One of the well-documented air pollution problems associated with intense anthropogenic activities is the urban heat island (UHI) phenomenon. At present, many studies have established the relationship between convergence and UHI. For example, a stronger convergence is due to stronger heat island intensities, which are in turn associated with anthropogenic heat flux (Khan & Simpson, 2001). A slight change in surface winds with convergence has formed over Paris and Tokyo due to the UHI effect (Fujibe, 2003; Lemonsu & Masson, 2002).
In urban areas, convergence worsens air quality. The convergence zone persists over the urbanized area, finally resulting in high concentrations of pollutants (Yoshikado & Tsuchida, 1996). Moreover, many epidemiological studies have shown an association between air pollution and various health issues (Lin et al., 2001; Llorca, Salas, Prieto-Salceda, Chinchon-Bengoechea, & Delgado-Rodriguez, 2005). Numerous studies have focused on the association between UHI and convergence, as well as the relationship between air pollution and human health. Associations among air pollution, hospital respiratory admissions, and convergence due to the UHI effect in urban settings, however, remain unclear.
According to data analysis, the air quality is worse in Dali than in other areas; in Dali, respiratory admission counts are higher than in other areas and many clear UHIs have appeared, especially in autumn. This county, however, is not near a major stationary air pollutant source, and its population count is not the largest in metropolitan Taichung. Hence, we suspect that the worsened air pollution is probably related to the UHI effect, and has resulted in higher hospital respiratory admission counts in Dali.
The study described here attempts to clarify the aforementioned ambiguous--or otherwise unknown--associations among the three parameters, by way of statistical analysis and observations of one experimental campaign.
The geographical location of Taiwan is on the border between the Eurasian continent and the Pacific Ocean; as such, it serves as a boundary between the earth's largest land and water masses, and it includes both temperate and tropical zones.
The Taichung metropolis, with nearly 2.5 million residents, is located in the Taichung Basin in central Taiwan (Figure 1: http:// www.envsci.thu.edu.tw/teacher/wlcheng/ fig1_5.pdf). Its UHI phenomenon has already been reported (Lin et al., 1999). Dali, a UHI center of the Taichung metropolis, has a population of 220,000 and serves active commercial and manufacturing functions.
In this study, Wufong was treated as a reference station, because it represents an isolated rural area in the Taichung metropolis. The balloon-borne measurements were collected at Tsaoton, a rural town near Dali and Wufong that is away from airplane routes, to obtain the vertical profiles of the wind field.
Materials and Methods
Data regarding the daily hospital respiratory admissions in the 29 political districts of the Taichung metropolis were obtained from the Taiwan Bureau of National Health Insurance (TBNHI) for between September 1 and November 30, 2003, as well as for 2004 (hereafter referred to as the "examined period"). All afflicted groups in the data were classified according to the International Classification of Disease, Ninth Revision (ICD-9; World Health Organization, 1998). To properly reflect the admission data in the events of poor air quality in the Taichung metropolis, only the data for hospital respiratory admissions (ICD-9 codes 460-519) were used for analysis. The admission data for influenza (ICD-9 code 487) were excluded, because they were treated as part of a flu epidemic. Four age groups, namely, 0-4 years (preschoolers), 5-14 years (children), 15-64 years (adults), and over 65 years (elderly), were categorized to account for different responses to poor air quality during UHI episodes.
Hourly air pollution data (i.e., concentrations of ozone [[O.sub.3]], sulfur dioxide [S[O.sub.2]]...