ASINT / agrobusiness
On May 25, 2026, the New York Times reported that Tsitakakantsa, Madagascar’s oldest and largest recorded baobab tree, is dying. The tree, estimated at 1,000 to 1,500 years old, with a trunk measuring nearly 29 metres in circumference and a height of approximately 25 metres, began showing signs of distress in August 2025 following Tropical Storm Jude in March of that year. By May 2026, half of the tree had fallen. Mould was present within the wood. Trunk cracks and foul-smelling seepage indicated advanced internal decay. Onja Razanamaro, a baobab specialist at Madagascar’s Tsimbazaza Zoo and Botanical Gardens, described the loss in terms that resonate beyond botany: the tree was a constant presence in the landscape, a subject of almost filial devotion for the surrounding community. Tsitakakantsa succeeded Tsitakakoike, another sacred baobab in the village of Andombiry, which died in 2018. Two sacred trees lost within eight years in the same community. The pattern is not local. Between 2005 and 2017, five of the six largest baobabs on the African continent either died or their oldest parts significantly deteriorated. The lead researcher, Adrian Patrut of Romania’s Babes-Bolyai University, described the die-off as “statistically very unlikely” and identified increased temperature and drought as the primary suspected drivers.
For the TMG series, Tsitakakantsa is not an environmental story. It is a climate risk signal. The question it poses is what the accelerating loss of millennial natural systems tells us about the physical risk environment in which every African asset documented in this series operates.
The mechanism that killed Tsitakakantsa is the same mechanism that threatens infrastructure, agriculture and extractive operations across the continent: climate-amplified extreme weather events creating conditions for biological and structural failure that were previously outside the historical probability distribution. Tropical Storm Jude in March 2025 delivered rainfall intensity that exceeded what the tree’s spongy, water-storing wood could absorb without triggering fungal proliferation. Baobabs survived for millennia precisely because their water-storage physiology was adapted to the cyclical droughts and moderate rainfall of Madagascar’s southwestern dry forests. When rainfall intensity increases beyond the historical range, the adaptation becomes a vulnerability. The wood that stored water now incubates fungus. The resilience that sustained the tree for 1,500 years fails within 14 months.
The parallel to infrastructure is direct. African infrastructure is designed for historical climate parameters. Roads are engineered for rainfall intensities observed over the past 50 to 100 years. Bridges are designed for flood return periods calibrated to historical data. Mines plan water management systems based on historical precipitation patterns. Power systems are designed for temperature ranges that determine transmission line capacity and cooling system specifications. When climate shifts push these parameters outside the design envelope, the infrastructure fails in ways that the design did not anticipate. The Akosombo substation fire documented in the Ghana nuclear article knocked out 960 MW of transmission capacity in a single event. The flooding at Kamoa-Kakula in 2024, preceding the seismic event, demonstrated how water ingress into mining operations creates cascading operational disruption. The Hormuz-amplified fuel crisis documented in the infrastructure report compounds with climate-driven agricultural losses to create multi-vector economic pressure.
A PLOS One study published in April 2026 by Estelle Razanatsoa and colleagues analysed baobab tree rings and isotope records in southwestern Madagascar, reconstructing a 700-year rainfall record. The research identified a wet period from 1350 to 1450, followed by centuries of cyclical variation within a range that the ecosystem adapted to. The current period is characterised by rainfall events that exceed the upper bound of that 700-year range in intensity, if not in total volume. The distinction matters: it is not that it rains more, it is that it rains harder. Concentrated, intense precipitation events trigger the fungal infections and structural failures that are killing the baobabs. The same concentrated intensity drives the flash flooding, soil erosion and infrastructure damage that affect every sector.
