C 55.2:AI 7/6 
 
 Air Quality Forecasting 
 
 A Review of Federal Programs and Research Needs 
 
 Penn 
 
 „lvanta State University 
 
 Air Quality Research Subcommittee 
 
 of the 
 Committee on Environment and Natural Resources 
 
 CENR 
 
 SEP! O2001 
 
 Documev 
 
 U.S. Depositoi < •-■ PJ 
 
 June 2001 
 
Air Quality Forecasting 
 
 A Review of Federal Programs and Research Needs 
 
 June 2001 
 
 The September and October 2000 meetings of the Air Quality Research Subcommittee of 
 the Committee on the Environment and Natural Resources (CENR) focused on a 
 discussion of federal programs related to air quality forecasting. This report provides a 
 brief summary of the current state of science as discussed at these meetings, with some 
 additional material that was not presented at the meetings due to time constraints. A 
 complete and comprehensive review of the science related to air quality forecasting is 
 clearly beyond the scope of this report. Rather, the report provides a brief overview of 
 the science, identifying key knowledge and capability gaps, and is intended as an 
 information piece to guide the development of future federal research programs relative 
 to air quality forecasting. This report is a scientific and programmatic document, as the 
 preceding Subcommittee reports have been, and it is not intended to represent 
 governmental policy. 
 
 Copies of this report are available from: 
 
 NOAA Aeronomy Laboratory 
 
 Office of the Director, R/AL 
 
 325 Broadway, Boulder Colorado 80305-3328 
 
 E-mail: aldiroff@al.noaa.gov 
 
 Phone:303-497-3134 
 
 Fax: 303-497-5340 
 
Air Quality Forecasting 
 
 A Review of Federal Programs and Research Needs 
 
 Air Quality Research Subcommittee 
 
 of the 
 Committee on Environment and Natural Resources 
 
 CENR 
 
 June 2001 
 
Table of Contents 
 
 1.0 Introduction 1 
 
 1 . 1 The Need for Air Quality Forecasts 1 
 
 1 .2 Air Quality Forecasting Techniques 2 
 
 1.3 Elements of an Air Quality Forecasting System 5 
 
 1 .4 Air Quality Forecasting as a Way to Improve Understanding 6 
 
 2.0 Existing Federal Programs to Forecast Air Quality 7 
 
 2. 1 Environmental Protection Agency (EPA) 7 
 
 2.2 National Oceanic and Atmospheric Administration (NOAA) 9 
 
 2.3 National Park Service (NPSJ 1 1 
 
 3.0 Research Needs to Support Air Quality Forecasting 13 
 
 3 . 1 Defining Customer Needs 1 3 
 
 3.2 Understanding Processes 14 
 
 3.3 Evaluating / Improving Models 15 
 
 3.4 Observing Systems 16 
 
 4.0 Conclusions and Next Steps 19 
 
 5.0 References 21 
 
Introduction 
 
 1.0 Introduction 
 
 Air quality in the United States has improved dramatically in the three decades since the 
 Clean Air Act was first promulgated. It is a measure of the Nation's commitment to clean 
 air that these gains came during a period of considerable growth (the population 
 increased by more than a third and the Nation's Gross Domestic Product increased by 
 almost a factor often) when air pollution would be expected to increase. However, there 
 is more work to be done. Air pollution is still a widespread problem in the United States, 
 with over 100 million individuals in 1 14 different areas exposed to levels of air pollution 
 that exceed one or more health-based ambient standards [U.S. EPA, 2000]. 
 
 A recent study sponsored by the Health Effects Institute [Kaiser, 2000] estimates that 
 exposure to high levels of particulate matter alone is responsible for more than 60,000 
 deaths each year in the United States. Haze from car and truck emissions, industrial 
 pollution, and wildfires obscures some of the most dramatic vistas in the country and can 
 pose a substantial hazard 1 . 
 
 Air pollution places a very real economic burden on the country. The American Lung 
 Association estimated [Cannon, 1 990] that air pollution related illness costs 
 approximately $100 billion dollars (1988 dollars) each year in the United States. High 
 levels of ozone can reduce the yields of economically important agricultural crops. These 
 losses have been estimated [Adams et ah, 1989] to be $1.7 billion dollars (1980 dollars) 
 each year. 
 
 1.1 The Need for Air Quality Forecasts 
 
 A system for forecasting future air quality cannot, by itself, solve the problems described 
 above. Forecasts, if they are reliable and sufficiently accurate, can however play an 
 important role as part of an air quality management system working in concert with more 
 traditional emissions-based approaches. The applications of air quality forecasts fall into 
 the following broad areas: 
 
 • Health Alerts - Many cities currently provide warnings to the public when air 
 pollution levels exceed specified levels. The more reliable the forecast is the 
 more effective it is. These warnings are directed at specific populations that 
 are particularly sensitive to air pollution (e.g., asthmatics). Interest in finding 
 innovative ways to protect these individuals has heightened in recognition of 
 
 1 In 1997, thick smoke from regional forest fires in Southeast Asia were blamed for the Garuda Airlines 
 Airbus crash near Sumatra, and the collision of two ships in the Strait of Malacca. A total of 263 were 
 killed. 
 
 June 2001 
 
Introduction 
 
 lack of a discernable health threshold for exposure to ozone or fine particles, 
 which implies that no level of emissions reduction will protect all individuals. 
 
 • 
 
 Supplementing Existing Emission Control Programs - In many parts of the 
 country, the air quality standards are exceeded only infrequently, a few days 
 out of the year. The availability of reliable air pollution forecasts affords local 
 environmental regulators the option of "on demand" or intermittent emission 
 reductions on those days, thus avoiding the high cost of continuous emission 
 control. This approach is currently being successfully employed in several 
 areas of the country and could be expanded were reliable forecasts available. 
 Many cities also offer free access to public transportation on "ozone alert" 
 days to reduce automobile emissions. The accuracy of these forecasts is 
 critical due to the high cost associated with these programs. 
 
