For Lagrangian measurements, several techniques have been developed, 

 tested, and used. Methods included are the use of dyes, or floats, 

 drogues, cards, confetti, and other free-floating objects. While drogues 

 are designed to eliminate measurement of all but the mean fluid motion, 

 dyes on the opposite end of the scale respond to microfluctuations in the 

 movement of water masses, and thus are true tracers of water particles 

 themselves. However, their use requires evaluation of both the advective 

 motion and the diffusive and dispersive properties of water masses. 



a. Rationale and Need . Dyes have been used as tracers of the quan- 

 tity V^, probably because of the difficulties experienced in designing 

 drogues which readily responded to wind drag and wave motion. Although 

 some of the reasoning for this and some of the techniques have been re- 

 ported in Teleki, White, and Prins (1973) and Teleki and Prins (1973), it 

 is best to enumerate the conceptual model required to carry out combined 

 Eulerian- Lagrangian experiments. 



Nearshore waters are generally turbulent which means considerable 

 mixing (momentum exchange) is taking place throughout the water column. 

 In this environment, simple measurement of the flow with a fixed probe 

 cannot provide an estimate of the eddy diffusion coefficient, E^, which 

 is the measure of dispersion of mass resulting from the turbulent veloc- 

 ity fluctuations and can be written as : 



3Ci 



E i37T = V i C l for C 1 = x > Y> z ) 

 Ei is related to the dispersion coefficient in: 



3Ci _ -tt-tt . „_ . _ , 3C 



f = u i c 'i + ( k i + c JT > 



where C^ = C^ - C^' is the deviation of the concentration of a conservative 

 tracer from the mean, U^ = u^ - u^ is the deviation of the velocity from 

 the mean, k^ is the molecular diffusion coefficient, and Ei is the 

 diffusion coefficient. The Eulerian velocity, u^, is used in the above 

 expression because its statistics are better known than that of V^. 



Fixed-probe measurements provide an easier solution of a given algo- 

 rithm than a probe moving with the velocity Vi , because the techniques 

 developed for the statistical manipulation of time-series data are well 

 understood, e.g., compared to the Lagrangian integral scale in which such 

 quantities as the mixing length and dissipation rate are needed but are 

 difficult to estimate. 



Near the coast and especially near engineering works, there may be an 

 interest in what is referred to as "general circulation." This is a series 

 of synoptic snapshots of currents past an object designed to train, divert, 

 or obstruct the same flow, or the sediment it carries. The engineer is 

 asked to evaluate the performance of the structure for which he needs to 

 know what the areal variations in waves and currents are. Measurement of 

 velocities in the Lagrangian domain is an answer. Unfortunately, present 

 techniques are designed for surface measurements only and not for the flow 



58 



