In this study, it was aimed to characterize temporal variations of air pollutants for determining contribution to pollution episodes and to obtain correlations between these pollutants. With this aim we used data analysis for measured sulfur dioxide (SO(2)), particulate matter (PM, black fume and PM10), nitrogen oxides (NO(x)), ozone (O(3)), carbon monoxide (CO), methane (CH(4)), and non-methane hydrocarbons (NMHC) recorded in Kocaeli, one of the most industrilizated cities of Turkey. Pollutant concentrations were the results of continuous and semi-automatic measurements. Semi-automatic measurements of SO(2) and PM (black fume) were enclosing period from 1987 to 2008 whereas continuous monitoring of all pollutants included years of 2007-2009. In the first stage of the study daily, monthly, annual, and seasonal variations of pollution were researched. Annual average concentrations were compared with limits set by Air Quality Protection Regulation (AQPR), Air Quality Evaluation and Management Regulation (AQEMR), World Health Organization (WHO), European Union (EU), and National Ambient Air Quality Standards (USEPA). In the following stage relationships between pollutants such as NO(2)-O(3), NO(x)-CO, NO(x)-NMHC, and NO(x)-SO(2) were investigated and correlation coefficients were determined as 0.87, 0.56, 0.51, and 0.69, respectively. R(2) values of regression models developed from these correlations were 0.78, 0.56, 0.34, and 0.72, respectively. Vehicle density of the traffic was evaluated with NO(x)-O(3) emissions and decrease was seen in NO(x) emissions due to decreasing vehicle density at weekends whereas O(3) concentrations increased. These correlations enable prediction of the parameters that cannot be measured which is important for providing improvement in early warning systems.