 • Operational plann ing - Regional haze can impair and even endanger activities 
 such as private and commercial aviation. Aerial photography and visits to 
 many National Parks are significantly impacted by the presence of haze. A 
 reliable visibility forecast could improve safety and efficiency by permitting 
 the scheduling of these activities during the most favorable periods. The U.S. 
 Forest Service (U.S.F.S.) is planning a 10- fold increase in prescribed burns. 
 Since these activities are regulated under the Clean Air Act, the U.S.F.S. will 
 have to demonstrate to local regulators that they can schedule these burns so 
 that no National Ambient Air Quality Standards will be violated, requiring 
 some form of air quality forecast. 
 
 • Emergency response - Wildfires consumed more than 4 million acres of forest 
 in the United States during 2000. The vast amount of smoke generated by 
 these burning forests affects the visibility in the area that can cause accidents, 
 increase traffic congestion and even jeopardize aviation safety. The 
 availability of reliable smoke forecasts offers rerouting options for 
 automobiles and air traffic to reduce the possibility of accidents. The 
 information provided by these forecasts can also provide an initial assessment 
 of the impact of these wildfires. 
 
 1.2 Air Quality Forecasting Techniques 
 
 A wide variety of techniques, ranging from the simple to the complex, have been used to 
 produce air quality forecasts. To date, most of these efforts have focused on producing 1- 
 to 3 -day ozone forecasts. The techniques that have been used to produce these forecasts 
 are described in a recent report [U.S. EPA, 1999]. The techniques used to forecast ozone 
 concentrations are representative of those that can, or could, be used for other pollutants. 
 They fall into three broad categories: 
 
 Climatolog y - The use of climatology to predict air quality is based on the assumption 
 
 June 2001 
 
Introduction 
 
 that the past is a good predictor of the future. This approach relies on the association of 
 elevated pollution levels with specific meteorological conditions. The application can be 
 as simple as assuming persistence (i.e., if pollution levels are high today they will also be 
 high tomorrow) or can involve the development of complex weather typing schemes (i.e., 
 identifying recurring weather patterns that are accompanied by high pollution levels) to 
 forecast air quality. These approaches are usually used to predict exceedances of specific 
 thresholds not ambient concentrations. These approaches do, however, have the 
 advantage of being reasonably simple and inexpensive to implement and operate. 
 
 Statistical Methods - The association between specific meteorological parameters and air 
 quality can be quantified using a variety of statistical techniques. In fact, these are 
 probably the most common techniques in use for ozone forecasting. In their survey, 
 EPA, 1999 has identified three statistical approaches that are in use: 
 
 • Classification and Regression Tree (CART) - This technique uses specialized 
 software to identify those variables (meteorological or air quality) that are most 
 strongly correlated with ambient pollution levels. These variables are then used 
 to predict future pollution levels based on current air quality and forecasted 
 meteorology. 
 
 • 
 
 Regression analysis - The association between pollutant levels and 
 meteorological and aerometric variables can be quantified by analyzing historical 
 data sets using standard statistical analysis packages. The resultant multi-variant 
 linear regression equation can be used to forecast future pollution levels. 
 
 • Artificial Neural Networks - Another way of analyzing historical data is to 
 identify atmospheric parameters that influence air quality and quantify that 
 association through the application of adaptive learning and pattern recognition 
 techniques, such as neural networks. Neural networks are intended to mimic the 
 way the human brain recognizes recurring patterns. Networks have been 
 developed that identify weather patterns that are associated with elevated ozone 
 levels (see U. S. EPA, 1999 and references therein). Presumably, the same 
 technique could be applied to other pollutants. 
 
 These approaches, while more complex than the ones discussed in the previous group, are 
 reasonably simple to develop and use, requiring only modest computing resources and 
 specialized knowledge. 
 
 Three Dimensional (3-D) Models - Although the techniques described above have many 
 strong points, they have a common weakness. They assume a certain amount of stability 
 in terms of the processes that affect air quality. Any change in emissions or climate 
 (short and long-term) will serve to diminish the skill of these techniques. One way 
 around this problem is to employ a more deterministic approach to the prediction of air 
 quality. Deterministic 3-D air quality models seek to mathematically represent all of the 
 
 June 2001 
 
Introduction 
 
 important processes that affect ambient pollution levels. These models are actually 
 comprised of several submodels that work together to simulate the emission, transport, 
 and transformation of air pollution. Examples of submodels include: 
 
 • Emissions models - These models simulate the time-dependent, spatially- 
 distributed emissions of the pollutant in question, and/or (in the case of secondary 
 pollutants such as O3) its precursors, from both anthropogenic and natural 
 sources. 
 
 • Meteorological models - These models forecast meteorological conditions that 
 determine transport and mixing and influence chemistry (solar intensity, 
 temperature, humidity, etc.), emissions (e.g. temperature), and deposition. 
 Trajectory models use the 3-D meteorology from these models in consort with 
 emissions data to forecast ambient levels of reasonably unreactive pollutants like 
 dust and smoke. 
 
 • Chemical models - These models use fundamental chemical kinetic rate 
 parameters, spectroscopic properties, and thermodynamic relationships to 
 simulate the transformation of primary (emitted) pollution into secondary 
 pollution, including the composition and morphology (size distribution and 
 optical properties) of aerosols. 
 
 Three-dimensional air quality models are classified as being either Lagrangian or 
 Eulerian depending on the method used to simulate the time-dependent distribution of 
 pollution concentrations. Lagrangian models follow individual air parcels over time 
 using the meteorological field to advect and disperse the pollutants. This approach 
 results in a computationally efficient system. However, it is difficult to properly 
 characterize the interaction of a large number of individual sources when nonlinear 
 chemistry is involved. Eulerian models use fixed grids (vertically and horizontally) and 
 solve the appropriate chemical equations simultaneously in all cells, including exchange 
 of pollutants between cells. Typically the computational requirements are reduced 
 through the use of nested grids, with a coarse grid used over rural areas (where 
 concentrations tend to be reasonably homogenous) and a finer grid used over urban areas 
 (where concentration gradients tend to be more pronounced). These models can also 
 accommodate a plume-in-grid treatment by performing a semi-Lagrangian calculation for 
 large point sources (e.g., power plants) during the early stages of plume dilution. These 
 models can produce three-dimensional concentration fields for several pollutants but 
 require significant computational power and expertise. 
 
