

NOTE: Weekly Average


51 51

Total Hourly Traffic Volume May 19, 2001Highway 6, Oxford, MS

Calculated from Automatic Counts

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,0007:

00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

23:0

0

0:00

1:00

2:00

3:00

4:00

5:00

6:00

Time, hours

Hou

rly

Traf

fic V

olum

e

Hwy 6 East, Outside LaneHwy 6 East, Inside LaneHwy 6 West, Outside LaneHwy 6 West, Inside LaneHwy 6 East, 2 Lanes CombinedHwy 6 West, 2 Lanes CombinedHwy 6 Total, All Lanes Combined

Cars Trucks Total 5,326 488 5,814 2,357 138 2,495 5,124 451 5,574 1,907 150 2,057 7,683 626 8,309 7,031 601 7,63114,714 1,227 15,941

May 19, 2001 May 20, 2001

Daily Traffic Data assuming same hourly distribution every day

NOTE:

May 19, 200110:00 PM, NO2 Measured 238 ppb

Figure 31. Air pollution data measured on May 19 and traffic data, MS Highway 6

Figure 32. Air pollution data measured on May 24 and 25 and traffic data, MS Highway 6

Total Hourly Traffic Volume May 24-26, 2001Highway 6, Oxford, MS

Calculated from Automatic Counts

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

7:00

9:00

11:0

0

13:0

0

15:0

0

17:0

0

19:0

0

21:0

0

23:0

0

1:00

3:00

5:00

7:00

9:00

11:0

0

13:0

0

15:0

0

17:0

0

19:0

0

21:0

0

23:0

0

1:00

3:00

5:00

Time, hours

Hou

rly

Traf

fic V

olum

e

Hwy 6 East, Outside LaneHwy 6 East, Inside LaneHwy 6 West, Outside LaneHwy 6 West, Inside LaneHwy 6 East, 2 Lanes CombinedHwy 6 West, 2 Lanes CombinedHwy 6 Total, All Lanes Combined

May 24, 2001 May 25, 2001

Daily Traffic Data assuming same hourly distribution every day

NOTE:

May 24, 200111:00 PM, NO2 Measured 37.4 ppb11:30 PM, O3 Measured

May 26, 2001

May 25, 2001, 10:00 AM NO2 Measured 943 ppb

6:00


52 52

Figure 33. Air pollution data measured on May 24 and 25 at MS Highway 6 and weather data

