DAM BREACH: 1-D VS 2-D MODEL AND IF 2-D IS FUN, WHY NOT CASCADE IT AND SUPER-SIZE YOUR PLEASURE?

摘要

In response to the U.S. Army Corps of Engineers (USACE) National Dam Safety Program, the Jacksonville District (SAJ) is producing breach flood inundation maps for the Herbert Hoover Dike Emergency Action Plan (HHD EAP). The HHD system comprises 143 miles of hydraulic-fill cohesionless levee/embankment surrounding Lake Okeechobee in south Florida. The floodplain to the south (currently, the area of most interest, i.e., risk) is 948 square miles of horizontally flat expanse whose vast majority of features consist of agricultural ditches and low, 2-6 feet secondary levees or dirt roads, with few major canals, levees, and highways. The region is bounded by major levees that effectively form an areal bowl with no gravity outflows; all conveyances are pump-controlled. In 2000, SAJ produced a Major Rehabilitation Evaluation Report (MRR), which applied the 1-D unsteady UNET model to route the breach floodwater. In 2007, SAJ produced interim flood maps that applied the 1-D HEC-RAS model for routing with improved breach parameters. After receiving new LiDAR topographic data in 2009, SAJ worked with Taylor Engineering, Inc. to determine whether 2-D modeling would provide more accurate and meaningful results than 1-D modeling. Knowing stakeholders' expectations would be for a quick turnaround deliverable, the key word really was "meaningful." Ultimately, the team decided to use a 2-D model-MIKE21 incorporated in MIKE-FLOOD-to route the breach floodwater. One part of this paper discusses and provides results from the 1-D versus 2-D model study. The team encountered another issue-given a 2-D model for the area of interest, should an investigator consider subsequent cascading failures (breaches) of secondary levees and / or elevated roads as floodwater rises and passes downstream? To find out, the team added cascading failures to determine the effects on model results. As model simulations with higher initial lake water levels showed that cascading secondary levee breaches had minimal impact, the USACE SAJ deemed inclusion of cascading breaches a "low priority" among the efforts in producing flood inundation maps. Secondarily, the team decided not to incorporate cascading failures because incorporation of cascading breaches would essentially result in thousands of potential combinations for the vast floodplain. The other part of this paper discusses and provides results from the cascading failure model study.