The study highlights the importance of protecting the Upper Zambezi Basin, where another recent study recorded significant forest loss over the past three decades.
A new study suggests that the Zambezi River, Africa's fourth-longest, is 11% longer than previously thought, with its most distant source lying in Angola, not Zambia. While the finding leaves the Zambezi as Africa's fourth-longest river -- after the Nile, the Congo and the Niger -- the redefinition of its most distant source underscores Angola's enormous contribution to sustaining the river best known for Victoria Falls.
This newly identified source in Angola's southern highlands sits 3,421 kilometers (2,126 miles) from the Zambezi's mouth at the Indian Ocean -- 342 km (213 mi) farther than its original source, a spring bubbling out of the ground in a marshy patch of woodland near the Zambian town of Mwinilunga.
The geographic definition of a river's source is the farthest point from its mouth, and based on that criterion, the team behind the study published in the journal Hydrology and Earth System Sciences has identified the source of the Zambezi as a small trickle oozing from a peatland bog in southern Angola, which is also the source of a river known as the Lungwebungu.
Lead author Rainer von Brandis of The Wilderness Project (TWP), a group of scientists and explorers that conducted the study, says the findings highlight the critical importance of the Angolan highlands, also known as the Angolan water tower, as the "big contributor" to the Zambezi.
"Having paddled [down the length of each river from] both of the sources -- the Zambian one and the Angolan one -- we figured out that 70% of the water that arrives in [Lake] Kariba [a giant hydropower reservoir on the Zambezi] or goes over Victoria Falls comes out of Angola," he says.
"That the Lungwebungu is further away from the end-point of the Zambezi than the traditional source in Zambia is just kind of an interesting fact; it makes the Zambezi longer, which gives more impetus to the river and gets all the downstream users to understand that this river essentially comes from even further [away], and the headwaters [in Angola] really are important."
Three other major Angolan tributaries -- the Chifumage, Luena and Luanginga -- also make significant contributions.
Von Brandis recalls how TWP researchers, traveling down the main stem of the Zambezi in 2023, started their journey from the spring near Mwinilunga, followed it for two days on foot, then paddled their canoes 100 km (60 mi) to where the river entered Angola.
Here, they discovered it was flowing at a rate of just 16 cubic meters (565 cubic feet) per second. A second team that paddled the Angolan section for another 400 km (250 mi) or so discovered that when the river reentered Zambia at a place called Chavuma, the Zambezi was now flowing at a rate of 900 m3/s (31,800 ft3/s) -- a more than 5,500% increase.
"That's how much the river grows during the Angolan section of the Zambezi," Von Brandis says.
The Angolan tributaries are fed by source lakes and peatlands up in the water tower, where deep Kalahari sands and unbroken expanses of miombo woodland aid the storage, percolation and filtration of massive amounts of rainwater.
A study two years ago by TWP researcher Mauro Lourenço, also a co-author of this latest study, estimated that the Angolan water tower receives annual rainfall equivalent to nearly the entire contents of Lake Erie. This water is stored not only in the lakes and peatlands, but in giant wetlands such as the Cameia on the eastern edge of the highlands, which is in turn drained by the Luena and Chifumage rivers.
The sand filtration means that water entering the Zambezi from the Angolan water tower is "crystal clear," Von Brandis says. "That is a massive ecosystem service for the downstream users."
Two separate expeditions, in 2022 and 2023, covered all 1,032 km (641 mi) of the Lungwebungu.
Although the team discovered evidence of unregulated diamond mining along the Lungwebungu's Angolan stretch, including at the source, they counted few people along its entire length.
"It was a very remote piece of river, and very much intact ecologically," Von Brandis says.
The same can't be said about the Kabompo, another Zambezi tributary equivalent in size to the Lungwebungu that flows out of Zambia's heavily industrialized Copperbelt province. The Kabompo is laden with pollutants, sediment and the seeds of a notoriously invasive South American shrub -- the giant sensitive plant (Mimosa pigra) -- that clogs waterways and outcompetes native riverine plants.