 Virtually all of the techniques described above start with a meteorological forecast. 
 Therefore, the reliability of the air quality forecast is dependent on the reliability of the 
 weather forecast. Weather forecasters use a number of tools to predict tomorrow's 
 weather. Local forecasters will typically use the output from several different models in 
 combination with local knowledge and experience to produce an accurate forecast. The 
 same must be true for an air pollution forecast. A skilled forecaster will combine several 
 of the techniques described above to ensure that the prediction is as accurate as possible. 
 
 June 2001 
 
Introduction 
 
 1.3 Elements of an Air Quality Forecasting System 
 
 As with a weather forecasting system, an air quality forecasting system must contain a 
 compatible combination of two components. These components are a suite of predictive 
 models/techniques tailored to specific needs of the customer community and an 
 observation network capable of providing real-time measurements of atmospheric 
 composition needed to initialize the models and evaluate the quality of the forecast. 
 
 Fig 1.1 Schematic showing the interrelation among the main elements of an air 
 quality forecasting system. A well designed forecast system includes both a process 
 for producing the forecast and an observing system to evaluate the quality of the 
 forecast and identify areas where improvement is needed. 
 
 A comprehensive observing system must be an integral part of any forecast system that is 
 developed to predict future air quality. It is only by comparing actual and forecast 
 pollution levels that we can assign confidence to future forecasts and identify areas where 
 improvements are needed. 
 
 Traditionally the evaluation of air quality models (whatever their form) has fallen into 
 two broad categories: 
 
 • Operational evaluation - An operational evaluation involves a direct comparison 
 between the forecast pollutant fields and the observed pollutant distribution. For 
 example, in the case of ozone the model predictions would be compared against the 
 concentrations measured in the regulatory network, and some skill score calculated 
 based on a point-by-point comparison at the monitoring sites. The existing regulatory 
 network for ozone is well suited to such an evaluation. Unfortunately the situation 
 
 June 2001 
 
Introduction 
 
 for other pollutants (e.g., fine particles) is not as good since those monitoring 
 networks are less dense and the instruments not as sensitive, or in some cases as 
 selective, as is desired. 
 
 • Diagnostic evaluation - An operational evaluation will tell you how close the model 
 came to the correct answer; a diagnostic evaluation will tell you if you got the right 
 answer for the right reason. As the name implies, a diagnostic evaluation requires the 
 measurement of parameters (both meteorological and chemical) that control pollutant 
 formation and distribution, not just the concentration of the pollutant that is being 
 forecasted. To perform a diagnostic evaluation, concentrations of pollutant 
 precursors and key intermediates need to be tracked to evaluate the performance of 
 the emissions model and the chemical processor (if one is used), while meteorological 
 parameters such as mixing height and winds aloft will aid in the evaluation of the 
 meteorological processor. 
 
 1.4 Air Quality Forecasting as a Way to Improve Understanding 
 
 As techniques for forecasting air quality improve and their use expands, we should not 
 overlook the opportunity to use this process to improve our understanding of the 
 processes that control the formation and distribution of air pollution. The meteorological 
 research community has benefited enormously from the experience gained through an 
 operational forecasting enterprise. The ongoing evaluation of the daily weather forecasts 
 is used to identify areas of insufficient understanding and guide research. 
 
 The same opportunity exists for advancing the understanding of atmospheric processes 
 that control ambient pollution levels. By evaluating the success of forecasts produced by 
 different techniques we have the opportunity not only to evaluate the relative merits of 
 these techniques but also to test our knowledge of key processes and identify areas where 
 more information is needed. 
 
 June 2001 
 
Existing Programs 
 
 2.0 Existing Federal Programs to Forecast 
 Air Quality 
 
 The large number of organizations throughout the United States that currently perform 
 operational forecasting of air quality is testimony to the interest in and need for this 
 information. The great majority of these programs focus on providing ozone forecasts, 
 primarily for urban areas during the summer months. The organizations performing the 
 forecasts include federal agencies, state agencies, universities, industry, and consulting 
 companies. The methods cover the range of complexity and sophistication from simple 
 statistical approaches to full-blown photochemical grid models. Several federal programs 
 to conduct, support, and disseminate these forecasts are described briefly below. 
 
 2.1 Environmental Protection Agency (EPA) 
 
 The forecasting program under EPA's AIRNOW Project was developed as part of a 
 multi-faceted program designed to provide the public with real-time and forecast 
 information on ozone air quality. The scope of the current forecasting effort is 38 states 
 and 154 cities (Figure 2.1) providing next-day ozone forecasts to EPA's AIRNOW 
 Project. Many States have been doing ozone forecasting for several years, but there had 
 been no coordinated effort to collect these forecasts and make them available to the 
 public on a National basis. That is the primary focus of the AIRNOW forecasting 
 program. 
 
 Figure 2.1 Cities (circles) and areas (shaded states) for which ozone forecasts are 
 available in 2001. 
 
 June 2001 
 
Existing Programs 
 
 In the process of developing this coordinated effort, EPA has established excellent 
 partnerships with the national media such as USA Today, The Weather Channel, and the 
 national Weather Service Providers (Weather Central, Weather Services International, 
 Kavouris, AccuWeather). It is through the partnerships with the State and Local Air 
 Agencies along with these media partnerships that EPA has been able to facilitate the 
 wide distribution of the air quality forecasts across the United States. The technical 
 aspects of the forecast program are relatively simple. 
 
 All of the forecasts are provided to EPA from State or Local air quality forecasters using 
 a variety of forecasting techniques. In some areas, these forecasts are made for 
 independent cities, while in other areas, the forecasts are a result of a regional 
 collaborative effort. Also, for the New England States, EPA receives city-specific 
 forecasts as well as a monitor-specific forecast that is interpolated and mapped. The key 
 role for EPA in this process is to acquire and compile the forecasts. 
 
