Understanding climate risks for water and agri-food systems: implications for smallholder farmers

Abstract

South Africa (SA) is projected to become warmer and drier, with increasing frequency and intensity of extreme climatological and hydrometeorological events. The changing climate affects the agri-food systems, compounds water insecurity, and impacts socio-economic and environmental systems. This study analyses climate risks on water and agri-food systems in three provinces, representing three Hydro-Climatic Zones (HCZs) of SA. The study used the Climate Moisture Index (CMI) to assess water stress and the Thermal Heat Index (THI) to identify heat stress across different HCZs and time periods. Past and projected rainfall and air temperatures, described as historical (1960–1990), present (2015–2044), and distant future (2070–2099) time horizons, were used to compute the CMI and THI. More severe changes in the CMI and THI, indicative of drier, hotter climates, are observed in HCZ 1. The results indicate a heterogeneous decrease in rainfall across the study sites alongside distinct increases in temperatures. These conditions indicate an increasing likelihood of water and heat stress, which will exceed the optimal thresholds for crops. Understanding the severity of climate risks is critical to mitigating their impact on socio-ecological and agri-food systems. The study’s results highlight the need for context-specific, transformative adaptive strategies. Simplifying crop stress indices and prioritising the adoption of crops with high thresholds to water and heat stress will support sustainable natural resource management and strengthen the resilience of agri-food systems, particularly in marginalised environments.

1 Introduction

Climate change exacerbates extreme events, affects agri-food systems, and compounds water risks across Sub-Saharan Africa (SSA), a region identified as a hotspot for climate-related vulnerabilities and risks (Edokpayi et al., 2020; Nhemachena et al., 2020; Engelbrecht et al., 2024). Climate change manifests as pronounced rainfall variability, rising temperatures, and other extreme weather events, a combination of the most limiting factors for food and water security, socio-economic and environmental systems in the region at large (Engelbrecht et al., 2015; Nhemachena et al., 2020; Edokpayi et al., 2020; Ayanlade et al., 2022; McBride et al., 2022; Nhamo et al., 2024; Nyoni et al., 2024; Touch et al., 2024). SSA’s vulnerability is further exacerbated by widespread poverty, limited adaptive capacity, and an extensive reliance on rainfed farming systems, particularly among marginalised smallholder farmers, exposing them to pronounced disruptions (Davis-Reddy and Vincent, 2017; Zenda et al., 2024; Ogundeji, 2022; Sikhwari et al., 2022; Ncoyini-Manciya and Savage, 2022; Okolie et al., 2023; Johnston et al., 2024).

Recent research indicates that between 2014 and 2023, climatological and hydrometeorological extremes, including record-breaking global temperatures, intensified storms, floods, droughts, and heatwaves, have become more frequent, worsening environmental and livelihood crises in the region (Davis-Reddy and Vincent, 2017; Ayanlade et al., 2022; Gosset et al., 2023; Franchi et al., 2024; Rouault et al., 2024; NOAA, 2024; World Meteorological Organization, 2024). Despite agriculture and water sectors being central to food security, livelihood, and economic stability in SSA, underdevelopment and climate-related stressors have weakened their performance, impacting over 33 million smallholder farmers who provide 70% of the region’s food supply (Mabhaudhi et al., 2018; Bjornlund et al., 2020; Ringler et al., 2022; Kapari et al., 2023; Maguire, 2024). Given that 95% of agriculture in the region is rainfed, sustainable natural resource management and climate-resilient strategies are essential for reducing risks and safeguarding the well-being of both humans and the natural environment (Biazin et al., 2012; Nhemachena et al., 2020; Nyoni et al., 2024).

Alongside the pronounced susceptibility to high potential evapotranspiration, South Africa (SA) is characterised by spatial rainfall variability and faces unprecedented hydrometeorological and climatological extremes with significant impacts (Edokpayi et al., 2020; Du Preez and Van Huyssteen, 2020; Scholes and Engelbrecht, 2021; McBride et al., 2022; Engelbrecht et al., 2024; Nhamo et al., 2025). The country is classified as semi-arid to arid, with common occurrences of dry spells and droughts that are often reinforced by the El Niño, primarily in the western parts (Rouault et al., 2019; Botai et al., 2021; Chikoore and Jury, 2021; de Perez et al., 2024; Letswamotse et al., 2024; Opiyo et al., 2025). While these occurrences can be linked to specific weather and climatic phenomena, they are increasing in frequency and intensity (Letswamotse et al., 2024; Opiyo et al., 2025). Drought and dry spells have immeasurable consequences for water resources, water quality, and availability, and, in turn, agricultural production and productivity, as well as socio-economic and environmental sectors in the country (Scholes and Engelbrecht, 2021; Department of Water and Sanitation, 2024; Opiyo et al., 2025).

SA is also prone to localised flooding due to the increasing frequency of extreme rainfall events (McBride et al., 2022; de Perez et al., 2024; Department of Water and Sanitation, 2024). The eastern side of the country, along the Indian Ocean, frequently experiences flash flooding and extreme rainfall events (Mashao et al., 2023; Moses et al., 2023; Eloff, 2024; Stoddard, 2025), especially in the recent decade. These typically occur in late summer-early autumn and are often associated with rain-bearing pressure systems such as the cut-off low and La Niña (Rouault et al., 2019; Mashao et al., 2023; Moses et al., 2023; de Perez et al., 2024; Stoddard, 2025). Extreme rainfall events are further projected to increase in frequency and intensity in some localities in South Africa (McBride et al., 2022).

On the other hand, parts of the country are projected to become warmer and drier in the future (Scholes and Engelbrecht, 2021; Engelbrecht et al., 2024). However, there is less agreement over the eastern escarpment areas of SA, where some models project increases in total rainfall (Lee et al., 2021; Abubakar et al., 2025). Increases in atmospheric temperatures are linked to the intensification and frequency of both drought and floods (Scholes and Engelbrecht, 2021; McBride et al., 2022; Pizzorni et al., 2024). The frequency and intensity of extreme temperature events, such as heatwaves, are projected to increase in South Africa due to anthropogenic climate change (Scholes and Engelbrecht, 2021; Engelbrecht et al., 2024). On the contrary, cold spells are expected to occur less frequently (Scholes and Engelbrecht, 2021). The increasing warming and aridity in the country will negatively impact people, livestock, and crop production, which rely on sensitive temperature thresholds for well-being, production, and productivity (Scholes and Engelbrecht, 2021; Engelbrecht et al., 2024).

Extreme weather events pose a significant risk to people and livelihoods by adversely affecting water resources, agri-food systems, environmental sustainability, and socio-economic systems. Together, they compound climate risks for resource-constrained smallholder farmers in marginal and underperforming agri-food systems, and, as a result, poverty and hunger (Harvey et al., 2014; Bjornlund et al., 2022; Williams et al., 2018; Giller, 2020; Johnston et al., 2024; Archer et al., 2018; Bjornlund et al., 2020; Engelbrecht et al., 2015, 2024; Scholes and Engelbrecht, 2021). Hydrometeorological and climatological extreme events are projected to intensify in SA, which makes addressing these risks a critical developmental and climate action requirement.

Although widespread literature exists on climate risks and their impacts on water and agriculture in SA (Archer et al., 2018; Nhemachena et al., 2020; Meza et al., 2021; Engelbrecht et al., 2024; Zenda et al., 2024; Nhamo et al., 2025; Johnston et al., 2024), most of these studies are conducted at a regional and national level. However, climate risks manifest disproportionately due to varying geographical, social, and economic vulnerabilities, posing significant risks to local communities, particularly smallholder farmers. This study examines climate risks and their effects on water and agri-food systems among smallholder farmers in selected Hydro-Climatic Zones (HCZs) of South Africa (HCZs 5, 2, and 1, represented by the Eastern Cape, KwaZulu-Natal, and Limpopo provinces). This is achieved by examining projected climate risks and impacts on water and agriculture using the Climate Moisture Index (CMI) and the Thermal Heat Index (THI). The study also explores ways to enhance climate action and adaptation planning to improve water and food security. The study provides insights to inform climate action in the water and agricultural sectors, as well as adaptation planning at appropriate scales.

