One usually thinks of the Parks as havens from civilization where nature prevails and scenic vistas and clean air are the norm. In recent years, we have found that air pollution haze blocks the views in some of the most scenic parks and air quality can be as bad as in the urban areas miles away. The NPS Inventory and Monito
Air pollution mixes in the atmosphere and has a persistence from a few hours to several days in the lower atmosphere. Generally, pollutant concentrations are highest near sources, but decrease with distance from a combination of dilution and removal from the air. For areas away from major pollutant sources, spatial Interpolation methods can be used to estimate concentrations, especially if periods of several years are included so that individual weather events are averaged out.

GIS spatial interpolation and mapping seemed like workable approach; it was proposed and got funded by the Inventory and Monitoring (I&M) Program. The Air Resources Division (ARD) has done network monitoring of air quality (Figure 1), atmospheric pollutant deposition, and visibility in the parks since the early 1980's in partnership with other agencies. The air quality GIS project used the network datasets in a cooperative effort between ARD staff and University of Denver researchers and students.

Data from 5-years of monitoring was gathered from all the national air monitoring networks for visibility (IMPROVE), for ozone (EPA AIRS database and CASTNet), for wet deposition (NADP), and for dry deposition (CASTNet). Processing and examination of the data sets for errors and formatting problems was the first step, statistical summaries in a format ready for entry into a GIS database followed. Inverse distance-weighted and kriging techniques were applied to make girded estimates of air quality for the contiguous 48 states. These estimates were then color-coded for display in the GIS system, ArcView. The result is color maps of air quality estimates with the locations of I&M park units shown for easy reference. Maps are available for several ozone statistics; the acidic wet deposition species of SO4, NO3, NH4, Cl, Ca, Na, K; visibility indicators; particulates such as PM2.5; and concentrations of SO4, NO3, and SO2 in the air. The locations of monitoring stations were also mapped, and the ones in prox
What is the Current Air Quality ?
Some examples show the broad utility of the GIS approach for answering air quality questions. Figure 2 illustrates the detail and patterns that can be obtained for an ozone statistic related to the national standard for ozone set for the protection of human health and natural resources. In general, ozone concentrations are high in the Eastern US and significant parts of California. An estimate of the ozone concentrations can be obtained for each park unit using these map layers and GIS tools. Two factors make this practical. First, air pollution is diluted and dispersed as it moves away from a source so that hot spots tend to occur only near sources. And, second, the monitoring networks are a fairly sparse set of sampling points. From a climatological view point, the results from the interpolation maps are a good starting point.

Haze that degrades scenic views is an important concern in the parks. Figure 3 presents the visibility as the amount of light extinction on the 20% worst days. Because the Eastern US tends to have more pollutant sources and be more humid, it also has the worst haze. Pictures of clear and hazy days at an eastern park and a western park show how much difference this can make. When visibility is good the colors are bright and distant peaks are visible; when it is hazy from pollution the views are flattened and dull.



Additional field work
Greater detail in the interpolation mapping may be possible. An intensive study with over 60 temporary monitoring sites was done in Great Smoky Mountains National Park with the help of about 50 volunteers who hiked the trails to do the sampling. The project was to see how ozone could be interpolated in complex topological terrain. Ozone distribution was found to have a varying background, but a fairly consistent gradient with elevation. Differences in the locations of pollutant sources and air flow patterns, likewise produced some localized variances in the ozone. The detailed map of ozone concentration (Figure 4) illustrates that closely related factors can be used to enhance the basic pollutant interpolations. The seasonal ozone concentration map provides detail on where in the park the highest exposures are likely to occur. When detailed maps of ozone sensitive vegetation are available, park resource managers will be able to predict which parts of the parks are most likely to have plant injury. Thus,
Significance to NPS and understanding air quality
The most immediate result of this procedure is a series of air pollutant maps that estimate concentrations for each of the 270+ parks that have significant natural resources. Tables of pollutant concentrations for each of the parks were obtained from the intersection of park boundary polygons and the pollutant layers. These concentrations can then be compared to thresholds and national standards for each park to assess relative expected effects of the air pollution. Estimating the effect of all the pollutants together, however, can be a more difficult task. We have attempted an overall view of air pollution by the calculation of an air quality condition (AQC). To calculate the AQC, each pollutant is indexed to its threshold of negative effects on resources and then the pollutant is given a relative weighting before summing all the indexed pollutants. The resulting AQC gives some idea of how much all of the air pollutants are affecting each park.

As with all resources that might be inventoried and monitored in a park, we are looking for the current condition, what effect it is having on resources, how much change is occurring, and how can we prevent undesirable consequences. We have initiated an ozone risk assessment that takes into account the presence of sensitive plant species, the annual and 5 year exposures of ozone, the climatology and drought index, and other factors to estimate the amount of ozone-caused plant injury. Similar processes may also be used to assess the degree of acidic deposition or the loss in visibility that degrades the natural scenic views. Using interpolated pollutant values from different periods of years as GIS layers, we can calculate and map the change in air pollutants.

The various GIS air pollutant layers also provide an important decision tool. For example (figure 5), parks and areas recognized from the mapping as exceeding the national standards or known thresholds may need further study and monitoring to verify the predictions and to assess the extent of resource damage. Areas that are under-represented by monitoring stations may have a greater uncertainty in the pollutant predictions. Other factors like rapid changes in local population or emissions sources, increasing trends in pollutant concentrations, the presence of known sensitive species, or absence of nearby monitors can be considered. These areas can be mapped along with park locations to help select sites for additional monitoring.

Making the Information Accessible
The GIS interpolation project has created many maps and tables of information, too much to easily use if the GIS software and expertise aren't available locally. Therefore, a web-based product was built using ArcIMS which we have called the Air Quality Air Atlas (http://www2.nature.nps.gov/ard/gas/airatlas-du/viewer_index.htm ). This mini-GIS mapping tool on the Internet makes the results from the project readily accessible. All the maps are viewable and the user can zoom into the region or park of interest. A query tool allows the user to get a table of the estimated air quality parameters for an individual park or a selection of several. The air quality inventory tables and links to closely related web sites are on the Air Atlas web site.

Future uses and possibilities
Other possibilities for this GIS approach to estimating air quality include future updates, trend estimates, and consideration of the interactive effects of multiple pollutants. Basically, the national mapping of networks and estimated air pollutants becomes a tool for further analysis. Thus, a park resource we can not normally see becomes a factor we can attempt to understand and manage.





Authors:
Dr. John D. Ray
NPS Air Resources Division
Denver, CO
John_d_ray@nps.gov
(303) 969-2820


Dr. Paul Sutton
Department of Geography
University of Denver
Denver, CO
Psutton@du.edu
(303) 871-2399


Andrew Bingham
University of Denver
Denver, CO
Drew_Bingham@nps.gov
(303) 969-2341