Daytime Convective Boundary-Layer Evolution on Three Fair-Weather Days in CASES-97

The daytime evolution of the fair-weather convective boundary layer (CBL) over a 60-km triangle enclosing the lower part of a watershed with rolling terrain and heterogeneous land cover is described for three days in the 1997 Cooperative Atmosphere-Surface Exchange Study (CASES-97). Based on 2-m data, morning transition isn’t complete until about 1500 UTC (0900. LST), when early morning horizontal variability of temperature and mixing ratio (driven by terrain and wind) reaches its daytime values and surface virtual-temperature fluxes are positive at all surface sites (driven by variability in Bowen ratio, largely associated with vegetation). Considering horizontal variability in CBL depth h₁ delays morning transition to as late as 1600–1700 UTC. Afternoon-evening transition starts as early as 2100 UTC but is significant after 2300. Hourly budgets of domain average CBL θ and q reveal changes mostly from flux divergence in the morning, with advection and then radiative heating becoming relatively more important as all decrease through the day. All-day (1400–2300 UTC) budgets using surface fluxes yield from their residuals relatively large θ_v fluxes at CBL top (w′θv′¯)h1 compared to surface values on the windier days, but uncertainty precludes meaningful hourly flux values. Thus, hourly (w′θv′¯)h1 estimates are based on a jump entrainment model, which results in -(w′θv′¯)h1 around 0.1–0.4 times its surface counterpart (w′θv′¯)s during the times of strongest fluxes (albeit with some uncertainty). Moreover, afternoon -(w′θv′¯)h1 persists between 2130 and 2300 UTC on 10 May (mean vertical velocity W(h₁) < 0) but decreases on 20 May (W(h1>0)); a difference possibly related to radiative cooling in the upper CBL on 10 May but not on 20 May.