“…The corresponding condition on z / L is that we must have (10) Thus the upper bound on zlL will not be a constant but will depend upon the height range 0 < z < zI, in which the profile is to be used. If we accept Webb's (1970) value of 5.2 for a we have (11) Taking z, = 1 cm and zr = 4 m gives zr/L 6 0.575. Table 1 contains upper limits for a range of values of z&,, and also the corresponding lower limits for L. Typical values of L for Webb's data are in the range 4 to 100 m whilst Zrlz, is typically 2,000.…”