2 Methods

2.1 Description of study areas

The study focuses on three HCZS in three provinces of South Africa, including HCZs 1, 2, and 5, as detailed by the Department of Water and Sanitation, South Africa (2022). HCZs are spatial units defined by a combination of factors, including current water-related climate features, such as seasonality, interannual variability, and the magnitude of variables such as rainfall and PET (Department of Water and Sanitation, South Africa, 2024). They also incorporate projected climate change signals and administrative boundaries (WMAs/municipalities) for water resources planning (Department of Water and Sanitation, South Africa, 2024). The study reports on Limpopo as HCZ 1, KwaZulu-Natal as HCZ 2, and the Eastern Cape as HCZ 5 (Figure 1). The three HCZs exhibit distinct geographical characteristics that contribute to their unique climate and weather patterns, thereby delineating specific climate regions, as noted by Kruger (2004). Their differing characteristics indicate that the HCZs will face distinct, context-specific climate change risks and impacts, requiring tailored mitigation and adaptation measures. Hydrological, meteorological, and climatological disasters are becoming a prominent risk to local agri-food systems, food and water security, and the environment across the three provinces (Ndlovu et al., 2021; Thoithi et al., 2022; Sikhwari et al., 2022; Mudefi, 2023). Since climate change impacts are multifaceted, they require context-specific solutions that strengthen localised coping and adaptation strategies (Braunschweiger and Ingold, 2023).

Figure 1

All HCZs experience specific wet, warm, dry, and cooler periods, corresponding to the summer and winter seasons. The Eastern Cape (HCZ 5) is a coastal province located at the southern boundary of the KwaZulu-Natal province, on the eastern south-central coast of South Africa. The province encompasses all seven ecological zones, highlighting the province’s extensive climatic and vegetative diversity. The province receives between 300 and 550 mm of rainfall annually, bi-modally (summer and winter). It has daily maximum air temperatures of up to 40 °C in summer and a mean annual temperature of 17.6 °C (Botai et al., 2021). Major agricultural activities include citrus (lemons, oranges), pineapples, deciduous fruit, chicory, and subtropical fruit, whilst field crops include maize, sorghum, soybeans, and sunflower seeds, while wool and mohair are significant farming products.

KwaZulu-Natal (HCZ 2) is a coastal province on the southeastern coast of South Africa, bordering the Indian Ocean. The province receives approximately 650 mm of annual rainfall in the eastern Grasslands and 900 mm in the central Bushveld, while the Coastal Bushveld receives approximately 1,400 mm (Ndlovu et al., 2021). During the austral summer months, from October to March, most rainfall occurs in the province (Rouault et al., 2013; Ndlovu et al., 2021). The province is a major agricultural hub in South Africa, dominated by large-scale sugarcane production, which accounts for roughly 80% of the country’s crop. Other key crops include timber, maize, soybeans, and subtropical fruits such as pineapples and bananas. The province is quickly diversifying its agriculture by exporting high-value crops such as macadamia nuts and avocados.

Limpopo, representing HCZ 1, is located in the northeastern part of South Africa, bordering Mozambique, Zimbabwe, and Botswana. The province experiences warm summer temperatures, averaging 27 °C and occasionally reaching 40–50 °C in some areas. The rainfall season in Limpopo begins in early or mid-summer, around mid-December, and lasts an average of 4 months, with an annual average of 500 mm (Cai et al., 2017; FAO, 2024). The province’s annual evaporation ranges from 1,600 mm to 2,600 mm/year, depending on local temperatures (FAO, 2024). Limpopo is an agricultural powerhouse in South Africa, responsible for a significant portion of the nation’s fruit and vegetable production, particularly tropical fruits, citrus, and tomatoes. Major crops include tomatoes (nearly 60% of SA’s crop), mangoes (~75%), avocados (~60%), papayas (~65%), citrus (lemons, oranges), bananas, macadamia nuts, and potatoes.

2.2 Data acquisition

Climate projections were carried out as part of the Water Research Commission’s National Assessment of Potential Climate Change Impacts on the Hydrological Yield of Different Hydro-Climatic Zones 9 (HCZs) of South Africa (Schütte et al., 2023). The rainfall and air temperature change projections are presented for three periods: present (1960–1990), present (2015–2044), and distant future (2070–2099). These changes are projected using an ensemble of six different computational Global Climate Models (GCMs): the ACC, CCS, CNR, GDF, and MPI (Table 1), at Quaternary Catchment resolution. The projections were forced by the low mitigation scenario (RCP 8.5) and downscaled using the regional climate model, the Conformal Cubic Atmospheric Model (CCAM) from CSIRO. The CCAM has locally calibrated physics options for SA and is a flexible and unified downscaling system. The chosen pathway, the RCP 8.5, remains the most feasible and realistic pathway for producing results that reflect the sensitivity of different systems beyond the mid-century. The projections were bias-corrected against historical observed data from 1961 to 1999 following the Quantile Delta Mapping method (Cannon et al., 2015; Schulze et al., 2023).

2.3 Indices

The identification of areas with more sensitivity to heat and water stress was assessed using the THI and CMI. The two indices were better suited as proxies of moisture and heat stress in long-term projections for this study.

2.3.1 Climate Moisture Index (CMI)

The study uses the Climate Moisture Index (CMI) to assess moisture conditions, a key indicator of aridity (water stress and degree of aridity or wetness), across the study areas over different time periods. The index, as highlighted in Vörösmarty et al. (2005), Nhamo et al. (2019), and Nhemachena et al. (2020), assesses the ratio of annual precipitation (P) to annual Potential Evapotranspiration (PET).

when there is more precipitation than PET or P = PET, while

when there is more PET compared to precipitation.

PET was substituted for Actual Pan Evaporation (APAN) in this study, making our equations:

when there is more precipitation than APAN or P = APAN, while

when there is more APAN compared to precipitation.

CMI values range from −1 to +1. Positive values of the CMI suggest that the study site has a wetter climate with sufficient precipitation for healthy vegetation (Vörösmarty et al., 2005). Negative CMI values, on the other hand, indicate drier conditions and are often seen in arid/semi-arid environments (Vörösmarty et al., 2005). The CMI for each HCZ was calculated using data from a WRC study (Schütte et al., 2023). The results were interpolated in ArcGIS Pro 3.0.0 to create visual representations of the CMI results. The CMI results for three time periods are presented as: historical (1960–1990), present (2015–2044), and distant future (2070–2099).

2.3.2 Thermal Heat Index (THI)

The study used the Thermal Heat Index (THI) to assess potential heat stress in agricultural crop production across the study sites using the following equation:

Where T is the maximum temperature (in °C), and RH is the relative humidity. RH is expressed as a decimal (e.g., 75% humidity is 0.75). Temperature data used for the computation of the THI were generated from an ensemble of six GCMs: ACC, CCS, CNR, GDF and MPI for three time periods: historical (1960–1990), present (2015–2044), and distant future (2070–2099). The results were interpolated in ArcGIS Pro 3.0.0 to create visual representations of the THI outputs. Du Preez et al. (1990), Nesamvuni et al. (2012), and Mpandeli et al. (2019) define the different categories of heat stress as (Table 2):

2.3.3 Data analysis

The mean annual precipitation results are presented in two forms: relative (%) change and absolute (mm) change approaches, as defined by Baker and Huang (2012). The changes in precipitation and air temperature were projected as an ensemble of six different GCMs. They were forced by a low mitigation scenario (RCP 8.5) across 5838 sub-catchments in South Africa. Changes in rainfall and air temperature were projected from the historical (1961–1990) to the present (2015–2044), and from the historical (1961–1990) to the distant future (2070–2099), at a Quaternary Catchment scale.

3 Results

3.1 Overview of projected changes in annual precipitation across the three HCZs at a Quaternary Catchment resolution

3.1.1 Limpopo: Hydro-Climatic Zone 1

Projected absolute changes (mm) in mean annual precipitation show subtle changes in Hydro-Climatic Zone 1, represented by Limpopo, across two time periods: historical to present and historical to distant future. Variations between <20 mm absolute change and 100–200 mm reduction are observed across Limpopo in the historical-to-present simulations, while variations ranging from >20 to 50–100 mm increase are observed in the historical-to-distant-future simulations (Figure 2). Relative changes (%) predominantly show reductions of 5–10% and 10–20%, except in some areas where changes of more than 5% are expected over the historical-to-present timelines. Similar results are observed across the historical-to-distant-future timelines, except in some areas on the northeastern side of the province, where a 10–20% increase is observed at a Quaternary Catchment resolution.

