بررسی تغییرات روزهای بارشی تحت تأثیر خرد اقلیم شهری در کلانشهر تهران

Introduction
Many researches done during the last twenty years has been detected a strong relationship between urban microclimate and precipitation. Urbanization is increasing in parallel with the increase in population. As a result of urbanization and population increase, the energy consumption has also been increasing due to heating, traffic and industrialization. These cause remarkable changes in the meteorological conditions, and climate of cities and neighboring areas. The most salient impact of urbanization on the climate can be observed on temperature. It is apparent that in all cities located in the tropics and in the Polar Regions, the temperature has increased. However, the impact of urbanization on precipitation varies with time and area. In various investigations, it was found that both, the number of precipitation days and intensity of precipitation, have changed due to urbanization especially in mid-latitude countries. Another important effect of the urbanization is the variation of the precipitation across the days of a week. In different studies, it was observed that the frequency of occurrence of precipitation in inter-week days is more than those of weekends. In this study, the relations between urbanization and precipitation in the warm season were investigated for Tehran.

Methodology
In order to study and determine the influences of Urbanization on precipitation process in Tehran, first of all long term daily precipitation (1966-2005) from 2 stations including Mehrabad (under urban properties) and Bilaghan (with rural properties) were compiled. Because the effect of urbanization is more effective on warm-season’s precipitation, so in this study daily precipitation data related to June until September for a given stations during 1966-2005 was investigated. Trend in data series were studied by using 2 methods: 1) the non-parametric Kruskal-Wallis H test, 2) the tau statistic of Kendall test. The non-parametrical Kruskal-Wallis H test under the null hypothesis "there is no change between precipitation values and the number of precipitation days over the years" was used in order to determine whether there are differences in certain periods in the data series. To determine the trends of the precipitation data, the method of Tau statistic of Kendall test was applied to the number of precipitation days. In this test, null hypothesis imply that data are coincidental, and trend doesn't exist, and exception of one hypothesis (rejection of null hypothesis) imply that trend exist in data series. Finally in order to determine the type and time of changes in data series, statistical graphical method of Mann Kendal was used. Sequential values u (t) and u' (t) from the progressive analysis of the Mann-Kendal test were determined in order to see change of trend with time. This test was found to be useful and widely used for detecting trends in climate and environmental sciences.

Results and Discussion
The evaluation of statistical test’s during the whole statistical period (1966-2005)showed that the assumptions of the data being accidental in 95% significance level in Mehrabad station is significantly rejected and also increasing trend of the number of precipitation days, especially in the recent years are significant. While in Bilaghan station non-significant trend was found in the 90% 95% confidence level.
Another important effect of the urbanization is the variation of the precipitation across the days of a week. In studying of variability of rainy days in all days of week, two statistical sub-periods in June until September were selected from the first and last 15-years statistical period, for both stations, Mehrabad and Bilaghan.
The study of the first 15 year data in both stations, Mehrabad and Bilaghan showed that to investigate the distribution of rainy days during a week doesn't obey any main pattern. In the study of the last 15 years, Bilaghan station had no pattern as the first 15 years, but in that period at Mehrabad station, there were seen the number of rainy days in the workdays of week increase and in the last days of week decrease.

Conclusion
The results presented in this paper are preliminary study on the impacts of urbanization on rain fall in Tehran, Iran. It is clear that rainfall has changed significantly over the study period in the Tehran metropolitan area. The records indicate that the number of precipitation days and light precipitation days in Tehran has a gradual increase around 1975s, which is considered as the start of an accelerating city expansion, hence increase of air pollution. These considerable changes in the trends of the precipitation day classes in station of interest are not just due to climate change, but it seems that these tendencies are related to influences of urban microclimate on precipitation process.