Goodwin Creek, Batesville, MS Air Temperature, May 24 - May 26, 2001

0

5

10

15

20

25

30

35

7:00

9:00

11:0

0

13:0

0

15:0

0

17:0

0

19:0

0

21:0

0

23:0

0

1:00

3:00

5:00

7:00

9:00

11:0

0

13:0

0

15:0

0

17:0

0

19:0

0

21:0

0

23:0

0

1:00

3:00

5:00

7:00

Time, hours

Tem

pera

ture

, °C


May 24, 2001 May 25, 2001 May 26, 2001

May 25, 200110:00 AM, NO2 Measured 943 ppb

Goodwin Creek, Batesville, MS Wind Velocity, May 24 - May 26, 2001

0

1

2

3

4

5

6

7

8

9

10

7:00

9:00

11:0

0

13:0

0

15:0

0

17:0

0

19:0

0

21:0

0

23:0

0

1:00

3:00

5:00

7:00

9:00

11:0

0

13:0

0

15:0

0

17:0

0

19:0

0

21:0

0

23:0

0

1:00

3:00

5:00

7:00

Time, hours

Win

d V

eloc

ity, m

/s


May 24, 2001 May 25, 2001 May 26, 2001

May 25, 200110:00 AM, NO2 Measured 943 ppb

ft / s32.81

0


53 53

4.3 CONCLUDING REMARKS Measurements of Ozone and NO2 were made using the DIAL technique in Oxford during May 2001. The major physical constraint was the short pathlength available for ranging at the test site. Greater measurement sensitivity for NO2 and Ozone could be obtained by using longer pathlengths. (In Tupleo, this option is currently being explored with the director of the Tupelo Regional Airport where a one-mile clear range is available in a safe area, parallel to the runway towards the outside airport fence. More than a one-mile pathlength is available at NCAT highway test track in Auburn. That site is included in the Year 2 measurement plan.) A method for the near-simultaneous measurements of Ozone and NO2 was demonstrated using the differential wavelength pairs by making only minor modifications of the tunable DIAL system. This work indicates that simultaneous measurements of both O3 and NO2 are feasible using the differential wavelength pairs presented by making only minor system modifications. Because of the rapidly varying background light levels, the daytime NO2 measurements were much more difficult than the nighttime measurements. Full daylight measurements of NO2 could be repeated with greater sensitivity using narrow bandpass filters and two different laser trains, as in the original Skyborne airborne Lidar system, so as to minimize the rapidly changing dc signal levels (scene changing) seen by the signal analyzer. This would result in much greater daytime sensitivity. Therefore, Lidar measurements could be made during the day by using the two laser system, and by splitting the return signal light impinging at the telescope’s focus into two different parts by a beamsplitter, and using two different photomultiplying tubes, each with their own narrow bandpass filters (one for 308 nm light, one for 450 nm light) to detect the light. The DIAL remote sensing measurements show that the daytime NO2 concentration 943 ppb is significantly higher than 37.4 ppb concentration measured at the nighttime. This can be contributed to a higher highway hourly traffic volume, and a higher air temperature in the daytime. The relationship between hourly traffic volume and NO2 concentration is highly nonlinear. This indicates that a good pollution prediction model should include the traffic volume and air temperature variables. The daytime NO2 concentration measurement at the Oxford test site is also much higher than the EPA annual average standard of 53 ppb. In addition to the emissions from vehicular traffic, several fixed-point sources in the Oxford area may potentially affect air pollution concentrations at this site. The emission inventory data from these point sources in the Oxford area, provided by the Air Division of the Office of Pollution Control of the Mississippi Department of Environmental Quality (DEQ), are shown in Table 8. A probable influence on the abnormally high NO2 concentration measurement on May 19 was the spreading of a large quantity of time-release fertilizer on the athletic fields early during the measurement set. A high concentration of atmospheric NO2 was measured at the Oxford test site on the evening of May 19. This hypothesis deserves to be investigated further. The chemical fertilizer manufacturing industry in Yazoo City, Mississippi has shown interest in these results, as related to the ingredients of the fertilizer used in the athletic field.


54 54

Table 8. NOX and VOC emission point sources in Lafayette County, Mississippi (courtesy of Air division, Office of Pollution Control, Mississippi DEQ)

VOC NOX

Facility (tpy) (tpy) Oxford Asphalt Co. 2.64 10.85 Georgia-Pacific Corp. 307.14 274.00 (Oxford Particleboard Factory) Jack King Asphalt Co. 4.82 31.97 University of Mississippi 0.00 48.97 (Central Heating Plant) Emerson Electric Co. 17.66 17.67 (Electric Motor) Total (tpy) 332.26 383.46

The natural background concentration of NO2 in rural areas varies widely, but is generally thought to be approximately 30 ppb. Elevated concentrations above this value at the Oxford site could result from local traffic emissions from the adjacent highway (MS Highway 6, approximately 750 feet from and parallel to the test site), from a variety of industrial emitters in the vicinity (Table 8), or from natural sources such as lightening discharge, or local burning. The natural background concentration of Ozone in suburban areas also varies widely, but is approximately 40-50 ppb. The annual means recorded at the Mississippi DEQ air monitoring station in DeSoto County, which is adjacent to metropolitan Memphis, are within this range for the years 1998, 1999, and 2000. The one-hour maximum values recorded at the Mississippi DEQ DeSoto County station for the years 1999 and 2000 were around 160 and 120 ppb. As discussed in Section 2.2, Mississippi falls in the region of high biogenic VOC sources; NO2 and VOC are both precursors of Ozone. The equation used to derive the concentration is a variant of the well-known Lambert-Beer Law of optical absorption. Errors in this technique arise primarily from differential scattering effects whereby aerosols and other randomly occurring and changing constituents of the atmosphere scatter the two wavelengths differently. The strength of the DIAL technique is its ability to perform in situ, real-time measurements of air pollutant gases without the need for specialized cannisters or sampling bags and post-sampling laboratory analysis (time delays and bag materials may change the concentrations through naturally-occurring chemical reactions). The real-time measurement allows the concentrations of trace gases to be measured as they naturally exist in the atmosphere, over pathlengths more representative of actual volume-averaged concentrations than point monitors.


55 55

5.0 PLANNING FOR YEAR 2 DIAL MEASUREMENTS 5.1 TUPELO AIRPORT The DIAL measurements of NO2 and O3 are planned at the Tupelo Regional Airport in Year 2 in cooperation with the Airport Director and Mississippi DEQ. It is proposed to make these measurements parallel to the runway across a one-mile pathlength, outside the normal hours of aircraft operations. The Federal Aviation Administration (FAA) Safety office in Atlanta is reviewing our request and proposed laser specifications shown in Table 9 to perform these measurements. Figure 34 shows the layout plan of the airfield at Tupelo Regional Airport [55]. Tupelo represents a major metropolitan area in Northern Mississippi with railroad traffic, and is surrounded by heavily trafficked US Highway 78, MS Highway 6, and US Highway 45.