Figure 2

3.1.2 KwaZulu-Natal: Hydro-Climatic Zone 2

Projections of absolute (mm) and relative (%) changes in mean annual precipitation for HCZ 2, represented by KwaZulu-Natal, for the historical to present and historical to distant future at Quaternary Catchment resolution are presented below. Projections of absolute changes show predominantly 20–50 mm reduction in much of the province, followed by <20 mm changes and exceptions of a combination of 50–100 mm reductions along the coast and some 20–50, 50–100 mm and 100–150 mm increases inland in the historical to present simulation. Similar results are seen in the historical to distant future timelines with an expansion of areas where 20–50, 50–100 mm, and 100–150 mm increases are seen both inland and in the south coast area. Projections for the same time frame show a >150 mm increase in mean annual precipitation across the Drakensberg region near the border. Projections of relative changes show predominantly <5% change and 5–10% reduction across both timelines. An exception of a 5–10% increase is observed inland towards the Drakensberg areas in the historical-to-present timelines. Reductions of 10–20% are seen in some areas in the southern inland of the province, while increases of 5–10%, 10–20 and >20% are seen in parts of the Midlands extending towards and in the Drakensberg region (Figure 3).

Figure 3

3.1.3 Eastern Cape: Hydro-Climatic Zone 5

The Eastern Cape (EC) represents HCZ 5. Projected changes (mm) in mean annual precipitation over the historical to present timeline show <20 mm change in much of the province, with some exceptions: inland areas show a 20–50 mm reduction, while areas near the Lesotho border show a 20–50 mm increase. Increases of 20–50, 50–100 and 100–150 mm inland towards the border are also observed in the historical to the distant-future timeline, while <20 mm change, 20–50 and 50–100 mm reductions are seen in much of the province, both inland and coastal areas. An exception is seen along the northern coast of the province towards the KZN border, where projections show increases of 20–50 and 50–00 mm from the historical to the distant-future timelines. Similar patterns are shown in the relative (%) change projection for both timelines. Increases of predominantly 5–10% are seen towards the Lesotho border, while a < 5% change is seen in most of the province, with some exceptions of 5–10% reductions in some inland areas in the historical to present timeline. In the historical to distant future timeline, <5% changes are predominantly observed across the province, with 5–10% and 10–20% reductions also observed inland. Some cases of 5–10% and 10–20% increases are observed along the Lesotho border, and some exceptions of 5–10% increase are seen in the province’s northern coast (Figure 4).

Figure 4

3.2 Projected changes in air temperature (daily minimum and daily maximum) in the three Hydro-Climatic Zones at a Quaternary Catchment resolution

3.2.1 Means of daily minimum air temperatures (Tmin)

Projected changes in the annual means of daily minimum air temperatures (°C) for the historical to present timelines were similar across all HCZs, showing increases of 1–2 °C (Figure 5). For the historical to the distant-future, projected changes in annual means of daily Tmin varied across the HCZs. HCZ 1, across Limpopo, showed increases of 3–4 °C and 5–6 °C in some localised areas, while increases of 4–5 °C were seen predominantly across the climatic zone (Figure 4). HCZs 2 (across KZN) and 5 (across EC) presented similar projections of 4–5 °C and 3–4 °C increases for the historical to distant future timelines. In both HCZ 2 and 5, the more extreme changes (4–5 °C) in air temperature are seen in the inland areas, while increases of 3–4 °C dominate along the coastline.

Figure 5

3.2.2 Means of daily maximum air temperatures (Tmax)

Hydro-Climatic Zone 1 across Limpopo showed more severe changes in both timelines than HCZs 2 and 5, where KwaZulu-Natal and the Eastern Cape are located, respectively. HCZ 1 showed varied results in both timelines, with daily Tmax increases ranging from 1–2 °C to 2–3 °C in the historical-to-present timeline, and more severe changes ranging from 4–5 °C to >6 °C in the historical-to-distant-future projections. Subtle changes are seen in HCZ 2 in the historical to present timeline, while more severe changes ranging from 4–5 °C to 5–6 °C are seen in inland areas, and changes of 3–4 °C are seen along the coastline in the historical to distant-future timeline. Similar to HCZ 2, changes of 1–2 °C are observed in HCZ 5 over the historical-to-present timeline. Historical to distant future projections in HCZ 5; however, show mixed results indicating moderate changes of 2–3 °C and 3–4 °C along the coastal areas and more severe changes ranging between 4–5 °C and 5–6 °C inland, with more severe changes towards the Lesotho border along the Drakensberg region (Figure 6).

Figure 6

3.3 Climate risks: CMI and THI

The CMI and THI were computed to assess the degree of aridity, moisture availability and heat stress across the HCZs (Du Preez et al., 1990; Vörösmarty et al., 2005; Nhamo et al., 2019). We present the CMI and THI results across three timelines: the historical (1960–1990), present (2015–2044), and distant future (2070–2099). The CMI results are presented as relative changes (mm) and absolute changes (mm). Results for the CMI and THI are projected using an ensemble of 6 GCMs, forced by a low mitigation scenario (RCP 8.5). The positioning of all the HCZs in different geographical locations and climatic regions contributes to their susceptibility and vulnerability to specific spatio-temporal climate risks. CMI values range from −1 to +1, with aridity indicated by negative CMI values and moisture availability indicated by positive CMI values in specific climates (Vörösmarty et al., 2005).

3.3.1 Climate Moisture Index (CMI)

The selected study sites were primarily associated with negative CMI values in the historical timeline. The CMI values averaged −0.62, −0.66, and −0.69 mm, respectively, across the historical-, present-, and distant-future at the study sites, indicating that the sites are generally dry. HCZ 5 showed outliers, with both the highest (CMI Max) and lowest (CMI Min) CMI values within a single hydro climatic zone. CMI max values across the study sites were +0.168, +0.103, and −0.008, respectively, across the timelines, and were found in inland HCZ 5 (positive values) and HCZ 2 (negative values) Sub-Catchments 4573 and 4840 (Figure 6). CMI min values, also found within the same zone, ranged from −0.929; −0.932 to −0.949 mm from the historical-, present- and distant-future, progressively increasing towards the 21st Century (Figure 7). The distribution and variability of the CMI within and across the HCZs are distinct. However, the degree of moisture availability progressed negatively (CMI −) in all HCZs, with the present period exhibiting higher moisture availability.

Figure 7

3.3.1.1 Projected absolute change in CMI

Projected absolute changes in CMI follow a trajectory similar to that of projected relative changes across the historical-to-present and historical-to-distant-future timelines. Moisture availability is projected to decrease drastically within and across the HCZs. The CMI values averaged −0.0357 in the historical-to-present projection and −0.0764 in the historical-to-distant-future timelines, exhibiting a −0.0407 change towards more aridity. CMI max values were higher in HCZ 5 (0.0148; SubCat 3635) in the historical-to-distant-future projections and in HCZ 2 (SubCat 4840) in the historical-to-distant-future projections (0.0429) (Figure 7). CMI min values, found in HCZ 1 (SubCats 5427 and 5767), ranged from −0.1299 in the historical-to-future projection to −0.2606 in the historical-to-distant-future projection, indicating increasing aridity with increasing CMI values (Figure 8).

Figure 8

3.3.2 Thermal Heat Index (THI)

3.3.2.1 Historical timeline (1961–1990)

Projections of heat stress under historical climate conditions (1961–1990) indicate varying and uneven heat stress across the study sites. Mild and severe heat stress (mean THI: 71.62 and max THI: 79.73) was prominent in HCZ 1 under historical climate conditions. Under the same conditions, the projections indicated largely mild to no heat stress in HCZ 2, with exceptions of very localised severe heat stress (THI max: 78.68) in northern KwaZulu-Natal (KZN) towards the borders with Eswatini and Mpumalanga. Projections for HCZ 5 suggest that there is primarily no heat stress in the area under current climate conditions. The mean Temperature-Humidity Index (THI) is 70.90, with a maximum THI of 76.22. This indicates mild heat stress, mainly in the region’s inland areas, though exceptions occur along the coastline (Figure 9).

Figure 9

3.3.2.2 Present timeline (2015–2044)

In the present timeline, projections indicate varying increases in heat stress across the HCZs. Mild to severe heat stress is projected in HCZ 1, with a THI max value of 81.61 and an increased mean THI of 75.61, an increase of 3.99 (5.57%) relative to the historical timeline mean. In the same timeline, projections indicate that predominant mild heat stress will be experienced in HCZ 2, peaking at THI max 79.90, with an average THI of 73.20. More severe heat stress also exists in the northern parts of KZN, bordering Eswatini, Mpumalanga, and Mozambique, as well as in some isolated coastal areas. Primarily, mild heat stress is indicated by projections for HCZ 5, both inland and along the coast. The THI max for the present timeline in this HCZ is 77.55, while the average is 71.58, indicating the lowest THI among all HCZs studied (Figure 10).

