Geomorphic Equations and Methods for Natural Channel Design

Abstract

Natural channel design in river engineering is the philosophy and practice of designing stream channels by copying or mimicking the geomorphology of stable, self-formed streams. This dissertation presents methods and equations for incorporating the principals of natural channel design into river and stream engineering in Kansas. Data from 123 reference reaches in Kansas are used to develop these methods and equations. An analysis of 46 gaged reference reaches indicates that the return period of bankfull flow (annual maximum series), ranges from 1.01 to 1.7 years, with an average of 1.2 years. This is significantly lower than the 2-year flow commonly used by engineers in naturalistic river designs. An equation is developed that predicts the 1.2-year flow as a function of watershed drainage area, mean annual precipitation, and the length of the longest flow path in the watershed. This equation is developed using data from 67 gaged streams with drainage areas less than 30 sq miles. Geomorphic measurements from the reference reaches are used to verify previously published relationships between bankfull discharge and bankfull width for streams with sand and gravel beds. A new relationship is developed demonstrating the relationship between bankfull discharge and bankfull width for streams with beds of cohesive clay. Equations are provided to predict the average meander wavelength from the bankfull width, and the pool depth from the depth at the adjacent riffle. Three stream design methods are presented: the Kansas Analytical Method (KAM), the Analytical Reference Reach Method (ARRM), and the Scaled Geomorphic Method (SGM). All three methods are based on natural stream processes and geomorphology. All three methods incorporate the Manning equation for flow resistance and the Meyer-Peter and Muller equation for sediment transport but differ in their use of geomorphic measurements from reference reaches. KAM uses a hydraulic geometry width equation which was developed from many reference reaches. ARRM uses the sinuosity, pool-depth ratio, and meander-width ratio from a single reference reach. SGM calculates a scaling factor that can be used to copy and scale additional cross-sections (pools, runs, and glides, as well as riffles) and planform features from the reference reach. The development of KAM and ARRM is presented in previously published reports. This dissertation presents the development of SGM in detail. KAM, ARRM, and SGM make a common assumption that the median size of sediment in the channel bed is an adequate surrogate for the entire gradation of bed sediments. The reasonableness of this assumption is verified by calculating the bankfull sediment transport capacity for seven ARRM designs. It is found that a channel designed for equilibrium transport of the median sediment size is reasonably designed for transport of the entire gradation of sediment sizes. The exception is when the median sediment size found on the bed is among the largest that are mobile at bankfull flow. These geomorphic relationships, equations, and design methods combine traditional hydraulic engineering and fluvial geomorphology in unique ways to provide practical tools to stream designers in Kansas.