Introduction

View the most recent version.

"Access to good quality water in sufficient quantity is essential for ecosystems, human health and economic performance. In Canada, water is used mostly by households and industries for electrical power generation, agriculture, manufacturing, oil drilling and mining. Water quality degradation can affect both aquatic life and human uses of water." (Canadian Environmental Sustainability Indicators, 2005, page 15.)

For the first time in December 2005, Canada published a surface water quality index for the protection of aquatic life.¹ This indicator was the Canadian Council of Ministers of the Environment (CCME) Water Quality Index (WQI). It summarizes complex data into a single overall quality rating for a given monitoring site and over a specific period of time, facilitating presentation of these data to the general public.

The CCME WQI was based on the water quality index developed by British Columbia in 1995. The index has since been examined and modified through research, testing and consultations and approved by a CCME working group (CCME, 2001).

This is the first time that an index used for water quality assessment has been applied systematically on a national scale. Given the complexity of the index and the fact that the selection of components that enter its calculation can vary significantly from one station to another or one province to another, it is vital to study its behaviour to better understand it.

Therefore, the objective of this study is to understand and explain how the index behaves and to determine its limitations. Initially, we will focus on the CCME Water Quality Index in order to familiarize ourselves with its functioning. We will then study four sets of data provided to us by certain provinces: Newfoundland and Labrador, Ontario, British Columbia and Quebec.

For each set of data, we will begin with a descriptive analysis that includes: a) a presentation of the raw data; b) analysis of the excursion; c) analysis of correlation to determine whether a strong relation exists between the parameters under study; and d) a principal components analysis to determine whether all parameters figure appropriately in the index calculation.

Then, we will perform a sensitivity analysis. This analysis will show us: a) how the index behaves when we increase the number of parameters and samples; b) the contribution of each term and the correlation between the terms in the index calculation; c) whether use of three samples instead of four per year influences the index findings; d) index variability within stations; and e) index behaviour when we factor the season into the index calculation (when available).

After having analysed and presented the findings of the four datasets, we will present the results of a previous study using simulated data.
 
Moreover, it is very important to understand that the goal of this study is not to compare the findings obtained among the various datasets, but rather, to learn and understand how the index behaves through these datasets.

Note

1. Statistics Canada, Statistics Canada, 2005, Canadian Environmental Sustainability Indicators, 2005, Statistics Canada Catalogue no. 16-251-XIE, Ottawa.