Rewilding 500 Acres

Conservation, done honestly, is mostly unglamorous. It is not a grand gesture or a ribbon-cutting; it is years of quiet, repetitive work, some of it tedious, some of it back-breaking, most of it invisible to anyone passing through. At Reserva Sete Quedas we protect and slowly heal a piece of the Atlantic Forest in the mountains above Paraty, and this page lays out exactly how we do it and how we pay for it. No slogans, no invented figures. Just the model, the trade-offs, and the reasoning behind both.

Reserve size

800 acres

Continuous protected forest

500 acres (100 owned + 400 adjoining parkland)

Land being rewilded

old pasture and farm edges within the reserve

Organic banana farm

~15 acres, roughly 300 years old

Water

3 streams + 4 springs feeding the seven waterfalls

Chalets, maximum ever

25 (one per two acres on 50 acres)

Biome

Mata Atlantica, Serra do Mar / Serra da Bocaina

Two different jobs: preserving and restoring

People tend to lump all forest work under one word, conservation, but on the ground it splits into two very different tasks that call for different tools, different timescales, and different states of mind. The first is preservation: keeping intact forest intact. The second is restoration, sometimes called rewilding: helping land that was cleared or degraded find its way back to forest. Both matter here, because our 800 acres are a mosaic of old-growth and disturbed ground, and neither job substitutes for the other.

Preservation sounds passive, as if you simply leave the trees alone and walk away. In practice it is the opposite. Intact forest that no one watches gets nibbled to death: a clearing widened here, a hunter's trail cut there, a fire that escapes a neighbor's pasture, a load of rubbish dumped where the road ends. Preserving standing forest means active defense of a boundary, season after season, and it is arguably the higher priority of the two. A mature tree that took a century and a half to grow cannot be replaced in a human lifetime. The single most valuable thing we do is make sure the old forest we already have never gets cut, burned, or fragmented. Everything else is secondary to that.

Restoration is the slower, more hopeful half of the work. Parts of the reserve were, at different points over three centuries, cleared for pasture, for subsistence crops, or for the farm. Some of that ground is still open or half-open. The question for every one of those acres is simple to ask and hard to answer well: what is the fastest, cheapest, most durable way to turn this back into functioning forest? Often the answer is to do very little and let the forest do the work itself. Sometimes it means active planting, weeding, and years of follow-up. Knowing which is which is most of the skill. Our broader philosophy is set out on the conservation page; this one gets into the how.

Why "rewilding" is the honest word

We use the word rewilding carefully. In some parts of the world it means reintroducing large animals to reshape a landscape, wolves, bison, and the like. We are not doing that, and we would be misleading you if we implied it. What we mean by rewilding here is more modest and more appropriate to the Atlantic Forest: letting a working, lived-in property tilt steadily back toward wildness, giving natural processes room to run, removing the things that block them, and accepting that the forest is a better designer of itself than we are. The goal is not a garden. The goal is a self-sustaining forest that needs us less every year.

The land, in plain numbers

It helps to be precise about what we are protecting, because the phrase "protected land" gets stretched to mean almost anything. The reserve is 800 acres in total. Within it, 100 acres are privately owned protected forest, held by us and kept off the market for any use that would damage it. Directly adjoining our boundary is a further 400 acres of protected parkland, held as natural Atlantic Forest outside the ownership line. Those two pieces join to make 500 acres of continuous protected forest, one unbroken block rather than a scatter of fragments. The protected forest page walks through the legal and physical side of that in more detail.

The rest of the 800 acres carries the human uses that keep the place alive and paying its way: the roughly 15-acre organic banana farm, about three hundred years old and threaded with donkey-and-mule roads laid down long before any engine reached these slopes; and the 50-acre "ecomix park," where we allow exactly one low-impact chalet per two acres, for a hard ceiling of 25 chalets, ever. That cap is not a marketing number we might quietly raise later. It is the load limit of the land, and it is described in full on the ecomix park page.

The land being actively rewilded sits mostly in the seams between these zones: old pasture edges, patches of ground cleared decades ago and never fully recovered, and the margins where the farm meets the wild forest. It is not glamorous acreage. It is exactly the kind of tired, in-between land that, across the Atlantic Forest, either creeps back into forest or gets swallowed by development. We are betting it can do the former.

Letting pasture become forest again

The most powerful restoration tool we have is patience, backed by the extraordinary regenerative appetite of this particular biome. The Atlantic Forest, when it is warm, wet, and close to a seed source, wants to grow back. Abandon a pasture near standing forest in this climate and it does not stay a pasture. Within a couple of wet seasons the first pioneers arrive, and within a few years, if nothing stops them, you have a young, scruffy, fast-growing thicket that is unmistakably becoming forest.

