SAVÉ VALLEY CONSERVANCY, ZIMBABWE — Sango Wildlife Conservancy today opened its ecological archive and aerial census records to external scientific review, publishing the data behind the hardest decision in its history: the reduction of its elephant population from roughly 950 animals toward an ecologically sustainable 180.

The responsibility rests not with the elephants, but with the humans who confined them.

What is at stake: 8.5 million tonnes of CO₂ and a functioning water catchment

Sango is 600 square kilometres of intact southern African lowveld: mopane woodland, baobab stands, perennial grassland and riverine systems within the Savé River catchment. Landscapes of this kind are among the least understood and most undervalued climate assets on the continent.

Sango has had its carbon stock independently audited and certified by TÜV Nord (Germany). The property holds and protects approximately 8.5 million tonnes of CO₂, held in a single privately managed block of African dryland.

That store is mostly below ground. Unlike rainforest, the carbon in a dry African savanna sits in deep perennial grass root systems, in woodland root mass, and in the soil those roots hold together. It is invisible, and it is destroyed silently. Ground that loses its grass roots and its tree canopy does not simply hold less carbon; it loses the structure that made it a store at all. Bare soil oxidises, erodes in the first hard rain, sheds water it would once have absorbed, and silts the rivers downstream.

The same intact vegetation is what buffers this landscape against drought. A woodland with canopy shade, litter cover and rooted soil holds moisture through a failed season. Bushland on eroded ground does not. In a region where rainfall is becoming less reliable, that difference decides whether the land recovers after a dry year or steps down permanently to a poorer state.

Sango sits in Natural Region V, an agro-ecological zone the Zimbabwean government classifies as unsuitable for sustainable rain-fed agriculture. That classification is the point. On land like this there is no productive alternative use that does not degrade it. Conservation is not the sentimental option here: it is the only land use that leaves the soil, the water and the carbon where they are. If a reserve of this size fails, it does not become a different kind of reserve. It becomes settlement, unmanaged grazing, charcoal and bare ground, and what it held is released.

That is the scale on which Sango asks this decision to be judged.

“Sango is a roaring success of a sustainable utilisation model returning to pristine wildlife what was before 70 years of cattle ranching severely disrupted nature. The success brings with it the challenge of a wildlife presence beyond a food base able to feed it. It further demands the decision to sacrifice a few for the survival of the whole. That’s a decision I took and I stand by it.”

Wilfried Pabst, Founder

The evidence: what the last twenty years have done to the land

Sango’s ecological database covers more than 25 years of continuous recording: soils, vegetation, rainfall, forage availability, fauna and flora; maintained to Kruger National Park standards and assessed by TÜV Nord (Germany) and under IUCN Green List processes as among the most comprehensive datasets held by any private conservation area in Africa.

It records the following:

• Grazing: Sango has lost the large majority of its grazing capacity over the past two decades, under the combined pressure of increasingly erratic rainfall and elephant density.
• Mopane woodland: roughly 30 percent has been reduced from woodland to bush.
• Baobabs: 93 percent of Sango’s baobabs, one of the densest concentrations of the species in the region, carry elephant-induced damage, ranging from minor to severe.

Baobabs show most clearly what “irreversible” means in practice. A mature baobab lost today is not replaced on any timescale that matters to anyone now living: not in a century, and in many cases not in a thousand years. There is no restoration programme for a loss of that kind.

Why elephant density does this, and why it is not the elephants’ fault

Elephants do not feed like cattle. They pull grass out with the roots, which prevents regrowth, and they ring-bark and push over mature trees at scale.

Toppling a tree looks, briefly, like abundance: the canopy opens and more browse becomes available at ground level. What disappears is the woodland. Repeated browsing on saplings stops young trees replacing the old ones, so the structure is never rebuilt. With the large trees go the cavities, shade, fruit and insect life that birds, reptiles, small mammals and pollinators depend on. The loss runs the length of the food web, and it runs downward into the soil.

At five times the sustainable density, the damage does not simply multiply by five. It compounds, because the vegetation is stripped again before it has recovered from the last impact, and each cycle leaves less to recover from.

