CLIMATE CHANGE IMPACTS ON RAINFALL DISTRIBUTION AND NUMERICAL STUDY OF MESOSCALE CONVECTIVE SYSTEM DEVELOPMENT OVER WEST AFRICA

Abstract

This study investigated the possible climate change impacts on seasonal and annual rainfall distribution and its factors (onset, cessation and length of rainy season) as well as the numerical study of mesoscale convective system (MCS) development over West Africa. In the former, seventeen years of daily rainfall of eight RCMs from the ENSEMBLE project, eleven years of daily rainfall from Global Precipitation Climatology Project (GPCP) and monthly rainfall from Climate Research Unit (CRU) were used. Performance evaluation of the regional climate models (RCMs) capability to simulate the rainfall distribution and its factors were carried out with CRU and GPCP. The result of each RCM compared to either CRU or GPCP shows that in ascending order of good performance, ICTP-REGCM3, METNO-HIRHAM, SMHI-RCA, KNMIRACMO2.2b, and MPI-M-REMO are best and thus can be used for climate studies over West Africa. Further evaluation of the two global climate models (GCMs) used for driving the RCMs for impact studies showed that three of the identified RCMs got their initial and boundary conditions from ECHAM5 did not produce the known “little dry season (LDS)” over the affected parts of West Africa while those models forced with HADCM3Q produced the LDS. This finding suggests that HADCM3Q is suitable for initializing RCMs for climate studies over West Africa. Climate change impact on rainfall distribution and factors over West Africa under scenario A1b also shows that, in the near future (2031-2050), there will be reduction of precipitation, various degrees of delay onset dates, early cessation dates and hence reduced length of rainy season over some parts of the three major climatic zones and entire West Africa. In the latter part of the study, a software for automatically detecting and tracking convective system was developed, tested and found to be very good. This software was used to assess the capability of Weather Research and Forecasting (WRF) model in simulating (MCS). Satellite imageries from European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) was used with model simulated MCSs. It is shown that WRF perfectly simulated the initiation, growth and propagation stages as well as splitting and merging processes of MCS while structural and morphological properties at each time slice of such tracked MCS were automatically determined and stored in a text file by the software. Threshold values of some derived environmental parameters at pre-initiation, initiation and growing stages of eight MCSs occurrences were established. As simulated by WRF model, an area capable of initiating a convective cloud cluster must first be thermodynamically activated by having an outgoing longwave radiation of 260Wm-2. The following threshold values must also be simultaneously satisfied as pre-initiation conditions; a convergence (CON) of 8 to 12 (x 10-5 s-1), a moisture flux convergence (MFC) of +15 x 10-5gkg-1s-1, vertically integrated moisture flux convergence (VIMFC) of +1600 x 10-5kgm-1s-1, moist static energy (MSE) of 306-310kJ/kg, a vertical velocity (omega) of -0.4Pa/s and a dipole of relative vorticity of -4 and +4 x 10-5s-1. At initiation, the thresholds are; CON of 12.0 x 10-5 s-1, MFC of +20 x 10-5gkg-1s-1, VIMFC of +1600 x 10-5kgm-1s-1, MSE of 306-310kJ/kg and vertical velocity between -0.4 and -0.6Pa/s. The growth and propagation are also within these threshold values. These thresholds were realised in at least seven of the eight cases investigated. These results can thus guide short-range forecast of where and when an MCS will be initiated over West Africa.