Articles | Volume 18, issue 9
https://doi.org/10.5194/acp-18-6543-2018
https://doi.org/10.5194/acp-18-6543-2018
Research article
 | 
08 May 2018
Research article |  | 08 May 2018

The use of hierarchical clustering for the design of optimized monitoring networks

Joana Soares, Paul Andrew Makar, Yayne Aklilu, and Ayodeji Akingunola

Data sets

Measuring the Influence of Individual Data Points in a Cluster Analysis R. Cheng and G. W. Milligan https://doi.org/10.1007/BF01246105

K-Means Clustering with Influence Detection R. Cheng and G. W. Milligan https://doi.org/10.1177/0013164496056005010

Mapping Influence Regions in Hierarchical Clustering R. Cheng and G. W. Milligan https://doi.org/10.1207/s15327906mbr3004_5

Government of Alberta Airdata warehouse http://airdata.alberta.ca/

Government of Canada Joint oil sands monitoring program emissions inventory report https://www.canada.ca/en/environment-climate-change/services/science-technology/publications/joint-oil-sands-monitoring-emissions-report.html

Government of Canada Emissions inventory files http://ec.gc.ca/data_donnees/SSB-OSM_Air/Air/Emissions_inventory_files/

Download
Short summary
Grouping data on the basis of (dis)similarity can be used to assess the efficacy of monitoring networks. The data are cross-compared in terms of temporal variation and magnitude of concentrations, and sites are ranked according to their level of potential redundancy. The methodology can be applied to measurement data, helping to identify sites with different measuring technologies or data flaws, and to model output, generating maps of areas of spatial representativeness of a monitoring site.
Altmetrics
Final-revised paper
Preprint