TABLE OF CONTENTS In Section 11.7.3 we saw that for longshore currents on a straight beach the lateral mixing coefficient could be written
| (13.1.1) |  |
where index x refers to mixing in the cross-shore direction, and it was found that a C x-value of 0.45 1.80 is required to obtain cross-shore variations of the longshore current that resembled the measured variations. The order of magnitude of this number has been verified repeatedly in the literature.
On the other hand the vertical variation of the cross-shore undertow profile investigated in Section 12.2 used a v t, which can be written as
| (13.1.2) |  |
requires a C z-value of only 0.01 0.02, at most, to achieve similarly reallistic comparisons to measured undertow velocities. Again this number has been verified in the litrature by several authors using different models.
Thus, there is an order of magnitude difference between the v t required to predict the measured cross shore variation of longshore currents and the measured vertical variation of undertow, respectively.
It is obvious that such a discrepancy is disturbing from a physical point of view. A natural hypothesis is to assume that the mixing is caused by the turbulence which is abundant due to the wave breaking. We may then express v t as
| (13.1.3) |  |
where k is the turbulent kinetic energy and ? a characteristic length scale for the turbulence. Based on measurements, Svendsen (1987) estimated 
, a result which was based on many laboratory experiments by independent authors.
