Semester

Summer

Date of Graduation

2006

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Civil and Environmental Engineering

Committee Chair

Robert N. Eli.

Abstract

The suitability of applying the NRCS Curve Number (CN) to continuous runoff prediction is examined by studying the dependence of the CN on several hydrologic variables. The continuous watershed model Hydrologic Simulation Program-FORTRAN (HSPF) is employed as a theoretical watershed in two numerical procedures designed to investigate the influence of soil type, soil depth, storm depth, storm distribution, and initial abstraction ratio value (lambda) on the CN. This study stems from a concurrent project involving the design of a computer modeling system to support the Cumulative Hydrologic Impact Assessments (CHIA) of over 230 watersheds throughout WV. A link between the CN and HSPF soil moisture parameters is proposed for continuous runoff simulation in surface mine affected watersheds in West Virginia. A soil physics model and numerical procedure have been developed to back calculate CN's at Antecedent Runoff Condition (ARC) II from synthetic rainfall input and simulated direct runoff. A second method of CN determination is also described to provide a reference to the calculated CN values. Each HSPF parameter set, determined through calibration and by the soil physics model, is treated as a unique hypothetical watershed. It was found that the calculated CN's are highly dependent on all of the computational variables, therefore the use of the CN in continuous modeling based on antecedent soil moisture or rainfall alone does not appear to be appropriate. Differences between lambda = 0.05 and lambda = 0.2 are seen predominantly in the lower storm depth calculations. It is suggested that a different symbol be used to distinguish classic CN's from continuous CN's.

Download

Share

COinS