United States Environmental Protection Agency	Office of Research and Development

National Exposure Research Laboratory
FY02 Research Abstract

Government Performance Results Act (GPRA) Goal 1
APM60

Significant Research Findings:

Preliminary Results of Modeling Airborne Toxic Pollutants
and PM at Neighborhood Scales

Air quality (AQ) simulation models provide a basis for developing and
implementing the National Ambient Air Quality Standards (NAAQS) and are
being used as tools for performing risk-based assessments of air toxics and
for developing and testing environmental management strategies. Spatial and
temporal patterns of fine particulate matter (PM2 5) and airborne air toxic
pollutants vary considerably across urban areas and under different
geographical and climatic regimes. For human exposure assessments, AQ
models should be capable of simulating the features observed on
neighborhood-sized scales (~1 km), especially for those situations where
pollutants exhibit a high degree of spatial and temporal variability.

Meanwhile, exposure models need to be able to handle concentration fields
that have been resolved to the neighborhood scale level when they are used
for addressing issues such as environmental justice, community-based risk
assessments, and hot spot analysis. Previous versions of the U.S.
Environmental Protection Agency's state-of-the-art Models-3/Community
Multiscale Air Quality (CMAQ) modeling system were able to simulate the
meteorology and complex chemistry associated with fine-particulate and
toxic pollutants at spatial resolutions ranging from 36 to 4 km (i.e., regional
to urban scales). However, human exposure models continue to be driven
using either limited amounts of monitoring data or crude dispersion modeling
tools. Thus, rigorous chemical transport models (like CMAQ) need to be
extended down to a spatial and temporal scale necessary for accurately
simulating neighborhood scale features (about 1 km) for use by human
exposure models.

Research	In this study, we explore the requirements for modeling air quality at

Approach	neighborhood scales to enable human exposure assessments at the community

level. Our approach is to investigate the science and operational
requirements for running CMAQ at ~1 km grid resolution. Additionally, we
suggest methods to handle the concentration variability due to within-grid
sources and atmospheric processes that are not directly resolved by the
model. Recognizing that much of the attention for exposure assessments will

National Exposure Research Laboratory — November 2002

Scientific
Problem and
Policy Issues


-------
be focused in urban areas, it is important to ensure that the flow fields in
urban areas are accurately simulated, taking into account the effects of
heterogeneous building features in urban areas.

In this study, a prototype urban canopy parameterization (UCP) is introduced
into the NCAR-PSU Mesoscale Model Version 5 (MM5) at a fine-scale
resolution (~1 km horizontal grid spacing). (The MM5 provides the
meteorological fields for the chemistry-transport model in CMAQ.) The
UCP is an approach to describe the impact that buildings and other urban
structures have on the winds and temperatures in each of the grids. The UCP
is applied to all grid cells in MM5 that have some fraction of land classified
as urban. For this investigation, a model domain and study area centered on
the Philadelphia metropolitan area was established, and the MM5/CMAQ
model system was run for simulations at horizontal grid dimensions of 36,
12, 4, and 1.33 km, the smallest grid size representing the neighborhood
scale.

Results and	The modeling results indicate that the UCP parameterization produces

Implications	significant differences in the predicted mean and turbulent wind fields within

the urban canopy, especially in areas characterized by a high density of tall
buildings. The vertical profiles of wind speed and turbulence derived with
UCP in MM5 are highly consistent with the results from wind tunnel studies.
These results illustrate the importance of accounting for the urban structures
in modeling the flow in urban areas. It provides a basis for more accurately
resolving the magnitude and spatial details of the modeled air quality fields,
especially for pollutant species that may have sharp spatial gradients that are
important for human exposure assessment. By adding UCP into MM5,
changes in spatial distributions from CMAQ were evident for several
pollutant concentration fields including nitrogen oxides (NOx), ozone, several
PM species, as well as the toxic species formaldehyde and acetaldehyde.
Furthermore, we found that the spatial representation of the concentration
fields varied by pollutant species and was dependent upon grid cell size.
Not surprisingly, for most pollutant species, decreasing grid cell size from 4
to 1.33 km accentuated the gradients of a pollutants concentration distribution
pattern. For example, NOx concentration features near dense mobile sources
and point sources were considerably sharpened both in terms of horizontal
gradients and concentration magnitude at 1.33 km resolution. Consequently,
the simulations showed a corresponding decrease in ozone due to the effect
of titration from the NO sources. Also, while the simulations for PM mass
were relatively insensitive to increased grid resolution, its constituents did
exhibit spatial texture. In addition, the simulation of both formaldehyde and
acetaldehyde demonstrated that modeling these (and other toxic pollutants) at
fine resolution (1.33 km) can lead to the identification of hot spots for toxic
pollutants.

National Exposure Research Laboratory — November 2002


-------
Contributions on developing and implementing the UCP parameterizations in
MM5 were made by Dr. Avraham Lacser from the Israel Institute of
Biological Research and by Dr. Sylvain Dupont, a Postdoctoral Fellow from
University Corporation for Atmospheric Research. Collaboration with Dr.
Jerry Herwehe of the Atmospheric Turbulence and Diffusion Division at the
National Oceanic and Atmospheric Administration's Air Resources
Laboratory is providing a basis for examining the variability in pollutant
concentrations due to coupled turbulence and photochemistry.

A report on this study has been produced towards meeting Annual
Performance Measure 60 in support of GPRA Goal 1 (Clear Air), Objective
1.2 (Eliminate risks from air toxics). A manuscript has been prepared for
submission to a peer-reviewed scientific journal, and a draft of this
manuscript can be viewed at www.epa.gov/asmdnerl/nscale.pdf.

Future Research Lidar scans and stereo photogrammetry data from airborne platforms are

being used to develop high-resolution, three-dimensional representations of
buildings and trees in a number of US cities. Such data can be incorporated
into improved UCPs to further enhance the simulation of both the air flow
and pollutant concentration patterns around urban areas. A set of detailed
UCPs based on this approach is being developed for Houston, Texas and
will be evaluated as part of this effort. Meanwhile, the development of
methods to describe the within-grid variability of pollutant concentrations
will continue, along with efforts to investigate the linkages to human
exposure models. Eventually, this research should lead to better models that
will allow predictive, as well as retrospective, assessments of human
exposure and to address issues relevant to Homeland security.

Questions and inquiries may be directed to:

Dr. Jason Ching

U.S. EPA, Office of Research and Development
National Exposure Research Laboratory (E243-03)

Research Triangle Park, NC 27711

Phone 919/541-4801
E-mail: ching.jason@epa.gov

Research
Collaboration
and Publications

Contacts for

Additional

Information

National Exposure Research Laboratory — November 2002


-------