| Abstrakt: |
The purpose of this research is to analyze the role of air pollutants (nitrate and nitrite oxide) in the changes of 24-hour temperature and precipitation elements in Tabriz synoptic station. The materials and data used in this research are from two different sources. The temperature and precipitation data were obtained from the tabriz synoptic meteorological station hourly for a period of 31 years and the data of Tabriz air pollutants (nitrate and nitrite oxide) were obtained from Tabriz environmental organization. In connection with the air pollutant data, it can be said that these data have been simulated by the multi-layer perceptron neural network machine learning R programming language. . In the logistic model, temperature and precipitation were selected as dependent variables and nitrate and nitrite oxide concentrations were selected as independent predictor variables. All data were included in the analysis and the logistic model was significant. The chi-square in nitrate and nitrite oxide was calculated as 348.01, which was significant at the error level of less than 0.05. The aforementioned independent variables have been able to correctly explain between 84 and 60 percent of the changes that led to the increase in temperature and decrease in precipitation. 78.2% of the months that had no changes were correctly classified, and 97.2% of the predictions about temperature and precipitation changes were correct. In total, 90.9% of the predictions have been estimated correctly. . The results showed that pollutants have a significant effect on temperature increase and precipitation decrease in Tabriz synoptic station. The highest and lowest levels of nitrate oxide were observed in September and March, nitrite in September and May, temperature in July and January, and precipitation in April and August, respectively. [ABSTRACT FROM AUTHOR] |