Table 9. Proposed tunable pulse Lidar specifications for consideration by FAA [47]

Tunable Pulse Laser Specifications for NO2 Measurement

Laser Type Lambda Physik FL3002 dye laser. Class IV. (Laser complies with DHHS performance radiation standards 21 CFR Chapter 1, Subchapter J).

Laser Pulse Energy 750 – 1,500 microjoules per pulse Laser Pulsewidth (FWHM) 20 nanoseconds Laser Pulse Rate 35 Hertz Laser Average Power Emitted 52 milliwatts Divergence 2 milliradians Wavelengths 4,500.0 and 4,478.5 Angstroms Beam Path Approximately 1 mile in distance Beam Height Approximately 10 feet

Notes: 1. The beam will travel straight from the back of the truck to a remote retroreflector and

then back to the truck again for detection. 2. At all times, the beam will be kept within a restricted area where personnel and planes are

not allowed. The risk of accidental exposure is slight. 3. Distraction. It is doubtful that the beam would even be visible to pilots and personnel on

the perimeter of the restricted area except at night. However, since we would like to take some measurements at nights there is the potential for distraction. At night, the potential of distraction is slight.

4. We will test the distraction-causing ability of the beam when we first set up by having one of our pilot friends who is active in the Civil Air Patrol land and take-off when the beam is on.


56 56

Figure 34. Layout plan of the airfield at Tupelo Regional Airport [55]


57 57

5.2 NCAT TEST TRACK, AUBURN UNIVERSITY, ALABAMA The DIAL measurement in Year 2 has been planned at the National Center for Asphalt Technology (NCAT) Highway Test Track, Auburn University, where 5 years of heavy truck traffic has been accomplished in one year, starting in Fall 2000 [56]. The location of the test track near Auburn, Alabama is shown in Figure 35. A site visit to the test track has been made, and the director of the test track has agreed to provide access to the truck fuel and emission data being collected at the track for the development of air pollution models in this study. The NCAT Asphalt Highway test Track is located about 15 miles from Auburn University and Opelika, Alabama. This heavy duty asphalt pavement test track is 1.7 miles long. The construction was completed in September 2000 and since then it has been subjected to accelerated heavy test truck traffic at about 5 million standard Equivalent Single Axle Load (ESAL) applications per year.

Figure 35. Layout plan of the pavement test track at Auburn, Alabama


58 58

6.0 AIR POLLUTION MODELING 6.1 LITERATURE REVIEW Air pollution is the result of burning fossil fuels at point industrial sources (estimated by standard EPA models as shown in Table 8 for the Oxford area), area sources (for example, emission from lawn mowers in a residential area), and mobile sources (transportation related). In this study we are primarily concerned with remote sensing measurements and modeling of air pollution resulting from transportation related mobile sources. Additionally, the biogenic source of VOC in southeastern states [15, 57], as discussed in Section 2 of this report, and reactive nitrogen compounds [58] can be significant factors in the production of tropospheric Ozone. An earlier Intelligent Transportation Systems (ITS) study on congestion management and air quality evaluated the use of the Lidar technology and found it ineffective and problematic for measuring air pollution [59]. The ITS study used near Infrared Lidar which is very effective for airborne remote sensing topographic surveys and terrain mapping [60], however, not effective for measuring air pollution because the absorption spectra of all major transportation related air pollutants fall outside the near Infrared band. This NCRST-E air quality study has demonstrated that the tunable differential absorption Lidar or DIAL technology is the best remote sensing method to measure in situ real-time air pollution. The air pollution data measured on selected test sites in this study will be used to develop and validate air pollution models, particularly for NO2 and O3 pollutants. The existing emission models used by the FHWA’s HPMS program, MOBILE4, and other recent models developed by the EPA are based upon vehicle-miles-traveled, vehicle characteristics, speed, road surface, and alignment [61, 62, 63]. The emission modeling is the first step in local and regional air pollution modeling considering the effects of meteorological conditions, photochemical reaction, dispersion of air pollution (for example, Gaussian models), air temperature, and urban sprawl [64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74]. Table 1 lists several air pollution models identified through an extensive literature and Internet search. 6.2 MODELING FRAMEWORK AND METHODOLOGIES Key steps involved in air pollution modeling and validation using remote sensing DIAL data are: �� Collect traffic data, air temperature and wind effects data, and DEQ air pollution data. �� Acquire independent air pollution data from DIAL measurements on selected sites. �� Develop air pollution concentration and dispersion models using historical pollutant data. �� Validate the pollution models using remote sensing DIAL data. The air pollution models can be improved by using the remote sensing data on climatic parameters, rural/urban/socio-economic factors, and more precise and traffic data collected many times in a year. The Test Track site in Auburn provides a unique opportunity to evaluate and calibrate vehicle emission and air pollution models. About 10-years heavy truck traffic volume will be applied within two years on the Auburn Test Track at an accelerated rate, starting July 2000.

