With climate change said to be affecting the intensity of rainfall, experts at the University of Pretoria (UP) have investigated if there are observable changes in the probability of significant to extreme daily rainfall across South Africa.

This research is especially relevant, as many areas in the country are prone to localised flooding. UP researchers found that the amount of rainfall in certain parts of South Africa have increased or become more extreme over the past 50 years or so.

These investigations were part of a study led by Charlotte McBride of the South African Weather Service and a PhD candidate in UP’s Department of Geography, Geoinformatics and Meteorology.

McBride explains that one of the consequences of human-induced climate change is the increase in frequency and/or intensity of some weather and climate extremes. These can include heatwaves, heavy rainfall, floods, droughts and tropical cyclones.

“South Africa is currently focused on flooding, as this is what is happening at the moment,” she says. “However, during 2018 and the Day Zero debate – when Cape Town was set to run out of water – the public’s attention was focused on drought. South Africa has a variable climate, with droughts and floods a common feature of this variability.”

As surface temperatures increase due to climate change, so the water content of the atmosphere changes. These increases in the water-holding capacity of the atmosphere equate to about 7% per degree of warming.

With more moisture available, the nature of rainfall events is likely to become more intense with increased rainfall rates. However, changes in extreme rainfall patterns are thought to be highly regionalised.

South Africa’s annual rainfall distribution is diverse, and increases from below 200mm in the west to above 1 200mm a year in the east. To investigate if rainfall patterns are changing, the researchers analysed the daily time series of 70 manual rainfall stations between 1921 and 2020; this was divided into an early subperiod (1921 to 1970) and a later period (1971 to 2020). After comparing the rainfall figures in these two periods, it became evident that most rainfall stations showed an increase in their 1 – 50- and 1 – 100-year return period values.

The two periods had more or less the same number of rain days (more than 1mm), but the rainfall amounts on any given rain day for certain parts of the country increased or became more extreme in the latter period.

“We then mapped the change between the two periods as a ratio for each station,” McBride explains. “This gave us an idea of where areas are experiencing higher or lower rainfall values for the specific return periods.”

The value for the second period was divided by the first to produce the difference as a ratio between the 1 – 50-year return period (A) and 1 – 100-year return period (B). The blue areas indicate where the return period values have become much greater over the analysis period (1921 – 2020). Brown areas show slight decreases. Source: Weather and Climate Extremes

Some stations over the eastern parts showed increases of more than 100mm in the later period compared to the early period, when considering the 1 – 50- and 1 – 100-year return period values. For example, the Letaba district rainfall station in Limpopo experienced an increase more than 35%. Another example is Hlobane in KwaZulu-Natal, where the 1 – 50- and 1–100-year values have essentially doubled.

This means that these areas and others highlighted in the research, such as the western interior and southern parts of the country, are likely to experience more extreme rainfall, which is probably a “feature of climate change over those areas”.

With reference to the blue areas in maps A (1 – 50-year return period) and B (1 – 100-year return period), a figure greater than 1.5, for example, indicates an increase in the return period value of about 50% or more.

“South Africa is projected to become warmer and thus experience an increase in the occurrence of droughts,” McBride says. “However, this does not mean that the risk of severe storms – including tropical cyclones and intense thunderstorms – will not be expected to occur. With the atmosphere heating, it can hold more water vapour. More water vapour means more rainfall. So we can expect the intensity of
rainfall to increase.”

“There is a lot of work being done internationally and in South Africa on climate change using Numerical Weather Prediction data,” says Professor Liesl Dyson, Associate Professor in Meteorology at UP. While these products are of great value and provide insight into what could happen in a future climate, they remain proxies of reality.

“The value of the presented research is that it uses real, observed data of rainfall over South Africa for an extended period,” she adds. “These results are based on what we know has happened over the past century, and show that in general, rainfall extremes are becoming more probable and therefore increasing in South Africa. Researchers who are using Numerical Weather Prediction data could also make use of these results to verify that the models they employ capture the current situation accurately, thereby placing a higher value on the reliability of projections in the future.”