COMPARISONS OF ACTUAL AND PREDICTED HYDROGEN SULFIDE EMISSIONS RATES AT WATER RECLAMATION FACILITIES IN CLARK COUNTY, NEVADA

To improve its air permitting and odor management programs, the Clark County (Nevada) Water Reclamation District (CCWRD) is continuously working to increase its ability to quantify and predict hydrogen sulfide emissions from its treatment processes. As part of this effort, CCWRD compared hydrogen sulfide emission rates from covered wastewater treatment processes with emission estimates developed using three software programs for estimating emissions of volatile compounds from wastewater treatment processes. The emission estimating programs evaluated were BASTE3, TOXCHEM+, and WATER9. Treatment processes evaluated included bar screens, grit basins, primary clarifiers, flow splitters, and screw pumps.Comparisons of measured and predicted emission rates showed significant variations among the modeling programs in predicted emission rates. The three modeling programs all provided comparable emission estimates for the primary clarifier, and the estimated emissions from the primary clarifier corresponded well with measured emissions. Variations in modeled estimates were higher for grit basins than for the primary clarifier, and all three models tended to overestimate grit basin emissions. For bar screens and flow in covered vaults, there were large differences among the models, and some of the estimates provided were far below measured emissions.Processes that have significant mixing or agitation will often have increases in dissolved oxygen in the effluent due to aeration associated with the mixing. The BASTE3 program can estimate hydrogen sulfide emissions from the changes in dissolved osygen. For some processes this may be the only feasible way to estimate emissions.The data from this study are too limited to draw conclusions about the overall reliability of various emission estimating models. Use of any model exclusively would have resulted in significant underestimates or overestimates of actual emissions. Whenever possible, models should be adjusted based on actual emission measurements.