Abstract:
Severe weather events associated with strong winds and flooding can cause fatalities, injuries and damage
to property. Detailed and accurate weather forecasts that are issued and communicated timeously, and
actioned upon, can reduce the impact of these events. The responsibility to provide such forecasts
usually lies with government departments or state-owned entities; in South Africa that responsibility lies
with the South African Weather Service (SAWS). SAWS is also a regional specialised meteorological
centre and therefore provides weather information to meteorological services within the Southern
African Development Community (SADC). We evaluated SAWS weather information using near real-time
observations and models on the nowcasting to short-range forecasting timescales during two extreme
events. These are the Idai tropical cyclone in March 2019 which impacted Mozambique, Zimbabwe and
Malawi resulting in over 1000 deaths, and the floods over the KwaZulu-Natal (KZN) province in April 2019
that caused over 70 deaths. Our results show that weather models gave an indication of these systems
in advance, with warnings issued at least 2 days in advance in the case of Idai and 1 day in advance for
the KZN floods. Nowcasting systems were also in place for detailed warnings to be provided as events
progressed. Shortcomings in model simulations were shown, in particular on locating the KZN flood
event properly and over/underestimation of the event. The impacts experienced during the two events
indicate that more needs to be done to increase weather awareness, and build disaster risk management
systems, including disaster preparedness and risk reduction.
Significance:
This paper is relevant for all South Africans and the SADC region at large because it provides information
on:
• the weather forecasting processes followed at the South African Weather Service,
• available early warning products in South Africa and for the SADC region made possible through the public purse,
• the performance of nowcasting and modelling systems in the case of predicting two extreme weather
events that had adverse impacts on southern African society, and
• the dissemination of warnings of future extreme weather events.
