Table of Contents | Previous | Next

Canada - United States Air Quality Agreement

2002 Progress Report

SECTION IV  Scientific Cooperation

PARTICULATE MATTER MONITORING, DATA ANALYSIS, MODELLING AND MAPPING

Transboundary PM Analysis and PM Workplan

As an outgrowth of the Joint Plan of Action to address transboundary PM issues, the governments of Canada and the United States are executing and evaluating regional air quality models with parallel ambient data analysis to characterize transboundary contributions of PM and precursors in border regions by the end of 2003. The report will be the focus of decision making on whether to develop a PM annex to the Air Quality Agreement.

As part of this undertaking, the governments have developed a PM workplan resulting from discussions at a second joint PM workshop held in the United States in November 2001. The first workshop was held in Canada in 1999. Key elements of the PM workplan are outlined below.

Canada and the United States plan to jointly analyze high PM_2.5 episodes in North America. The largest component of PM_2.5 in the summer is sulphate; nitrate can be a significant component of PM_2.5 in the winter. Organic carbon is a large component of PM_2.5 in all seasons.

A joint 1995/1996 Canada–U.S. emissions inventory for 36 km grids is being developed and processed for the Regional Modelling System for Aerosols and Deposition (REMSAD) and the Community Multiscale Air Quality model (CMAQ) with a target completion date of August 2002. Projected Canada–U.S. emission inventories to future years are targeted for completion by fall 2002. An approach for resolving confidentiality issues associated with Canadian data was identified and is being used in the development of these inventories. Considerable work is already underway in both countries using the air quality transport model AURAMS (A Unified Regional Air Quality Modelling System) in Canada and the CMAQ model in both the United States and Canada to investigate the impact of emission changes on PM and ozone air quality. Over the next six months to a year, the models will move through their evaluation phases and begin to be applied against emission reduction scenarios investigating the potential for transboundary impacts.

Joint Canada-U.S. modelling will take place for the full year of 1996 (MM5 meteorology) with REMSAD at 36 km horizontal grid resolution run as a base case with the 1995/1996 emissions inventory. The REMSAD modelling domain is being expanded northward and eastward to cover most of Canada. Episodes for July 1995 with MM5 meteorology will be run with the CMAQ and AURAMS models. Projection years (e.g., 2010, 2020) will also be run with the models. The joint 1995/1996 base case modelling is scheduled to be completed before the third U.S.–Canada workshop. The workshop, which will be held in Toronto, Canada in fall 2002, will focus on assessing progress and refining measures to complete the transboundary impact assessment.

The Canadian National Atmospheric Chemistry (NAtChem) Database And Analysis System, NAPS, CASTNET, IMPROVE, and Aerometric Information Retrieval System (AIRS) air quality databases also are being used to illustrate transboundary transport by combining the PM data with meteorological data and trajectories. Over the next year, specific ambient data analysis tasks will be identified and carried out.

This undertaking will be one of the first major policy applications of the PM regional air quality models. The combined approach of using regional air quality models and ambient data analysis is essential for PM. Using the weight of evidence derived from both approaches will result in the best policy guidance based on current scientific understanding.

PM Monitoring Comparison

Several different monitoring methods are used within the Canadian and U.S. networks. They operate on various sampling schedules, from continuous hourly measurements to 24-hour average measurements taken once every six days.

Monitoring agencies in the United States and Canada are operating both filter-based and continuous fine PM samplers. In the United States, a robust network of Federal Reference Method (FRM) samplers is augmented by a smaller number of continuous monitors. Most U.S. monitoring agencies use the Tapered Element Oscillating Microbalance (TEOM) as their continuous monitor; a smaller number of agencies use other technologies. However, many of the TEOMs operated are using different method approaches with respect to operating temperature and use of a particle separation device. In Canada, most city and provincial monitoring agencies operate the dichotomous sampler at a limited number of sites to obtain filter measurements collocated with the TEOM continuous monitor for a real-time PM signal.

A larger number of Canadian sites operate the TEOM without a collocated filter-based measurement. Canadian monitoring agencies are also operating the TEOMs in varying ways with respect to temperature and particle separation device. Since each country primarily relies on different methods for filter-based measurements and differences within a method with respect to the TEOM, an understanding of the comparability of these methods is required for data to be appropriately analyzed across the two countries. To provide an understanding of methods employed in each country, monitoring representatives from Canada and the United States have been cooperating by sharing information on methods used.

