Time series clustering for estimating particulate matter contributions and its use in quantifying impacts from deserts

Abstract

Source apportionment studies use prior exploratory methods that are not purpose-oriented and receptor modelling is based on chemical speciation, requiring costly, time-consuming analyses. Hidden Markov Models (HMMs) are proposed as a routine, exploratory tool to estimate PM10 source contributions. These models were used on annual time series (TS) data from 33 background sites in Spain and Portugal. HMMs enable the creation of groups of PM10 TS observations with similar concentration values, defining the pollutant's regimes of concentration. The results include estimations of source contributions from these regimes, the probability of change among them and their contribution to annual average PM10 concentrations. The annual average Saharan PM10 contribution in the Canary Islands was estimated and compared to other studies. A new procedure for quantifying the wind-blown desert contributions to daily average PM10 concentrations from monitoring sites is proposed. This new procedure seems to correct the net load estimation from deserts achieved with the most frequently used method.

Effects of Strong East Asian Cold Surges on Improving the Air Quality over Mainland China

“…As a special type of clustering, time series clustering has been used in diverse scientific fields to determine patterns and extract valuable information from complex and large datasets [29,30]. We applied the method to analyze API datasets related to the number of days preceding or following cold surge occurrences in 47 major cities throughout China to determine the days with the lowest API values after cold surge occurrences for these different cities.…”

Estimation of Particulate Matter Contributions from Desert Outbreaks in Mediterranean Countries (2015–2018) Using the Time Series Clustering Method

“…It represents a model-based strategy for clustering by assuming that each cluster of data is described by a different probability distribution (components of the mixture). This methodology is explained in Gómez-Losada et al [14,20] and illustrated graphically in Figure 1. A complete definition and elements of HMM are provided in SM1.…”

“…Once the pollution profiles in the TS data are determined, a meaning for each profile must be then assigned to estimate the PM 10 and PM 2.5 contributions from desert outbreaks. The following definitions are provided for the average values of each profile [14], applied to the remote stations studied in this work and considering as an example, the PM 10 TS from Figure 1: Some useful quantities can be then roughly estimated using these definitions, namely: It must be noted that the above definitions coincide from a conceptual point of view with those of horizontal profiles given by Lenschow et al [27] when studying the PM 10 regimes of concentrations in the Berlin area. These definitions have to be adapted then to the geographical location under study and adapted to each monitoring network.…”

“…There are several statistical techniques that were applied to quantify the contribution of this natural source. Gómez-Losada et al [14] applied the Hidden Markov Model (HMM) to estimate PM 10 contributions using the data of 33 background sites in Spain and Portugal. The model results were then compared with the method widely accepted to estimate the daily African PM 10 load: the 40th percentile (P40) method introduced by Escudero et al [15].…”