The natural capital dimension connects this directly to the AfDB’s $1.3 trillion financing gap report documented in this series. The AfDB identified natural capital as one of seven mechanisms for closing the gap, arguing that Africa’s mineral reserves, forestry resources, agricultural assets and renewable energy potential are undervalued in national accounting frameworks and underutilised as collateral for sovereign borrowing. Baobabs are natural capital in the most literal sense. Madagascar’s baobab forests generate tourism revenue, support community livelihoods through fruit harvesting, provide water storage in arid regions, and contribute to soil stability and biodiversity. Six of Madagascar’s seven baobab species are found nowhere else on Earth. The IUCN classifies several as endangered. When natural capital deteriorates at the rate the baobab die-off demonstrates, the asset base that the AfDB’s framework relies on is itself depreciating.
The insurance and risk pricing dimension is where Tsitakakantsa’s decline connects to the investment environment. Climate risk is increasingly priced into sovereign debt, project finance and insurance premiums across Africa. Moody’s, S&P and Fitch all incorporate climate vulnerability assessments into their sovereign rating methodologies. The Africa infrastructure report documented that ODA fell 23.1% in 2025, the largest contraction on record, and that domestic capital must step up. But domestic capital, whether pension funds, insurance companies or commercial banks, prices risk based on historical loss data. If the physical risk environment is shifting faster than the historical data captures, the models underestimate risk. A 1,500-year-old tree dying from a fungal infection triggered by a single tropical storm is a data point that sits outside any historical model. It is, in statistical terms, a tail event becoming a central tendency.
The agricultural transmission is the most economically consequential channel. Madagascar’s economy is approximately 25% agricultural. Across sub-Saharan Africa, agriculture employs the majority of the workforce and is the sector most directly exposed to climate variability. The agritech article in this series documented $168 million in funding in 2025, a 19% decline, with debt replacing equity. The WAEMU review documented food price deflation that “flatters consumers but quietly drains the producers who carry the economy.” The West Africa health market article documented a $11 billion market where 70 to 95% of medical supplies are imported, funded by export revenue that depends on agricultural and extractive output. Each of these economic chains is exposed to the same physical risk that killed Tsitakakantsa: extreme weather events that exceed the parameters within which existing systems were designed to function.
The mining sector exposure is specific. Water is the single most critical operational input for mining after energy. Too little water halts processing. Too much water floods pits and destabilises tailings. The copper deficit article documented Kamoa-Kakula’s seismic event and revised guidance. The Burkina Faso Kiaka article documented operations in a security-constrained environment where climate variability adds a further operational layer. The South Africa mining budget article documented fuel costs jumping 38% on the Hormuz spike, a cost that compounds with climate-driven water management expenses that are rising across every mining jurisdiction. The blended finance article documented $479 million to close a solar-plus-storage project in Egypt. If climate risk reprices insurance and financing costs for African energy and mining assets upward, the blended finance architecture documented in this series must absorb that repricing, adding another variable to an already complex capital stack.
For investors and operators, the Tsitakakantsa signal is not about a tree. It is about the rate of change. A system that survived 1,500 years of climate variability failed within 14 months when conditions exceeded its adaptive range. The question for every asset documented in this series, from Kamoa-Kakula’s smelter to the Lobito railway to Kasi Cloud’s data centre to Diamba Sud’s open pit, is whether the design parameters account for the rate at which the physical environment is changing. The IRP 2025 that Eskom Green is built on assumes certain solar irradiance and wind profiles. The PPA structures that finance renewable projects assume certain generation profiles. The mine water management plans assume certain rainfall distributions. If any of these assumptions are calibrated to a climate that is shifting faster than the models project, the assets built on those assumptions carry a physical risk that is not fully priced.
The structural question is whether Africa’s infrastructure investment pipeline, the $1.3 trillion annual gap that the AfDB has quantified, incorporates climate resilience as a design parameter rather than as an afterthought. The projects documented across 64 articles in this series are being designed, financed and built in a physical environment that is demonstrably less stable than the one in which their design parameters were calibrated. A 1,500-year-old baobab that survived droughts, cyclones, fires and human encroachment could not survive 14 months of climate-amplified fungal infection. The question for the infrastructure being built today is whether it is designed for the climate it will operate in for the next 25 years, or for the climate that existed when the feasibility study was written. The answer determines whether the $1.3 trillion closes a gap or builds assets that the changing physical environment degrades faster than the investment recovers.