This density is human-made. The land now inside the Savé Valley Conservancy was cattle ranching, carrying no wildlife, before it was restored. Elephants were brought here: around 500 translocated from Gonarezhou National Park in the mid-1990s. That founder population now numbers more than 3,000 across the Conservancy, and newer camera-based counting methods suggest the true figure is higher still than manual aerial counts indicate. Sango holds about 23 percent of the Conservancy’s area and close to a third of its elephants.

“This density is human-caused, not elephant-caused. We brought these animals here. That is why the responsibility to sacrifice some for the survival of the whole is ours.”

Wilfried Pabst

The numbers

Aerial surveys by trained ecologists, supported by drone and camera monitoring, counted approximately 950 elephants on Sango in September 2025. Sango has flown aerial counts for more than twenty years; the trend is consistent and documented.

Sango’s ecological assessment, covering soils, vegetation, rainfall, forage availability and regional scientific data, and reviewed by independent specialists, puts sustainable carrying capacity at approximately 0.3 elephants per square kilometre: about 180 animals for the property.

The population is therefore more than five times what the land can carry. On Sango’s own projection, if the density is not brought down, the reserve as a functioning ecosystem has less than a decade. That is why the decision is being taken now rather than deferred again.

Since May 2025, about 367 elephants have been culled. Sango estimates the current standing population at approximately 600. Removals do not translate directly into a lower count: Sango is not fenced against the rest of the Conservancy, and animals move into the space they create.

“This isn’t a manufactured crisis, it’s arithmetic. And arithmetic can be checked. We’re providing the numbers for exactly that. I put this decision off for more than ten years, and I am to blame for not making it earlier. It is hard, and it is emotionally draining, but it has to be done. We are the ones who put fences around wildlife. We are the ones who have to manage what that causes, if we want to protect and grow the whole of Sango.”

Wilfried Pabst

The decision, and why family groups are taken whole

The part many people find hardest is that entire family groups are removed rather than individual animals. The reason comes from the research most often cited against culling.

In the 1980s and early 1990s, population-reduction operations in South Africa’s Kruger National Park killed adult elephants and spared calves, which were then relocated to Pilanesberg National Park. A 2013 study in Frontiers in Zoology (Shannon et al., “Effects of social disruption in elephants persist decades after culling”, Front Zool 10:62) compared those animals decades later against a largely undisturbed population in Kenya’s Amboseli National Park. The Pilanesberg survivors were still unable to reliably distinguish callers by social familiarity or age-related dominance, social judgements the Amboseli population made readily.

Sango states the limits of that finding plainly. The study documents lasting harm in animals that survived the loss of their family. It does not compare methods of removal, and it takes no position on whether elephants should be removed at all. What follows from it is one point: it matters whether survivors are left behind. A family group is therefore removed in its entirety, within minutes, so that no animal is left to witness the loss of its herd.

The approach that sounds gentler, taking individuals and sparing the rest, is the approach that produced Pilanesberg. It is easier to accept from a distance. The animals left behind are the ones who carry it.

“There is no version of this without suffering. There is only a choice of what actions to take to allow the whole to survive and flourish. If you haven’t got the guts to make hard decisions, don’t attempt to bring wildlife back to where it once was.”

Wilfried Pabst

What this is not

Culled animals are not hunted. The operation generates no hunting revenue and produces no trophies.

Ivory is held under lock at Sango, at Sango’s own expense, because the national parks authority does not cover the cost of safekeeping. It is not sold.

The operation runs on aircraft, helicopters, TLBs, crane trucks and a substantial vehicle fleet. Meat and hide sales offset those costs; any surplus in meat or proceeds is shared with local chiefs. Full financial records are available on site for review.

Alternatives, and why each one is not

Translocation. Sango has translocated elephants in the past, has offered a further 400 for translocation, and has searched for receiving areas across a 3,000-kilometre radius. Around 200 animals were moved to the Zambezi Valley in recent years; a further 50 are scheduled for the coming year. In 15 past translocations Sango has sent surplus predators, plains game and rhino to areas in need across Zimbabwe. Transport runs to roughly US$2,500 to US$3,000 per elephant over 800 to 900 km. Veterinary advice caps transport at three days, roughly 3,000 km, and Sango applies a stricter limit of two days and roughly 2,000 km. Within that range, receiving areas have run out: landmines in former conflict zones, heavy poaching pressure for ivory and bushmeat, shrinking wildlife areas under human pressure, and recovering elephant populations across their former southern African range.

