The Trends of Biologically Active Ultraviolet Radiation Exposure in Taipei, Taiwan

The decreasing trend of total ozone in Taipei is presened, analyzed and discussed. Seasonal trends of the total column ozone show that winter has the sharpest decreasing trend in the past. Summer shows the slightly increasing trend from the TOMS data. The result from trend analysis of total ozone levels provides the information for the estimation of UV-B reached the surface in Taipei. The UV-B model predicts that the future increasing exposure of UV-B due to the decreasing ozone level varies dependent upon the legislation enacted in Taiwan.


INTRODUCTION
Ozone depletion is one of the major concerns in earth history. Many researchers have documented the discovery of a decreasing trend of ozone concentrations in Antarctic regions (Michaels and Stookskury, 1992;Hamill and Toon, 199 1). Ozone, serving as the shield for the Earth from the overreaching of UV radiation, has declined as much as 8% over the past decade in between 40° and 50° (WMO, 1992). Several researchers have noted the negative trend of total ozone measurements based on monthly data (Niu et al., 1992). At low latitude, the changes in total ozone levels were found to be near zero but and negative at middle and high latitudes in both hemispheres (Stolarski et al., 1991). Researchers also analyzed the seasonal trend of total ozone. It was found that in the summer time there was about a 2% dec�ease per decade over middle latitudes, but an even greater decrease in winter.
The decrease in total ozone causes the intrusion of more UV light, which leads to several potential effects such as cataracts , skin cancer, and damage to immune systems as well as to the DNA structure (Elmer-Desitt, 1992). Abnor1nal UV radiation would also alter the routine mechanism of marine life (Carmichael, 1992).
Since November 1978, total ozone has been measured on a nearly global basis by the ·Total Ozone Mapping Spectrometer (TOMS) from the Nimbus 7 satellite. The TOMS measures the earth's ultraviolet albedo at several wavelengths near 300 nm. The main problem with maintaining a long-term ozone record with the TOMS has been the degradation of the diffuser plate used to make the solar observation for the deter1nination of albedo. Ozone data from the TOMS before 1983 included a correction for diffuser plate darkening (Fleig et al., _ 1986). Recent measurements of ozone data, however, have been corrected through an improved method. Thus, a precision of 1.3 percent is the final estimation of ozone data relative to the beginning of the record.
Satellite measurements of the vertical profile of stratospheric ozone have been made by the Stratospheric Aerosol and Gas Experiment (SAGE). SAGE I data began in February 1979 and extended through November 1981. SAGE II was the instrument of experiment on the Earth Radiation Budget Satellite (ERBS). and it has been continuously operated since October 1984. The technique employed in SAGE is solar occultation, and profile measurements are obtained at sunrise and sunset on each orbit. This measurement technique is insensitive to drift in instrumental calibration, but there may be systematic differences between the data from SAGE I and SAGE II. Since the data used in this study were derived from SAGE II only, such differences do not pose a problem in final representat�on of the data. Data from the SAGE II contain the vertical profile of ozone, temperature, and pressure data plus other measurements such as N02, aerosols and water vapor. Spatial coverage of the SAGE II ranges from 80°N to 80°S (for 5 km above or cloud tops). Resolution is 250 km by 250 km horizontally and 1 km below 25 km and 5 km above 25 km vertically. The SAGE II provides an ozone concentration profile of the stratosphere and troposphere (above 5 km), as derived from solar occultation radiometric measurements.

The UV-B Calculation Model
The prediction of UV irradiances and photolysis prototype was based on the model developed by Bruhl and Crutzen (1989). However, a modified model was performed to obtain the estimates of UV-B exposures for Taipei. This section outlines the fundamental theories behind this computer model.
In order to compute the photolysis rates, the requ.i red actinic fluxes are derived from the irradiances (Madronich, 1987): is the radiance from zenith angle() ' and azimuth8 ' .
The above outline of the configuration of computer models is used for calculating the UV-B and other related parameters.

