Definitions and Nomenclature

Figure 1. Antidunes are symmetric, transverse bedforms that form in rapid, shallow flows. The water surface above antidunes is deformed into a waveform that is in phase (crest of the waveform coincides with the antidune crest) with the bedform. Antidunes are so named because they typically migrate upstream, opposite the migration direction of dunes and ripples. Most terminology developed for bedforms is applicable to antidunes. Like other bedforms, the stoss and lee sides refer to the upstream and downstream side of the bedform, respectively.

Kennedy (1961) investigated the mechanics of flow over antidunes, and focused on the close match between the bedform and the surface waveform. With this point of view, he modified Gilbert's definition of antidunes to include any symmetric bedform that is in phase with waves in the water surface above the bed. Kennedy argued that these bedforms resulted from the waves at the surface deforming the bed, and that the migration direction of the bedform was relatively unimportant.

This study defines antidunes as a symmetric bed configuration that forms in flows where the Froude number (the ratio of the flow velocity (U) squared to the product of the flow depth (d) and the acceleration of gravity (g)),

exceeds unity (Fr > 1). As we shall see, this definition forces the waveform at the water surface to be in phase with the bedform and constrains the bedform to migrate in the upstream direction (unless the minimum bed shear stress of the flow exceeds the critical shear stress for transport of all grains in the bed).

Since antidunes migrate upstream, they are an oddity among the predominantly downstream-migrating bedforms. The nomenclature for bedforms has been developed in reference to downstream-migrating bedforms (dunes and ripples) and can be confusing when applied to antidunes. The antidune literature adheres to the connotations of stoss and lee, meaning the up-flow and down-flow sides of an object, respectively. This nomenclature provides continuity with discussions of low flow regime bedforms. In this report stoss-side and upstream-side will be used interchangeably.

There is some confusion in antidune literature on the use of foreset and backset to describe inclined laminae produced by antidunes. Several workers have described upstream-dipping cross-laminae produced by antidunes as backsets (Barwis and Hayes, 1985; Langford and Bracken, 1987). This usage is based on the idea that the cross-laminae are dipping counter to flow direction. The term foreset is equally applicable, since the upstream-dipping cross-laminae are inclined in the direction of bedform migration. Throughout this discussion, however, I will avoid this problem by referring to structures as upstream-dipping cross-laminae or downstream-dipping cross-laminae.

Finally, several workers use the term "breaking antidunes" to describe the collapse of the waveform at the water surface above the antidune (Allen, 1966; Hand, 1974). Langford and Bracken (1987) used the term for the formation of a hydraulic jump above the antidune crest. Although the term breaking antidunes is a misnomer, in the sense that the water waveform is collapsing and not the bedform, it is a useful description when applied to the collapsing surface waveform. The collapsing waveform at the water surface is similar in appearance to breaking ocean waves. The term should not, however, be used to describe other phenomena, like the formation of hydraulic jumps. In this study "breaking antidunes" will be used interchangeably with surface waveform collapse or breaking. The transformation of flow states implied by the formation of a hydraulic jump above the antidune crest will be referred to as "transforming antidunes".