نشریه نیوار
پیاپی 72-73 (بهار و تابستان 1390)

Ever-increasing in climatic observations, appearing advanced computers and changing of outlook towards natural phenomenon have caused that multivariate statistical methods to be accepted by climatic experts. The importance of the subject is clarified by utilizing classification in identifying climatic homogenous regions. In order to classify the precipitation's characteristics based on 260 stations over observation period of 1966 - 2005, cluster analysis technique has been used. Cluster analysis is a multivariate statistical method that classifies a group of cases according to difference or similarity scales. This method has been used much in climatology and indicates more exact classification in comparison with traditional methods. Precipitation annual mean in northwest of country is 360.9 mm that it has a range equal to 587.3 mm and it's spatial coefficient of variation is 24 percent. To achieve results of research, softwares SPSS / win, surfer / win and Matlab was used. By using cluster analysis in 5374 pixels, finally 4 precipitation regions were identified in northwest part of Iran. These regions with different dimension and different distribution contain:- Low concentrated precipitation region with moderate snowfall- Moderate precipitation region with tendency to low distributed precipitation and much snowfall- Moderate precipitation region with tendency to much distributed precipitation and moderate snowfall- Much concentrated precipitation region with moderate snowfall

Many natural resources studies and surveys are faced to problem due to lack of meteorological stations coverage. Estimation of the rainfall is needed for the region located between the stations. To solve the problem there are some methods such as interpolation methods, which has different accuracy. In this research, the accuracy of the interpolation methods was evaluated for estimating of the rainfall for preparation of Isohytal maps in Chaharmahal-Bakhtiari province, Iran. The cross validation technique (CVT) was employed. Statistical analysis of the model was done with the root mean square errors (RMSE) and correlation coefficient (R2). The results show that the minimum curvature method is the most accurate one for the interpolation. The correlation coefficient (R2) for the method was grater than 0.928. The RMSE value was 12 and 50 mm for monthly and annually, respectively. Finally, monthly and annual Isohytal maps for the province were generated.

In this paper, modeling of temperature of the Caspian southern coasts has been done by sARIMA or seasonal Autoregressive integrated moving average. Therefore monthly mean temperature related to Anzali, Ramsar and babolsar synoptic stations with long term dataset has been studied through (1955 – 2008). Statistical surveys of climate change usually analyze three phenomena as homogeneity, trend and jump in the climatic series. Hence in order to study the climate change process in the region, three phenomena; homogeneity, trend and discontinuity of the series were analyzed. Then temperature time series have prepared for performance of seasonal ARIMA model. In preparation of the time series to use the ARIMA model, at first the time series have transformed to normal and stationary series using differencing method. Then, after selection of some suitable models and estimation of parameters by maximum likelihood method, independence and normality of model residuals (t a) were considered. At the end, Akaike information criterion (AIC) was used to select one among the various alternative models. So, sARIMA (1,0,0) (0,1,1) 12 was selected for Anzali and Babolsar and sARIMA (0,0,2) (0,1,1) 12 was selected for Ramsar mean monthly temperature. Performance of seasonal ARIMA model for each time series has well suited, in comparison with the actual data in four years 2005 to 2008 as the gauge by sARIMA model. Correlation coefficient between the actual and fitted data was nearly 0.97 and the absolute and relative errors were very small.

In this research a non-linear model presents for modeling the tidal currents in shallow sea areas. This pattern is based on the solution of shallow water equations, effects of earth rotation, topographic changes and influences of the bottom friction. For facilitating the calculations a program was written in MATLAB and C++ computer language. Finally the tide current in Kangan port in Persian Gulf in different seasons was obtained by this program and a graph was provided. This research shows that the maximum tide current’s speed amount is in spring which its amount is 0.17 m/s and the minimum amount is in summer with the amount of 0.132 m/s. The maximum current is in the low depth sea side zones which decrease in other areas. Also the maximum amplitude changes in the tide current is in spring which shows extensive instability in the tide currencies in this season compared to other seasons. A yearly administration in Kangan port shows the average speed of tide currencies of 0.146 m/s and an average change 0.001 to 0.996 m/s.

Oil pollution, is the act of release a hydrocarbon liquid in the sea environment that comes from processes such as, spill through offshore oil fields, accident from tankers and piping. Persian Gulf for having the world's major oil and gas fields exposed to contaminants such as oil pollution and the pollutants those are associated with them. In this study the distribution of pollution in various conditions is simulated by a hydrodynamic 3-d model to determine the behavior of the oil pollutants spill into the sea and the effect of the physical processes such as wind, heat fluxes and wind stress on the distribution of these contaminants. Using numerical model in the case of oil pollution is known as a good way to provide reasonable consistent for the distribution of oil pollution. COHERENS model were used to simulate the oil pollution transport and diffusion, which coupled with contaminant and biological modules and has the ability of solving transport equation by use sigma coordinate in the vertical direction and Cartesian coordinate in horizontal. Model results show that oil spill in the northern Persian Gulf moves towards the northern Persian Gulf and Bushehr coast along the Iranian coast and then due to counterclockwise currents of that region moves toward southwest of gulf along the Arabian coast under all forces. The model results also indicate change in how the distribution of pollutant and major decrease of pollutant concentration under the effect of wind.

One of the most important features in analyzing the climatology of any region is to study the precipitation as the most variable climatic element. The main objective of this study is spatial and temporal analysis of inter-annual variability of frequency of month-to-month precipitation over Iran using harmonic analysis. To do so, long monthly precipitation time series data from 33well-scattered synoptic stations are studied with basic harmonic analysis in two statistical periods (1980-1992, 1993- 2005) in order to identify the effects of climate change and climate variability on precipitation. First, relative frequency of precipitation for each month is calculated and finally characteristics of monthly precipitation over the whole country are analyzed in form of six harmonics. The harmonic analysis showed that the first three harmonics are the most important ones for explanation of inter-annualvariability of precipitation over the country. As the first harmonic represents a single annual cycle with the largest amplitude in comparison to the other harmonics, so that this harmonic revealed more than 70 percent of monthly precipitation variability in most of regions except Northwest and Southeast of the country in both studied statistical periods. Also, it is observed that the rate of spatial and temporal changes of first Harmonic in both statistical periods is less than second and third harmonic. This represents higher significant of first harmonic and it’s less variability in explanation of precipitation changes over the country. In contrast, comparing the results of two statistical periods indicate spatial and temporal displacement in second and third harmonics caused by the effects of climate change and climate variability which represents more variability in these two harmonics in comparison with first harmonic.

Ozone is one of the main pollutants in the agricultural sector and each year many of agricultural products have damaged due to exposure of this gas in terms of qualitative and quantitative, however the formation of ozone as a secondary pollutant is affected by ambient atmospheric conditions. Although due to insufficient instrument for measuring ozone in agricultural and rural places in Iran, presenting an equation for predicting ozone is necessary. In this study, various statistical models based on the values of Meteorological variables and atmospheric concentrations of pollutants (as input variables) to predict changes in the concentration of Troposphere ozone in scale of time and an hour a day is presented. These researches shows that the equations contains two parameters, main square of sun radiation and main square of carbon monoxide concentration could expresses the 35% of ozone fluctuations during a day. How ever, it seems that the regression models and complex Orthogonal functions have high efficiency to expressing ozone fluctuations and predicting the concentrations of it, but many input parameters of them cause that these models are not useful for applied fields.