59 59

Primary modeling products of this air quality research project include: modeling of wind speed and direction; pollution models using artificial neural network (ANN) and nonlinear regression techniques; time series analysis; implementation of these models and geographical information system (GIS) maps for assessing environmental impacts; and evaluation and calibration of these models using bench mark data obtained from DIAL measurements and published EPA air quality data. Figure 36 shows historical Ozone data measured by the DEQ monitoring station located at Tupelo Airport. A trend of increased Ozone level during July-August is observed from 1991 to 2000. This seasonal process can be best modeled by autoregressive integrated moving average (ARIMA) time series analysis [75]. If analyzed on the basis of long-term life cycle costs, traffic volume data and emission levels impact social and societal costs. This approach will help highway planners to make better evaluation of alternative strategies for corridor planning, design, and construction. The models developed in this air quality project will assist in rational decision-making for achieving enhanced environmental compatibility of transportation projects.

Table 10. List of air pollution dispersion models

�� ADAM (Air force dispersion assessment model) �� ADMS-3 (Atmospheric dispersion modeling system) �� ASPEN (Assessment system for population exposure nationwide) �� BLP (Bouyant line and point source model) �� CALINE4 (Steady-state Gaussian dispers�� CDM2 (Climatological dispersion model)�� CMAQ (Community modeling air quality�� CTDMPLUS (Complex terrain dispersio stable situations)�� DEGADIS (Dense gas dispersion model)�� EKMA (Empirical, city-specific model) �� EMS-HAP (Missions modeling system fo�� ISC3 (Industrial source complex model) �� MESOPUFF II (Short term, regional sca�� OBODM (Open burn/open detonation mo�� OCD (Offshore and coastal dispersion mo�� OZIPR (One-dimensional photochemical�� PLUVUEII (Model for estimating visual r c discoloration) �� RAM (Gaussian-plume multiple source a�� REMSAD (Regulatory modeling system �� RPM-IV (Reactive plume model) �� SDM (Shoreline dispersion model) �� SLAB (Model for denser-than-air releases�� SMOKE (Emission data processing for po sources) �� UAM-IV (Urban airshed model) �� UAM-V (The UAM-V photochemical mo

ion model) ) n model plus algorithms for un

r hazardous pollutants)

le puff model) del) del) box model) ange reduction and atmospheri

ir quality algorithm) for aerosols and deposition)

) int, area, mobile, and biogenic

deling system)


August 2001


60 60

Figure 36. Time series of Ozone concentration history at Tupelo Airport, Mississippi [6]

Maximum Day Value for Ozone, 8-hour ConcentrationTupelo, Lee County, MS (1991 - 2001)

0

25

50

75

100

125

150

Janu

ary

Apr

ilJu

lyO

ctob

erJa

nuar

yA

pril

July

Oct

ober

Janu

ary

Apr

ilJu

lyO

ctob

erJa

nuar

yA

pril

July

Oct

ober

Janu

ary

Apr

ilJu

lyO

ctob

erJa

nuar

yA

pril

July

Oct

ober

Janu

ary

Apr

ilJu

lyO

ctob

erJa

nuar

yA

pril

July

Oct

ober

Janu

ary

Apr

ilJu

lyO

ctob

erJa

nuar

yA

pril

July

Oct

ober

Janu

ary

Apr

ilJu

lyO

ctob

er

Year

Ozo

ne c

once

ntra

tion,

ppb

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

NOTE: Concentration values calculated considering AQI based on 8-hour ozone values

Unhealthy

EPA Air Quality Standard = 85 ppb85

Data source: EPA Air Graphics. Office of Air Quality Planning and Standards


61 61

7.0 REFERENCES (1) NCRST: Remote Sensing and Spatial Information Technologies in Transportation. Synthesis

Report 2001. A Collaborative Research Program of U.S. Department of Transportation and National Aeronautics and Space Administration, 2001.

