Analysis of Java Island’s ozone layer and ultra violet index variability based on satellite data

. The ozone layer has a very important role in our atmosphere because it can protect every living on the surface of the earth against harmful Ultra Violet-B radiation. This study aims to discuss and analyze the linkages of ozone layer and Ultra Violet index (UVI) in Java Island and using of FFT to find the period that dominates the ozone layer and UVI variation. The results obtained are characteristic of ozone layer and UVI as well as linkage of UVI to ozone layer in Java Island. Using data of the Ozone Monitoring Instrument (OMI) sensor on AURA satellites from 2005 — 2016 has been obtained monthly, seasonal and annual characteristics for ozone and UVI and the period dominating the variation of the ozone layer and UVI. The ozone layer varied between 238 DU to 277 DU, the annual variation pattern peaked in October and the minimum in January. The UVI varies between 7.8 and 13.6. The annual variation of UVI peaked in October and the minimum in June. Linear regression of the UVI with ozone in December, January and February (DJF) showed a negative correlation coefficient of 0.77 which means there is a strong correlation between decreasing of ozone concentration with increasing of UVI. Variability of Java Island’s ozone layer is dominated by six month, 12-months and 28-months cycles. While UVI most dominated by the cycle of 6 months and 12 months.


Introduction
Life system on earth can be protected from Ultra Violet radiation due to the absorption of UV radiation by ozone, especially ozone in the stratosphere layer. Most of the ozone in the atmosphere is in the stratospheric layer (90 %) therefore called the ozone layer. The remainder (10 %) is in the troposphere layer and other layers. The ozone layer is the concentrated ozone in the stratosphere layer at 10-50 km above the earth's surface. The thickness of this layer illustrates the amount of ozone concentration in the coating. Total ozone is a vertical density (number of ozone molecules of broad unity) at 0 °C and 1 atmospheric pressure. The unit used is DU (Dobson Unit). 1 DU = 2.69 × 10 16 molecules/cm 2 . Total ozone is used to represent the ozone layer condition. In this paper the ozone layer data also uses the Dobson Unit (DU) unit. The solar radiation that reaches the Earth's surface is significantly affected by ozone changes, high UV radiation can affect human health [1]. Therefore, studies of temporal variability of the ozone layer and UV radiation become a priority for scientific research [17]. The UV index (UVI) is designed to represent UV weighted erythema radiation in a simple form. So, we can get an overview of the UVI [2]. Long-term observation and monitoring results show that ultraviolet radiation reaching the Earth's surface is increasing in response to the depletion of the ozone layer. The results show that: the amount of UV-B radiation that reaches the surface of the earth at a particular location is highly dependent on the amount/condition of the ozone layer in the area. When this ozone-depleting condition occurs in the stratosphere or in the troposphere, the total amount of ozone is reduced and the amount of ultra violet radiation reaching the earth's surface will increase proportionally. The relationship between total ozone and UV-B radiation has been demonstrated in many locations by taking direct measurements of ozone and UV-B radiation. Another thing that causes UV-B radiation changes to the surface of the earth depends also on several factors such as sun position that changes daily and seasonally cycles, local cloud conditions, altitude location, the amount of ice or snow cover and the amount of atmospheric particles (aerosols) in the atmosphere above that location. Changes in cloud and aerosol conditions are also related to the condition of air pollution and GHG emissions from human activities [3,4]. The global distribution of UV radiation and its seasonal change compared to long-term changes in atmospheric composition obtained significant correlation results between time series of ozone and UV indices i.e. decreasing ozone resulting in an increase in UVI in New Zealand. The UVI is defined as a unit with no linearly related unit to erythema dose rate, i.e. the size of human skin relevant to the intensity of UV radiation on the Earth's surface. 1 UVI is equal to 25 mW/m 2 [5]. Longterm observations and monitoring indicate that ultraviolet radiation reaching the Earth's surface will increase in response to the depletion of the ozone layer. UV radiation is also stronger in summer and weak in winter. Clouds, particulates, aerosols and contaminants can absorb and disperse some UV radiation, thereby reducing the amount of UV radiation reaching the Earth's surface [4]. Specific studies show that clouds, ozone and atmospheric aerosols are atmospheric compounds that mainly attenuate the UV that reaches the surface of the earth [6]. The study of ozone and UV indices in Java aims to investigate and analyze changes in the UVI due to variations in the ozone layer. This research is part of a series of researches that support and realize the development of atmospheric compositions using satellite data [7] which is very useful for databases and to support the study of atmospheric characteristics in several regions of Indonesia.

