In the second year of the project at the University of Minnesota, limited research data suggests that air dispersion models are an accurate means of tracking odor emissions from livestock and poultry operations.
A team of agricultural engineers at the school compared odor emission data compiled by a computer model with a group of seven "nasal rangers" who subjectively collected data in the field.
Comparing data on odor one-on-one between the model and the human sniffers will reveal some differences. But overall, the preliminary results show the procedure used was rational and useful in predicting livestock and poultry odor dispersion, if proper field verification is carried out.
Although there are many models currently available for predicting livestock odor dispersion, few have gone through an extensive validation process using field measurements, point out the University of Minnesota researchers.
That makes this research to develop an odor-rating system of particular significance. It paves the way for the use of an air dispersion model in predicting the spread of livestock odor from farms.
In the first round of trials, moderate-sized farms studied had confinement facilities that are mechanically and naturally ventilated, plus animals in open feedlots. These operations feature a variety of manure handling and storage systems including earthen storage basins, deep bed/litter pack, above ground tanks and deep pits.
It was decided that the second trial should involve farms that have large odor sources. Among the farms added was a farm with 3,000 finishing hogs.
For further verification of the air dispersion model, the nasal rangers sampled additional hog farms that were similar to what the model measured, but different in size. Four finishing operations were specifically monitored. These farms have the same manure handling and storage systems (deep pits) and same size curtain-sided buildings. The only difference was that the number of barns varied from site to site.
In this way, the influence of farm size on odor dispersion can be measured for distance and quantity, and a better understanding of how the dispersion model works can be achieved.
For starters, for the model to be verified requires large quantities of field measurements. Seven trained sniffers conduct on-site odor plume measurements. They rank the intensity of odors they sniff based on a numerical scale of 0 (no odor) to 5 (intense odor). If a sniffer could not distinguish between two levels, they would record a half level. The odor sniffers are required to wear a protective charcoal mask at all times except for the times they are actually sniffing for odors.
In this study, measurements are taken every 10 seconds during a 10-minute period, for a total of 60 readings per individual. Distances between 83 ft. to 1,320 ft. (depending on site and strength of odor source) were marked off at the centerline of the downward odor plume. Sniffers were positioned at points perpendicular to this centerline from 17 ft. to 66 ft. apart (See Figure 2) to cover the plume width.
Before each sniff, the nasal rangers would calibrate their noses (sensory system) by sniffing a static scale of N-butanol supplied to the group. To avoid a reduction in the sensitivity of the sniffers, a medium-range odor concentration was used in the first measurement at a distance of 330 ft. A second measurement was taken at 660 ft. and a third one at 1,320 ft. Weather information was recorded by an on-site, portable weather station at a 10-second interval to match the frequency of downwind odor data collection by the nasal rangers.
To calculate the odor intensity experienced by each panelist, average odor readings were converted to an "odor unit" by which a comparison can be made between the output of an odor dispersion model, expressed in odor units, to the levels detected by the nasal rangers in the field.
Researchers: Larry D. Jacobson, Richard E. Nicolai, David R. Schmidt and Jun Zhu, University of Minnesota, St. Paul, MN. Phone Jacobson at (612) 625-8288.