Wastewater Collection System VOC Emissions Modeling: A Comparison of INTERCEPTOR, TOXCHEM+ and WATER9 Models

Three models were compared to determine their ability to accurately predict emissions from two wastewater collection system components; gravity flow reaches and drop structures. Models investigated included the INTERCEPTOR Model, TOXCHEM+ and WATER9.Data used to perform the analysis were obtained from published literature. A total of 128 individual runs per model were completed in performing the analysis; 60 runs for the reach evaluation and 68 runs for the drop structure assessment. Compounds and range of operating conditions utilized inthe analyses were selected to ensure a broad coverage of compound volatilities and operating conditions. In addition, only data meeting stringent mass closure requirements, i.e., acceptable rangebetween 75–125% were deemed acceptable for inclusion in the analyses.Results indicate that on average, WATER9 under-predicted reach emissions by 44% whereas the INTERCEPTOR and TOXCHEM+ models over-predicted emissions by 32 and 36%, respectively. For the drop structure analyses, WATER9 over-predicted by 103 and 108% (depending on how drop structures were configured) vs. INTERCEPTOR and TOXCHEM+ which under-predicted by 6 and 36% respectively.An error analysis was completed to determine if the results were statistically significant. A Duncan's Multiple Range Test, a procedure to determine which treatment means are different than others, was used to perform this assessment. Based on the statistical assessment completed, it was determined that the results generated by WATER9 are significantly different than the field-measured data and the INTERCEPTOR and TOXCHEM+ models.In order to better understand the reasons for the differences between the models, an in-depth analysis of the algorithms, assumptions and compound property databases used by each model was completed.