Trend Analysis
Niu et al. have researched the trend of total ozone measurements based on TOMS data (Niu et al., 1992). In this study, the variation of ozone levels in different categories such as month, latitude and longitude were discussed and presented. A twenty-nine percent decrease in the total ozone per decade was found in the region of 70°S and 20°W-100°W. A maximum decrease in total ozone' was found to be seven percent in the northern hemisphere. They also proposed a specific time series model to estimate the seasonal variation of total � ozone levels.
464 TAO, Vol. 6, No. 3, September 1995 One of the major purposes of the present study is to find the ozone trend in one specific location, Taipei, and to predict the UV-B that reaches the surface in this area. Taiwan is an island located at 24°N and 12 1°E of the Pacific Rim (Figure 1). With a population of 20 million people living in a 36,000 square-kilometer area, this country has gone through a rapid industrialization process over the past ten to twenty years. Recent studies have shown the surface ozone problem in Taipei with higher surface ozone levels during the summer season (Liu et al., 1990). To understand the existing trend, the authors start from the global visual image of total ozone, then focus on ozone trends in Taipei and fi nally predict solar radiation change in the future.  globally. This presentation also shows the time serial decreasing trend during the past decade. Figure 3 shows the thirteen-year daily data of total ozone over Taipei. from the TOMS. A total of 4,834 daily records is shown in this figure. The all time low total ozone level is found in the winter of 1984/85, but this is not found in Figure 2, indicating that the detailed analysis is necessary to obtain the infor1nation of ozone trend in a specifi c location.

