A conceptual model of the role of excess energy in the maintenance of a riffle–pool sequence

Abstract

There is much debate over the role of hydraulic ‘reversal’ in the maintenance of riffle–pool sequences in gravel-bed channels. It is, however, generally acknowledged that the pool has the greatest maximum energy during flood flows allowing scour and pool maintenance. Little emphasis has been placed on the reversal period in many previous studies despite its potential importance being noted in the seminal study of Keller [Geol. Soc. Am. Bull. 82 (1971) 753]. This is potentially significant, as reversal may only occur for short periods at the peak of flood hydrographs; hence, the hydraulic argument behind riffle–pool maintenance is based around what is often a marginal energy excess in the pool for a very short period of time. This study uses the magnitude frequency concept of Wolman and Miller [J. Geol. 68 (1960) 54] and applies it to an investigation of the temporal dominance in morphologic unit stream power between a series of four pools and three riffles derived from measured flow and calculated energy slopes on the upland gravel-bed River Rede, Northumberland. The geomorphologically effective flow regime is used to define temporal balances between riffle and pool energy levels. The geomorphic work carried out by riffles and pools, as described by the excess stream power ( Ω− Ω c) capable of transporting the bed sediments, displayed a broad balance for the upstream four units before reducing significantly at pool 3 and rising again for the subsequent two units which appear particularly energetic. These findings contrast with hydraulic theories of riffle–pool maintenance, e.g., Keller [Geol. Soc. Am. Bull. 82 (1971) 753]. For sediments to be transferred through a riffle–pool sequence whilst maintaining a quasi-equilibrium form, the pools should carry out the same amount of work as the riffle upstream in order for it not to fill with sediment. Although stream power reversal did occur for some riffle–pool units, the magnitude and longevity of the reversal did not balance the energy exerted at riffle locations. Armouring and shallow gravel depth at riffle 3 appear to indicate scouring in response to high excess stream power found in this area. However, there was no evidence of sub-dominant pools 3 and 4 filling in response to low stream power.