(2) NCRST-E: Building the Future of Environmental Assessment in Transportation. Program Brochure, Mississippi State University, 2001.

(3) Eccleston, Charles H. The NEPA Planning Process: A Comprehensive Guide with Emphasis on Efficiency. John Wiley & Sons, Inc. New York, 1999.

(4) Transportation Air Quality: Selected Facts and Figures. U.S. Department of Transportation, Federal Highway Administration, Publication No. FHWA-PD-99-015, January 1999.

(5) Aviation’s Effect on the Global Atmosphere. GAO/RCED-00-57. General Accounting Office, 2000.

(6) http://www.epa.gov/agweb/ EPA Air Graphics. Office of Air Quality Planning and Standards.

(7) Environmental Protection Agency, 40 CFR Part 58, [FRL-6409-7], RIN 2060-AH92, Air Quality Index Reporting. Federal Register, Vol. 64, No. 149m Wednesday 4, 1999.

(8) http://www.epa.gov/airnow/publications.html Air-Quality: Air Quality Index (AQI), EPA-454/R-99-010.

(9) Texas Transportation Researcher. Vol. 36, No. 4, 2000. (10) Clean Air Quarterly. Houston-Galveston Area Council’s Clean Air Action. Winter 1999 (11) The Air Pollution-Transportation Linkage. A Report from the State of California Air

Resources Board, Office of Strategic Planning, 1989. (12) Lindsay, Ronald W. and William L. Chameides. High-Ozone Events in Atlanta, Georgia, in

1983 and 1984. Environmental Science Technology, Vol. 22, No. 4, April 1998, pp. 426-431.

(13) Imhoff, R.E., R. Valente, J.F. Meagher, and M. Luria. The Production of O3 in an Urbane Plume: Airborne Sampling of the Atlanta Urbane Plume. Atmospheric Environment, Vol. 29, No. 17, September 5, 1994, pp. 2349-2358.

(14) http://www.airhead.org/AirPollution/gases.php The Skinny on Pollutants: Breaking it down.

(15) Ryerson, T.B., M. Trainer, J.S. Holloway, D.D. Parrish, L.G. Huey, D.T. Sueper, G.J. Frost, S.G. Donnelly, S. Schauffler, E.L. Atlas, W.C. Kuster, P.D. Goldan, G. Hubler, J.F. Meagher, and F.C. Fehsenfeld. Observations of Ozone Formation in Power Plant Plumes and Implications for Ozone Control Strategies. Science, Vol. 292, April 27, 2001, pp. 719-723.

(16) W.L. Chamiedes, R.W. Linbsay, J. Richardson, C.S. Kiang, Science 241, 1473 (1988). (17) M. Trainer et al., Nature 329, 705 (1987). (18) Gillani, Noor V., James F. Meagher, Ralph J. Valente, Robert E. Imhoff, and Roger L.

Tanner. Relative Production of Ozone and Nitrates in Urban and Rural Power Plant Plumes. Journal of Geophysical Research, Vol. 103, No. D17, September 20, 1998, pp. 22,593-22,615.

(19) Rethinking the Ozone Problem in Urban and Regional Air Pollution. Report. National Academy of Sciences, Washington, D.C., 1991.

http://www.epa.gov/agweb/

http://www.epa.gov/airnow/publications.html

http://www.airhead.org/AirPollution/gases.php


62 62

(20) http://eande.lbl.gov/HeatIsland/ Heat Island Group. (21) http://www.elmhurst.edu/~chm/onlcourse/chm110/outlines/nitrogencycle.html Nitrogen

Cycle. (22) Nitrogen in the Nation’s Rain. National Atmospheric Deposition Program. NADP Brochure

2000-01a (Revised), 2000. (23) Healey, Jayne and Stork, Christine M. Breathless Farmers and Hockey Players. The

CNYPCC Toxicology Letter, Central New York Regional Poison Control Center, Vol. VI, No. 1, pp. 1-3.

(24) http://www.electrifyingtimes.com/mag.html March 13, 2001. New Scientist Magazine Excerpts. Compiled by Remy Chevalier.

(25) http://www.mmc.co.jp/english/business/cement05.html March 13, 2001. NOx -Absorbing Blocks. Mitsubishi Materials.

