10.1 The Urban Airshed Model - UAM-IV

10.1.1 Conceptual Overview

The Urban Airshed Model is a three-dimensional photochemical grid model designed to calculate the concentrations of both inert and chemically reactive pollutants by simulating the physical and chemical processes in the atmosphere that affect pollutant concentrations [178]. The basis for the UAM is the atmospheric diffusion equation (ADE). This equation represents a mass balance in which all of the relevant emissions, transport, diffusion, chemical reactions, and removal processes are expressed as follows:

$$\frac{\partial C_i}{\partial t} + \frac{\partial (UC_i)}{\par... ...tial (VC_i)}{\partial y} + \frac{\partial (WC_i)}{\partial z}$$

$$= \frac{\partial }{\partial x} \left( K_H \frac{\partial C_i... ...partial z} \right) + {\cal R}_i + {\cal S} + {\cal L}_i \hfill$$ (10.1)

where,

$i$	=	chemical species $i$ (where $i$ = 1, 2,..., $i$,..., $N$)
$C_i$	=	pollutant concentration of species $i$
$u, v, w$	=	horizontal and vertical wind speed components = $f(x, y, z, t)$
$K_H, K_V$	=	horizontal and vertical turbulent diffusion coefficients = $f(x, y, z, t)$
${\cal R}_i$	=	rate of formation of $i$ by chemical reactions = $f(C_1, C_2, \ldots C_i, \ldots, C_N)$
${\cal S}$	=	emission rate of all precursors $P_1, P_2, \ldots, P_p$,
	=	$S(x, y, z, t, P_1, P_2, \ldots, P_i,\ldots, P_p)$
${\cal L}_i$	=	net rate of removal of pollutant $i$ by surface uptake processes
	=	$f(x, y, z, t)$

This model is typically applied to an 8- to 72- hour period to model air quality ``episodes'' - periods during which adverse meteorological conditions result in elevated pollutant concentrations of the chemical species of interest. UAM is mainly used to study the photochemical air quality, which pertains to air quality with respect to ambient ozone concentrations. High ozone concentrations in the ambient environment lead to adverse health effects [78,181]. Ozone is primarily formed in the atmosphere through a complex chemical mechanism involving oxides of nitrogen (NO$_x$) and volatile organic compounds (VOCs) in the presence of sunlight.

The major factors that affect photochemical air quality include [223]:

• The spatial and temporal distribution of emissions of NO$_x$ and VOC (both anthropogenic and biogenic),
• The composition of the emitted VOC and NO$_x$,
• The spatial and temporal variations in the wind fields,
• The dynamics of the atmospheric boundary layer, including stability and the level of mixing,
• The chemical reactions involving VOC, NO$_x$, and other important species, (the Carbon Bond Mechanism-IV described in detail in Table C.2 is implemented in the UAM),
• The diurnal variations of solar radiation and temperature,
• The loss of ozone and ozone precursors by dry deposition, and
• The ambient background of VOC, NO$_x$, and other species at the immediate outside of the boundary of the region of study, and above the region of study.

The UAM is typically run with a horizontal resolution of 2-5 km and a vertical resolution of 4-6 layers up to about 2 km above ground level. The region to be simulated is divided up into a three-dimensional grid covering the region of interest. Horizontal grid cells are rectangular with constant lengths in the x- and y-directions. Vertical layer thicknesses are defined by the user based on the diffusion break, the top of the region, the number of layers below and above the diffusion break and the minimum layer thickness. The diffusion break corresponds to the top of a mixed layer, either an unstable convective layer during the day (i.e. the mixing height) or a shallow mechanically mixed layer at night. The region top is usually defined at or slightly above the maximum daily diffusion break. The UAM solves the atmospheric diffusion equation (ADE) using the method of fractional steps. The master or advection time step, determined by numerical stability criteria (grid size vs domain-maximum wind speed), is typically on the order of 5 min. In each advection time step, the terms in the equation that represent the different atmospheric processes (e.g. advection, chemistry or vertical diffusion) are solved separately using the most efficient numerical integration technique for the given process. The advection time step must be divided into multiple chemistry and vertical diffusion time steps to maintain numerical stability.

