Surface water quality data by principal component analysis
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
This study used multivariate techniques for data analysis in order to determine the natural and anthropogenic factors that contribute to the spatial and temporal variations of water quality in urban watersheds of Caxias do Sul, Brazil. Principal Component Analysis (PCA) was used to analyze data collected at 30 points between September 2012 and January 2014. Monitoring was conducted bimonthly in six urban basins, where a total of 21 physical, chemical and biological parameters were analyzed. We found that PCA can explain 71.3% of the total variance in water quality, and that domestic and industrial pollution are the main contributors to the water quality variation in the region, especially from the galvanic manufacturing sector. Furthermore, we observed a trend of self-attenuation of pollutants in water downstream from urban areas and great anthropogenic influence as the pressure from urbanized areas decreases.