 AEPA 
 
 jU/OAQPS 
 
 EPA developed a simple web- 
 based system to facilitate the 
 submittal of air quality 
 forecasts by State or Local Air 
 Quality Agencies. This 
 system is designed to 
 accommodate multiple 
 pollutants (ozone, particulate 
 matter, carbon monoxide, 
 nitrogen oxides and sulfur 
 dioxide) and multiple day 
 forecasts (up to 4 days). All 
 of the forecasts are then 
 merged into one comma- 
 delimited text file and 
 delivered to a variety of 
 locations. Initially, all of the 
 forecasts are posted on the 
 AIRNOW website 
 ( www.epa.gov/airnow ) under 
 the "Air Quality Forecasts" 
 
 and "Where I Live" pages. In addition, the forecast fdes are then electronically delivered 
 to the media partners for distribution. The forecast system is designed to allow for 
 updates by the local forecaster at any time during the day. A current version of the 
 forecast file is delivered to the AIRNOW website and media partners every hour from 
 6:00 AM until midnight. This allows for updated forecasts to be delivered in a timely 
 fashion should the initial forecast change. 
 
 While EPA does not issue the actual forecasts, they are involved in supporting the 
 forecast programs at the State and Local level. Through EPA's Environmental 
 Monitoring for Public Access and Community Tracking (EMPACT) program, the Office 
 
 Figure 2.2 Ozone forecasts are provided for 154 cities on 
 EPA's AIRNOW web site ( www.epa.gov/airnow ) 
 
 June 2001 
 
Existing Programs 
 
 of Air Quality Planning and Standards has provided resources to support the forecasting 
 efforts in several areas of the United States. EPA developed the guidance for developing 
 an ozone forecasting program and provided that guidance to the State and Local Air 
 Agencies. EPA will be developing similar guidance for forecasting particulate matter 
 during fiscal year 2001. Finally, EPA has supported the distribution of the forecasts to 
 the public and media through website development and partnerships with external 
 weather groups. It is this national "top-down" support that has increased the number of 
 forecast areas in the United States. 
 
 Now that the forecasting program is operational, it would be useful to suggest some 
 enhancements that could improve the quality and coverage of the forecasts. While the 
 final word on any air quality forecast will always reside with the State or Local Air 
 Agency, it would be very helpful to have some type of national forecast model that could 
 assist areas that do not have the capability to generate their own forecasts. Even areas 
 with experienced forecasters would benefit from a forecast model. Air quality 
 forecasting is similar to weather forecasting. The models make a prediction and the local 
 forecasters adjust that prediction based on local knowledge and information. An air 
 quality forecast model would be utilized in a similar manner. 
 
 Overall, the forecasting approach developed under AIRNOW has been very successful. 
 It has succeeded because it continues to give the authority for the forecast to the State or 
 Local agency. A Federal forecast tool will be beneficial, but we cannot overlook the 
 importance of having the local agency have the final word with regards to the actual 
 forecast. Local conditions (thunderstorms, topography, etc) can have a significant impact 
 on air quality conditions and we must always make sure that component is covered. In 
 addition, the credibility of a forecast made at the local level will always be greater than 
 one made at the national or Federal level. 
 
 2.2 National Oceanic and Atmospheric Administration (NOAA) 
 
 NOAA has a goal under its forecasting mandate to extend, over the next decade, the 
 current weather forecasting capability to include predictions of future air quality. During 
 the past several years, as the first steps in this process, NOAA has developed several 
 experimental air quality forecast products and made them available on the World Wide 
 Web ( http : //www . arl . noaa . gov/ready . htm 1 ) . The air quality forecasts are generated using 
 high-resolution meteorological forecast models coupled with a sophisticated air-mass 
 trajectory analysis (HY-SPLIT) and, in the case of ozone, complex, photochemical grid 
 models. These forecasts are being used primarily as a learning tool to evaluate current 
 capability and identify areas where improvements are needed. Some of these forecasts 
 are provided on a routine basis (e.g., ozone forecasts during the summer months), some 
 are provided in response to specific events (e.g., natural disasters), and some are provided 
 to support air quality research (e.g., intensive field campaigns). 
 
 June 2001 
 
Existing Programs 
 
 Ozone Forecasts 
 
 In a cooperative effort, scientists from 
 NOAA and the Pennsylvania State 
 University (PSU) have developed the 
 Hybrid Single-Particle Lagrangian 
 Integrated Trajectories with a 
 generalized non-linear Chemistry 
 Module (HY-SPLIT CheM) model to 
 calculate the spatial and temporal 
 distribution of different chemical 
 species in the troposphere over a 
 regional scale [Stein et al, 2000]. The 
 simulation accounts for the advection, 
 dispersion, chemical transformation, 
 and deposition of the different 
 pollutants. The model uses the 1995- 
 1996 Ozone Transport Assessment 
 Group (OTAG) emissions inventory 
 [EPA, 1999] and Carbon Bond IV gas- 
 phase chemistry coupled with a 
 Lagrangian transport scheme to forecast 
 ambient ozone concentrations. Ozone 
 forecasts are provided on a grid 
 resolution of 0.5° x 0.5° for the 
 Northeastern US (Figure 2.3). 
 
 80 100 110 
 
 Ozone Concentration (ppb) 
 
 Fig. 2.3 Comparison of ozone forecast using the 
 PSU/NOAA hybrid model (isopleths) with data from 
 the EPA AIRS network (colors). 
 
 NOAA is also performing ozone forecasts on an experimental basis using a high- 
 resolution coupled meteorology/chemistry model [Grell et ah, 2000]. This Eulerian grid 
 model uses the PSU / NCAR nonhydrostatic mesoscale meteorological model MM 5 
 coupled with the RADM-2 gas-phase chemical reaction scheme. To date, the model has 
 been used primarily to support flight planning during intensive field campaigns designed 
 to study atmospheric processes related to air quality. 
 
 The comprehensive data sets acquired in intensive field campaigns, which include 
 measurements from ground based instrumentation, remote sensing technology, and 
 highly instrumented aircraft, are being used to evaluate the meteorological and chemical 
 components of the NOAA forecast models. 
 
 Smoke from Forest Fires 
 
 NOAA is actively engaged in the prediction of smoke dispersion from forest fires. The 
 NOAA HY-SPLIT model is being used in this application. The initial development was 
 in the context of fires in Mexico, Brazil, and parts of the USA. However, these 
 
 June 2001 
 
 10 
 
Existing Programs 
 
 applications have recently been extended to the recurring fires in Indonesia, with plans to 
 include Singapore in FY 2001. NOAA also provides a quasi-operational smoke forecast 
 system that is used by the Idaho National Engineering and Environmental Laboratory in 
 its fire-fighting and control-burn programs. A comparison of a smoke transport forecast 
 for a forest fire in eastern Idaho during August of 2000 with a satellite image is provided 
 in Figure 2.4. 
 
