Preponderance of Evidence: Advances in Using Distributed Temperature Sensing to Locate and Quantify Sources of I/I

At WEFTEC 2021, Brown and Caldwell (BC) presented a paper on the first United States-based study of a sanitary sewer system using Distributed Temperature Sensing (DTS) along a fiber optic cable to identify specific locations of significant I/I sources active during wet weather events with cold I/I conditions. BC has now performed additional studies in the City of Wauwatosa, Wisconsin, with funding from the Milwaukee Metropolitan Sewerage District (District). The District's Wauwatosa study was performed in the summer of 2020 during several significant storm events when I/I was warm compared to background wastewater conditions. The District study provided sufficient data for quantifying actual I/I discharges and proved the technology was capable of working successfully in warm I/I conditions.
Description of Technology
The principle of DTS operation is fairly simple: when laser light passes down a fiber optic cable, the bulk of the energy is passed through; however, a small percentage is reflected back to the source by defects within the fiber optic cable. These defects occur throughout the length of a fiber optic cable. Most of this reflected energy is equal to the transmit frequency, but a small portion is shifted in frequency due to interaction between the laser light and the defect. Some energy is shifted higher in frequency, and some to a lower frequency. The intensity ratio of higher frequency energy to lower frequency energy is temperature dependent, allowing an accurate method for measuring temperature along the length of the cable. Figure 1 provides an illustration of DTS technology. The DTS technology represents an advanced approach for localizing significant sources of I/I, which can lead to cost-optimized spending on rehabilitation programs. As the DTS method will identify temperature gradients in the sewers, it highlights where infiltrated rainwater is influencing the wastewater being carried to the downstream treatment plant. Figure 2 shows a simplified approach for setting up DTS to study I/I sources in sanitary sewers and is representative of the approach used in this study. Figure 3 is a photograph of the equipment installation manhole (MH).
Project Setting
The District established a regional Private Property Infiltration and Inflow (PPII) Reduction Program in 2011 after experiencing major rainstorms that caused massive sewer overflows and thousands of sewer backups. The District PPII program provides funding to every municipality served by the regional system, and funds can be used for a wide variety of I/I reduction activities so long as the funds are exclusively used for private property measures. The City of Wauwatosa is one such community in Wisconsin with many homes that have connected foundation drains. As a result, the City experiences significant I/I and occasionally has sewer backups during large storm events. The City has been working diligently throughout its system to combat I/I, taking full advantage of District PPII Program funds. The City work, however, has largely focused on addressing problems with sewer laterals and not yet tackled connected foundation drains. Eagle Street in the City of Wauwatosa is one such area with connected foundation drains. The City has cured-in-place-pipe (CIPP) lined the sanitary sewers and rehabilitated manholes using City funds and has rehabilitated sewer laterals with CIPP using District PPII funds. Despite the extensive rehabilitation completed to date, significant I/I persists in this area. As a result, the most suspected sources of I/I were the connected foundation drains and the un-lined portions of the lateral under the homes. Previous flow monitoring data collected by the District showed that flow temperatures in this system responded during I/I events in both cold winter and spring conditions as well as warm summer conditions.
Field Deployment
The field test utilized a DTS system capable of measuring the temperature of water at 1 meter intervals in a long reach of sewers; approximately 1,200 meters in total length for this project. During the field monitoring period, data would be collected for 48-hours of dry weather and three separate periods of wet weather data, when the rainfall event total amounted to at least 25.4 millimeters (1.0 inches). Data analysis would include plotting the temperature profile of the monitored sewers and how the profile varied during rainfall events. Additional data analysis would identify specific locations where temperature changes appear to indicate the presence of I/I inflow sources. After the monitoring system was installed, a dry weather period of data was collected to confirm that the system was working correctly. Dry weather system data collection occurred from May 28 to 30, 2020. Three wet weather events were recorded on June 10-11, July 9-10, and July 15-16. Additional events were also monitored in case the total rainfall exceeded 12.7 mm (0.5 inches), but ultimately these events provided insufficient data to analyze.
Analysis
Color-coded temperature plots (waterfall plots) were prepared for each monitored event. Each of these plots show points of interest along the monitored sewer, including the location of manholes, laterals, and the DTS unit. Plots of flow and rainfall are synchronized to, and plotted with, the DTS temperature data. Observations can be made from reviewing these wet weather event data and the distinct temperature changes that coincide with measured rainfall. Consistent behavior at specific locations, across multiple rain events, provides strong evidence that an I/I source is present and deserving of follow-up field investigations. An example plot is shown in Figure 4 during the July 9-10 rainstorm. To estimate flow along the cable, the study utilized a methodology proposed by Beheshti and Saegrov (Water, 2018), in which a temperature and flow mass balance is performed stepwise in a downstream fashion starting at the upstream end of the cable. Figure 5 presents a comparison of one event between temperature based I/I estimates and the I/I model. Figure 6 shows the accumulation of flow along the pipe during the peak of this same event.
Compared to other methods of locating and quantifying I/I source inputs, DTS could be considered expensive unless considering the extent of information this method generates. BC develop an estimated Return on Investment for using the DTS approach to identify actionable I/I sources. The process used for estimating ROI and the outcome will be detailed in the full paper.
Conclusions
This study showed the capabilities of the DTS method for finding potential significant sources of I/I in a sanitary sewer system. The methodologies developed by Beheshti and Saegrov (Water, 2018) for quantifying I/I flow rates based on DTS data were evaluated and show promise, though further studies are needed to better understand the capabilities and limitations of their approach.
The full paper and presentation will provide further details on the complete study.