Sango is not alone in reaching that conclusion. Ezemvelo KZN Wildlife, the state conservation authority in KwaZulu-Natal, stated publicly in April 2026 that live removal remains its preferred option but that after two decades of searching, suitable destinations for elephants have become critically scarce, and issued an open appeal to NGOs and landowners to identify any that remain.

Fertility control. Immunocontraception does not remove a single animal already on the ground. It addresses future growth, not present damage, and the timescales do not match: to halt the habitat loss now underway, numbers have to come down fast. At 950 elephants across 600 km², contraception would require large numbers of cows to be individually identified, reached by helicopter and re-treated repeatedly over years, in herds that range widely and change composition, while the woodland continues to be lost throughout.

The clearest test of that argument is not theoretical. Ezemvelo KZN Wildlife has run long-term contraception programmes across its elephant reserves, alongside sustained efforts to expand protected areas. In April 2026 it reported that Hluhluwe-iMfolozi Park, Tembe Elephant Park, Ithala Game Reserve and uMkhuze Game Reserve are nonetheless all carrying elephant populations above their estimated capacity, and that more than a thousand animals now need to be removed. Contraception slows growth. It does not reverse an existing overpopulation.

Sango’s monitoring of herds following whole-group removal indicates no lasting disruption to the structure of the remaining population.

Fence removal. Subsistence farming families live immediately beyond the boundary. Sango is legally required to maintain its eastern and western fences under problem animal control obligations. This protects neighbours’ livestock and crops, and makes poaching harder in return. Opening the fence would give elephants and predators access to crops, livestock, grain stores, homes and people; it would also make the protection of Sango’s wildlife, and in particular its critically endangered rhino, impossible to guarantee, and would open the reserve to unmanaged logging, grazing, settlement, cultivation and land degradation. The carbon and the catchment would go with it.

Zimbabwe’s country-wide elephant situation

The Zimbabwe Elephant (Loxodonta africana) Management Plan 2026–2036, published by ZimParks, puts the national population at 91,768 elephants, based on consolidated 2022 to 2024 aerial surveys. The plan notes this is nearly three times the national target population of about 35,000 envisaged in the early 1980s, and represents an 11.8 percent rise on the 2020 estimate of 82,091. It also states that the core state protected area of roughly 45,000 km² currently carries close to 80,000 elephants, about 1.8 per km², against a policy target density of 0.3 to 0.5 per km².

Sango note: with a net 5 percent annual progeny, the total number today stands closer to 100,000 elephants.

Sango is not an isolated case: the same arithmetic across the region

The situation on Sango is not unique to Sango, to private land, or to Zimbabwe. It is the predictable outcome of successful elephant conservation inside a fence, and state conservation agencies in South Africa are now confronting it on a larger scale.

Madikwe Game Reserve, North West Province. Madikwe is a 750 km² fenced reserve created on former farmland, restocked from 1992 under Operation Phoenix, one of the largest wildlife reintroductions ever undertaken. Its founder elephant population of roughly 220 animals came in part from Gonarezhou National Park, the same source as the Savé Valley herds. That population now exceeds 1,600. During the 2024 drought, according to the North West Parks and Tourism Board and the National Council of SPCAs, some 75 to 80 elephants died of starvation. The Board has since put forward a reduction programme, and the matter has been before South Africa’s parliamentary environment committee, where the process has drawn substantial criticism.

KwaZulu-Natal. Ezemvelo KZN Wildlife reported that as of 31 March 2026 the province held an estimated 2,862 elephants across 22 properties, 2,089 of them in reserves it manages. It has stated that its major elephant reserves are all above estimated carrying capacity despite long-running contraception programmes, that more than a thousand animals require removal, the largest single figure being 650 at Hluhluwe-iMfolozi Park, and that its reserves are all under 100,000 hectares and surrounded by dense human settlement. Ezemvelo has been explicit that no final decision on culling has been taken and that live removal is its preferred route; it has also said that culling could become necessary if destinations cannot be found.

Sango draws two conclusions from this, and no more than two.

The first is that a fenced reserve with restored elephants, no predation on adults, permanent water and no migration route out, will exceed its carrying capacity. This is arithmetic, not ideology, and it applies to state and private land alike.