Figure 10

3.3.2.3 Distant future timeline (2070–2099)

The distant-future projections indicate increasing heat stress across all study areas. Severe heat stress is indicated prominently across HCZ 1. The THI max in this area under this timeline peaks at 83.25, and the average THI across the region is 78.91. HCZ 2 projections also signal increasing mild and severe heat stress in the region. The mean THI in HCZ 2 under this timeline is projected at 76.68, while THI values peak at 82.12. Projections for HCZ 5 also indicate increasing mild heat stress across the region. The mean THI in the region under the distant future timeline is 74.41, while the maximum THI is 79.93, indicating severe heat stress in specific isolated localities within the same HCZ (Figure 11).

Figure 11

4 Discussion

Water and crop production challenges are among the major outlets through which climate change impacts will manifest, particularly in water-scarce environments such as SA. However, the degree of climate-driven impact will differ. Recent evidence, such as that by Rouault et al. (2024) and Engelbrecht et al. (2024), highlights variations in the degree of changes in rainfall and air temperature regimes across SA. Rouault et al. (2024) further highlight the increasing influence of localised pressure systems and natural climate patterns, such as the ENSO phenomenon, on interannual variability and the onset and demise of the rainfall season. Over the past decade, the country has been subject to increasing extreme weather events, changes in rainfall distribution and variation and altered air temperatures (Ndlovu et al., 2021; McBride et al., 2022). All these changes have presented and exacerbated challenges within agri-food systems, including risks related to heat, water, and food and nutrition, particularly in smallholder farming systems (Du Preez and Van Huyssteen, 2020; Mirzabaev et al., 2023).

4.1 Rainfall and air temperature

Limpopo (HCZ 1) is among the driest provinces in the country (Phophi et al., 2020; Maposa et al., 2021; Matimolane et al., 2024; Nhamo et al., 2025) with a high risk of dry spells and droughts (Phophi et al., 2020; Maposa et al., 2021; Matimolane et al., 2024; Nhamo et al., 2025). Projected reductions in rainfall (Figure 2) indicate that intensifying dry spells and droughts will likely undermine water availability and crop productivity. In the absence of appropriate coping and adaptation strategies, this will negatively impact smallholder farmers’ livelihoods and food and nutrition security outcomes. Changes in mean annual precipitation in HCZ 2 across KZN (Figure 3) are consistent with findings by Engelbrecht et al. (2024), highlighting that rainfall increases are projected in some parts of the province. A study by Ndlovu et al. (2021) also projected a decrease in annual rainfall days along the KZN coastline, while an increase in rainy days was seen in parts of central KZN. This aligns with the study’s results, which highlight the degree of variability in the rainfall regime across the KZN climatic zones. Under HCZ 5, projections for the Eastern Cape also showed mixed rainfall trends, indicating a combination of rainfall reduction and increases across the two timelines, with unclear changes and increasing rainfall reduction primarily in the inland parts of the province (Figure 4).

Projected means of minimum (Tmin) and maximum (Tmax) daily air temperatures systematically increase across all HCZs, indicating a clear warming trend in future climates. More extreme changes in Tmax from the historical to present and historical to distant-future timelines were observed in HCZ 1 (Limpopo), followed by HCZ 5 (Eastern Cape) (Figures 5, 6). Limpopo is already at a heightened risk of heat waves and heat stress (Phophi et al., 2020; Maposa et al., 2021). The projected changes indicate a potential increase in both the intensity and frequency of heat-related disasters in this region. However, the overall study’s objective of focusing only on mean annual precipitation and mean Tmin and Tmax is limited and risks overlooking the dynamics of extreme weather events across all HCZs. Both in their increasing frequency and intensity (Engelbrecht et al., 2024; Jubase and New, 2025; Nhamo et al., 2025), extreme rainfall and temperature events drive some heat and water stress in agricultural production and risks to socio-ecological systems.

4.2 Water and heat stress indicators: Climate Moisture Index (CMI) and Thermal Heat Index

South Africa is a water-scarce country with a semi-arid climate and is characterised by high potential evapotranspiration, which poses a great threat to agricultural productivity and production under rainfed conditions (Du Preez and Van Huyssteen, 2020; Bonetti et al., 2022). The CMI and THI were computed to examine heat and water stress in agriculture across 1961–1990, 2015–2044, and 2070–2090 time periods in three HCZs. The application of the CMI and THI indices in the context of agricultural crop production ensures a multi-perspective link of agro-climatology indicators under climate change to improve food production and agri-food systems productivity in SA.

4.2.1 Degree of moisture availability and aridity: Climate Moisture Index (CMI)

The mean annual CMI for the chosen study sites ranges from −0.623 in the historical timeline to −0.659 in the present and −0.700 in the distant-future. The minimum CMI values across the study sites range from −0.929 in the historical timeline to −0.932 and −0.949 in the present and distant-future timelines, respectively (Figures 6, 7). The CMI values vary spatially and temporally across the three HCZs. However, they are primarily negative across the HCZs across all timelines, indicating moisture scarcity at the selected study sites. However, water scarcity and availability vary due to intra-annual rainfall variability, leading to moisture availability that also varies within and across the study sites. Despite this, Vörösmarty et al. (2005) classify CMI values into five distinct categories: CMI <−0.6 = is considered arid; −0.6 < CMI < 0 = signifies semi-arid; 0 > CMI < 0.25 = is classified as sub-humid, and CMI > 0.25 = denotes humid. In this context, the results suggest that our study sites are characterised by increasingly arid to semi-arid climates (Nhamo et al., 2019).

Decreasing CMI values may trigger natural feedback, such as soil degradation, reduced vegetation cover, changes in crop suitability, and challenges to the long-term productivity of agri-food systems across the study sites (Nhemachena et al., 2020; Furtak and Wolińska, 2023). The spatial heterogeneity in CMI trends across the study sites underscores a growing need for context-specific interventions in natural resource use and management. For example, decreasing CMI values, particularly in inland HCZ 5 in the Eastern Cape, which is projected to be at increased risk of day-zero-type droughts (Archer et al., 2022; Engelbrecht et al., 2024), indicate increasing aridity in the region and a need for water-adaptation plans. Similarly, the decreasing CMI values across the three timelines in HCZ 1, under which Limpopo falls, as expected, indicate increasing widespread aridity in the region under future climates and emphasise a need for context-specific adaptation plans, particularly since Limpopo is among the driest and drought-prone provinces in SA (Matimolane et al., 2024; Nhamo et al., 2025). In areas as arid as the chosen study sites, agri-food systems, which are primarily rainfed, need progressive interventions to sustain agricultural production under increasing aridity (Mpandeli et al., 2019; Nhemachena et al., 2020).

4.2.2 Degree of heat stress

The THI results indicate increasing heat stress spatially and temporally across the study sites (Table 3, Figures 9–11). Heat stress severity is primarily observed in HCZ 1 across the study sites, while the least heat stress is observed in HCZ 5. This corresponds with previous studies such as Archer et al. (2021) and Jubase and New (2025), which highlighted that Limpopo is among the region’s most vulnerable to temperature extremes and heat stress in SA. Heat stress in KwaZulu-Natal is predominantly mild, with increasing severity from the historical to the distant-future. The increasing heat stress alongside rainfall reduction in Limpopo (HCZ 1) suggests compounding climate stress in the region, which will harm socio-ecological and agri-food systems, and natural resource availability, underscoring a need for intentional climate risk assessments and water and heat adaptation strategies (Engelbrecht et al., 2024; Jubase and New, 2025). An urgent need for adaptation planning also applies to HCZs 2 and 5, although projected changes do not exhibit the same extremes as those in HCZ 1. Ensemble and scenario uncertainties could oversimplify the projected spatial variabilities in the regions, and the study overlooked the probability of extreme weather events within the given timelines.

4.2.3 The application of the THI in agricultural crop production

The THI index combines air temperatures and relative humidity to quantify heat stress. However, linking the THI to agricultural crop production is unconventional. The index is often linked to livestock production (Du Preez et al., 1990; Nesamvuni et al., 2012; Mpandeli et al., 2019). However, crops also have temperature thresholds and can experience both heat and water stress (Mugiyo et al., 2021; Nzimande et al., 2025); therefore, the index can be applied in crop production. Both heat and water stress affect crop phenology and yield, thereby influencing the productivity, particularly under rainfed conditions (Hatfield and Prueger, 2015; Nzimande et al., 2025).

Thermal indices, such as the THI, can play a crucial role in agricultural crop production by providing accurate, simplified forecasts throughout crop development. This can safeguard crop production under short-term weather and long-term climate scenarios and serve as a decision-support tool, strengthening coping and adaptation strategies and building resilience. Indices such as the THI can be adapted to crop-specific physiological thresholds for anticipating potential heat stress and related losses, and crop-calibrated THI thresholds can indicate heat stress across different seasons. For example, integrating crop-calibrated THI with remote sensing and real-time weather feeds offers an easy platform for real-time stress management in water-scarce environments.