Frank Nyoni, a Zambian co-author of the study, former acting director in Zambia's Water Resources Management Authority and now regional manager with TWP, tells Mongabay that regardless of where its true source lies, the Zambezi remains critical to Zambia, and not just for hydropower, agriculture and tourism.
"The Zambezi is an important river when it comes to our cultural heritage," Nyoni says.
In western Zambia, for instance, the Lozi community holds its annual Kuomboka barge ceremony on the Barotse floodplain around April. But if the river drops too low during severe drought, the ceremony is cancelled, depriving the community of vital tourism revenue.
It's a striking example of how this river at the heart of Zambia's identity is vulnerable to the impacts of climate change, but also dependent on responsible stewardship by others who share its headwaters, its course and its massive basin: namely, Angola, Botswana, Malawi, Mozambique, Tanzania and Zimbabwe.
"It means that countries need to work together more to avert the water stresses, which are induced because of drought," Nyoni says.
Kawawa Banda, an associate professor of water resources at the University of Zambia, who wasn't part of the TWP study, says it makes a strong case for redefining the Lungwebungu as the true source of the Zambezi. But more importantly, he adds, it underscores the need to protect a key part of the Zambezi's catchment, currently threatened by deforestation in both Angola and Zambia.
Last year, Banda and colleagues published research showing that nearly 4.4 million hectares (10.9 million acres) of forest -- about 14% of the Upper Zambezi Basin -- has been degraded over the past three decades, mainly through conversion to croplands. Banda and his colleagues predict that another 2.8 million hectares (6.9 million acres), or 8%, will likely be lost over the next decade, threatening ecosystems, hydropower generation and agriculture in the lower Zambezi.
"You can see from the data that there is so much sedimentation coming from upstream in the Angolan highlands," Banda says. "All these materials [from runoff] are ending up in the Barotse floodplain, which is acting as a sediment trap at the moment."
Von Brandis says TWP, through its Angolan sister organization, Fundaçao Lisima, is building a small conservation base and laboratory in the town of Tempue, located at the top of the water tower. Here, young local researchers will conduct long-term monitoring studies into biodiversity, climate and hydrology.
"We have a whole different branch of social scientists working with the local communities there to establish community protected areas," he says.
"People up there know conservation; they understood, very well, how to protect their land, so it's not a new concept to them."
In July, at a Ramsar Convention on Wetlands held in Victoria Falls, TWP presented its case for a large part of the Angolan water tower to be declared a Wetland of International Importance, to boost its recognition and protection. Approval for the site, known as Lisima lya Mwono, or "the source of life," is expected later this year or early next.
Banner image: The river survey team near the Zambian border. Image courtesy of Jen Guyton /The Wilderness Project/NGOWP.
New study pushes for protection of one of Africa's 'least understood treasures'
All conservation is local: Interview with Angolan conservationist Kerllen Costa
Bicolored waterberry: The overlooked tree shaping Zambia's rivers
Defending the hidden forest gems of Zambia's Copperbelt Province
Lourenço, M., Boyes, R. S., Cotterill, F. P., Flügel, T., Nyoni, F. C., Ramatlapeng, G. J., & Von Brandis, R. (2025). Assessment of source regions of the Zambezi river: Implications for regional water security. Hydrology and Earth System Sciences, 29(18), 4557-4583. doi:10.5194/hess-29-4557-2025
Lourenço, M., & Woodborne, S. (2023). Defining the Angolan Highlands Water Tower, a 40 plus-year precipitation budget of the headwater catchments of the Okavango Delta. Environmental Monitoring and Assessment, 195(7), 859. doi:10.1007/s10661-023-11448-7
Zimba, H. M., Banda, K. E., Mbewe, S., & Nyambe, I. A. (2024). Integrated use of the CA-Markov model and the Trends.Earth module to enhance the assessment of land cover degradation. Environmental Systems Research, 13(1). doi:10.1186/s40068-024-00355-6
This article is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.
You may republish this article, so long as you credit the authors and Mongabay, and do not change the text. Please include a link back to the original article.