Malaria at the margins: climatic extremes, migration, and urban informal settlement risk in South Africa

Malaria remains a significant public health issue in sub‑Saharan Africa, influenced by climate variability, population movement and urbanisation. In South Africa, efforts to eliminate malaria operate within a dynamic environment where climatic fluctuations and population movements complicate risk patterns. However, existing approaches to analyzing climate-malaria relationships, often rely on seasonal averages, ignoring the potential influence of rare climatic extremes and migration flows. This study investigates how climatic extremes, climatic variability, and migration flows interact to shape malaria risk and inform adaptive surveillance and control strategies. Climate data (monthly minimum/maximum temperature, rainfall and relative humidity) for Limpopo, Mpumalanga and KwaZulu‑Natal (2000–2024) were obtained from the South African Weather Service. Laboratory‑confirmed malaria cases were sourced from provincial surveillance reports, and in‑migration data into Thembelihle and selected parts of Soweto (2017–2024) were obtained from the Soweto and Thembelihle Health and Demographic Surveillance System (SaT‑HDSS). Seasonal climate trajectories were evaluated using Ordinary Least Squares regressions. Climatic seasonality was assessed using diagnostic metrics, humidity plateaus, rainfall extremes and thermal season‑length indices. Shock‑year dynamics (KwaZulu‑Natal 2022 floods; Limpopo 2017 Cyclone Dineo) were analysed using monthly anomalies against 2000–2010 climatological envelopes. Pearson correlations examined same‑year and lag‑1 climate-malaria associations. Climate-malaria-migration co‑movement was assessed using min–max‑normalised rainfall, humidity, temperature, incidence and migration data (0–1). Malaria tail events were defined as years at or above the provincial 90th percentile. All statistical analyses were conducted using IBM Statistical Package for the Social Sciences (SPSS), Version 30; p < 0.05 was considered significant. Climate trends across the three endemic provinces showed consistent minimum-temperature warming, declining warm-season humidity, and rainfall patterns increasingly dominated by short-duration extremes. Province-specific seasonality diagnostics revealed that KwaZulu-Natal experienced frequent multi-month humidity plateaus, Limpopo showed pronounced lagged hydrological effects following extreme rainfall events, and Mpumalanga displayed prolonged warm-season durations in recent years. Shock-year analyses (KwaZulu-Natal 2022 floods; Limpopo 2017 Cyclone Dineo) demonstrated compound climatic anomalies, combinations of intense rainfall, sustained humidity, and sub-seasonal cooling that extended environmental suitability beyond expected seasonal boundaries. Same-year humidity was positively associated with malaria incidence in KwaZulu-Natal and Mpumalanga, whereas lag-1 rainfall showed a significant relationship with malaria in Limpopo, reflecting hydrological carry-over effects. Following the April–May 2022 floods, migration into Thembelihle increased markedly, and a co-movement analysis indicated temporal alignment between malaria tail years, elevated rainfall, higher minimum temperatures, and increased migration during climatologically disrupted periods. Climatic extremes and post‑shock mobility reorganise malaria risk in South Africa. Tail‑aware, lag‑sensitive and mobility‑integrated surveillance is essential for adaptive elimination planning.