The second is that the receiving capacity for elephants that Sango is told to use does simply not exist. When a provincial conservation authority with two decades of research and search behind it publicly appeals for somewhere to put a thousand elephants, the shortage Sango describes is not a private excuse.

Reduction on Sango alone cannot hold the line. Sango is not fenced against the rest of the Conservancy, and neighbouring properties carry the same surplus, so animals move into the space that removals create. The position Sango is managing toward is a sustainable stocking rate across the Savé Valley Conservancy as a whole, of around 900 elephants at 0.3 elephants per km².

Sango takes no position on how any other reserve should proceed. Those decisions belong to their custodians and their regulators. What Sango asks is that the debate about its own decision be conducted with the knowledge that this problem is regional, structural, and arising everywhere at once.

The Kasungu precedent: why translocation is not a cost-free alternative

Translocation is the first option Sango reaches for every time, but only where the receiving area can genuinely take the animals. Where it cannot, the price is paid by the farmers on the boundary. Kasungu is what the recommendation looks like when it is carried out at scale under the wrong conditions.

In 2022, 263 elephants were moved from Liwonde National Park to Kasungu National Park in Malawi. At release, 47 kilometres of the park’s roughly 267-kilometre boundary were fenced. Within 24 hours, two bulls left the area and killed two men. Reported deaths in the surrounding communities since the translocation stand at more than ten.

Retrofitted fencing now covers about half the park, with no plan for the remainder. According to reporting by bioGraphic, the responsible organisation states that the Zambian side is to remain permanently unfenced as part of a wildlife corridor initiative. A group action brought by affected communities was initiated in December 2024, according to the claimant firm.

The point is not that translocation is always wrong. Sango has done it repeatedly and is doing it again this year. The point is that moving elephants into an area that cannot hold them does not solve the problem. It transfers it onto rural people who had no say in the decision.

The people on the boundary

Sango employs approximately 200 permanent staff, 98 percent Zimbabwean and around 90 percent from immediately neighbouring villages, plus roughly 30 seasonal workers. On a standard estimate of ten dependants per job, those positions support an estimated 2,300 people.

Grazing land supporting local cattle, sheep and goats is shrinking sharply under recurring drought. Meat from population-management operations is distributed to surrounding communities. Sango does not present this as a solution to food security, but under these conditions it is a real source of protein.

Opening the record

Sango is making available for external scientific review:

• Aerial census records spanning more than twenty years, with count protocols and methodology
• The carrying-capacity assessment and its underlying soil, vegetation, rainfall and forage data
• The full 25-year ecological database
• The TÜV Nord carbon audit and its underlying measurements
• Financial records for the population-management operation

Access is on site and open to qualified experts.

Sango further invites independent researchers to conduct a full carbon and soil assessment of the property, so that what this landscape holds, and what is lost when dryland woodland collapses, can be measured rather than asserted.

The photograph

Sango is aware of a widely shared social media post accusing founder Wilfried (“Willy”) Pabst of running an industrial-scale elephant killing operation for financial gain, illustrated with a photograph of a culling operation.

Sango disputes the characterisation. It does not dispute the image.

The photograph was taken on Sango on 23 October 2025 by a press photographer working for a British daily newspaper, whom the conservancy had invited to document the operation. The newspaper published an article, but not the photograph now in circulation; that image has been submitted to the World Press Photo Contest 2026.

“Nobody is pretending this image is easy to look at, least of all me. We gave access anyway. A decision whose images you’d need to hide is not a decision you should be making.”

“Economic gain at the expense of the ecology is not permitted in Sango’s business model. And there is no meaningful profit to be made in an operation like this in any case. The records are open.”

Wilfried Pabst

Sango’s challenge to its critics

Sango’s position is that the criticism should be testable in the same way its own numbers are. It therefore invites its critics to bring forward a destination that meets the conditions any receiving area has to meet:

• Suitable habitat with the forage and water to carry the animals
• Safe for the elephants on arrival
• Accepted by the local population who will live alongside them
• Accessible to 30-tonne transport vehicles
• Within 2,000 km, or two days’ drive, of Sango

And to bring the funding to move them.

Sango has searched for such a destination for more than a decade and has not found one within reach. Ezemvelo KZN Wildlife is currently making the same appeal, for more than a thousand animals, and reports the same lack of destinations. If such a destination exists, Sango will use it. Until then, the alternatives on offer are not realistic alternatives.

ENDS