Selecting a Rainfall that Results in a Peak RDII Condition for Design - Using a Standardized Procedure

Engineers and sewer system operators need tools for predicting peak flow rate conditions in sewers. This paper describes a general approach for projecting peak RDI/I (rainfall dependent I/I) flow rates from flow and rainfall monitoring data to compare the peak flow rates resulting from different types of storms. This approach represents an improvement over previous practices because it includes simple statistical tests for quality control and determining the level of confidence. Therefore, users of peak flow rate information will know the level of confidence for the rate projections and can make better decisions about the level of risk for SSOs (Sanitary Sewer Overflows) and estimating sewer design capacity. This approach was applied to data collected at 14 long-term flow monitors from Atlanta, Brentwood, and Nashville. The results from these southeastern cities indicated that in most cases the 24-hour storm was more useful in for predicting the peak hour I/I rate than shorter, more intense storm events. On average, estimates of peak-hour RDII flow rates using 3-hour storms were 25% less than estimates using the 24-hour storm events. In every case, the 95% confidence interval was tighter when using the 24-hour event data — indicating a higher level of confidence in the final result. That means that the proper duration rainfall event may be selected with a known level of confidence for hydraulic modeling and capacity management in those sewer systems. This study incorporated a non-proprietary procedure using standard spreadsheets and commonly accepted statistical parameters.