A two-dimensional finite-element surface-water model was used to study the effects of proposed modifications to the State Highway 203 corridor (proposed Elba Bypass/relocated U.S. Highway 84) on water-surface elevations and flow distributions during flooding in the Pea River and Whitewater Creek Basins at Elba, Coffee County, Alabama. Flooding was first simulated for the March 17, 1990, flood, using the 1990 flood-plain conditions to calibrate the model to match measured data collected by the U.S. Geological Survey and the U.S. Army Corps of Engineers after the flood. After model calibration, the effects of flooding were simulated for four scenarios: (1) floods having the 50- and 100-year recurrence intervals for the existing flood-plain, bridge, highway, and levee conditions; (2) floods having the 50- and 100-year recurrence intervals for the existing flood-plain and levee conditions with the State Highway 203 embankment and bridge removed; (3) floods having the 50- and 100-year recurrence intervals for the existing flood-plain, bridge, and highway conditions with proposed modifications (elevating) to the levee; and (4) floods having the 50- and 100-year recurrence intervals for the proposed conditions reflecting the Elba Bypass and modified levee. The simulation of floodflow for the Pea River and Whitewater Creek flood of March 17, 1990, in the study reach compared closely to flood profile data obtained after the flood. The flood of March 17, 1990, had an estimated peak discharge of 58,000 cubic feet per second at the gage (just below the confluence) and was estimated to be between a 50-year and 100-year flood event. The estimated peak discharge for Pea River and Whitewater Creek was 40,000 and 42,000 cubic feet per second, respectively. Simulation of floodflows for the 50-year flood (51,400 cubic feet per second) at the gage for existing flood-plain, bridge, highway, and levee conditions indicated that about 31 percent of the peak flow was conveyed by the State Highway 203 bridge over Whitewater Creek, approximately 12 percent overtopped the State Highway 203 embankment, and about 57 percent was conveyed by the Pea River flood plain east of State Highway 125. For this simulation, flow from Pea River (2,380 cubic feet per second) overtopped State Highway 125 and crossed over into the Whitewater Creek flood plain north of State Highway 203, creating one common flood plain. The water-surface elevation estimated at the downstream side of the State Highway 203 bridge crossing Whitewater Creek was 202.82 feet. The girders for both the State Highway 203 and U.S. Highway 84 bridges were partially submerged, but U.S. Highway 84 was not overtopped. For the 100-year flood (63,500 cubic feet per second) at the gage, the simulation indicated that about 25 percent of the peak flow was conveyed by the State Highway 203 bridge over Whitewater Creek, approximately 24 percent overtopped the State Highway 203 embankment, and about 51 percent was conveyed by the Pea River flood plain east of State Highway 125. The existing levee adjacent to Whitewater Creek was overtopped by a flow of 3,200 cubic feet per second during the 100-year flood. For this simulation, flow from Pea River (6,710 cubic feet per second) overtopped State Highway 125 and crossed over into the Whitewater Creek flood plain north of State Highway 203. The water-surface elevation estimated at the downstream side of the State Highway 203 bridge crossing Whitewater Creek was 205.60 feet. The girders for both the State Highway 203 and U.S. Highway 84 bridges were partially submerged, and the west end of the U.S. Highway 84 bridge was overtopped. Simulation of floodflows for the 50-year flood at the gage for existing flood-plain and levee conditions, but with the State Highway 203 embankment and bridge removed, yielded a lower water-surface elevation (202.90 feet) upstream of this bridge than that computed for the existing conditions. For the 100-year flood, the simulation indi