 Figure 2.4 Comparison of smoke forecast using NOAA's HYSPLIT model (black dots) with 
 aerosol column measurements from NASA's TOMS satellite (colored grids - hotter colors 
 represent higher aerosol loadings) for 18Z on August 10, 2000 for a fire in eastern Idaho. 
 
 Volcanic Ash 
 
 NOAA also conducts research and development on volcanic ash transport and dispersion 
 modeling. Since 1992, the National Weather Service (NWS) has used the Volcanic Ash 
 Forecast Transport And Dispersion (VAFTAD) model to protect aircraft from hazards 
 associated with volcanic emissions. NWS has provided volcanic ash forecasts using 
 VAFTAD output about 150 times in each of the past three years. 
 
 2.3 National Park Service (NPS) 
 
 As a Federal Land Manager and steward of some of the Nation's most pristine areas, the 
 NPS works to ensure that the lands under their jurisdiction are protected from the impacts 
 of pollution. To quantify levels of air pollution within the many National Parks and 
 
 June 2001 
 
 11 
 
Existing Programs 
 
 Wilderness Areas the NPS has, for many years, operated an air quality monitoring 
 network. These efforts to quantify air pollution exposure and deposition have been 
 expanded in recent years to include the issuance of ozone advisories when ozone 
 pollution reached or was forecasted to reach unhealthy levels (Figure 2.5). 
 
 NATIONAL PARK SERVICE 
 8-HOUR OZONE HEALTH ADVISORY GUIDANCE 
 
 Ozone Value 
 
 Descriptor 
 
 - 64 ppb 
 
 Good 
 
 65 -84 ppb 
 
 Moderate 
 
 85 -104 ppb 
 
 Unhealthy for 
 Sensitive Groups 
 
 105 ppb 
 or greater 
 
 Unhea 
 
 Unhealthy for 
 General Public 
 
 
 Groups Notified 
 
 none 
 
 none 
 Sensitive Groups - 
 
 people with increased risk to ozone n 
 effects (active children and people > 
 who regularly engage in outdoor 
 activities and people with 
 preexisting respiratory diseases) 
 
 General Public - level 
 
 may cause irritation to lungs and 
 discomfort in breathing for 
 general population and 
 more pronounced symptoms 
 for sensitive group individuals 
 
 Actions 
 
 #• 
 
 Limit prolonged 
 outdoor exertion. 
 
 Ill^ 
 
 Sensitive group 
 individuals should 
 avoid prolonged 
 outdoor exertion. 
 Everyone else 
 should limit 
 prolonged outdoor 
 exertion. 
 
 Predictive Values: 
 
 Sensitive Groups - when three consecutive hours within the range of 85-104 ppb occurs. 
 
 General Public - when three consecutive hours of 105 ppb or greater occurs 
 
 Figure 2.5 National Park Service guidance for the issuance of ozone advisories. 
 
 The NPS currently issues ozone advisories for the following National Parks: 
 
 • Acadia 
 
 • Great Smoky Mountains 
 
 • Mammoth Cave 
 
 • Sequoia - Kings Canyon 
 
 • Shenandoah 
 
 The ozone forecasts for the National Parks are a mix of in-house efforts by the NPS staff 
 and collaboration with other organizations, primarily through the EPA AIRNOW 
 program. An advisory is issued for that day when in-park monitoring data indicate that 3 
 hourly values have exceeded 85 ppb. Often these advisories are issued too late in the 
 day, and a better system is needed so that a forecast is available the day before. 
 
 June 2001 
 
 12 
 
Research Needs 
 
 3.0 Research Needs to Support Air Quality 
 Forecasting 
 
 The interest in and need for a robust capability to predict future air quality is clear. A 
 wide variety of techniques are currently being used with mixed results. The majority of 
 current efforts focus on ozone. A number of the approaches have been carefully 
 evaluated and some have demonstrated impressive skill, while others have been less 
 rigorously evaluated and the quality of the predictions is essentially unquantified. The 
 current ozone forecasts are largely a collection of disparate efforts, primarily the work of 
 State and Local air regulatory programs, with very little consistency in approach or 
 methods used to perform evaluations of forecast skill. The work to date, while 
 encouraging, would clearly benefit from a better coordinated national effort, including a 
 program of focused research and development, that would facilitate and augment work at 
 the local level. This support is particularly important as interest grows in forecasting 
 pollutants other than ozone (e.g. particulate matter, visibility, carbon monoxide, etc.) and 
 extending the forecasts to rural and remote areas (e.g. National Parks and Wilderness 
 Areas). 
 
 3.1 Defining Customer Needs 
 
 In designing an air quality forecasting program and identifying the research that is needed 
 to support the effort it is essential that the needs of the user community be understood and 
 considered. A number of potential applications of air quality forecasts have been 
 discussed earlier in this report. In order to determine the relative importance of each and 
 maximize the usefulness of the forecasts and provide focus for any supporting research 
 we must look to the user community for answers to several basic questions related to 
 forecast needs/requirements, including: 
 
 • For what pollutants are forecasts needed? What are their relative priorities? 
 
 ozone, particulate matter, carbon monoxide, visibility, sulfur dioxide, 
 air toxics, others? 
 
 • What is the time scale for the desired forecast? 
 
 How far in advance? 
 
 For what period; hours, days, weeks? 
 
 • How accurate does the forecast need to be and what form should it take? 
 
 Probability of exceeding a standard level. 
 
 Predicted concentration range. 
 
 Ambient concentration with estimated uncertainty. 
 
 • What is the spatial scale? 
 
 city, region, other 
 
 June 2001 13 
 
Research Needs 
 
 The applications of air quality forecasts are rapidly evolving. As the reliability and 
 coverage (number of pollutants for which forecasts are produced and the temporal, and 
 spatial extent) grows, new uses will be identified that we cannot now foresee. Therefore, 
 a close and effective relationship among the user community, operational forecasters, and 
 those involved in evaluating and improving forecasting methodologies must be 
 maintained. Thus, the process of defining customer needs and requirements must be 
 ongoing with effective communication between developers and users. 
 