(26) www.epa.gov/air/data/info.html EPA Air Data Graphics. (27) Eckert, J.A. and R.B. Evans. Remote Sensing of Air Pollutants. Proceedings of the

Eleventh International Symposium on Remote Sensing of Environment, Vol. 1, April 25-29, 1977, pp. 353-357.

(28) Pokharel, Sajal S. Gary A. Bisop, and Donald H. Stedman. Fuel-Based On-Road Motor Vehicle Emissions Inventory for the Denver Metropolitan Area. Proceedings, One Atmosphere, One Inventory, Many Challenges, 10th International Emission Inventory Conference, Denver, Colorado, May 1-3, 2001, Sponsored by Emission Factor and Inventory Group, Emission Monitoring and Analysis Division, Office of Air Quality Planning and Standards.

(29) Wolf, Paul R. and Bon A. Dewitt. Elements of Photogrammetry With Applications in GIS, 3rd Edition, McGraw-Hill, Inc. New York, 2000.

(30) Barringer, A.R. and J.H. Davis. Remote Sensing for Air Pollution Control. Proceedings of the Twelfth International Symposium on Remote Sensing of Environment. Ann Arbor, Michigan, Volume 1, April 1978, pp. 257-259.

(31) Smith, Craig A., John S. Gergely, James A. Trainham, Samir A. Ahmed, and Fred Moshari. Solid State DIAL for the Remote Sensing of Ambient Air Pollutants. Proceedings SPIE, Vol. 2112, pp. 142-146.

(32) Levine, Jeffry S., Michael Guerra, and Ali Javan. Development of Tunable High Pressure CO2 Laser for Lidar Measurements of Pollutants and Wind Velocities – October 1977 Through January 1979. NASA Contractor Report 3274, April 1980.

(33) Brenner, Pietro, Thomas Trickl, Klaus Schäfer, Ulrich Weisensee, and Dieter Kalab (1996) – Development of a Mobile Wide-range Lidar System for Fast Observation of Ozone Smog in the Lower Troposphere. Optical Remote Sensing for Environmental and Process Monitoring. Proceedings. Air & Waste Management Association, San Francisco, CA, September 25-27, 1995, pp.114-118.

(34) Kolsch, H., J. Kolenda, P. Rairoux, B. Stein, D. Weidauer, J.P. Wolf, L. Woste. Remote Sensing of Air Pollution over Large European Cities by Lidar. Lidar for Remote Sensing, Vol 1714, December 1992, pp. 22-31.

(35) http://www.elight.de. Elight Laser Systems GmbH. Germany, August 2001. (36) Killenger, Dennis K., and Norman Menyuk. Remote Probing of the Atmosphere Using a

CO2 DIAL System. IEEE Journal of Quantum Electronics, Vol. QE-17, No. 9, September 1981 pp. 1917-1929.

http://eande.lbl.gov/HeatIsland/

http://www.elmhurst.edu/~chm/onlcourse/chm110/outlines/nitrogencycle.html

http://www.electrifyingtimes.com/mag.html

http://www.mmc.co.jp/english/business/cement05.html

http://www.epa.gov/air/data/info.html

http://www.elight.de/


63 63

(37) Carr, Lewis, Russell Warren, Clinton Carlisle, Sylvie Carlisle, David Cooper, Leland Fletcher, Steve Gotoff, and Felix Reyes. Multiple Volatile Organic Compound Vapor Chamber Testing with a Frequency-Agile CO2 DIAL System: Field Test Results. Optical Instrumentation for Gas Emissions Monitoring and Atmospheric Measurements, Proceedings SPIE, Vol. 2366, McLean, Virginia, November 1994, pp. 291-302.

(38) Zhao, Y., R.D. Marchbanks, and R.M. Hardesty. ETL's Transportable Lower Troposphere Ozone Lidar and Its Applications in Air Quality Studies. 42nd SPIE Annual meeting, Application of Lidar to Current Atmospheric Topics II, Proceedings, SPIE, Vol 3127, July 1997, pp. 53-62.

(39) http://www2.etl.noaa.gov/ndyag_lidar.html Ozone Profiling Atmospheric Lidar (OPAL) Photograph. Environmental Technology Laboratory. September 2001.

(40) Ozdemir, Phillip. An Airborne DIAL Lidar System for Pollution Detection and Mapping. Optical Instrumentation for Gas Emissions Monitoring and Atmospheric Measurements, Proceedings SPIE, Vol. 2366, McLean, Virginia, November 1994.

