Case Study:

Envision Utah

Methodology

Air Quality

Modeling Approach

The air quality analysis was conducted by the Utah Department of Environmental Quality, Division of Air Quality (DAQ). DAQ considered the use of a regional airshed model to model ambient air quality conditions under each scenario. DAQ determined that this class of models was not a practical alternative for the Envision Utah analysis, however, because of the long lead time necessary to prepare the necessary data inputs. Instead, DAQ developed a modeling system which is highly simplified in some respects, yet captures some of the meteorological parameters involved in regional airshed models.

The resulting model, QMOD, is a simplified air quality planning model that tracks the movement of pollutants but does not model chemical reactions (for example, ozone formation). DAQ has been using a GIS to support its regional airshed modeling since 1994. The idea for QMOD resulted from the strong integration developed between the GIS and two important components of the regional airshed model. The first component is a model that simulates wind speed and direction over the area on an hourly basis. The second component is the capability to spatially disaggregate emissions from county-wide totals to an overlay of a regularly spaced (four-kilometer) grid. The essence of QMOD is to track the movement of pollutants, by grid cell and hour, during some hypothetical day in the future and then compare the air quality implications for each development scenario against the baseline scenario. The relationship between QMOD and the underlying GIS data is shown in Figure 6.

Figure 6. Integration of GIS and QMOD Regional Airshed Modeling

Source: Based on Utah Governors Office of Planning and Budget (1999).

The process for applying QMOD is as follows:

1. Develop the emissions inventory from mobile, point, and area sources;

2. Assign the emissions to four-kilometer grid cells;

3. Use GIS to track hourly emissions based on windfield analysis; and

4. Tabulate the results.

In addition to differences in mobile source emissions, the various scenarios in the Envision Utah analysis also differed to some extent in point and area sources. While large emitters - such as power plants and major industries - were held fixed, the location of small emitters was allowed to vary in proportion to the amount of industrial development. Area sources dependent upon population (for example, lawn and garden equipment) were also distributed in proportion to population.

Data Requirements

Data available to DAQ to estimate the air quality implications of each growth scenario included:

• Daily emissions from both mobile and stationary sources;
• The geographic location of those emissions; and
• The two-dimensional wind conditions for a day of "unfavorable" meteorological conditions.

During the model run an hourly activity adjustment, based on VMT profiles obtained from UDOT, was made at the beginning of each hourly iteration to characterize a more true-to-life temporal profile of emissions. The resulting hourly pollutant concentrations by four-kilometer grid cell were measured for total ozone precursors, the sum of volatile organic compounds (VOC) and oxides of nitrogen (NOx); carbon monoxide (CO); and particulate matter (PM).

Impact Measures

Typically, estimates of pollution concentration derived from airshed models are compared to national health standards to develop regulatory policy. Since QMOD estimates the spatial distribution of emissions which contribute to the concentration of air pollution, rather than the concentration of pollution in the ambient air, the evaluation of a scenario's effect on air quality cannot be directly measured against the national health standards for air pollution.

As an alternative, the Envision Utah analysis produced a matrix of three evaluation criteria for individual pollutants for each scenario. The performance ranking of each scenario became an ordinal measure of that scenario's effect on air quality. The three metrics include:

1. Total emissions inventory;

2. Inequality of distribution; and

3. Coincidence of population and pollution.

The first metric is simply the sum of emissions across the entire region. The second metric quantifies the localization of build-up of pollutants. This was developed by ordering the four-kilometer grid cells from lowest to pollutant emissions density, then plotting the cumulative percentage of total land area against the cumulative percentage of pollutants (when plotted graphically, this produces a "Lorenz curve"). For example, in Scenario A, 58 percent of the average hourly emissions occur in 10 percent of the geographic region. In Scenario D the share of pollution increases to 66 percent of the emissions in 10 percent of the region. The underlying hypothesis is that a greater geographic concentration of pollutants is more likely to lead to an exceedance of standards in some particular area.

The third evaluation metric compares the proximity of people to air pollution. Metric 3 was created by overlaying a grid of population density onto a grid of average daily emissions after the air quality model was run. Population density by cell was then multiplied by the daily average emissions density, and the values summed across all cells. The higher the final value of this metric, the greater the likelihood that high population density and high pollution density occur in the same proximity.

Level of Effort

QMOD was developed in-house by DAQ. According to DAQ, the model took about three person-months to develop, although this included "lots of talking" to figure out what to do. A similar model could be developed in less time by another air quality agency or MPO assuming they have staff experienced in air quality modeling and also have the required GIS data and skills. DAQ is also potentially willing to provide the QMOD algorithms to others who may be interested. For comparison, DAQ believes that applying state-of-the-science Model 3 Airshed Model, which is capable of modeling ozone and particulate formation, would have required 10 to 100 times more effort.

[TOP]