Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Evaluation of Air Dispersion Models Using Swine Odour Plume Measurement Data

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:   No Citation available.
Authors:   Y Xing, H Guo, John Feddes, Stan Shewchuck
Keywords:   Air dispersion model, swine, odour, Fractional Bias

Air dispersion models have been used to predict livestock odours downwind of livestock operations in the last several years. It is important that these models be properly evaluated before their predictions can be used with confidence. The model evaluation involves comparison of the models predictions with measured field data, which very limited work has been done. In this study, four air dispersion models, ISCST3, AUSPLUME, CALPUFF, and INPUFF2, were evaluated using odour plume measurement data downwind of two swine operations. The model predicted odour intensities were compared with the measured odour intensities. Considering all the measurements, four models agreement results using the conversion equation from University of Alberta achieved 56% to 62% for all distances and 67 to 76% for distance of 500 to 100m which were better than the obtained using the University of Manitoba conversion equations. However, if the measurements with intensity zero (no odour) were excluded, the agreements for these models with the conversion equation from University of Manitoba were higher than the results with University of Alberta conversion equations. Agreements reached 29 to 35% for all distances and 36 to 48% for distance of 500 to 1000 m. Furthermore, the agreements between the model predictions and measured values could not be improved considerably by using scaling factors. The odour intensity and concentration conversion equation was the main cause of this low improvement. Using ASTM standard guide for air dispersion model evaluation is a try in this study. The selected four models fractional biases were all in the acceptable range from -0.67 to 0.67. ISCST3 performs the best with the lowest bias in matching field measured odour intensity followed by AUSPLUME. CALPUFF and INPUFF2 also performed well within FB value lower than 0.67. However, CALPUFF over-predicted by bias of average intensity while INPUFF2 under predicted by a value of -0.66 of the bias of average.

(Download PDF)    (Export to EndNotes)