Flow and solute-transport models for the New River in the New River Gorge National River, West Virginia

Abstract

This report presents the results of a study by the U.S. Geological Survey (USGS), in cooperation with the National Park Service, to apply flow and solute-transport models to the New River in the New River Gorge National River, West Virginia. Relations between crosssectional area and discharge developed from input parameters of an unsteady-flow model were compared to relations between cross-sectional area and discharge developed from a steady-flow model output. The study reach, 53 miles of the lower New River from Hinton to Fayette, is characterized as a pool-and-riffle stream that narrows, steepens, and deepens in the downstream direction. Three subreaches Hinton to Meadow Creek, Meadow Creek to Sewell, and Sewell to Fayetterepresented similar slopes, geometries, and roughness of the study reach. A USGS steady-flow model, WSPRO (Water Surface PROfile), was applied to the study reach. The model was calibrated by use of relations developed between river stages and discharges. Cross-section configurations were determined by means of aerial photography, topographic maps, rating curves, and water-surface and streambed profiles. The model was verified by comparing random predicted water-surface elevations at a discharge of 2,000 ft3/s (cubic feet per second) to those of a surveyed profile. The model was more sensitive to changes in Manning's roughness coefficients than to changes in the hydraulic-depth breakpoints corresponding to Manning's roughness. A USGS unsteady-flow model, DAFLOW (Diffusion Analogy FLOW), and a USGS solutetransport model, BLTM (Branch Lagrangian Transport Model), were also applied to the study reach. Difficulty in calibration required development of separate models for discharges greater than or equal to 8,000 ft3/s (highdischarge model) and less than or equal to 8,000 ft3/s (low-discharge model). The DAFLOW models were calibrated by use of relations between river discharges and traveltimes of the change in discharge at the leading edge of waves. The DAFLOW models were verified by predicting discharges at the streamflow-gaging station at Thurmond using discharges from the Hinton station. The BLTM models were calibrated by use of relations between traveltime of peak concentration and discharge, and peak concentration and traveltime of peak concentration. The BLTM models were verified by predicting peak concentrations and traveltimes of peak concentrations for two unsteady-flow and one steadyflow dye measurements. Relations between cross-sectional area and discharge, developed from calibration parameters for the steady-flow and unsteady-flow models, were compared. No explanation could be determined for the poor comparisons.