Africa’s solar power pools show rising synchronization risk under climate change

African regional power pools are built on an assumption that is weakening under climate change: that solar generation failures across member countries occur largely independently, allowing one country’s surplus to cover another’s deficit. A new study quantifies how far that assumption already strains, and how much further it may erode.

“Climate-driven synchronization of solar extremes threatens the resilience of Africa’s regional power pool,” recently published in npj Clean Energy, analyzes synchronized photovoltaic low-output events – days when countries representing half or more of a pool’s projected solar capacity simultaneously experience output below their historical 10th percentile – across all five African power pools under moderate and high emissions scenarios through 2100.

The authors describe their projected figures as indicative planning diagnostics rather than precise forecasts, given known biases in the climate models used, and emphasize relative differences between pools and the direction of change as their most robust findings.

Synchronization risk

The paper identifies a three-tier vulnerability hierarchy across the five pools. The West African Power Pool (WAPP) and Central African Power Pool (CAPP) show the largest projected increases in the ensemble mean: synchronized low-output days rise from roughly 21 and 24 per year historically to more than 100 per year under the high-emissions scenario (SSP5-8.5).

The Eastern African Power Pool (EAPP) shows substantial increases across most members, though the magnitude varies significantly by country. The Southern African Power Pool (SAPP) shows limited change, with synchronized days rising from 25.4 to 29.0 under high emissions – a pattern the paper attributes to its structural geographic characteristics.

The Maghreb Committee for Electricity (COMELEC) shows high but more slowly increasing synchronization; when Algeria and Mauritania simultaneously experience low output, the conditional probability of pool-wide synchronized conditions exceeds 95%.

“The assumption lives inside the reserve margin itself,” said Paul Adigun, a researcher at the University of Tsukuba and the paper’s lead author. “African power pools generally operate on reserve margins of 15% to 20%, and those margins are calibrated to historical variability under the expectation that weather-driven shortfalls occur largely independently across member countries. Our analysis shows that in several pools this independence does not hold and is weakening under warming.”

The physical mechanisms differ by region. In WAPP and CAPP, synchronization is driven primarily by within-month weather anomalies — Saharan dust transport, Harmattan winds, and mesoscale convective systems – rather than predictable seasonal cycles. The paper references field observations reporting irradiance reductions of 18 to 50% during moderate Harmattan dust episodes across West African locations, and up to 79% during extreme events.

Because these events reflect within-month anomalies rather than seasonal cycles, the paper frames them as a distinct planning challenge from the resource minima that operators already account for through seasonal backup sizing. COMELEC’s elevated synchronization reflects its compact geographic footprint of approximately 2,200 km, where Mediterranean blocking patterns and Saharan dust transport can influence all five members within 24 to 48 hours.

SAPP’s more limited projected increase is rooted in its latitudinal span: equatorial members such as Tanzania sit in atmospheric regimes governed by intertropical convergence zone-driven convection, while subtropical members such as South Africa and Namibia are controlled by the South Atlantic Subtropical Anticyclone – a physical separation the paper projects will persist under warming.

“What protects SAPP is not its grid design – it is the spread of its members across independent climate systems,” Adigun told pv magazine. “You cannot replicate that with wires alone, because transmission can move power between regimes but it cannot create regime independence where the geography doesn’t provide it.”

The paper identifies specific cross-border country pairs that contribute disproportionately to pool-wide synchronized event probability (SEP). In WAPP, the Nigeria–Burkina Faso pair accounts for 82.6% of SEP under SSP5-8.5, with a conditional probability of 85.7% for pool-wide synchronized conditions when both countries simultaneously experience low output. The Burkina Faso–Benin pair accounts for 76.0% of SEP, with a conditional probability of 92.8%. In COMELEC, the Algeria–Mauritania pair accounts for 96.2% of SEP, with a conditional probability of 96.3%. These figures reflect corridors where climatic coupling is strongest and where, the paper suggests, reserve sizing and storage assessments are most consequential – though the authors are explicit that full system reliability conclusions require integration with wind, hydro, and storage assessments.

Physical drivers

The study decomposes the drivers of projected amplification into thermal and radiative components. Thermal degradation – projected warming of 3 C to 5 C translating to roughly 12 to 18% efficiency loss from silicon’s temperature coefficient – is a universal effect across all pools. Radiative changes diverge by region: WAPP faces projected increases in dust-driven surface irradiance reductions, while SAPP shows some projected atmospheric clearing that partially offsets thermal losses.

“Thermal degradation sets the continent-wide baseline you cannot diversify your way out of,” Adigun said. “Synchronization determines the spatial structure of the residual risk on top of that baseline. One is a hardware problem solved at the panel; the other is a network problem solved at the grid.”

The paper’s inter-pool correlation analysis identifies continental-scale diversification potential that current planning frameworks do not fully capture. WAPP and SAPP show a negative correlation of minus 0.31 in solar low-output occurrence, and SAPP and EAPP show minus 0.25, indicating that deficits in one pool tend to coincide with more favorable conditions in another. WAPP’s peak synchronized risk falls in summer and autumn, while SAPP conditions are generally more favorable in those seasons. The paper notes that transcontinental transmission connecting pools with negatively correlated solar output could provide diversification benefits beyond what intra-pool integration can deliver, though the authors acknowledge these are the hardest corridors to finance and govern across sovereign boundaries.

One such corridor is under construction. The Tanzania–Zambia interconnector, a 620 km line linking SAPP and EAPP, was launched in 2025 with a 2030 completion target and a €268 million ($309.2 million) financing package led by the World Bank, with contributions from the EU’s Global Gateway initiative and the United Kingdom. The paper identifies Tanzania–southern SAPP member connections as holding particular diversification value: Tanzania sits in an equatorial atmospheric regime decoupled from the subtropical dynamics governing South Africa and Namibia, and the two sub-regions show negative solar output correlation – consistent with the paper’s finding that connections between climatically distinct regimes are where transmission delivers genuine diversification rather than linking already-correlated systems.

The study’s scope is solar-specific. It does not model storage, demand response, or compensation from wind and hydropower. The authors note that Harmattan dust episodes in WAPP coincide with strong wind seasons in Niger and Chad, which could partially offset the solar deficits identified, and that countries such as the DRC and Ethiopia – where hydropower dominates planned capacity – experience these dynamics primarily as backup providers to solar-dependent neighbors rather than as direct managers of solar risk. The absolute magnitudes of all projected figures should be read as indicative planning diagnostics: the CMIP6 models used underestimate photovoltaic low-output occurrence by 40% to 89% relative to observational data, and the authors emphasize relative differences between pools and the direction of change as their most robust findings.

The paper estimates that approximately $50 billion in renewable energy generation and transmission investment is planned across African power pools over the next decade, with solar expected to dominate new capacity additions. Infrastructure decisions being made now are projected to shape system exposure for 30 to 40 years.