|
Activity Number:
|
343
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #310119 |
|
Title:
|
Spatiotemporal Bayesian Maximum Entropy Modeling of Hydrogen Sulfide Concentrations Using Data Collected at Different Observation Time Scales Near Swine Operations
|
|
Author(s):
|
William Allshouse*+ and Marc Serre and Devon Hall and Katherine Mills and Steve Wing
|
|
Companies:
|
The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill and Rural Empowerment Association for Community Help and The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill
|
|
Address:
|
Department of Environmental Sciences and Engineering, , ,
|
|
Keywords:
|
hydrogen sulfide ; Bayesian Maximum Entropy ; CAFO ; hog farms ; multiple time scales
|
|
Abstract:
|
Hydrogen sulfide (H2S) is an air pollutant produced by industrial hog operations that provides a unique problem for space/time modeling. Spraying hog waste on adjacent fields is believed to be the main source of the compound, creating temporal variation based on the spraying events and spatial variation based on distance to source. Active samplers recorded 15 minute H2S measurements, while passive samplers made a two-week measurement. The difference in measurement duration required that "soft" 15 minute H2S data be created from the passive samplers. The variance of the soft data was derived by accounting for the difference in time scales in the covariance models. The Bayesian Maximum Entropy (BME) method was used to integrate general and site-specific knowledge from the two data types to produce space/time estimates of H2S concentrations.
|