 3.2 Understanding Processes 
 
 The coupling of emissions to ambient pollution levels is controlled by a number of key 
 atmospheric processes (mixing, transport, chemical transformation, deposition, etc.). The 
 prediction of ambient pollution levels relies on an understanding of how these processes 
 influence the formation and distribution regardless of whether the approach to forecasting 
 involves a parametric or deterministic approach. Therefore, efforts to improve the 
 reliability of air pollution forecasts will benefit from an improved understanding of these 
 atmospheric processes. Areas where improved understanding would benefit efforts to 
 develop and/or improve air quality forecasting systems include: 
 
 Linkages between weather and air quality: Ambient levels of pollution (e.g., ozone 
 and fine particles) are clearly dependent on the emissions of precursor pollutants like 
 NO x , VOCs, and SO2. Meteorological parameters like solar intensity, temperature, wind 
 speed, and mixing height greatly influence the day-to-day and seasonal-to-interannual 
 variations in atmospheric concentrations. These associations form the basis of many of 
 the statistical models used to forecast air quality. A better understanding of the couplings 
 between meteorology and air quality can point to approaches that will improve the 
 quality of these techniques. 
 
 Nighttime transport and chemistry: Because of the important role that photochemistry 
 plays in the formation of ozone and other secondary air pollutants, the vast majority of 
 the intensive atmospheric process studies conducted to date have focused on the daylight 
 hours. Analysis of data from intensive field studies [e.g., Banta et ai, 1998] has shown 
 that nighttime mixing and transport plays a significant role in the redistribution of urban 
 pollution throughout the region. During the day, under the stagnant conditions associated 
 with most pollution episodes, only limited horizontal transport of the pollution plumes is 
 expected. However, at sunset a shallow nocturnal inversion layer may form that can 
 isolate the boundary layer pollution from the surface, reducing the loss of ozone, PM and 
 their precursors via deposition. The formation of this layer also reduces the frictional 
 resistance for the air masses above the nocturnal layer, allowing this pollution to be more 
 effectively transported over larger distances. Moreover, the emissions from smokestacks 
 of major point sources, such as power plants, may enter the atmosphere at night above the 
 nocturnal inversion directly into the nighttime residual mixed layer. 
 
 June 2001 14 
 
Research Needs 
 
 Nitrogen oxide and VOC emissions continue to react with ozone at night, leading to 
 additional reactions involving the nitrate radical. These processes can significantly affect 
 the reactivity and concentration of the pollutant mix that populates the atmosphere during 
 the next diurnal cycle. Thus, the accurate prediction of tomorrow's air quality requires a 
 better understanding the chemistry and transport in the nighttime atmosphere. 
 
 Heterogeneous processes: Aerosols participate in a variety of chemical and physical 
 processes in the troposphere. On a regional scale, these processes are associated with 
 regional air quality as related to visibility and the effects of fine particles on human 
 health. In this regard, there is a natural synergism between pollutants like ozone and 
 sulfur dioxide and fine particles. A basic scientific understanding of the chemistry and 
 physics of the atmosphere is prerequisite for an effective predictive capability and that 
 the concentrations of pollutants in the atmosphere are often co-dependent because of 
 interacting chemical reactions. 
 
 With this in mind, new research is needed to elucidate: 
 
 • how chemical processing on aerosols influences ozone formation 
 
 • how the atmospheric oxidation leading to ozone formation leads to aerosol formation 
 
 • how atmospheric chemistry influences the growth and chemical composition of 
 aerosols. 
 
 Emissions: Day-to-day changes in emission rates need to be taken into account to 
 accurately forecast air quality. The potential effects of changes in temperature include 
 increased biogenic emissions of VOC, increased VOC emissions from motor vehicles, 
 and increased nitrogen oxides emissions due to increased power demands. Variations in 
 activity levels at major industrial facilities will also impact air quality. Research is needed 
 to develop methods to adjust seasonal emission inventories to reflect the probable 
 emission levels during the forecast period. 
 
 3.3 Evaluating / Improving Models 
 
 The models used for forecasting air quality can be divided into two broad categories, 1) 
 parametric models and 2) deterministic models. Regardless of the form of the model it is 
 important to evaluate model performance so that there is some way to quantify forecast 
 reliability and identify areas for improvement. The wide variety of approaches and the 
 form of the forecast produced in current forecast models make it very difficult to 
 objectively and quantitatively compare different techniques. Also, the factors that control 
 pollution levels vary from one region to another and mean that a model/technique that 
 works well in one region may not perform well in another. 
 
 In the case of statistical models, a systematic evaluation of the various techniques and 
 their relative performance would provide a useful guide for areas that are considering 
 developing a forecast capability. 
 
 June 2001 15 
 
Research Needs 
 
 In the case of deterministic models (e.g., photochemical grid models), there is a need to 
 evaluate model performance in a number of key areas, including: 
 
 Meteorology: Many of the most severe air quality episodes occur during air mass 
 stagnations when there is very little synoptic forcing. There is a need to evaluate model 
 performance under these conditions. There is also a need to evaluate the models for 
 specific geographic areas where there are terrain-induced effects on air quality (e.g., 
 coastal areas where land/sea-breeze recirculation can play a role, and mountainous areas 
 where the mixing and transport is very complex). 
 
 Chemistry: Many of the newest photochemical grid models include modules to simulate 
 particle emission, production, transport and deposition. The development of fast 
 response instrumentation for the characterization of particle composition and morphology 
 affords the opportunity to evaluate the performance of these models for a variety of 
 emission mixes and meteorology. 
 
 Emissions: Most air quality modeling efforts conducted to date have relied on emission 
 inventories developed by state regulatory agencies. In most cases, these inventories have 
 not been rigorously evaluated. Also, there has not been a comprehensive study to 
 examine the coupling between emissions and meteorology. The quality of the predictions 
 from emissions-based air quality models is directly proportional to the quality of the 
 emissions used in the model. There is also a need to evaluate the day-to-day variability 
 in biogenic and anthropogenic emissions and their dependence on meteorological 
 variables, and to determine the sensitivity of air quality models to this variability. 
 
