7Be spatial and temporal pattern in southwest of Europe (Spain): Evaluation of a predictive model. (August 2018)

• Record Type: Journal Article
• Authors: Chham, E.; Piñero-García, F.; Brattich, E.; El Bardouni, T.; Ferro-García, M.A.
• Is Part Of: Chemosphere. Volume 205(2018)
• Page Start: 194
• Page End: 202
• Publication Date: 2018-08
• Subjects: 7Be -- Radionuclides -- Multivariate statistical analysis -- Artificial neural networks -- Meteorological variables -- Sunspot number; Pollution -- Periodicals; Pollution -- Physiological effect -- Periodicals; Environmental sciences -- Periodicals; Atmospheric chemistry -- Periodicals; 551.511
• DOI: 10.1016/j.chemosphere.2018.04.099
• Languages: English
• ISSNs: 0045-6535

Abstract

Abstract: This study presents a comprehensive statistical analysis of the cosmogenic radioisotope 7 Be measured in surface air in ten stations over Spain for a 9 years period (from January 2006 to December 2014). Besides the analysis of 7 Be seasonal and inter-annual variability, 7 Be frequency distributions and its correlations with meteorological variables observed in the 10 sampling sites were investigated. The second part of this paper focuses on a feasibility study for the application of the Artificial Neural Networks (ANNs) to predict monthly 7 Be activity concentrations using meteorological variables, PM10 concentrations and the sunspot number as input parameters. Notwithstanding the low correlations found between 7 Be and input parameter, the performance of the ANNs, as evaluated by the relevant statistical parameters, demonstrates their capability to correctly predict 7 Be monthly activities in the 10 Spanish sampling sites. Highlights: 9 years of monthly measurements of 7 Be in 10 sampling sites were investigated. Temporal, spatial, and frequency distribution of the 7 Be activity were treated. The ANN method was proposed to forecast 7 Be activity. The ANN model provides robust forecasts of 7 Be air concentrations for all stations.