Research methods
The data used in this study is the ozone layer data (DU) and UVI in Java Island area (5.5° S-8.75° S and 105° E-115° E) as an observation result of Ozone Monitoring Instrument (OMI) from the AURA satellite obtained from NASA's data provider website. The data obtained is a monthly global average data with 1° latitude × 1° longitude grid. The data period used in this study is from 2005 to 2016 [1,8]. The research methodology is to extract ozone data and UVI for Java Island region from global OMI-AURA data in Net CDF format for period of 2005 to 2016. Convert data from Net CDF format for ozone data and UVI using HDF view and EXCEL. Then the data is processed with time series analysis to obtain seasonal and annual variations. Statistical analysis is also performed to obtain UVI and ozone correlation that will explain the effect of UVI on ozone variation. Fast Fourier Transform (FFT) analysis is also used to find the dominant period affecting the ozone layer and UVI variation in Java Island [9].

Results and discussions
Result of the ozone layer and UVI temporal variation from 2005-2016 in Java Island is presented in Figure 1.  To illustrate the differences in the characteristics of ozone and the UVI on the island of Java are made in Figure 2 which shows the deviation of temporal variation to the mean value of 2005-2016.    The deviation of ozone seasonal variation and UVI is presented in Figure 5. The deviation value of seasonal variation gives information on ozone seasonal variation and UVI which is higher or lower than the 2005-2016 seasonal average. Relationship between the seasonal variations of ozone layer with UVI in Java Island based on the scatter plot is presented in Figure 6. Scatter plots between UVI and the ozone layer result linear regression equations. The equations resulting from the plot of the ozone layer concentration and the UV indices can be seen in Table 1. where y is the UVI and x is the ozone layer.
The coefficient of determination obtained is 0.601 and the correlation coefficient between the UVI and the ozone layer is 0.775. From the obtained equation, it means that 60.10 % of UVI variable can be explained by ozone layer variable at the equation, while the rest cannot be explained by ozone layer variable because it depends on other variable. For other seasons of MAM, JJA and SON correlation coefficient are 0.77, 0.45, 0.38 and 0.25. Statistical analysis of the correlation between seasonal variation of UVI and ozone layer in Java give the result of negative correlation. It is mean that the increase of UVI is due to the decreasing (depletion) of ozone layer or, in other word, the thicker the ozone layer the lower the UVI. The highest correlation occurred in the DJF season with a correlation coefficient of 0.77.
Result of the Fast Fourier Transform (FFT) analysis to find the dominant period affecting the ozone layer and UVI variation in Java Island is shown in Figure 7.   (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016). It was found that ozone layer data are sensitive to the 6 months, 12 months and 28 months variability. In Figure 7 (b), the FFT result of UVI data are sensitive to the 6 month, 12 months and the 28 months variability show lower. Comparison of the FFT results for ozone and UVI shows there is a similarity in the dominant period of 12 months. These two variables also have a 6-month period of dominance. If only the ozone variable to be considered, the period of 28 months (or almost 2.5 years) also shows that there is a dominant occurrence with this period. However, this is not indicated in the UVI variable.
Analysis of the ozone layer in Java Island in 2005 to 2016 obtained the variation from 238 DU to 277 DU. These values correspond to the ozone layer in the equator region that is varying from 240 DU to 270 DU. The ozone at the equator is lower than in the mid and high latitudes. The correlation value between ozone layer and Java Island UV index in each season has different value. According to [4,10], it is indicated that the sun's position and season factor over an area affects the value of the ozone layer and the UV index. Aerosol and air pollutant factors can also absorb and disperse some UV radiation and thus reduce the amount of UV radiation that can reach the surface of the earth. This is influenced also by the local and season factors. Since the level of UV radiation that reaches us each day is impacted by a number of factors, this can be the reason why the UV level fluctuates from day to day, or why you get sunburnt when you think that you are safe. The total annual variation pattern of ozone and the UVI in Java Island obtained has a different pattern than that obtained in [7,5,10].
The UVI in Java Island is between 8 to 13.6. This value is already classified as very high to extreme [14] (Figure 4). Based on the observation results it can be informed that the long-term conditions of ozone and UVI in Java, the importance of accurate data on solar UV radiation is required for UV trends study. This data is also required for photochemical studies of atmospheric and human health protection against UV effects. To obtain UV data in various areas, UV monitoring networks need to be established in many countries [11].

Conclusion
Results of the ozone layer and UVI analysis in Java Island based on OMI-AURA data from 2005-2016 gave the variation of ozone layer between 238 DU to 277 DU. The values obtained correspond to the ozone layer values in the equator region which is varying from 240 DU to 270 DU. The UVI in Java Island is between 8 to 13.6, which is categorized as a very high to extreme UV index. It will be very risky for us when doing activities outside without using sun protection, since the higher the UVI the greater the rate of skin dosage and the eyes are damaged by UV irradiation.
The correlation results indicate that there is a strong correlation between the UVI and the ozone layer. The relationship between the UVI and the ozone layer shows a negative relationship. If the ozone layer decreases, the UV index will increase and vice versa. FFT result of the ozone and UVI shows that the 6 months, 12 months and 28-month period affecting both ozone and UVI variation even the power spectrum is different.
The author would like to thank the atmospheric composition data providers of the AURA satellite in particular the NASA MIRADOR website also to the Center for Atmospheric Science and Technology of LAPAN that has supported this research.