Although THI is a straightforward index, due to crop-specific heat and water thresholds and additional requirements (Nzimande et al., 2025), it may lack sufficient sensitivity. However, these crop-specific thresholds can be developed and identified on platforms such as AquaCrop (Steduto et al., 2009) to define the onset and tipping points of heat-stress risk in warming environments. This can also improve the model’s sensitivity to predicting the impact of extreme heat on crop production. Following the identification of crop thresholds, crop-sensitive THI classifications can be utilised, enabling the accurate application of THI to crop production as a straightforward, cost-effective method for measuring temperature stress.

4.2.4 Implications for crop production

Due to climate change, air temperatures are increasingly likely to exceed optimal thresholds for crop growth (Hatfield and Prueger, 2015), directly impacting crop productivity and, by extension, human and environmental health. The study projected increasing THI values and moisture scarcity, indicating potential heat and water stress, which will pose undeniable challenges for crop production across the study areas. Studies such as those by Chapman et al. (2020) and Chemura et al. (2022) have highlighted that coupled water and heat stress lead to ecological shifts that enforce changes in crop species distribution and suitability.

Mainstream crops (maize, wheat, soybeans, and potatoes) are becoming increasingly unsuitable in these changing environments across broader Sub-Saharan Africa, including SA (Chapman et al., 2020). However, neglected and underutilised crops (NUS) have proven to be heat- and water-stress-adapted and to have higher temperature and water-stress thresholds than mainstream crops (Mabhaudhi et al., 2017; Nzimande et al., 2025). In environments marked by increasing aridity and temperature stress, such as those in this study, NUS gain a competitive advantage and offer a sustainable solution to building resilient agri-food systems by improving diversity and derisking rainfed agri-food systems (Mabhaudhi et al., 2017; Mugiyo et al., 2021; Ndlovu et al., 2024). However, realising the full potential of NUS requires overcoming significant socio-economic and institutional barriers (Ndlovu et al., 2024). To leverage NUS strategically as a resilience strategy, it is crucial to integrate climate risk research with agri-food systems and crop production, as well as breeding and agronomy research, policy, market support, and extension services, to position them as a plausible adaptation strategy within local agri-food systems under current and projected climatic conditions.

The increasing aridity and water stress conditions in the study areas are triggering a cascading series of natural feedback mechanisms that degrade agricultural environments, including soil degradation and fertility loss, reduced vegetation cover, a vicious cycle of drought and desertification, and shifts in microbial activity. The impacts on agri-food systems being experienced in the study areas include reduced crop suitability and yields, disrupted crop cycles, and long-term productivity challenges. Restoring degraded land and changing land-use intensity (e.g., reverting to grassland or forestry) are required in the study areas to mitigate these effects and improve soil moisture content. The natural feedback mechanisms resulting from increasing aridity and worsening water stress, along with their impacts on agri-food systems in the study areas, are indicated in Table 4.

The findings from the study indicate the need to integrate the results with existing climate information services and decision-support systems (DSS) primarily through the co-production processes, which transform scientific data into tailored, actionable products, such as localised early warning systems or agricultural advisories. This integration bridges the “usability gap,” that is, the disconnect between what research produces and what decision-makers need, by creating iterative, trusted partnerships between producers and users of information. This is key in transforming research findings into practical, actionable, and transformative products.

5 Recommendations: improving adaptation planning for improved water and food security

Improving adaptation planning for water and food security under drying conditions requires a contextualised transition from reactive, crisis-based management to proactive, long-term strategies that integrate nature-based solutions, technological innovation, and community participation. Effective adaptation involves aligning policy with on-the-ground agricultural needs, such as water-saving irrigation, drought-resistant crops, and enhanced soil moisture management, to build resilience against increasing dry spells. The study underscores the importance of context-specific adaptation planning informed by climate risk assessments. The following recommendations are derived from the study:

• Integration of climate-risk assessments and social-vulnerability research to identify and prioritise areas that are vulnerable to food and nutrition insecurity: In areas vulnerable to water stress and food insecurity, there is an urgent need to develop localised climate-risk overlay maps that incorporate projected CMI and THI outputs with high-resolution food insecurity data. The maps would enable visual identification and prioritisation of the most vulnerable food insecurity hotspots requiring immediate transformative adaptation.

• Development of crop-calibrated THI thresholds: The application of the THI to crop production could be a straightforward, economical tool for measuring crop temperature stress. However, to get to that point, future research is needed to develop crop-calibrated THI thresholds using crop-specific physiological data for priority crops, which could help address the food insecurity pandemic in the country and beyond. This action would transform the THI into a practical and sensitive tool for risk detection and management in crop production.

• Transformative adaptation: a system-wide shift in response to climate change that alters the structural, economic, and social systems themselves, rather than just adjusting existing practices (Biesbroek and Nalau, 2026). It is a system that addresses root causes of vulnerability, often aiming for deeper sustainability, equity, and resilience when incremental changes are insufficient. Adaptive planning targeting the study sites should focus on promoting contextualised, climate-informed, transformative adaptation frameworks and strategies for climate action and adaptive planning, rather than reinforcing incremental adaptation measures (Biesbroek and Nalau, 2026). This entails, for example, adopting locally adapted risk-detection tools and water-efficient crop species in local agri-food systems to sustainably improve natural resource management and enhance food and nutrition security outcomes. Examples of transformative adaptation include shifting from fossil-fuel-based agriculture to regenerative, locally driven food systems; altering urban planning to prioritise ecosystem-based solutions over hard infrastructure; and implementing fundamental policy changes to promote social equity. Context-based adaptation is key to addressing local challenges, as climate change impacts are felt more locally than nationally, regionally, or globally. National, regional, or global adaptation strategies are only indicative, as each community faces unique challenges that require tailored adaptation strategies. However, it is not being widely adopted due to deeply ingrained political, economic, and practical barriers.

• Integrated basin management (IBM) and water storage: The study’s results indicate that climates across the study sites will become drier and hotter. This necessitates urgent conjunctive ground- and surface-water management to sustain farmers’ productivity during dry spells and droughts. These should be supported by technological innovations that enhance water-use efficiency, including Internet of Things (IoT)-driven automated control systems, soil moisture sensors, and satellite-based data analysis. These technologies, supported by Artificial Intelligence (AI) and cloud-based models, enable precise, real-time irrigation scheduling based on crop water requirements, reducing waste while increasing yield. The uptake of these technologies is quite slow in the study areas, particularly among smallholder farmers, due to limited adaptive capacity and limited resources.

• Coordination and collaboration: When initiating climate action and adaptation initiatives, ensure that key stakeholders, particularly smallholder farmers, are involved and that the initiatives are feasible and contextually relevant. Such stakeholders can include government officials, researchers, policymakers, extension services, and farmers. This collaboration could ensure alignment in adaptation planning, policy, and development, water management, climate risk research, and local agriculture towards context-specific, sustainable, and resilient adaptation frameworks and pathways.

6 Conclusion

This study examined climate risks and their effects on water and agri-food systems using the Climate Moisture Index (CMI) and the Thermal Heat Index (THI) across three Hydro-Climatic Zones in South Africa. Understanding the spatio-temporal severity of climate-related risks is critical to managing their impacts on resource availability and the broader socio-ecological and agri-food systems. The decreasing CMI alongside increasing THI in the chosen study sites indicates a general, progressive increase in moisture and heat stress from the historical to the future climate. This parallel trend of decreasing moisture and rising heat suggests that study areas will face harsher, more arid environments, where higher temperatures combined with reduced water availability significantly increase ecological and socio-economic vulnerabilities. The results provide pathways to develop context-specific adaptation strategies to extreme heat and arid conditions, particularly with a focus on smallholder farmers. These could include nature-based solutions that strengthen natural systems while derisking agri-food systems, and prioritising the adoption of strategic, nutrient-dense crops with high water and heat stress thresholds. The trend indicates that future climates are expected to be simultaneously drier and hotter, increasing overall atmospheric and soil moisture demand while reducing the capacity of living organisms to cool. This intensification of land-atmosphere coupling is leading to extreme droughts and reduced ecosystem cooling capacity, increasing atmospheric and soil moisture demand and creating a feedback loop in which dry conditions aggravate heat extremes and reduce the ability of both natural ecosystems and human-managed environments to mitigate heat. The trend indicates that current water management and conservation strategies, particularly in agriculture and water supply, need to be revised to address the growing challenges of water scarcity.

