The air pollution due to tropospheric ozone (O3) is one of the most serious problems of large industrialized cities in the world. The excessive increase in O3 has a negative impact on the population health. Consequently, researchers have focused their efforts establishing measures to characterize the statistical analysis of spatio-temporal data. This work shows a study based on seasonal analysis of spatio-temporal data through second order structure function to and the scale behavior in power law by using the Hurst exponent (H) and analyzing the trend of fluctuations associated with O3 pollution concentration records at four monitoring stations in the Metropolitan Area of Mexico City (MAMC) considering the four season of the year. The records were consulted from the database of the Automatic Atmospheric Monitoring Network (AAMN) in Mexico City from 2010 to 2018. The results show the differences in behavior of ozone according to the seasons of the year in this megacity. The behavior of statistical persistence predominates in spring, with 63.89% of occurrence over the total of the samples analyzed. In winter, the observed regime is antipersistent, with 80.56%. The three regimes: persistence, randomness and antipersistence were observed in summer and autumn, with a similar proportion of occurrence of 33% ± 3%. Given the above, the climatological characteristics of each season could be associated with the regimes of persistent, random and antipersistent behavior of the fluctuation of the concentration of the pollutant O3

En este trabajo se muestra la tendencia de escala de las fluctuaciones de registros de concentración contaminante del aire tomados de la estación de monitoreo Merced (MER) de la zona metropolitana del valle de México (ZMVM). La función de estructura de orden dos es usada como una medida para establecer la tendencia de los registros mediante el exponente de Hurst H. Se utilizaron cuatro series temporales de  por año durante el período de 2010 a 2018.  Se encontró que las fluctuaciones de  obedecen un comportamiento de ley de potencia, donde se obtiene perstistencia estadística para los tres primeros trimestres y antipersistencia para el último trimestre.