An Area-Based Approach to Replacing Single R-Value Thresholds for RDII Evaluation

R-value, defined as rainfall-derived inflow and infiltration (RDII) volume divided by rainfall volume, is widely used to classify sanitary sewer wet-weather performance and to guide rehabilitation planning. Industry literature provides typical R-value thresholds; however, these thresholds often do not account for key regional controlling factors such as seasonal groundwater conditions, meteorological and climatic variability, and sewer age and material. In addition, RDII response to rainfall is inherently nonlinear. Small and medium storm events experience varying abstraction conditions throughout the year, while RDII response becomes more uniform as storm size increases and abstraction effects diminish. During large storm events, soil infiltration capacity, crack dimensions, and pressurized pipe conditions limit the amount of RDII that can enter the system. Furthermore, many flow monitoring programs are designed to target known problem areas, resulting in datasets that may be biased toward higher R-values, while large basins may exhibit artificially low R-values that mask localized defects. Consequently, a single “typical” R-value threshold derived from arbitrarily monitored basins can over-classify small basins as poor condition and under-classify larger basins with concentrated RDII sources.

This paper presents an improved, scalable method to define utility-consistent R-value ranges that explicitly account for contributing sewered area, enabling more focused rehabilitation prioritization aligned with asset management objectives. Using calibrated hydrologic and hydraulic (H&H) models, the proposed method develops an area-normalized geometric mean relationship derived from large sets of hydraulically independent basins to ensure unbiased aggregation. The approach was developed for the City of Columbus analyzing 116 meter-basins and North Texas Municipal Water District (NTMWD) analyzing 52 meter-basins. In each study, basins were calibrated to multiple historical storm events. The Columbus model uses a groundwater module to calculate RDII, accounting for soil infiltration capacity and seasonal groundwater conditions. The NTMWD model applies the RTK approach, with median monthly RTK values used to represent seasonal response. RDII volumes were computed using multiple design storms to illustrate the influence of storm intensity on R-value behavior, and RDII response was evaluated over a three-day simulation period with storms applied at the start of the model run.

Two complementary characterizations are compared: (1) an equal-weighted, direct median Rvalue approach, which demonstrates bias toward smaller basins; and (2) an area-dependent geometric mean envelope that removes basin-size bias and provides a defensible benchmark for relative categorization of sewersheds. The resulting curves define performance categories ranging from Very Low to Very High using half- and full-standard-deviation envelopes about the geometric mean. These envelopes yield a practical decision framework in which basins exceeding the half-standard-deviation curve indicate elevated RDII, while those exceeding one standard deviation represent high-priority rehabilitation targets with increased return on investment. Separate winter and summer geometric-mean R-value relationships are developed.

This approach replaces a single threshold with an objective, scalable criterion tied to contributing area, improving comparability across monitored systems and strengthening defensibility for RDII screening, acceptance, and rehabilitation planning.

 Click here to download a static PDF version of the presentation.