A supplemental illustration of the generalized effect of turbulent diffusivity diurnal variation is depicted in Fig. 4, where frictional decoupling of the 950–800-hPa wind after sunset cause the simultaneous streamwise acceleration of the flow over large areas across the domain (Fig. 4d), whereas the activation of turbulent mixing during daytime induce the overall deceleration of the wind (Fig. 4a, b). In a like manner, the diurnal evolution of the potential temperature contours and mean wind speeds at different locations (Fig. 6, 8, and 10) show that once the OLLJ has developed during nighttime (Fig. 6d–f, 8d–f, and 10d–f), the growth of the daytime convective boundary layer erodes the LLJ from below diminishing its mean wind speed (Fig. 6a–c, 8a–c, and 10a–c).

Interestingly, despite the turbulent diffusivity diurnal variation triggering the inertial oscillation, so causing the clockwise rotation of the horizontal wind at each of the core locations (not shown), the Blackadar mechanism does not strictly apply to all of them. The 950–800-hPa streamwise and crosswise diurnal cycle of the mean horizontal, geostrophic, and ageostrophic wind (Fig. 15), for the same locations as in Fig. 14, show that the maximum streamwise wind speed is the result of the combined streamwise geostrophic and ageostrophic winds at the C1 and C3 regions only (Fig. 15a, c), whereas the maximum wind speed in the C2 and C4 regions (Fig. 15b, d) results from a balance between the geostrophic and ageostrophic winds. In general, the horizontal wind in the C2 and C4 regions reflects such balance during the entire diurnal cycle. Moreover, the minimum wind speed is subgeostrophic at the C1 and C2 regions only, whereas it is supergeostrophic in the C3 and C4 regions. Strictly speaking, the minima wind speeds should be subgeostrophic everywhere, according to Blackadar’s theoretical predictions.

Hence, the diurnal variation of turbulent diffusivity and subsequent inertial oscillation contribute to the OLLJ formation, although it plays a secondary role, especially close to the equator where the Coriolis parameter is so small that ageostrophic effects (e.g., changes in PGF due to terrain heating) drive the wind. The streamwise ageostrophic wind is, in general, of higher magnitude than the streamwise geostrophic wind in the C3 and C4 regions (Fig. 15c, d).