West African monsoon (WAM) diurnal cycle has a major role in the monsoon water budget as well as those of aerosols and trace gases (Parker et al. (2005)). One of the main features of the diurnal cycle is the decrease of turbulence at night leading to the formation of a nocturnal low-level jet (NLLJ). The NLLJ has been described as the major responsible for moisture transport to the northern Sahel. A comparison between model and observations in the NBL showed that the Met Office Unified Model underestimates the wind speed (including the NLLJ) in the NBL as well as the moisture flux (Bain et al. (2010)). Surface energy balance is also poorly represented in current model (Milton et al. (2008), and Haywood et al. (2005)). This error is mainly due to cloud radiative forcing errors. Indeed, the formation of low-level clouds (200m-400m) (Schrage et al. (2007)) which reduce solar radiation are not taken into account by the model leading to this error. It could represent up to 90W m-2 in the mean daily surface solar radiance in this region in global climates models (Knippertz et al. (2011)). The formation of these clouds has been identified as a consequence of the wind shear turbulence underneath the NLLJ which mixes moist surface air upward generating clouds. The low level cloud layer could last the whole morning before being erode by the sun (Knippertz et al. (2011)). Further studies have been made to explain the parameters acting on the formation and the maintenance of these clouds. The low-level stratus layer is a consequence of turbulent vertical mixing, advection of cool air, and forced lifting from the mountain winds. Each parameter contribution in the cloud formation process depends with the distance to the coast. Small changes in turbulent fluxes and advected air masses lead to bias in the cloud cover, affecting SW radiation balance and temperature at the surface (Schuster et al. (2013) and Schrage and Fink (2012)).
Different mechanism where proposed to explain the formation of a LLJ. The oldest one was introduced by Blackadar in 1957 (Blackadar (1957)). It infers that a wind inertial oscillation induced by a disequilibrium between pressure gradient and Coriolis force is at the origin of the jet. This imbalance is the consequence of the decoupling of the residual boundary layer from the surface boundary layer. In regions where there is a change in surface characteristics such as coast lines the difference of sensible and latent heat flux produces strong low-level baroclinicity. In this situation the jets are parrallel to the low-level horizontal temperature gradient. These jets are considered as LLJs in region where fluxes have a diurnal component (Stensrud (1996)). It suggests that LLJs have different behaviour regarding to the distance to the coast and the sea breath strength. Aged convective cold pools can as well trigger NLLJ (Knippertz and Todd (2012)). Cold pools can glide over a stable boundary layer leading to local pressure gradients triggering the NLLJ (Heinold et al. (2013)).
2 NLLJ inertial oscillation conceptual models
To identify the origin of the jet, conceptual inertial oscillation models were run. Blackadar developed a model (named here B57) for the development of low level jets due to inertial oscillations assumes that the geostrophic winds perform an oscillation with a period 2π/f (where f is the Coriolis parameter) and the amplitude around the ageostrophic velocity component at sunset. However, friction being largest near to the ground causes the ageostrophic component to become too large to apply the B57 model at the NBL's bottom (Van de Wiel et al. (2010)). Van de Wiel et al. proposed a model extension of the B57 model where he includes an equilibrium friction (i.e the friction if the equilibrium between the geostrophic wind and the friction force was made and constant with time).
Comparison between these run and ceilometer results in the morning will show the consequences of the inertial oscillation on cloud formation.
3 Cold pool outflows Convective activity is chase through satellite imagery. Stability and speed profile Comparison between nights with and the one without cold pools will be used to describe the modification into the inertial oscillation induced by the cold pool. The understanding of the cold pool NLLJ and the inertial NLLJ will be used to produce a Low level clouds predictive model based on nocturnal condition.