Spatio-Temporal Data Mining for Air Pollution Problems

Spatio-Temporal Data Mining for Air Pollution Problems

Seoung Bum Kim (The University of Texas at Arlington, USA), Chivalai Temiyasathit (The University of Texas at Arlington, USA), Sun-Kyoung Park (North Central Texas Council of Governments, USA) and Victoria C.P. Chen (The University of Texas at Arlington, USA)
Copyright: © 2009 |Pages: 8
DOI: 10.4018/978-1-60566-010-3.ch277
OnDemand PDF Download:
List Price: $37.50


Vast amounts of data are being generated to extract implicit patterns of ambient air pollution. Because air pollution data are generally collected in a wide area of interest over a relatively long period, such analyses should take into account both temporal and spatial characteristics. Furthermore, combinations of observations from multiple monitoring stations, each with a large number of serially correlated values, lead to a situation that poses a great challenge to analytical and computational capabilities. Data mining methods are efficient for analyzing such large and complicated data. Despite the great potential of applying data mining methods to such complicated air pollution data, the appropriate methods remain premature and insufficient. The major aim of this chapter is to present some data mining methods, along with the real data, as a tool for analyzing the complex behavior of ambient air pollutants.
Chapter Preview


In 1990, under the Clean Air Act., the U.S. Environmental Protection Agency (EPA) set the National Ambient Air Quality Standards (NAAQS) for six pollutants, also known as criteria pollutants, which are particulate matter, ozone, sulfur dioxide, nitrogen dioxides, carbon monoxide, and lead (US EPA, 1990). Any exceedance of the NAAQS results in non-attainment of the region for that particular pollutant.

Well-known consequences of air pollution include the green house effect (global warming), stratospheric ozone depletion, tropospheric (ground-level) ozone, and acid rain (Wark, Warner, & Davis, 1998). In this chapter, we present applications on tropospheric ozone and the less publicized air pollution problem of particulate matter. High concentrations of tropospheric ozone affect human health by causing acute respiratory problems, chest pain, coughing, throat irritation, or even asthma (Lippmann, 1989). Ozone also interferes with the ability of plants to produce and store food, damages the leaves of trees, reduces crop yields, and impacts species diversity in ecosystems (Bobbink, 1998; Chameides & Kasibhatla, 1994). Particulate matter is an air contaminant that results from various particle emissions. For example, PM2.5 (particulate matter that is 2.5 micrometers or smaller in size) has the potential to cause adverse health effects in humans, including premature mortality, nose and throat irritation, and lung damage (e.g., Pope et al., 2002). Furthermore, PM2.5 has been associated with visibility impairment, acid deposition, and regional climate change.

To reduce pollutant concentrations and establish the relevant pollution control program, a clear understanding of the pattern of pollutants in particular regions and time periods is necessary. Data mining techniques can help investigate the behavior of ambient air pollutants and allow us to extract implicit and potentially useful knowledge from complex air quality data. Figure 1 illustrates the five primary stages in the data mining process in air pollution problems: data collection, data preprocessing, explanatory analysis and visualization, model construction, and model evaluation.

Figure 1.

Overview of data mining in air pollution problems.

Complete Chapter List

Search this Book: