Exploratory data analysis in the study of 7 Be present in atmospheric aerosols

Purpose Exploratory data analysis (EDA) is applied in this research to study the behavior of radioactive aerosols present in the surface atmosphere of Granada, using 7 Be as radiotracer. The reason for this study is to reduce the large number of parameters involved in understanding their behavior, given the complexity of the atmosphere. Methods Aerosol particles were collected weekly in Granada (Spain) over a 5-year period. Low-background gamma spectrometry was used to determine concentrations of 7 Be-aerosol activity. The variables studied were: 7 Be concentration, cosmic ray intensity, temperature, temperature interval, rainfall, relative humidity, and Saharan intrusions. Least significant difference test (LSD), hierarchical cluster analysis (HCA), and principal component analysis (PCA) with varimax rotation have been applied to study the datasets. Results and discussion The results of our study reveal that aerosol behavior is represented by two principal components which explain 86.23 % of total variance. Components PC1 and PC2 respectively explain 74.61 and 11.62 % of total variance. PC1 explains the cyclical and seasonal pattern of the samples, while PC2 is related to the production of 7 Be. In addition, PCA and HCA show good distribution of the samples by families with two groups, summer and winter, at the extremes and spring–autumn in the middle. This result corroborates that there are no differences between spring and autumn in the climate of Granada. Conclusions EDA has been found to be quite useful in studying the behavior of radioactive aerosols in the surface atmosphere of a city with the climate and geographical characteristics of Granada.