References

• 1

  AbubakarH. B.ScholesM. C.EngelbrechtF. A. (2025). Spatio-temporal trends in daily precipitation extremes over the Enkangala escarpment of South Africa: 1961–2021. Theor. Appl. Climatol.156:164.

  • Google Scholar

• 2

  ArcherE.Du ToitJ.EngelbrechtC.HoffmanM. T.LandmanW.MalherbeJ.et al. (2022). The 2015–19 multi-year drought in the eastern cape, South Africa: its evolution and impacts on agriculture. J. Arid Environ.196:104630. doi: 10.1016/j.jaridenv.2021.104630

  • CrossRef
  • Google Scholar

• 3

  ArcherE.EngelbrechtF.HänslerA.LandmanW.TadrossM.HelmschrotJ. (2018). “Seasonal prediction and regional climate projections for southern Africa,” in Climate Change and Adaptive Land Management in Southern Africa—Assessments, Changes, Challenges, and Solutions, eds. RevermannR.KrewenkaK. M.SchmiedelU.OlwochJ. M.HelmschrotJ.JürgensN. (Göttingen & Windhoek: Klaus Hess Publishers), 14–21. (Biodiversity & Ecology, 6)

  • Google Scholar

• 4

  ArcherE. R. M.LandmanW. A.MalherbeJ.MalulekeP.WeepenerH. (2021). Managing climate risk in livestock production in South Africa: how might improved tailored forecasting contribute?Clim. Risk Manag.32:100312. doi: 10.1016/j.crm.2021.100312

  • CrossRef
  • Google Scholar

• 5

  AyanladeA.OluwarantiA.AyanladeO. S.BorderonM.SterlyH.SakdapolrakP.et al. (2022). Extreme climate events in Sub-Saharan Africa: a call for improving agricultural technology transfer to enhance adaptive capacity. Clim. Serv.27:100311. doi: 10.1016/j.cliser.2022.100311

  • CrossRef
  • Google Scholar

• 6

  BakerN. C.HuangH. P. (2012). A comparison of absolute and relative changes in precipitation in multimodel climate projection. Atmos. Sci. Lett.13, 174–179.

  • Google Scholar

• 7

  BiazinB.SterkG.TemesgenM.AbdulkedirA.StroosnijderL. (2012). Rainwater harvesting and management in rainfed agricultural systems in Sub-Saharan Africa—a review. Phys. Chem. Earth A/B/C47-48, 139–151. doi: 10.1016/j.pce.2011.08.015

  • CrossRef
  • Google Scholar

• 8

  BiesbroekR.NalauJ. (2026). Defining transformational adaptation and why it matters. Nat. Clim. Chang.16, 253–254. doi: 10.1038/s41558-025-02550-4

  • CrossRef
  • Google Scholar

• 9

  BjornlundV.BjornlundH.Van RooyenA. F. (2020). Why agricultural production in Sub-Saharan Africa remains low compared to the rest of the world–a historical perspective. Int. J. Water Resour. Dev.36, S20–S53. doi: 10.1080/07900627.2020.1739512

  • CrossRef
  • Google Scholar

• 10

  BjornlundV.BjornlundH.van RooyenA. (2022). Why food insecurity persists in sub-Saharan Africa: a review of existing evidence. Food Secur.14, 845–864.

  • Google Scholar

• 11

  BonettiS.SutanudjajaE. H.MabhaudhiT.SlotowR.DalinC. (2022). Climate change impacts on water sustainability of South African crop production. Environ. Res. Lett.17:084017. doi: 10.1088/1748-9326/ac80cf,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12

  BotaiC. M.BotaiJ. O.de WitJ. P.NcongwaneK. P.MurambadoroM.BarasaP. M.et al. (2021). Hydrological drought assessment based on the standardized streamflow index: a case study of the three cape provinces of South Africa. Water13:3498. doi: 10.3390/w13243498

  • CrossRef
  • Google Scholar

• 13

  BraunschweigerD.IngoldK. (2023). What drives local climate change adaptation? A qualitative comparative analysis. Environ. Sci. Policy.145, 40–49.

  • Google Scholar

• 14

  CaiX.MagidiJ.NhamoL.van KoppenB. (2017). Mapping Irrigated Areas in the Limpopo Province, South Africa, vol. 172. Pretoria, South Africa: International Water Management Institute (IWMI).

  • Google Scholar

• 15

  CannonA. J.SobieS. R.MurdockT. Q. (2015). Bias correction of GCM precipitation by quantile mapping: how well do methods preserve changes in quantiles and extremes?J. Clim.28, 6938–6959.

  • Google Scholar

• 16

  ChapmanS.E BirchC.PopeE.SalluS.BradshawC.DavieJ.et al. (2020). Impact of climate change on crop suitability in Sub-Saharan Africa in parameterized and convection-permitting regional climate models. Environ. Res. Lett.15:094086. doi: 10.1088/1748-9326/ab9daf

  • CrossRef
  • Google Scholar

• 17

  ChemuraA.NangombeS. S.GleixnerS.ChinyokaS.GornottC. (2022). Changes in climate extremes and their effect on maize (Zea mays L.) suitability over Southern Africa. Front. Clim.4:890210. doi: 10.3389/fclim.2022.890210

  • CrossRef
  • Google Scholar

• 18

  ChikooreH.JuryM. R. (2021). South African drought, deconstructed. Weather Clim. Extrem.33:100334. doi: 10.1016/j.wace.2021.100334

  • CrossRef
  • Google Scholar

• 19

  Davis-ReddyC. L.VincentK. (2017). Climate Risk and Vulnerability: A Handbook for Southern Africa. 2nd Edn Pretoria: Council for Scientific and Industrial Research.

  • Google Scholar

• 20

  de PerezE. C.AndersonW.HanE.MasukwedzaG. I. T.MphonyaneN. (2024). Detectable use of ENSO information on crop production in southern Africa. Clim. Serv.36:100514. doi: 10.1016/j.cliser.2024.100514

  • CrossRef
  • Google Scholar

• 21

  Department of Water and Sanitation. (2024). National State of water report 2023. Integrated water resource studies report number WII/IWRS/NSoW 2023 PRETORIA, South Africa

  • Google Scholar

• 22

  Department of Water and Sanitation, South Africa. (2022). Update of the climate change status quo analysis for water resources and the development of a climate change response strategy for the water and sanitation sector in South Africa: Hydro-climatic zones of South Africa reviewed. Project Number WP11353. DWS Report No: P RSA 000/00/23221/4

  • Google Scholar

• 23

  Department of Water and Sanitation, South Africa. (2024). Climate change status quo analysis for water resources and the development of a climate change response strategy for the water and sanitation sector in South Africa: first draft climate change response strategy for water and sanitation sector. WP11353. DWS Report No: P RSA 000/00/23221/20

  • Google Scholar

• 24

  Du PreezJ. H.GieseckeW. H.HattinghP. J.EisenbergB. E. (1990). Heat Stress in Dairy Cattle under Southern African Conditions. II. Identification of areas of potential heat stress during summer by means of observed true and predicted temperature-humidity index values.

  • Google Scholar

• 25

  Du PreezC. C.Van HuyssteenC. W. (2020). Threats to soil and water resources in South Africa. Environ. Res.183:109015. doi: 10.1016/j.envres.2019.109015

  • CrossRef
  • Google Scholar

• 26

  EdokpayiJ. N.MakungoR.MathivhaF.RiversN.VolenzoT.OdiyoJ. O. (2020). “Influence of global climate change on water resources in South Africa: toward an adaptive management approach,” in Water Conservation and Wastewater Treatment in BRICS Nations, Eds. P. Singh, Y. Milshina, K. Tian, D. Gusain, J. P. Bassin (Oxford, United Kingdom: Elsevier), 83–115.

  • Google Scholar

• 27

  EloffH. (2024). Climate Change Wreaks Havoc: Major Floods in KwaZulu-Natal and….NSRI. Available online at: https://www.nsri.org.za/2024/06/climate-change-wreaks-havoc-major-floods-in-kwazulu-natal-and-eastern-cape/ (Accessed September 7, 2025).