 The greatest uncertainty in forecasting air quality impacts from natural disasters such as 
 volcanic eruptions and wildfires is associated with the source term. A better 
 quantification of emission rates could dramatically improve the reliability and therefore 
 the usefulness of such forecasts. 
 
 There are a number of data sets that have been accumulated as part of comprehensive 
 field campaigns in which both the physical and chemical condition of the atmosphere has 
 been characterized in some detail for the study period. Several of the NARSTO data sets 
 (e.g., Southern Oxidants Study - SOS campaigns in Nashville and Texas, NARSTO-NE, 
 and Southern California Ozone Study - SCOS) provide a rich resource for model 
 evaluations such as those described above. 
 
 3.4 Observing Systems 
 
 A well-designed observing system is essential to the success of any air quality forecasting 
 system. The need for information on the chemical constituents of the atmosphere is 
 analogous to the meteorological observing system that supports the current weather 
 forecasting system. Information on atmospheric pollution is needed to both initialize and 
 evaluate the forecast. For example, today's ozone levels will have a strong influence on 
 tomorrow's levels. Specific research needs include the following: 
 
 June 2001 16 
 
Research Needs 
 
 Data to initialize models and perform operational evaluation: Most air quality 
 forecast models use some form of dynamic initialization, assimilating observed chemical 
 data in a manner akin to the way weather forecast models are initialized using 
 meteorological observations. This process requires near real-time access to air quality 
 data. The United States has a reasonably dense network of ozone monitors that are 
 operated for both regulatory and other purposes. With this network it is possible to 
 quantify both urban and rural levels in most parts of the country. However there are 
 areas of interest, most notably National Parks and Wilderness Areas, where there are no 
 monitors and therefore no way to evaluate forecast skill. The situation for other 
 pollutants is not nearly as good. If short-term (1-3 day) forecasts of these pollutants are 
 needed, fast-response (1-hr or less) near real time data will be needed for model 
 initialization and evaluation. For some pollutants (e.g., particulate matter) there are 
 currently no reliable fast-response measurements, while for others (e.g. carbon monoxide 
 and sulfur dioxide) reliable data are available only for urban or industrial areas. New 
 monitoring systems will have to be developed if forecasts for these pollutants are needed. 
 In some cases satellite data may be able to fill in some of the gaps through integration 
 data from sparse surface networks. In other instances useful surrogates may be found 
 that can substitute for the desired data. For example, visibility data from the 900+ airport 
 meteorology network (ASOS) may serve as a surrogate for fine particle mass [Richards et 
 al, 1996]. 
 
 Data for diagnostic evaluation of complex grid models: A diagnostic evaluation of 
 model performance requires information on the fundamental physical and chemical 
 processes responsible for the formation and distribution of air pollution. While much of 
 this information is available at research sites located throughout the United States, these 
 stations are few and far between and access to the data is difficult. An integrated network 
 of research monitoring stations that are capable of characterizing the atmosphere both in 
 terms of meteorologically and chemically important parameters (including chemical 
 precursors and key intermediates) is needed to support a national forecasting effort. Such 
 a network could build on the existing efforts by augmenting measurements at existing 
 sites and adding stations in key areas where such data are currently not available. 
 
 Information on the vertical distribution of pollution: The formation and collapse of 
 daytime and nocturnal inversion layers serves to isolate the surface from the residual 
 mixed layer and the free troposphere. Transport of pollution above the daytime mixed 
 layer and the stable nocturnal layer can have a strong influence on next day pollution 
 levels. Current monitoring networks are blind to this pollution, and vertical pollution 
 distributions are made only infrequently and usually only during intensive field 
 campaigns. New remote sensing instruments (i.e., DIAL lidars) and advances in sonde 
 technology offer the opportunity to characterize the vertical pollution distribution on a 
 more frequent basis and over a wider area of the country. The development of reliable air 
 quality forecasts will require access to such data. By way of comparison, imagine the 
 difficulty of performing weather forecasts while denied access to the daily soundings and 
 profiler data. 
 
 June 2001 17 
 
Research Needs 
 
 Access to data in a timely manner: Most of the air quality data in the United States are 
 collected for regulatory purposes. In most cases, these data are not available for several 
 months after they are collected. Access to this data in near real-time would facilitate and 
 improve the air quality forecasting process. Some states and the EPA's AIRNOW 
 program are providing limited access to data from part of the ozone network in real time. 
 This capability needs to be extended to other pollutants for which forecast information is 
 needed. 
 
 June 2001 18 
 
Conclusions and Next Steps 
 
 4.0 Conclusions and Next Steps 
 
 The need for air quality forecasts is clear. A growing number of State and Local Air 
 Agencies (currently 35 states and 135 cities) are using summertime ozone forecasts to 
 alert sensitive populations and reduce precursor emissions through reduced vehicular 
 traffic on "ozone action days". A similar approach is being used to reduce carbon 
 monoxide levels in Denver. Plans are being developed to extend these forecasts to 
 include other pollutants (e.g. particulate matter). The National Park Service uses ozone 
 forecasts to protect their workforce and inform park visitors. Air quality forecasts are 
 also an important part of the Federal process for dealing with natural disasters (wildfires 
 and volcanic eruptions). All indications are that uses for air quality forecasts will 
 continue to grow as reliability and coverage improves and new applications are 
 identified. 
 
 Federal activities related to air quality forecasting are currently limited to 1) EPA efforts 
 to facilitate the collection and dissemination of real time data and the posting of locally- 
 produced ozone forecasts and 2) efforts by NOAA to develop and evaluate a number of 
 experimental forecasting techniques and 3) the National Park Service's program to use 
 ozone forecasting as a proactive approach to resource management. To date, these efforts 
 are largely uncoordinated with only model resources available to improve and extend air 
 quality forecasting methodologies and supporting infrastructure (i.e., observing systems, 
 data ingest, etc.). The Subcommittee has identified a number of activities to improve air 
 quality forecasts: 
 
 Improved coordination - A number of Federal agencies are currently engaged in some 
 aspect of air quality forecasting; others have related activities that could contribute to a 
 national effort. A more formal process for coordinating Federal activities would ensure 
 that limited resources are used to maximum advantage. The Air Quality Research 
 Subcommittee could facilitate such coordination and NARSTO, the tri-national (Mexico, 
 United States, and Canada) research consortium focused on ozone and fine particles, 
 could serve as a connection to the private sector. 
 