10.1.2 Treatment of Physical/Chemical Processes

10.1.2.1 Advective pollutant transport

Pollutants are transported primarily by advection, that is by the mean or bulk motion of the air. Advection in the UAM is treated by specifying horizontal wind fields (i.e. $u$ and $v$ wind components in each grid cell) for each vertical layer. The vertical wind velocity in the UAM terrain-following coordinate system can then be calculated from the conservation of mass equation. Proper specification of the hourly and three-dimensional varying winds is one of the key steps to successful application of the UAM. The winds influence how different emissions are mixed together, advected downwind and diluted.

10.1.2.2 Turbulent diffusion

Turbulent diffusion (dispersion) is assumed to be proportional to the rate of change of concentration in space (i.e. the concentration gradient). The proportionality factor is termed the eddy diffusivity coefficient ( $K_{x}, K_{y}$ and $K_{z}$) in the ADE). Because it has been difficult to obtain precise measurements of the eddy diffusivity coefficients, theoretical estimates have been used. Control theory techniques are employed in conjunction with the results of a planetary boundary layer model to generate optimal diffusivity coefficients in the UAM [139,138].

10.1.2.3 Atmospheric Chemistry

The UAM employs the Carbon Bond Mechanism (CBM-IV) for solving chemical kinetics of atmospheric chemistry. As implemented in the UAM, the CBM-IV contains 86 reactions and 35 species. The differential equations that describe the CBM-IV are a ``stiff'' system, that is, the equations contain wide variations in time (reaction rate) constants. Solving these equations with a ``stiff'' numerical integration scheme, such as the one developed by Gear [77], would result in prohibitively expensive computer time. Thus, the solution of the CBM-IV in the UAM uses quasi-steady-state assumptions (QSSA) for the low-mass fast-reacting species (i.e. the stiff species) and the more computationally efficient Crank-Nicholson algorithm for the remainder of the state species.

10.1.2.4 Surface Removal Processes

Many types of pollutants, including nitrogen oxides, ozone, carbon monoxide and sulfur compounds, are removed from the surface layer by such features as vegetation through the process of dry deposition. In the UAM, dry deposition is assumed to occur in a two-step process: the transfer of pollutants through the atmosphere to the surface and the uptake of the pollutants by vegetation and other materials at the surface. This process involves a resistance to mass transport and a resistance to surface removal. The transport resistance is estimated from theoretical considerations of turbulent transfer in the atmospheric boundary layer. The surface resistance is obtained from experimental data on the uptake of pollutants by various surface features.

10.1.3 Applications of the UAM

Because the UAM accounts for spatial and temporal variations as well as differences in the reactivity/speciation of emissions, it is ideally suited for evaluating the effects of emission control scenarios on urban air quality. This is accomplished by first replicating a historical ozone episode to establish a base case simulation. Model inputs are prepared from observed meteorological, emission, and air quality data for a particular day or days. The model is then applied with these inputs and the results are evaluated to determine its performance. Once the model results have been evaluated and determined to perform within prescribed levels, the same meteorological inputs and a projected emission inventory can be used to simulate possible future emission scenarios. That is, the model will calculate hourly ozone patterns likely to occur under the same meteorological conditions as the base case.

The first regulatory use and practical applications of the UAM were carried out for the Denver area on behalf of the Colorado Division of Highways and EPA's Region VII in 1978. The UAM was used to evaluate whether various transportation plans and programs were consistent with the SIP and to evaluate the effects on Denver's air quality of urban growth and development that might result from the construction of proposed wastewater treatment facilities.

In the late 1970s, EPA's OAQPS initiated a program to examine the applicability and practicality of the UAM in routine ozone attainment demonstrations required by the SIP process. Data collection, emission inventory development, model performance evaluation and application were major elements of this nation-wide program. Building off the St. Louis UAM applications and an extensive series of UAM sensitivity studies designed to provide guidance concerning the types and amounts of data required to support the UAM application, data for an application of the UAM, supported by OAQPS, were collected in Tulsa, Philadelphia/New Jersey, Baltimore/Washington and New York.

Routine use of the UAM is an emerging practice. Since its first use for air quality planning in 1978, the UAM has been or is being used by the EPA, the Exxon Corporation, British Leyland, the TNO in the Netherlands, Pacific Gas and Electric, Southern California Edison Company, Arizona Department of Health, the South Coast (California) Air Quality Management District, the New York Department of Environmental Protection, the New Jersey Department of Environmental Protection, the California Air Resources Board, and the cities of Taipei and Kaohsiung in Taiwan.