Bulkley Valley Research Centre - Science in the Public Interest

Below the Method Detection Limit:  Problems, Definitions, Data, and Solutions

Project Reference Number: 2009-02

Project Status: Complete

Led by: Carolyn Huston, Statistics and Actuarial Science, Simon Fraser University

Funder: Workshop Fees

Data sets containing values below the method detection limit commonly arise in a variety of activities that require chemical analysis. This can include air quality data, water quality data, or even in evaluating the cells of plants or animals for the presence of contaminants.

Valuable information is contained in the values below the method detection limit, but despite this they are often deleted from an analysis or treated incorrectly through the use of substitution. The planned workshop wil focus on introducing both the statistical software, and statistical methodology, that is appropriate for analyzing this special type of data. This workshop will focus on introducing R as a powerful statistical computing tool for the first half of the course. The second half of the course will focus on summarizing and understanding data containing values below the method detection limit.

Performing statistics is rarely greeted with cheer and good feeling. Quite the opposite! Extra anxiety might also be present at the suggestion to use R, a new and potentially unfamiliar software program. Realize that far from having the intent to cause uncertainty, this course and statistics in general are meant to be a way of thinking that will assist in quantifying and addressing uncertainty to facilitate decision making.



On Feb. 24, the Bulkley Valley Research Centre hosted Below the Method Detection Limit: Problems, definitions, data, and solutions. This workshop, taught by Simon Fraser University statistician Carolyn Huston, better acquainted 22 participants to R software and statistics methodology for data analysis.

Taking place at Northwest Community College in Smithers, the workshop hosted 11 employees from the Ministry of Environment and 11 other local government employees and consultants. Attendees travelled from as far as Prince George to attend the workshop, which garnered positive feedback from participants.

“The workshop and accompanying guidance document provide Ministry of Environment staff with new tools that could revolutionize the way we analyze water quality data. We will explore and apply the techniques we learned in this course over the next year to better understand their potential” says Greg Tamblyn, an Impact Assessment Biologist with the Ministry of Environment.



Introduction and Why use R

What is R?


Information for Section 2:  hgtdata.csv

Creating a common lexicon

Odds and Ends

The Nitty Gritty!!

Information for Section 3:  GuidanceFunctions.R

Probability plots and review

Information for Section 4:  Savona.csv



Exercises 1


Exercises 1 Solutions

Exercises 2


Exercises 2 Solutions

Exercises 3

Exercises 3 Solutions

Exercises 4


Exercises 4 Solutions