(41) Lange, R., W. Diehl, P. Dinges, M. Fiedler, E. Golusda, and K. Luhmann. Monitoring Air Pollutants with Compact and Mobile CO2 –Laser-Based LIDAR. Optical Instrumentation for Gas Emissions Monitoring and Atmospheric Measurements, Proceedings SPIE, Vol. 2366, McLean, Virginia, November 1994, pp. 285-290.

(42) Yamanaka, Chiyoe, Sadao Nakai, M. Matoba, Hisanori Fujita, Yoshiyuki Kawamura, Hiroyuki Daido, Mitsuo Inoue, Fumio Fukumaru, and Kiyohisa Terai. The LEKKO VII CO2 Gas Laser System. IEEE Journal of Quantum Electronics, Vol. QE-17, No. 9, September 1981, pp. 1678-1970.

(43) Moskalenko, Irina V., Shecheglov, Djolinard A., Rogachev, Aleksei P., Avdonin, Aleksandr A., Molodtsov, Nikolai A. Mobile LIDAR System for Monitoring of Gaseous Pollutants in Atmosphere over Industrial and Urban Area. Conference on Air Monitoring and Detection of Chemical and Biological Agents. SPIE Vol. 3533. Boston, MA, January 1999, pp. 241-249.

(44) Agishev, Ravil R., Bajazitov, Ravil A., Galeyev, Marat M. Spatial-Resolvable Remote Sensing and Detection of Hydrocarbons Based on CW-Low-Power-Lasers. Proceedings SPIE, Vol. 2782, 1996, pp. 820-827.

(45) Lori, Todd A., Michael G. Yost, and Ram A. Hashmonay. Trends and Future Applications of Optical Remote Sensing and Computed Tomography to Map Air Contaminants. Environmental Monitoring and Remediation Technologies. Proceedings SPIE, Vol 3534, Boston, MA, November 1999, pp. 399-404.

(46) Kagann, Robert H., William T. Walter, C. David Wang, Dale K. Kotter, and Gary J. McManus. Field Test on Signal Processing Methods for Passive FTIR Remote Sensing Measurements. Optical Instrumentation for Gas Emissions Monitoring and Atmospheric Measurements, Proceedings SPIE, Vol. 2366, McLean, Virginia, November 1994, pp. 263-285.

(47) Uddin, Waheed. Personal Communications with Phillip Ozdemir of Skyborne, Inc. May 2001.

(48) Brimblecomb, Peter. Air Composition and Chemistry. Cambridge University Press, New York (1986).

http://www2.etl.noaa.gov/ndyag_lidar.html


64 64

(49) Uchino, O. et. al., Measurement of Stratospheric Vertical Ozone Distribution with a Xe-Cl LIDAR; Estimated Influence of Aerosols, Applied Optics, Vol. 19, No. 24, 15 December 1980.

(50) Inn, Edward C.Y., Tanaka, Yoshiro. Absorption Coefficient of Ozone in the Ultraviolet and Visible Regions. Journal of the Optical Society of America, Vol. 43, No. 10, October 1953

(51) Ahmed, Samin, Remote Monitoring of Pollutants Using Differential Absorption LIDAR, EPA-600/2-80-049, 1980.

(52) Hypoxia, Fertilizer, and the Gulf of Mexico. Science, Vol. 292, 25 May 2001, pp. 1485-1486

(53) Pianin, Eric. Bush plans shifts EPA enforcement role to states. The Berkshire Eagle. Pittsfield, Massachusetts, Monday, July 23, 2001.

(54) http://www.srrb.noaa.gov/surfrad/goodwing.htm SURFRAD Network: Goodwin Creek, MS.

(55) http://www.airnav.com/airport/KTUP TUP - Tupelo Regional Airport. (56) Brown, E. R. and R. B. Powell. A General Overview of Research Efforts at the NCAT

pavement test track. CD Proceedings of the Second International Symposium on Maintenance and Rehabilitation of Pavements and Technological Control. Auburn, Alabama, July 29 – August 1, 2001.

(57) Chameides, W.L., R.W. Lindsay, J. Richardson, C.S. Kiang. The Role of Biogenic Hydrocarbons in Urban Photochemical Smog: Atlanta as a Case Study. Science, Vol. 241, No. 4872, September 16, 1988, pp. 1473-1475.

(58) Schell, B., H. Feldmann, M. Memmesheimer, A. Ebel. Effects of Initial and Boundary Values of Reactive Nitrogen Compounds and Hydrocarbons on the Ozone Concentration in the Free Troposphere. Air Pollution Modeling and its Application XII, NATO-Challenges of Modern Society, Vol. 22, 1998, pp. 97-105.