Monitoring contacts at the Federal, State, and regional levels from the United States have participated with Canadians in workshops designed to work toward better data comparability within each country and between the countries. Additionally, EPA is working with monitoring agencies in the United States and Canada to test mapping of data from continuous PM monitors.

Through the work of NEG/ECP there are collocated TEOMs and FRMs in Nova Scotia, New Brunswick, Quebec, and Prince Edward Island.

PM Monitoring

In Canada, the PM_2.5 and PM₁₀ monitoring initiated in the mid-1980s is continuing to be updated to meet current scientific and policy needs. In 2002, the regional CAPMoN network is planning to add up to five PM_2.5 and PM₁₀ gravimetric monitors, in addition to all the gaseous precursors (SO₂, NO_x, VOCs, and NH₃) at two of the five sites. Currently, in the primarily urban NAPS network the federal and provincial governments are operating 106 gravimetric PM₁₀ monitors and 50 TEOM PM10 monitors. Similarly, there are 102 PM_2.5 TEOM sites and 21 gravimetric monitors. PM_2.5 chemical monitors being set up at five sites in the NAPS network this year will identify particle sulphate, nitrate, ammonium, organic carbon, black carbon, and soil contributions to total PM_2.5 mass.

In the United States, deployment of new monitoring networks for PM_2.5 is supplying PM_2.5 compliance monitoring data and Air Quality Index data for reporting to the public. Specific monitoring network data include:

• Approximately 1,100 Federal Reference Method PM_2.5 sites, which were installed throughout 1999 and into 2000. Data from this compliance portion of the network are available publicly. Work has begun to identify design values for upcoming designation decisions.
• Approximately 267 continuous ambient monitors, which are particularly needed for public data reporting and mapping work that is being planned currently. EPA is focusing on getting near real-time data reporting through the AIRNOW system in the 36 metropolitan areas that are carried by various media sources, including USA Today.
• A network of sites that provide chemically speciated PM data, including 54 urban trends sites operated by state and local agencies, 213 supplemental sites important for SIP development and also operated by state and local agencies, 28 rural sites operated by state and local agencies that use the regional haze program's IMPROVE protocols, and 110 sites operated by Federal land managers in Class 1 areas for the IMPROVE program. Also within the state and local agencies' urban networks, EPA expects to have up to 15 chemical speciation sites that use continuous speciation technologies for nitrates, sulphates, and carbon. EPA is investigating these continuous methods at five state and local agency locations prior to releasing states to purchase equipment for all sites. Tribal agencies are also providing additional fine PM data through the use of both filter-based Federal reference methods and the IMPROVE protocols.
• Eight supersites with useful data in 2001, specifically from monitoring sites located in New York, Pittsburgh, Baltimore, St. Louis, Atlanta, Fresno, Los Angeles, and Houston. The supersites were developed as platforms for research and measurement methods by various universities in these eight areas. The measurement equipment and analysis techniques used at each supersite were often experimental or more advanced than those typically used in the routine PM networks. Data and information from each supersite are being analyzed both by the individual universities and through a data analysis program designed to evaluate data centrally from all eight supersites.
• CASTNet data

  [D]

Figure 19 shows the annual average PM_2.5 concentrations for the period 1999-2000 measured by these networks.

  [D]

Sulphate and Nitrate Concentrations in the Eastern United States and Canada

Particle sulphate concentrations in eastern North America are highest in the central portion of the eastern United States in an area encompassing southeastern Ohio, southwestern Pennsylvania, Kentucky, Virginia, West Virginia, Tennessee, northern Georgia, and northwestern Alabama (see figure 18). Almost all states south of the Great Lakes (except Florida) show moderately high concentrations. Southwestern Ontario has the highest concentrations in eastern Canada.

In contrast, particle nitrate concentrations are highest in the northwest part of the eastern United States in an area that includes Illinois, Indiana, Ohio, and southwestern Ontario. The lowest concentrations occur in the southern Appalachian area, New England, and northern and Atlantic Canada.

