Surge Analysis for the OARS Tunnel, Columbus, Ohio

A critical component of the Long Term Control Plan (LTCP) for the City of Columbus, Ohio is the proposed OSIS Augmentation Relief Sewer (OARS). The OARS is proposed to be a tunnel that serves the Combined Sewer Overflow (CSO) outfalls along the Scioto River. The OARS will store and convey the combined sewer overflow to the City’s Wastewater Treatment Plants so that overflow occurs infrequently in the downtown Columbus area.

Most significantly, the overflow from the Whittier Street Storm Standby Tank (WSST), which comprise 85% of all annual overflows, will be eliminated at the Whittier Street site for the typical year storm event. The OARS was initially proposed to be an 18-ft diameter tunnel, about 23,000 ft long. The OSIS would be relieved at three different locations with the flow introduced into OARS by means of off-line drop shafts. An additional offline drop shaft will convey overflow from a local regulator to the tunnel. A large downstream shaft is proposed with a high level overflow to the river, dewatering pump stations, and conveyance conduits to the treatment plants.

A city-wide SWMM 4.4H model was developed over a decade and has been utilized for the planning and design of the City’s LTCP components, including the OARS. The SWMM 4.4H model provides the input hydrographs to the OARS for the typical year and the 10 year design storm events. However, it was realized that SWMM 4.4H is not capable of modeling surges resulting from the fast-filling hydrographs that are predicted to occur, particularly for the 10-y storm event. The Transient Analysis Program (TAP) developed by Applied Science, Incorporated, of Detroit, Michigan, was utilized to model the surges that result from fast-filling hydrographs and determine the ventilation requirements for the OARS.

Surge analysis using TAP indicated that significant changes in the 30% design concept were required in order to handle the open channel bore and subsequent closed conduit surges predicted to occur during the filling process within the tunnel. In addition, TAP indicated that there were issues related to conveyance as a result of significant losses in the drop shafts and related appurtenant structures (tangential inlet, approach channel, etc.) during surcharged flow-through conditions. The design changes included increasing the diameter tunnel to 20 feet, increasing some drop shaft and adit diameters, and providing dedicated surge overflow chambers, vent shafts, and bypass channels parallel to the approach channels at the drop shafts to prevent upstream migration of the surges.

The TAP model is not well suited for the analysis of the entire City of Columbus wastewater collection system because of the long computational time that would be required as a result of the very small time steps typically associated with the calculations. The time step in TAP is automatically determined by the Courant conditions and generally is a tenth of a second or less. TAP was used only to model the OARS tunnel, drop shafts and appurtenant structures, critical overflow structures such as the WSST, and the surge relief devices. An interesting part of the project was to establish the appropriate interface between the SWMM 4.4H and TAP models at the boundaries of the TAP model. In addition, in order to obtain an acceptable match between the results of the two models at the boundaries, an iterative procedure between the SWMM 4.4H and TAP models was developed. The use of TAP enabled the design team to develop a configuration for OARS that met all of the goals of the project.