  • Google Scholar

• 28

  EngelbrechtF.AdegokeJ.BopapeM. J.NaidooM.GarlandR.ThatcherM.et al. (2015). Projections of rapidly rising surface temperatures over Africa under low mitigation. Environ. Res. Lett.10:085004. doi: 10.1088/1748-9326/10/8/085004

  • CrossRef
  • Google Scholar

• 29

  EngelbrechtF. A.SteinkopfJ.PadavatanJ.MidgleyG. F. (2024). “Projections of future climate change in Southern Africa and the potential for regional tipping points,” in Sustainability of Southern African Ecosystems under Global Change: Science for Management and Policy Interventions, Eds. G. P. von Maltitz, G. F. Midgley, J. Veitch, C. Brümmer, R. P. Rötter, F. A. Viehberg, et al. (Cham: Springer International Publishing), 169–190.

  • Google Scholar

• 30

  FAO. (2024). Drought impact mitigation and prevention in the Limpopo River Basin. Available online at: https://www.fao.org/4/y5744e/y5744e05.htm (Accessed September 7, 2025).

  • Google Scholar

• 31

  FranchiF.MustafaS.ArizteguiD.ChirindjaF. J.Di CapuaA.HusseyS.et al. (2024). Prolonged drought periods over the last four decades increase flood intensity in southern Africa. Sci. Total Environ.924:171489. doi: 10.1016/j.scitotenv.2024.171489,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32

  FurtakK.WolińskaA. (2023). The impact of extreme weather events as a consequence of climate change on the soil moisture and on the quality of the soil environment and agriculture–a review. Catena231:107378. doi: 10.1016/j.catena.2023.107378

  • CrossRef
  • Google Scholar

• 33

  GillerK. E. (2020). The food security conundrum of sub-Saharan Africa. Glob. Food Sec.26:100431.

  • Google Scholar

• 34

  GossetM.Dibi-AnohP. A.SchumannG.HostacheR.ParisA.ZahiriE. P.et al. (2023). Hydrometeorological extreme events in Africa: the role of satellite observations for monitoring pluvial and fluvial flood risk. Surv. Geophys.44, 197–223. doi: 10.1007/s10712-022-09749-6

  • CrossRef
  • Google Scholar

• 35

  HarveyC. A.RakotobeZ. L.RaoN. S.DaveR.RazafimahatratraH.RabarijohnR. H.et al. (2014). Extreme vulnerability of smallholder farmers to agricultural risks and climate change in Madagascar. Trans. R. Soc. Lond. B Biol. Sci.

  • Google Scholar

• 36

  HatfieldJ. L.PruegerJ. H. (2015). Temperature extremes: effect on plant growth and development. Weather Clim. Extrem.10, 4–10. doi: 10.1016/j.wace.2015.08.001

  • CrossRef
  • Google Scholar

• 37

  JohnstonP.EgbebiyiT. S.ZvobgoL.OmarS. A.CartwrightA.HewitsonB. (2024). Climate Change Impacts in South Africa: What Climate Change Means for a Country and its People. Cape Town, South Africa: University of Cape Town.

  • Google Scholar

• 38

  JubaseP.NewM. (2025). The trends in heatwaves and associated synoptic conditions across the upper Limpopo Basin of Botswana and South Africa. Environ. Dev.55:101211. doi: 10.1016/j.envdev.2025.101211

  • CrossRef
  • Google Scholar

• 39

  KapariM.Hlophe-GinindzaS.NhamoL.MpandeliS. (2023). Contribution of smallholder farmers to food security and opportunities for resilient farming systems. Front. Sustain. Food Syst.7:1149854. doi: 10.3389/fsufs.2023.1149854

  • CrossRef
  • Google Scholar

• 40

  KrugerA. (2004). Climate of South Africa: Climate Regions. South African Weather Service: Pretoria, 19.

  • Google Scholar

• 41

  LeeJ. Y.MarotzkeJ.BalaG.CaoL.CortiS.DunneJ. P.et al. (2021). Future global climate: scenario-based projections and near-term information.

  • Google Scholar

• 42

  LetswamotseT. V.ArshadS.BashirB.AlsalmanA.HarsányiE.Al-DalahmehM.et al. (2024). Integrating maize yield and agricultural drought analysis for sustainable food security: a provincial study in South Africa (1993–2022). Food Energy Secur.13:e70006. doi: 10.1002/fes3.70006

  • CrossRef
  • Google Scholar

• 43

  MabhaudhiT.ChimonyoV. G.ModiA. T. (2017). Status of underutilised crops in South Africa: opportunities for developing research capacity. Sustainability9:1569. doi: 10.3390/su9091569

  • CrossRef
  • Google Scholar

• 44

  MabhaudhiT.MpandeliS.NhamoL.ChimonyoV. G.NhemachenaC.SenzanjeA.et al. (2018). Prospects for improving irrigated agriculture in southern Africa: linking water, energy and food. Water10:1881. doi: 10.3390/w10121881

  • CrossRef
  • Google Scholar

• 45

  MaguireG. (2024). Supporting climate-resilient African smallholder farmers through carbon markets policy briefing 301. Available online at: https://saiia.org.za/wp-content/uploads/2024/11/SAIIA_PB_301_ClimateResilientAfricanFarmers.pdf (Accessed 23 April, 2025)

  • Google Scholar

• 46

  MaposaD.SeimelaA. M.SigaukeC.CochranJ. J. (2021). Modelling temperature extremes in the Limpopo province: Bivariate time-varying threshold excess approach. Nat. Hazards107, 2227–2246.

  • Google Scholar

• 47

  MashaoF. M.MothapoM. C.MunyaiR. B.LetsoaloJ. M.MbokodoI. L.MuofheT. P.et al. (2023). Extreme rainfall and flood risk prediction over the East Coast of South Africa. Water15:50. doi: 10.3390/w15010050

  • CrossRef
  • Google Scholar

• 48

  MatimolaneS.StrydomS.MathivhaF. I.ChikooreH. (2024). Evaluating the spatiotemporal patterns of drought characteristics in a semi-arid region of Limpopo Province, South Africa. Environ. Monit. Assess.196:1062. doi: 10.1007/s10661-024-13217-6,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  McBrideC. M.KrugerA. C.DysonL. (2022). Changes in extreme daily rainfall characteristics in South Africa: 1921–2020. Weather Clim. Extrem.38:100517. doi: 10.1016/j.wace.2022.100517

  • CrossRef
  • Google Scholar

• 50

  MezaI.RezaeiE. E.SiebertS.GhazaryanG.NouriH.DubovykO.et al. (2021). Drought risk for agricultural systems in South Africa: drivers, spatial patterns, and implications for drought risk management. Sci. Total Environ.799:149505. doi: 10.1016/j.scitotenv.2021.149505,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  MirzabaevA.KerrR. B.HasegawaT.PradhanP.WrefordA.von der PahlenM. C. T.et al. (2023). Severe climate change risks to food security and nutrition. Clim. Risk Manag.39:100473.

  • Google Scholar

• 52

  MosesO.BlameyR. C.ReasonC. J. C. (2023). Extreme rainfall events over the Okavango river basin, southern Africa. Weather Clim. Extrem.41:100589. doi: 10.1016/j.wace.2023.100589

  • CrossRef
  • Google Scholar

• 53

  MpandeliS.NhamoL.MoeletsiM.MasuphaT.MagidiJ.TshikolomoK.et al. (2019). Assessing climate change and adaptive capacity at local scale using observed and remotely sensed data. Weather Clim. Extrem.26:100240. doi: 10.1016/j.wace.2019.100240

  • CrossRef
  • Google Scholar

• 54

  MudefiE. (2023). Disaster management ‘deeds’ in the context of April 2022 KwaZulu-Natal floods: a scoping review. Int. J. Disaster Risk Reduct.98:104122.

  • Google Scholar

• 55

  MugiyoH.ChimonyoV. G.SibandaM.KunzR.NhamoL.MasemolaC. R.et al. (2021). Multi-criteria suitability analysis for neglected and underutilised crop species in South Africa. PLoS One16:e0244734. doi: 10.1371/journal.pone.0244734,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 56

  Ncoyini-ManciyaZ.SavageM. J. (2022). The assessment of future air temperature and rainfall changes based on the statistical downscaling model (SDSM): the case of the Wartburg community in KZN midlands, South Africa. Sustain.14:10682.

  • Google Scholar

• 57

  NdlovuM.ClulowA. D.SavageM. J.NhamoL.MagidiJ.MabhaudhiT. (2021). An assessment of the impacts of climate variability and change in KwaZulu-Natal. Province, South Africa.12:427.