(59) Jasper, Keith. Intelligent Transportation Systems Field Operational Test Cross-Cutting Study: Emissions Management Using ITS Technology. Report FHWA-JPO-99-039. Federal Highway Administration, ITS Joint Program Office. September 1998.

(60) Uddin, Waheed, Li Yiqin, and Lucy D. Phillips. Integration of Remote Sensing and Geospatial Technologies for Managing Transportation Infrastructure Assets. CD Proceedings of the Second International Symposium on Maintenance and Rehabilitation of Pavements and Technological Control. Auburn, Alabama, July 29 – August 1, 2001.

(61) User’s Guide to MOBILE4.1 (Mobile Source Emission Factor Model). Report Number EPA-AA-TEB-91/01; EPA/SW/MT-91/098A; Environmental Protection Agency, Ann Arbor, MI. Test and Evaluation Branch. July 1991.

(62) Science Algorithms of the EPA Models-3 Community Multiscale Air Quality (CMAQ) Modeling System. Office of Research and Development. EPA/600/R-99/030, March 1999.

(63) Randall, Patrick C. and Arthur Diamond. Air Quality Analysis Tools (AQAT-3). California Air Resources Board, 1990.

(64) Pryor, S.C. and R.J. Barthelmie. Evaluation of an Aerosol Model: Responses to Meteorology and Emission Scenarios. Air Pollution Modeling and its Application XII, NATO-Challenges of Modern Society, Vol. 22, 1998, pp. 37-42.

(65) Hanna, Steven R., Joseph C. Chang, Mark E. Fernau, and D. Alan Hansen. Estimates of Sensitivities of Photochemical Grid Models to Uncertainties in Input Parameters, as

http://www.srrb.noaa.gov/surfrad/goodwing.htm

http://www.airnav.com/airport/KTUP


65 65

Applied to UAM-IV on the New York Domain. Air Pollution Modeling and its Application XII, NATO-Challenges of Modern Society, Vol. 22, 1998, pp. 89-96.

(66) Taha, Haider. Modeling Impacts of Increased Urban Vegetation on Ozone Air Quality in the South Coast Air Basin. Atmospheric Environment, Vol. 30, 1996, pp. 3423-3430.

(67) Taha, Haider. Modeling the Impacts of Large-Scale Albedo Changes on Ozone Air Quality in the South Coast Air Basin. Atmospheric Environment, Vol. 31, 1997, pp. 1667-1676.

(68) Taha, Haider, Steven Konopacki, and Hashem Akbari. Impacts of Lowered Urban Air Temperatures on Precursor Emission and Ozone Air Quality. Journal of the Air & Waste Management Association, Vol. 48, 1997, pp. 860-865.

(69) Taha, Haider, Sharon Douglas, and Jay Haney. Mesoscale Meteorological and Air Quality Impacts of Increased Urban Albedo and Vegetation: Energy and Buildings 25 (1997), pp. 169-177.

(70) McNider, Richard T., William B. Norris, Daniel M. Casey, Jonathan E. Pleim, Shawn J. Roselle, and William M. Lapenta. Assimilation of Satellite Data in Regional Air Quality Models. Air Pollution Modeling and its Application XII, NATO-Challenges of Modern Society, Vol. 22, 1998, pp. 25-35.

(71) Goward, Samuel N. Thermal Behavior of Urban Landscapes and the Urban Heat Island. Physical Geography, Vol. 2, 1981, pp. 19-23.

(72) Pottier, Joanne L. Application of the Urban Airshed Model in the Fraser valley of British Columia, Canada and Implications to Local Ozone Control Strategies. Air Pollution Modeling and Its Application XII, 1998, pp. 79-87.

(73) Langner, Joakim, Christer Persson, and Lennart Robertson. High Resolution, Long-Period Modeling of Photochemical Oxidants Over Europe. Air Pollution Modeling and its Application XII, NATO-Challenges of Modern Society, Vol. 22, 1998, pp. 71-77.

(74) Kondratyev, K. Ya. Radiation in the Atmosphere. Academic Press, New York, 1969. (75) Uddin, Waheed, Frank B. McCullough, and Melba Crawford. Methodology for Forecasting

Air Travel and Airport Expansion Needs. In TRR 1025. Transportation Research Board, National Academy of Sciences. Washington, D.C., 1985. pp. 7-14.


66 66

ATTACHMENTS


67 67

Air Quality Project REMOTE SENSING LASER … quality project remote sensing laser measurements of...

Documents

Transcript of Air Quality Project REMOTE SENSING LASER … quality project remote sensing laser measurements of...