Functional exploratory data analysis for high‐resolution measurements of urban particulate matter

摘要: In this work we propose the use of functional data analysis (FDA) to deal with a very large dataset atmospheric aerosol size distribution resolved in both space and time. Data come from mobile measurement platform town Perugia (Central Italy). An OPC (Optical Particle Counter) is integrated on cabin Minimetro, an urban transportation system, that moves along monorail line transect town. The takes sample air every six seconds counts number particles aerosols diameter between 0.28 μm 10 classifies such into 21 bins according their diameter. Here, adopt 2D representation for each spatiotemporal series. fact, unidimensional since it measured as distance base station Minimetro. FDA allows reduction dimensionality accounts high space-time resolution data. Functional cluster then performed search similarities among channels terms pattern. Results provide good classification relatively small groups (between three four) season year. Groups including coarser have more similar patterns, while those finer show different behavior period Such features are consistent physics highlighted patterns useful ground prospective model-based studies.