 Identification of customer needs - This is a rapidly evolving area with a need for 
 ongoing communication between the user and research communities. EPA has close 
 contact with the State and Local Air Programs that are the core of the AIRNOW 
 program; NOAA has conducted a number of workshops to identify current and future 
 needs. These efforts need to continue and be integrated to improve communication. A 
 coordinated Federal program must be responsive to the needs of the user community. 
 
 Evaluate existing technology - Current approaches to air quality forecasting vary from 
 the simple to the complex. Some have been carefully evaluated, while others have not. 
 There is a lot to be learned from the work that has already been done. A systematic and 
 objective evaluation of the methods that are currently in place is the first step in 
 identifying areas for improvement and opportunities for extension to other pollutants. 
 
 June 2001 19 
 
Conclusions and Next Steps 
 
 Program development - A forecasting system (Section 1.3) is needed to support 
 operational air quality forecasting. The system must include a method to continuously 
 evaluate and improve the forecast consistent with the needs of the users. A program plan 
 should be put in place that identifies operational, research, and developmental needs and, 
 where possible, agency resources that can be used to address these needs. The 
 information provided in Section 3 of this report is intended as a first step at identifying 
 the key elements of the research portion of such a program. 
 
 June 2001 20 
 
References 
 
 5.0 References 
 
 Adams, R.M., J.D. Glyer, S.L. Johnson, and B.A. McCarl, A reassessment of the 
 
 economic effects of ozone on U.S. agriculture, J. Air Pollution Control Assoc, 
 39, 960-968, 1989. 
 
 Banta, R.M., C.J. Senff, A.B. White, M. Trainer, R.T. McNider, R.J. Valente, S.D. 
 
 Mayor, R.J. Alvarez II, R.M. Hardesty, D. Parrish, and F.C. Fehsenfeld, Daytime 
 buildup and nighttime transport of urban ozone in the boundary layer during a 
 stagnation episode, J. Geophys. Res., 103, 22519-22544, 1998. 
 
 Cannon, J.S., The Health Costs of Air Pollution: A Survey of Studies Published 1984- 
 1989, American Lung Association, 1990. 
 
 Grell, G.A., S. Emeis, W.R. Stockwell, T. Schoenemeyer, R. Forkel, J. Michalakes, R. 
 Knoche, and W. Seidl, Application of a multiscale, coupled MM5/chemistry 
 model to the complex terrain of the VOTALP valley campaign, Atmos. Environ., 
 34,1435-1453,2000. 
 
 Kaiser, J., Evidence Mounts That Tiny Particles Can Kill, Science, 289, 22-23, 2000. 
 
 Richards, L.W., T.S. Dye, M. Arthur, and M.S. Byers, Analysis of ASOS data for 
 
 visibility purposes, Sonoma Technology, Inc., Report No. STI-996231-1610-FR, 
 September, 1996. 
 
 Stein, A.F., D. Lamb, and R.R. Draxler, Incorporation of detailed chemistry into a three- 
 dimensional Lagrangian-Eulerian hybrid model: Application to regional 
 tropospheric ozone, Atmos. Environ., 34, 4361-4372, 2000. 
 
 U. S. Environmental Protection Agency (EPA) Green Book - Nonattainment Areas for 
 Criteria Pollutants ( http://www.epa.gov/air/oaqps/greenbk/index.html ) Data as of 
 July 31, 2000. 
 
 U. S. Environmental Protection Agency (EPA), Guidelines for Developing an Ozone 
 Forecasting Program, U.S. EPA, Office of Air Quality Planning and Standards, 
 Report No. EPA-454/R-99-009, Research Triangle Park, NC, July 1999. 
 (available on the web at http://www.epa.gov/ttn/oarpg/tl/memoranda/foreguid.pdf 
 
 U. S. Environmental Protection Agency (EPA), 1999. 
 
 ftp:/www. epa.gov/pub/scram001/modelingcenter/NOxSIPcall/emissions 
 
 June 2001 21 
 
Committee on Environment and Natural Resources (CENR) 
 Subcommittee on Air Quality Research 
 
 The CENR is charged with improving coordination among Federal agencies involved in 
 environmental and natural resources research and development, establishing a strong information- 
 transfer link between science and policy, and developing a Federal environmental and natural 
 resources research and development strategy that responds to national and international issues. 
 There are five research subcommittees under the CENR: 
 
 • Air Quality 
 
 • Ecological Systems 
 
 • Global Change 
 
 • Natural Disaster Reduction 
 
 • Toxics and Risk 
 
 The Air Quality Research Subcommittee has articulated two major goals in its Strategic Plan: 
 
 • to enhance the effectiveness and productivity of U.S. air quality research, and 
 
 • to improve information exchange between research and policy on air quality issues, 
 including the scientific knowledge base for air quality standards and assessing 
 compliance 
 
 Dan Albritton of NOAA chairs the Subcommittee, which includes representatives from the following 
 departments and agencies: 
 
 Department of Agriculture - Agricultural Research Service 
 
 Department of Agriculture - Cooperative State Research, Education, and Extension Service 
 
 Department of Agriculture - Natural Resources Conservation Service 
 
 Department of Agriculture - U.S. Forest Service 
 
 Department of Commerce - National Institute of Science and Technology 
 
 Department of Commerce - National Oceanic and Atmospheric Administration 
 
 Department of Defense 
 
 Department of Energy 
 
 Department of Health and Human Services - Center for Disease Control and Prevention 
 
 Department of Health and Human Services - National Institutes of Health 
 
 Department of Housing and Urban Development 
 
 Department of State 
 
 Department of the Interior - National Park Service 
 
 Department of the Interior - U.S. Geological Survey 
 
 Department of Transportation - Federal Aviation Administration 
 
 Department of Transportation - Federal Highway Administration 
 
 Environmental Protection Agency 
 
 National Aeronautics and Space Administration 
 
 National Science Foundation 
 
 Office of Management and Budget 
 
 Office of Science and Technology Policy 
 
 Tennessee Valley Authority 
 

 PENN STATE UNIVERSITY LIBRARIES 
 
 ADDDDMSMlbEflb