  • Google Scholar

• 58

  NdlovuM.ScheelbeekP.NgidiM.MabhaudhiT. (2024). Underutilized crops for diverse, resilient and healthy agri-food systems: a systematic review of Sub-Saharan Africa. Front. Sustain. Food Syst.8:1498402. doi: 10.3389/fsufs.2024.1498402,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 59

  NesamvuniE.LekalakalaR.NorrisD.NgambiJ. W. (2012). Effects of climate change on dairy cattle, South Africa. Afr. J. Agric. Res.7, 3867–3872. doi: 10.5897/AJAR11.1468

  • CrossRef
  • Google Scholar

• 60

  NhamoG.ChapunguL.MutandaG. W. (2025). Trends and impacts of climate-induced extreme weather events in South Africa (1920–2023). Environ. Dev.55:101183. doi: 10.1016/j.envdev.2025.101183

  • CrossRef
  • Google Scholar

• 61

  NhamoL.MabhaudhiT.ModiA. T. (2019). Preparedness or repeated short-term relief aid? Building drought resilience through early warning in southern Africa. Water SA45, 75–85. doi: 10.4314/wsa.v45i1.09

  • CrossRef
  • Google Scholar

• 62

  NhamoL.MpandeliS.LiphadziS.DirwaiT. L.MugiyoH.SenzanjeA.et al. (2024). Why do farmers not irrigate all the areas equipped for irrigation? Lessons from Southern Africa. Agriculture14:1218. doi: 10.3390/agriculture14081218

  • CrossRef
  • Google Scholar

• 63

  NhemachenaC.NhamoL.MatchayaG.NhemachenaC. R.MucharaB.KaruaiheS. T.et al. (2020). Climate change impacts on water and agriculture sectors in southern Africa: threats and opportunities for sustainable development. Water12:2673. doi: 10.3390/w12102673

  • CrossRef
  • Google Scholar

• 64

  NOAAMonthly Global Climate Report for Annual 2023. (2024). National Centers for Environmental Information. Available online at: https://www.ncei.noaa.gov/access/monitoring/monthly-report/global/202313 (Accessed April 11, 2025).

  • Google Scholar

• 65

  NyoniR. S.BruelleG.ChikowoR.AndrieuN. (2024). Targeting smallholder farmers for climate information services adoption in Africa: a systematic literature review. Clim. Serv.34:100450. doi: 10.1016/j.cliser.2024.100450

  • CrossRef
  • Google Scholar

• 66

  NzimandeT. N. M.ChimonyoV. G.WimalasiriE. M.MabhaudhiT. (2025). “Using AquaCrop, DSSAT and the SIMPLE to estimate water use of underutilised cereal in South Africa,” in Enhancing Water and Food Security through Improved Agricultural Water Productivity: New Knowledge, Innovations and Applications, Eds. T. Mabhaudhi, V. G. P. Chimonyo, A. Senzanje, P. P. Chivenge (Springer Nature: Singapore), 201–221.

  • Google Scholar

• 67

  OgundejiA. A. (2022). Adaptation to climate change and impact on smallholder farmers’ food security in South Africa. Agriculture12:589.

  • Google Scholar

• 68

  OkolieC. C.Danso-AbbeamG.OgundejiA. A. (2023). Livelihood vulnerability to the changing climate: the experiences of smallholder farming households in the Free State Province, South Africa. Clim. Serv.30:100371.

  • Google Scholar

• 69

  OpiyoS. B.PienaarL.PikethS. J.BurgerR. P.ChikooreH.HavengaH. (2025). A comparative study of past and projected future dynamics of drought hazards across the districts of the northern Cape Province, South Africa: analysis of CRU TS4 data and CORDEX-Africa RCM simulations. S. Afr. Geogr. J.108, 1–28. doi: 10.1080/03736245.2025.2490599

  • CrossRef
  • Google Scholar

• 70

  PhophiM. M.MafongoyaP.LotteringS. (2020). Perceptions of climate change and drivers of insect pest outbreaks in vegetable crops in Limpopo province of South Africa. Climate8:27.

  • Google Scholar

• 71

  PizzorniM.InnocentiA.TollinN. (2024). Droughts and floods in a changing climate and implications for multi-hazard urban planning: a review. City Environ. Interact.24:100169. doi: 10.1016/j.cacint.2024.100169

  • CrossRef
  • Google Scholar

• 72

  RinglerC.AgbonlahorM.BarronJ.BayeK.MeenakshiJ. V.MekonnenD. K.et al. (2022). The role of water in transforming food systems. Glob. Food Sec.33:100639. doi: 10.1016/j.gfs.2022.100639

  • CrossRef
  • Google Scholar

• 73

  RouaultM.DieppoisB.TimN.HünickeB.ZoritaE. (2024). “Southern Africa climate over the recent decades: description, variability and trends,” in Sustainability of Southern African Ecosystems under Global Change: Science for Management and Policy Interventions (Cham: Springer International Publishing), 149–168.

  • Google Scholar

• 74

  RouaultM.MonyelaB.KoungeR. A. I.StellaA.NjouodoN.DieppoisB. (2019). Ocean Impact on Southern African Climate Variability and Water Resources. Pretoria: Water Research Commission.

  • Google Scholar

• 75

  RouaultM.RoyS. S.BallingR. C. (2013). The diurnal cycle of rainfall in South Africa in the austral summer. Int. J. Climatol.33, 770–777. doi: 10.1002/joc.3451

  • CrossRef
  • Google Scholar

• 76

  ScholesR.EngelbrechtF. (2021). Climate Impacts in Southern Africa During the 21st Century. Report for Earthjustice and the Centre for Envrionmental Rights. Pretoria, South Africa: Global Change Instiute, University of Witwatersrand.

  • Google Scholar

• 77

  SchulzeR. E.SchütteS.GwenaK. R.NxumaloN. (2023). A National Assessment of Potential Climate Change Impacts on the Hydrological Yield of Different Hydro-Climatic Zones of South Africa (No. 2833/3, p. 22).WRC Report.

  • Google Scholar

• 78

  SchütteS.SchulzeR. E.ClarkD. J. (2023). A National Assessment of Potential Climate Change Impacts on the Hydrological Yield of Different Hydro-Climatic Zones of South Africa: Report 1 – Methodology and Results. WRC Report No. 2833/1/22. Pretoria, RSA: Water Research Commission, 271.

  • Google Scholar

• 79

  SikhwariT.NethengweN.SigaukeC.ChikooreH. (2022). Modelling of extremely high rainfall in Limpopo Province of South Africa. Climate10:33.

  • Google Scholar

• 80

  StedutoP.RaesD.HsiaoT. C.FereresE.HengL. K.HowellT. A.et al. (2009). Concepts and applications of AquaCrop: The FAO crop water productivity model. In Crop modeling and decision support. (pp. 175–191). Berlin, Heidelberg: Springer Berlin Heidelberg.

  • Google Scholar

• 81

  StoddardE. (2025). La Niña is over as South Africa’s southwest braces for an unseasonably dry winter. Daily maverick. Available online at: https://www.dailymaverick.co.za/article/2025-03-31-la-nina-is-over-as-sas-southwest-braces-for-an-unseasonably-dry-winter/ (Accessed September 23, 2025).

  • Google Scholar

• 82

  TouchV.TanD. K.CookB. R.Li LiuD.CrossR.TranT. A.et al. (2024). Smallholder farmers’ challenges and opportunities: implications for agricultural production, environment and food security. J. Environ. Manag.370:122536. doi: 10.1016/j.jenvman.2024.122536,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 83

  ThoithiW.BlameyR. C.ReasonC. J. (2022). April 2022 floods over East Coast South Africa: interactions between a mesoscale convective system and a coastal meso-low. Atmos.14:78.

  • Google Scholar

• 84

  VörösmartyC. J.DouglasE. M.GreenP. A.RevengaC. (2005). Geospatial indicators of emerging water stress: an application to Africa. AMBIO34, 230–236. doi: 10.1579/0044-7447-34.3.230

  • CrossRef
  • Google Scholar

• 85

  WilliamsP. A.CrespoO.AbuM.SimpsonN. P. (2018). A systematic review of how vulnerability of smallholder agricultural systems to changing climate is assessed in Africa. Environ. Res. Lett.13:103004.

  • Google Scholar

• 86

  World Meteorological Organization (2024). Africa Faces Disproportionate Burden from Climate change and Adaptation Costs. World Meteorological Organization. Available online at: https://wmo.int/news/media-centre/africa-faces-disproportionate-burden-from-climate-change-and-adaptation-costs (Accessed September 7, 2025).

  • Google Scholar

• 87

  ZendaM.RudolphM.HarleyC. (2024). The impact of climate variability on the livelihoods of smallholder farmers in an agricultural village in the Wider Belfast Area, Mpumalanga Province, South Africa. Atmosphere15:1353. doi: 10.3390/atmos15111353

  • CrossRef
  • Google Scholar