Statistical Analysis of Total Ozone Levels
A total of thirteen-years of data is� a,vailable from the TOMS as mentioned above. Starting from daily data, the total ozone level trend shows a slightly decreasing trend over the past decade. Figure 3 presents all the data points collected from the TOMS. A 5.63 Julian Day Fig. 3. Daily measurements of total ozone over Taipei, 1978Taipei, -1992 Dobson unit decrease in total ozone was noticed during this period, a decrease equal to a 1.99% decrease over the 13.25-year span. Thus, a 1.5% decrease per decade was shown in this figure. Figure 3 presents the daily variation of total ozone measurements over the 13-year period. Overall, the decreasing trend was determined to be 1.5% in the past decade. This type of data isn't useful in analysis; , thus, further integration of data is needed and provided here. A monthly pattern of ozone variation is shown in Figure 4. The comparatively low ozone level in winter as shown in Figure 4 suggests that the study of the integrated data of seasonal change would be more useful. The integrated data in Figure 5 presents the seasonal variation over the past decades .for ozone measurements. Winter is defi ned as the season from December through February, Spring as the season from March through May, Summer as the season from June through August and Fall as the season from September through November. Again, the specific pattern of low winter ozone is seen evident. As indicated    Taipei from 1979Taipei from -1992 in Figure 5, each season shows the particular trend that can not be discussed in terms of yearly trends. Figures 6, 7, 8, and 9 display the variation in ozone measurements over the past decade for four seasons. Winter has the largest decreasing trend as shown in Figure 6. A total 4.7% decrease was found in this study, which is equal to 12.5 Dobson Units (D.U.) of ozone depletion. In 1985, the total ozone level was observed at an average of 242 D.U. whi�h is the record low of the past decade. Recently, from 1988 up to 1992, a slightly decreasing trend was measured, but still, the overall ozone level during this period has been decreasing accordingly. A pattern of random walk was recognized in observing the ozone level of Spring. This season presented a slightly decreasing trend in terms of the overall level (Table 1). Summer has had a very steady ozone level over the past decade as shown in Figure 8. A low ozone level span was fouf\d from 1983 through 1986. From 1983 to 1988, ozone was found at lower levels in Fall. In general, Winter and Spring had larger decreasing trend compared to those of Summer and Fall.   Li et al.       . . .    Year Seasonal ozone variations for Spring.
: : 0 . : : : .0 0 : : : :  . .  Year Fig. 9. Seasonal ozone variations for Fall.  An increase in tropospheric ozone due to the greenhouse effect in the industrialized northern hemisphere can overcompensate for increased UV-B radiation resulting from ozone depletion due to chlorine-catalyzed reactions in the stratosphere. Such a phenomenon is common, especially in urban areas. Taipei is one such city where a decreasing trend in total ozone and an increasing ozone concentration at the troposphere are shown.
Due to the positive effects of decreased ozone in total ozone levels and negative effects from the increase in surface ozone, the research of UV-B radiation is interesting and complex.
Bruhl and Crutzen 11 showed the counter-effectiveness of these two phenomena by using estimated and observed data. The results from their study showed that, based on data from 1966 to 1986, a 0.5% decrease in the UV-B was observed . Other researchers also presented the seasonal UV-B flux variation and radiation models (Chou, 1992;Prasad et al., 1992). Due to the lack of information on cloud amount and sunshine duration, the factors influencing the UV-B flux reaching the surface of Taipei, this paper only discusses the change of UV-B fl ux due to ozone concentrations.
A 1.5 percent decrease in total ozone per decade is confi1·med for the urban area of Taipei (Figure 3). Data from the SAGE II and simulated models provide the ozone profi les for the analysis of past trends, from which the authors selected the years 1978/79 and 1988/89 for the starting and ending points of the past decade. Figure 10 shows the estimated seasonal changes of the UV-B in the past decade based on the total ozone data from the TOMS. In order to simulate future scenarios, the models used parameters modified from urban areas in the Eastern United States. The estimated UV-B (280 to 320 nm) flux based on the ozone trend in Taipei shows the decreasing trend over the last decade. The high percentage change shown in the summer is due to the assumption that there was much higher surface ozone in 1989 than in 1979. A significant decrease in the UV-B (10%) was found in sumrri.er ( Figure  11 ). Winter had the largest decrease in the total ozone compared to other seasons (Figure 11 ), but it did not show the influence of ozone depletion. This result matches the observations reported by Scotto, et al. (1988). An average 7.75% decrease in UV-B flux reaching the surface of Taipei is shown in Figures 10 and 11. There are two scenarios studied in this case. The polluted scenario assumes that the current pollution trend extends into the future, but, on the other hand if clean air legeslation is enacted in Taiwan, an unpolluted scenario make for the future prediction. The two scenarios were defi ned to compare the polluted and unpolluted scenarios in the Spring of 2010. For the polluted scenario, a 10% increase in tropospheric ozone and 1.5% decrease in total ozone per decade was assumed and, for the unpolluted scenario, an unchanged tropospheric ozone level was assumed. Final simulation showed that a fourteen percent increase in the UV-B was expected in the unpolluted scenario compared to a near zero percent change on the UV-B due to the polluted source and ozone depletion ( Figure 12).
In this research, major findings regarding the ozone trends on the Taipei area are pre sented, discussed and analyzed statistically. Satellite measured data were analyzed and a UV-B prediction model was used. The regional trends of ozone levels in Taipei are presented and discussed, and the results from ultraviolet radiation models are discussed. The major conclusions are outlined as follows: • For the 20° to 30°N latitude band area, total ozone is plotted and discussed. A de creasing trend was observed while outliers show the extreme cases of ozone levels.
• Total ozone level data obtained from the TOMS show a decreasing trend in the Taipei area.
• Infonnation integration processes used in analyzing data present the differences in ozone in the past decade.
• Monthly percentage changes per decade indicate the seasonal trends in ozone. More than a 4% decrease per decade in ozone is observed in winter, while summer shows a very slightly increasing trend in the data.
• Daily, monthly, yearly, and seasonal variation of total ozone measurements are pre sented, compared and analyzed.   . .