This article evaluates the ability of three models (Longuet-Higgins, Komar and Inman, and a modified Longuet-Higgins approach) to replicate observed longshorecurrent velocities. These models were based on the concept of local driving forces induced by radiation stress. Observation of longshore-current velocities was part of a field investigation of the nearshore system at Sandy Hook, New Jersey. Correspondence between observed and predicted velocities was assessed statistically. The modified Longuet-Higgins approach provided the best estimate of both site-specific and general velocities. LONGSHORE currents are generated by energy dissipation in breaking waves. Largely confined to the surf zone and the adjacent nearshore, these currents may flow parallel to the shore for relatively long distances or may occur as a part of a cellular circulation associated with rip currents. The flow is caused by an oblique angle of wave approach, an alongshore variation in wave height, or both. Despite these variations in their origins and form, longshore currents retain several basic attributes of importance to coastal geomorphologists. The role in sediment transport has particular interest, because the advection of sediments by a longshore current is fundamental to the development of different beach morphologies. Reliable models of the behavior of longshore currents are central to an understanding of nearshore sediment transport. Much research has been devoted to the description of these currents and to the development of theoretical explanations of their generation. There have been four main approaches to the modeling of longshore currents: energy and momentum equations (Putnam, Munk, and Traylor 1949), mass-continuity equations (Brunn 1963), empirical equations based on statistical approaches (Harrison and Krumbein 1964; Nummedal and Finley 1979), and radiation-stress equations (Komar and Inman 1970; Longuet-Higgins 1972; Sherman 1982). Although the radiation-stress equations are also momentum based, the theory differs radically from the traditional momentum approach. At present, most attention remains focused on the use of the radiation-stress concept to model the forcing of longshore currents. The purpose of this article is to discuss the radiation-stress-based models of Longuet-Higgins (1972) and Komar and Inman (1970) as well as a modification of the Longuet-Higgins equation (CERC 1984) and to assess their efficiency in replicating prototype measurements of longshore-current velocities. * Fieldwork for this study was funded by a grant (principal investigator, Karl Nordstrom) from the National Park Service. Data analysis was supported by National Science Foundation grant SES8521344 to Mr. Sherman. * Department of Geography, University of Southern California, Los Angeles, California 90089-0663. This content downloaded from 207.46.13.115 on Sat, 08 Oct 2016 05:02:06 UTC All use subject to http://about.jstor.org/terms LONGSHORE-CURRENT MODELS SIMPLE LONGSHORE-CURRENT MODELS Bowen (1969) was first to describe the manner in which radiation-stress gradients might force nearshore circulation. Longuet-Higgins and Stewart (1964) defined radiation stress as "the excess flow of momentum due to the presence of the waves." They show that the radiation stress for linear waves propagating normal to the shoreline in shallow water is a proportion of the wave energy