Cross-border climate risks threaten Malawi’s drinking water, study finds

Malawi’s water security hinges on a lake it only partly controls. The Lake Malawi-Shire River system receives roughly 60% of its inflows from catchments lying in Tanzania and Mozambique, yet the drinking water infrastructure downstream answers almost entirely to Malawi. A new integrated study, published in Frontiers in Earth Science, quantifies how that asymmetry, combined with a sharp hydrological threshold and fragmented governance, leaves the water supply of Blantyre — a city of more than 1.2 million people — exposed to climate shocks that originate well beyond the country’s borders.

The research team, led by Miriam Dalitso Kalanda Joshua of the University of Malawi with Cosmo Ngongondo and Raymond Abudu Kasei, combined four lines of evidence: hydrological analysis of 40 years of observational data from 1970 to 2013, climate projections from 20 CORDEX-Africa regional climate models under the RCP4.5 and RCP8.5 emission scenarios, a systematic analysis of Malawi’s National Water Policy, and semi-structured interviews with 28 stakeholders across the four riparian states of Malawi, Tanzania, Mozambique and Zambia. Vulnerability was assessed using the IPCC exposure-sensitivity-adaptive capacity framework, focused on Walkers Ferry Dam, the primary raw water abstraction point from which the Blantyre Water Board draws and treats water for the city.

The system’s central vulnerability is a documented critical threshold at 474.1 metres above sea level. Below this lake level, outflows to the Shire River cease, and with them the water supply pathway to downstream infrastructure. Sensitivity analysis showed that each 1 °C increase in annual temperature reduces the mean lake level by 0.3 metres and mean annual outflow by 17%. The hydrological regime of the Shire River could shift from perennial to seasonal flow under combined warming beyond 1.5 °C and rainfall declines beyond 20%, or warming beyond 3.5 °C with rainfall declines beyond 15%. The most severe historical recession, between 1989 and 1997, saw lake levels fall roughly two metres, breaching the threshold and halting Shire River flows entirely.

The transboundary dimension is starkly asymmetric. Tanzania occupies only 21% of the lake’s catchment area yet contributes approximately 52.9% of total inflows, reflecting the high precipitation and runoff efficiency of its highland catchments; the Ruhuhu River alone supplies about 18% of all inflows. Mozambique contributes 7.1% and Malawi 40%. In practical terms, climate variability in a country with no direct stake in Malawi’s water infrastructure exerts dominant influence over whether that infrastructure functions — a structural dependency the study argues existing governance frameworks do not address.

Compiling records from humanitarian reports, disaster management archives, meteorological bulletins and peer-reviewed literature, the researchers identified 43 cascading climate events between 1990 and 2023: 24 drought-type and 19 flood-type cascades, each defined by a documented hydrological driver, propagation through the lake-river system, and measurable disruption to water supply or hydropower. Mann-Kendall trend analysis revealed a statistically significant acceleration in event frequency, with Tau = 0.48 and p < 0.01. The 1991–1995 drought sequence caused the most severe historical impacts, including prolonged cessation of Shire River flows and water rationing in Blantyre, while the 2015–2016 El Niño event cut hydropower output by roughly 30% and forced emergency rationing.

Future projections sharpen the picture. Drought intensity, measured through the Standardized Precipitation Evapotranspiration Index, is projected to increase by 25%–50% during 2021–2050 and by 131%–388% during 2071–2100 under RCP4.5 and RCP8.5 respectively, with three to five additional drought months per year in the near term and seven to eight by century’s end. Projected droughts under RCP8.5 are 1.7 times more severe than under RCP4.5. Lake levels are projected to fall by between 0.4 and 1.2 metres depending on scenario and period, outflows to decline by 7% to 35%, and hydropower production — which supplies over 95% of Malawi’s electricity — to drop by 1%–2.5% by mid-century and 5%–24% by 2071–2100.

The study also documents that risk runs across the full hydrological spectrum. A 2024 high-water event at Walkers Ferry Dam disrupted intake operations and raw water abstraction for Blantyre’s treatment system, illustrating a flood-type cascade. The authors are careful to note, however, that this flood-type finding rests on a single personal communication from Blantyre Water Board staff and has not been independently corroborated in published records; they present it as an observation warranting systematic investigation rather than an established result. Still, it suggests that infrastructure and protocols optimized for drought may be poorly suited to rapid swings toward flood conditions — a pattern consistent with so-called hydrological whiplash under warming.

The governance findings are among the most striking. Scoring Malawi’s National Water Policy against six dimensions of transboundary climate risk management, the analysis found severe gaps: transboundary coordination scored 2.1 out of 10, data-sharing protocols 1.8, and cascading risk recognition just 1.5. Climate change integration scored 3.2, reflecting general adaptation language without provisions for threshold-based risks. The SADC Revised Protocol on Shared Watercourses establishes principles of equitable utilization but lacks binding mechanisms for real-time hydrological data exchange. Stakeholder interviews reinforced the operational consequences: Blantyre Water Board operators frequently learn of critical lake level changes through informal channels rather than formal notification systems.

The vulnerability assessment places Walkers Ferry Dam in the high-exposure, high-sensitivity quadrant even under baseline conditions, with adaptive capacity rated moderate-low due to limited finances, dependence on external expertise, and the absence of formal transboundary data-sharing. Under RCP8.5 projections for 2071–2100, the system is projected to operate in conditions of extreme vulnerability in approximately 40% of years, with estimated service disruption rising from 18–35 days per year currently to 60–120 days. Malawi’s exposure is compounded by near-total dependence on a single abstraction point with no alternative raw water sources.

From the triangulated evidence, the authors propose five resilience strategies implementable within a decade: transboundary early warning systems integrating real-time lake telemetry, upstream precipitation monitoring in Tanzanian tributary catchments, and probabilistic threshold-exceedance forecasts at 30–90 day lead times; adaptive engineering of intake structures to operate across projected lake and river levels of roughly 470–480 metres; formalized data-sharing agreements among the riparian states, using the UNECE Water Convention as a legal template; nature-based solutions such as reforestation and watershed rehabilitation in upstream headwaters to sustain dry-season baseflows; and regional climate risk governance frameworks with pre-agreed response triggers. The authors acknowledge limitations, including data scarcity in the observational record, the coarse resolution of the climate ensemble, and policy scores based on documents that may miss informal practices. Comparative benchmarks — Cape Town’s Day Zero crisis and repeated El Niño-driven hydropower disruptions in the Zambezi basin — underline their central conclusion: infrastructure operating near a critical threshold demands fundamentally different, cooperative governance than systems with larger safety margins.

Subject of Research: Cascading cross-border climate risks to drinking water infrastructure in the Lake Malawi-Shire River system

Article Title: Building resilience of water infrastructure to cascading cross-border climate risks: a case study of Malawi’s drinking water infrastructure

Article References: Joshua, M. D. K., Ngongondo, C., & Kasei, R. A. (2026). Building resilience of water infrastructure to cascading cross-border climate risks: a case study of Malawi’s drinking water infrastructure. Frontiers in Earth Science, 14, Article 1903297. https://doi.org/10.3389/feart.2026.1903297

Image Credits: AI Generated

DOI: 10.3389/feart.2026.1903297

Keywords: Lake Malawi, Shire River, Malawi, transboundary water management, climate risk, drought, water infrastructure, Blantyre, hydrological threshold, CORDEX-Africa, climate projections, water governance