प्रदूषण प्रभाव और नियंत्रण जर्नल
खुला एक्सेस

अमूर्त

Functional Clustering of Particulate Matter less than 10 Microns: A Case of Africa

Musopole A

Particulate matter (PM) air pollution is a challenge that is endangering the environment and human health in Africa. Most countries in Africa are poor, hence monitoring of PM-related aspects is a challenge. Clustering of aspects of PM can ease the burden. In this paper, time series (TS) of PM of less than 10 microns (PM10) for 51 African countries are clustered. The data are represented in functional form and complete-linkage hierarchical clustering algorithm is used to cluster the coefficients of the functional data, thereby clustering the original TS. The functional form has the advantage of looking at the trajectories as a whole. 2 clusters are extracted from the data. 2014 Cross-sectional data of PM of less than 2.5 microns (PM2.5) for African countries are also clustered and 3 clusters are obtained. Adjusted Rand Index (ARI) computed for clusters resulting from the two data sets is 0.588, indicating some agreement on the clusters resulting from the two data sets.