Initiation of deep convection during the early -monsoon sahelian convective boundary layer: an observational study

The initiation of deep convection in Sahel remains poorly understood, and still difficult to model. The aim of our study is to better understand the transitions from dry convective boundary-layer to shallow convection, and from shallow to deep convection, by use of the AMMA dataset. Based on the MIT radar, UHF wind profiler, satellite data, surface flux and meteorological stations in the area of Niamey, we have characterized the atmospheric convection that occurred over Niamey during the period of 6-31 July 2006, and investigated the mechanisms responsible for deep convection initiation.

For this period of Monsoon pre-onset and onset, we have classified the type of convection observed each day within a radius of 100 km of the MIT location. Four types of convection have been identified: (1) dry convection all day, (2) shallow convection, (3) afternoon deep convection initiated locally and (4) propagating deep convection. In the early morning, the boundary layer convective structures are organized in bands for all cases of classes (1), (2) and (3), except one. Cells or heterogeneous convective structures are observed for class (4). Convergence lines are frequently found to be precursors of local deep convection at the beginning of the monsoon. We found 3 days where deep convection was locally triggered on a convergence line.

All cases of locally-initiated deep convection except one generated a circular gust front. The density currents are almost always involved in the generation of new cells that have a life span of about one hour. For two days, we estimated the wake available potential energy and demonstrated the capability of wakes to trigger deep convection.

Neither early morning CAPE alone, nor other convective indexes such as the CTP (Convective Triggering Potential) alone allow to correctly predict the onset of deep convection. Indeed, CAPE and CTP are favourable to deep convection most of the time, while convective inhibition is typically quite large.

Our results show that the daytime growth of the atmospheric boundary layer plays a key role on the occurrence and type of moist convection observed during that period. Land-surface temperature and moisture heterogeneities also appear to play an important role, that we will further study.