Project: MLMMI 00-02-11
Objective
To compare the odour concentration measured by the Olfactometer to the odour intensity assessed by certified Nasal Rangers™.
Performer
Q. Zhang, University of Manitoba
J Feddes, U of Alberta
Details
Status: Completed
Started: 2001-02-01
Completed: 2002-02-20
Funding Partners: who have contributed to MLMMI in support of this project:
ARDI - $25,100
Manitoba Pork Council and industry groups - $25,100
Amount Funded: $50,200.00
Performer Funded: $0.00
Total Cost: $50,200.00
Activity
First Interim Report received May 16/01.
Second progress report received Aug 29/01
Third Progress Report received Nov. 29/01
Final report due February 1, 2002; received Feb. 20, 2002.
Summary
Measurement of odour emissions from swine operations is a difficult task. Olfactometers are currently the most accepted method for odour measurement. However, obtaining downwind odour samples that are representative to what is actually “felt” by those in the field is almost impractical. The goal of this study was to evaluate a potentially more satisfactory method of evaluating odour directly in the field. This method was developed by St. Croix Sensory Inc. (Stillwater, MN) to use trained human odour assessors (Nasal Rangers) to quantify odour intensity according to n-butanol reference scales. The specific objective of this study was to establish a relationship between odour intensity assessed by the Nasal Ranger technique and odour concentration measured with olfactometers.
Four swine production sites were selected for this study, two located in Southern Manitoba and two in Central Alberta. A total of 154 samples were collected in 10 L Tedlar bags from the four sites between June and October, 2001. While these samples were being collected, odour intensity of the ambient air was assessed (the field odour intensity) by two or more Nasal Rangers using the 8-point n-butanol reference scale. Odour intensity of bagged samples was also measured in the laboratory by the Nasal Rangers. Odour concentrations of bagged samples were determined by using a single-port olfactometer at the University of Manitoba and an 8-port olfactometer at the University of Alberta. Thirteen strong odour samples collected from barn exhaust fans were diluted to 88 sub samples and odour intensity of these sub samples was then measured to determine the rate of decrease in odour intensity with the dilution level. This rate of decrease indicates the odour persistence, which is an important odour parameter in determining the “hang” time of odour in the air.
The field odour intensity measured by the Nasal Rangers did not correlate well with the odour concentration of bagged samples measured with olfactometers. The overall coefficient of determination (R2) was only 0.049. When a high odour intensity was detected by the Nasal Rangers in the field, the odour concentration measured from the bagged samples changed little. In other words, the composite odour samples taken in Tedlar bags could not capture any instantaneous bursts of strong odour in the field that were detected by the Nasal Rangers.
Odour intensity of bagged samples correlated well with the odour concentration measured with olfactometers, with an overall coefficient of determination of 0.61. The relationship between odour concentration measured with olfactometers and odour intensity assessed by Nasal Rangers using the 8-point n-butanol intensity scale could be adequately predicted by two commonly used models: the Weber-Fechner model and the Stevens model. These models would allow the odour intensity assessed by Nasal Rangers to be compared with “standard” olfactometer measurements and provide a tool for researchers and regulatory agencies to relate odour concentrations predicted by dispersion models to odour intensity levels, and ultimately to odour annoyance levels.
When a strong source odour was diluted, its intensity decreased linearly with the dilution level (log scale). For 13 odour samples that were diluted between 8 to1088 times, the average decreasing rate in odour intensity was 1.65 on the 8-point scale for every 10-fold dilution.