Simulating time-varying cave flow and water levels using the Storm Water Management Model

2002 
Abstract The Storm Water Management Model (SWMM) is an Environmental Protection Agency code used to estimate runoff through storm water drainage systems that include channels, pipes, and manholes with storage. SWMM was applied to simulate flow and water level changes with time for a part of Stephens Gap Cave in Jackson County, Alabama. The goal of the simulation was to estimate losses from a surface stream to the cave. The cave has three entrances that can remove water from the surface stream. These entrances connect through several passages to an 8-m (27-ft) high waterfall in a dome room. After a storm, the walls of this dome room had leaves on the wall as high as 4.6 m (15 ft) above the floor. The model showed that the height of the leaves did not represent a water level that could have occurred following any recent storm. Campbell, Livingston and Garza in 1997 developed the CLG model to estimate losses from karst surface streams. This model treats losses as pipe flow from a reservoir and gives the loss flow rate as ∼ h 0.5 where h is the depth of flow in the surface stream. Losses to Stephens Gap Cave calculated with SWMM varied as h 1.8 . This depth dependence is more characteristic of flow over a weir than of pipe flow. The SWMM-calculated losses to Stephens Gap Cave showed no hysteresis, that is, the rising and falling limbs of the stage-discharge plot followed the same curve. Loss curves with significant hysteresis are difficult to simulate with simple models such as CLG or a weir flow model. However, an SWMM model of a simple hypothetical cave demonstrated that storage in Stephens Gap Cave is far below that required to cause hysteresis. Losses from many karst surface streams can probably be adequately estimated with a calibrated weir flow model. The utility of SWMM for analyzing cave flows was established. SWMM produced stable solutions with very low continuity errors for this cave.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    2
    References
    44
    Citations
    NaN
    KQI
    []