Impact of variable emission rates on odour modelling at WwTW’s

The use of dispersion modelling as a means of regulating the impact of odours from WwTW’s on surrounding areas is increasing. Although dispersion modelling is an invaluable tool in assessing odour impacts, its accuracy depends primarily on the emission rates used in the model. Current practice is to use static emission rates, either from look-up tables, or site sampling, despite the known variability of emissions. If dispersion modelling is to provide the accuracy required of a regulatory tool, an assessment of the impact of variability of emissions on the results obtained needs to be undertaken. This paper presents a case study of variable emission rate modelling using ODOURsim®, and the impact of variable emissions on dispersion modelling results. ODOURsim® is a novel odour emission modelling software package that has been developed for accurately predicting odours from sewage works by determining the rate of generation at source. A biological model derived from the well-known ASM2 kinetic model is used to calculate the formation of H2S in the liquid phase. The emission rates from the liquid to gas phase are calculated using a suite of process-specific mass-transfer models. The calculated variable emission rates can then be included in proprietary dispersion models as an hourly emission file. Methods and techniques used for the calibration and validation of the ODOURsim® emission rate model are described. Inputs to the ODOURsim® emission model for the case study sites were standard wastewater quality measurements (TCOD, SCOD, pH, Temperature, Soluble Sulphide), allowing diurnal variation profiles to be simulated. The diurnal profile was used in combination with one years hourly flow data to produce the years’ hourly influent file for the model. Emission rates obtained from the ODOURsim® model were used in the dispersion model to obtain contour plots for odour dispersion from the case site. Graphical dispersion results of a diurnal simulation of a medium sized WwTW are presented. Dispersion model results, both instantaneous and yearly percentile contour plots, are compared for both variable and static emissions. Significant differences between the yearly percentile contour plots for variable and static emission rates have been observed at a medium sized WwTW. Implications for design of odour control equipment with regards to data analysis of the simulation results for calculation of peak to mean ratios are discussed.