PM Modelling

At the Meteorological Service of Canada (MSC), work is underway on a "unified" regional air quality modelling system to address multiple-pollutant, multiple-issue applications, including PM, ozone, acid rain, and air toxics–AURAMS . The four foundation components for this new modelling system are the Environment Canada CEPS (Canadian Emission Processing System), MC2 (Mesoscale Compressible Community Multiscale Meteorological Model), CHRONOS (Canadian Hemispheric and Regional Ozone and NO_x System), Eulerian regional photochemical model, and a sectional size distributed aerosol module. The latest version (March 2002) of AURAMS will be evaluated with available data. Current model development and evaluation uses data sets from the 1990 emission inventory and various studies. Visibility modelling in Canada is planned via application of the CMAQ methodology based upon anomalous diffraction theory within the AURAMS framework post 2001. Also at MSC, CHRONOS, a tropospheric ozone model whose domain now includes all of Canada and most of the United States, is being enhanced to include methodologies from AURAMS for PM processes and ultimately to feed public air quality advisory programs.

Environment Canada and Indiana University have been applying a Lagrangian aerosol model, ACDEP (Atmospheric Chemistry and Deposition model), to simulate aerosol levels in the Lower Fraser Valley airshed in southwestern British Columbia and northwestern Washington State. Model-predicted inorganic aerosol fields have been evaluated against 1993 event-specific field data (the Pacific 1993 field campaign). The model has been extended to describe the formation of secondary organic aerosols and has been renamed the Inorganic and Secondary Organic PARTicle model (ISOPART). Also for this region, MC2/UAM-VPM (Urban Airshed Model V Particulate Matter) is under development by Environment Canada and other government and industrial partners. In cooperation with the University of Washington and Environment Canada, CMAQ is being applied over the Pacific Northwest on a 12 km grid.

The Ontario Ministry of the Environment, the University of Waterloo, and Environment Canada are working with the Models-3/CMAQ platform. The model domain extended from 300 km west of Chicago to Quebec City, with Tennessee as the southernmost state. At the National Research Council of Canada, the Models-3/CMAQ modelling system has been modified, evaluated, and applied to the Lower Fraser Valley (LFV), which includes southwest British Columbia and northwest Washington State. To generate better boundary conditions for the LFV study, model runs have also been conducted for a larger area extending southward into the State of Oregon. The modifications to the Models-3/CMAQ system include the development and integration of a new secondary organic aerosol modelling component, development of a new CMAQ postprocessor for converting CMAQ model PM results into size-resolved PM concentrations, and adjustment to the CMAQ parameters for primary PM speciation.

The United States is proceeding with development, testing, application, and evaluation of REMSAD and CMAQ. REMSAD has been revised to include updates recommended by a scientific peer review panel. EPA is planning to use Version 7 of REMSAD. The latest version of CMAQ was issued in June 2002. Both models have been applied for the continental United States, southern Canada, and northern Mexico for every hour of a full year using 1996 MM5 meteorology generated at 36 km horizontal grid square resolution. REMSAD has also been applied using 1998 Rapid Update Cycle (RUC) meteorology. The emission inventory for the United States is the 1996 National Emissions Inventory (NEI).

The United States has been using Canadian data for REMSAD applications. Canadian emissions for the REMSAD applications have been 1990 point, area, and mobile sources. The Canadian emissions for the CMAQ applications have been 1995 area and mobile sources. The latest 1995 Canadian point, area, and mobile emissions will be incorporated into future applications of both models. REMSAD has been run for future years (2010, 2020, and 2030) to evaluate the effects of regulatory initiatives, including the heavy-duty engine rule and President Bush's proposed Clear Skies Initiative. The REMSAD modelling domain is currently being expanded northward and eastward to include most of Canada. REMSAD 1996 base case modelling using the latest Canadian emissions and the expanded modelling domain are scheduled for completion by October 2002.

PM Mapping

In 2001, initial work was completed to begin the transfer, as with ozone, of real-time PM_2.5 data from six Canadian provinces and several northeastern states to begin exploring the concept of mapping PM_2.5. The plan is to test the real-time delivery of the data and explore the feasibility of mapping this parameter.

Effects of Air Pollution

In Annex 2, the Parties also specifically agreed "to cooperate and exchange information" concerning the monitoring of ecosystem effects such as those evident in aquatic ecosystems, visibility, and forests and concerning effects of atmospheric pollution in general on human health and ecosystems, including research on health effects of acid aerosols and research on the long-term effects of low concentrations of air pollutants on ecosystems, "possibly in a critical loads framework." This section focuses on Canadian and U.S. progress in meeting this commitment.