Tuesday, 11 May 2010
Arizona Ballroom 7 (JW MArriott Starr Pass Resort)
Handout (1.3 MB)
The rainfall diurnal cycle of the African Monsoon is one of the most intriguing physical features of tropical convection along with its phase locked occurrence. Global models do not reproduce its spatial and temporal characteristics, while the rainfall cycle is clearly identified by satellite-based precipitation patterns. In order to provide additional information to the dynamical description of monsoon mechanisms during the warm rainy season and a better comprehension of the convection coherence, a regional reanalysis strategy has been developed based on Regional Atmospheric Modelling System (RAMS), forced by NCEP/DOE Reanalysis (R-2) dataset. A sensitivity study has been performed with several model configurations in order to analyse the different contributions of principal forcing mechanisms acting on the phase locked rainfall diurnal cycle. Six (2004-2009) extended boreal summer seasons, from April to November, have been simulated, testing two model grid spacing resolutions (i.e., at 30km and 60km), and two different sea surface temperature datasets (i.e., HADSST at 1°x1° of horizontal resolution and the 8 days MODIS SST at 36km x 36km of horizontal resolution). The proposed regional modelling strategy has been evaluated analysing atmospheric fields with respect to NCEP/DOE Reanalysis and precipitation fields, using a classical statistical skill scores analysis, with respect to the CMORPH/NOAA rainfall estimates. Furthermore footprint of RAMS convection diurnal cycle has been compared with one computed by MSG satellite based precipitation patterns. A general good accordance of spatial and temporal precipitation distribution and convection diurnal cycle has been achieved by the regional modelling strategy: the higher model resolution the better characterisation of maxima of rainfall diurnal cycle, both in time and in space; furthermore MODIS SST dataset greatly improved the overall description of convection features along the season.
![](images/flag.png)
![](images/blue_ribbon.png)