Association of low-middle level winds with March-May seasonal rainfall over Rwanda

Abstract

The occurrences of low middle-level winds and rainfall extremes in Rwanda cause severe socioeconomic hardships to individuals, communities and the government. Rainfall early warning can be used to mitigate and maximize on the adverse impacts. In this study, attempts are made to investigate the variability of low-middle levels winds over Rwanda and their linkage to extreme rainfall in the March-May season in Rwanda. This research aims to examine the atmospheric Patterns linked with rainfall anomalies during March-April-May (MAM) season. Monthly rainfall (mm) from Rwanda Meteorological Agency and IRI ECMWF wind reanalysis data at 850mb (low level) and 500mb (Middle level) for the period from 1981 to 2016. Time series analysis of monthly rainfall, inter-annual variability, composite analysis, rainfall anomaly indices, wind vector analysis and correlation methods were used. The inter-annual variability average indicates that the major wettest years were 1997,2006,2002,2003 and 2004 while the driest years were 1984, 2008, 2009 and 2014. During MAM, mean rainfall increases from eastern (Nyagatare, Gatsibo, Ngoma and Rwamagana) to western (Nyamasheke, Rusizi) with a range of between 327mm to 478mm. Much of the rainfall is observed over the highlands. Mean wind vector and streamlines show strong easterly winds of 5m/sec at middle level and relatively weaker at low level with speeds of up to 1.6m/sec over Rwanda. At low level, south easterlies and north westerlies converge at low levels over Lake Victoria and diverge at middle level. During the wets years (1997 and 2006), stronger easterlies were observed with internsity of 5.5m/sec at middle level whereas during dry years (1984, 2014), stronger easterlies were observed at middle level and weak (2m/sec) at low level. Wind anomalies were found to be westerly during wet years and easterly during dry years at middle level. The study found a negative correlation (< 0.5) between rainfall and wind speed and highest in Nyagatare (0.55) and the lowest in Byimana (0.10). The study recommends that in depth analysis be carried out with wind and derived circulations at several levels associated with depressed and enhanced rainfall during MAM season in addition to the use of derived circulation properties such as divergence/convergence and vorticity.