This process has a name, ecological succession, and watching it happen on your own ground is one of the quiet pleasures of this work. The first colonizers are the pioneer species: fast, sun-loving, short-lived plants and trees that thrive in the open and hard light of an old pasture. Around here the embauba, the trumpet tree with its pale trunk and big hand-shaped leaves, is a classic early arrival, and its fruit pulls in birds and sloths, which in turn drop the seeds of the next wave. Pioneers grow quickly, shade out the pasture grasses, cool and shelter the ground, and then, having done their job, get overtopped and replaced by slower, longer-lived trees that can only germinate in the shade the pioneers created. The forest builds itself in layers, each generation preparing the ground for the next.

Our default, wherever a patch of open ground is close enough to intact forest and the soil has not been too badly beaten down, is to let this happen on its own. Natural regeneration is cheaper than planting, it is more genetically diverse than anything we could plant, and the trees that arrive are the ones the local birds and animals actually chose to bring, which means they belong here and they know how to survive here. We keep grazing animals out, we keep fire out, we knock back the worst of the invasive grasses so the tree seedlings are not smothered, and then we mostly get out of the way. It is astonishing how much forest you can grow by simply defending a piece of ground and waiting.

When we let it come back, and when we step in

Passive regeneration is not a cure for everything, and pretending otherwise leads to failed restoration and wasted years. It works when three things line up: a nearby seed source, animals to carry those seeds, and soil that is still alive. Take any of those away and the process stalls. A clearing far from mature trees may sit for years as a sea of exotic grass because no seeds arrive. Ground where the topsoil has eroded or compacted may resist tree roots. And a landscape stripped of its seed-dispersing birds and mammals loses the very machinery that moves the forest around.

So we read each site before deciding. Where regeneration is clearly working, we protect it and leave it alone. Where it has stalled, we ask why, and we intervene as narrowly as possible to fix the specific bottleneck rather than bulldozing the whole approach. Sometimes the fix is removing an invasive grass. Sometimes it is planting a scatter of hardy pioneer trees to create shade and draw in seed-carrying birds, a technique often called nucleation: instead of planting a whole field in neat rows, you seed small islands of vegetation that then spread outward and knit together on their own. It mimics how forests actually recolonize open ground, and it costs a fraction of blanket planting.

Natural regeneration versus planting

There is a reflex, when people picture forest restoration, to imagine rows of saplings and volunteers with spades, a tree-planting day with a photograph at the end. Planting has its place, and we do it where it is genuinely the right tool. But we are wary of treating it as the whole of restoration, because planting is expensive, labor-hungry, and surprisingly easy to get wrong. A field of nursery saplings that no one weeds, waters through the first dry season, or protects from grass and browsing is mostly a field of dead sticks a year later. The photograph looks good; the forest never arrives.

Natural regeneration, where conditions allow it, beats planting on almost every measure that matters. It is far cheaper, since the trees arrive for free. It produces a richer mix of species than any planting list a person would draw up, and a more natural spacing and structure. And crucially, the trees are locally sourced in the truest sense: they are the offspring of trees already thriving on this exact mountain, adapted to this rainfall, this soil, these pests. We would have to work hard to design something as well suited to the site as what the forest supplies for nothing.

Where we do plant, we plant with intent. We favor pioneer and early-succession species that establish fast and create the shaded, humid conditions later trees need. We source seed and seedlings as locally as we can, to keep the genetics of our forest local rather than importing stock from distant populations. And we plant to solve a problem, to bridge a gap between forest patches, to jump-start a stalled clearing, to stabilize an eroding slope, rather than planting for the sake of a number. A restored acre that survives and self-seeds is worth more than ten acres planted for a headline and then abandoned.

The seed-carriers do the real work

None of this succession happens without animals, which is the part people most often miss. Trees cannot walk. A forest recolonizes open ground almost entirely on the backs, in the guts, and in the beaks of its wildlife. Toucans and other fruit-eating birds swallow seeds and drop them across the clearings. Bats carry others by night. Agoutis and other rodents bury seeds and forget some of them, which is precisely how large-seeded trees get planted. Where the forest still has howler monkeys and a healthy cast of frugivores, the seed rain into nearby clearings is heavy and varied, and regeneration is fast. Where those animals are gone, hunted out or cut off, the seeds stop arriving and the forest cannot rebuild itself no matter how much bare ground is available.

This is the deep reason our wildlife and our restoration are the same project, not two projects. Protecting the seed-dispersing animals is protecting the forest's ability to grow itself. A juçara palm, for instance, produces fruit that a whole guild of birds and mammals depend on, and in turn those animals are the ones that plant the next generation of palms. Break that loop, by hunting the animals or by stripping out the palms, and both halves collapse together. Keeping the loop intact is worth more to the future forest than any planting program we could run.

Removing what does not belong

If patience is the tool for growing forest, then steady, unromantic labor is the tool for defending it, and nowhere more than in the fight against invasive species. Centuries of farming, ornamental planting, and traffic up and down the mountain left this landscape, like most of the Atlantic Forest, seeded with aggressive non-native plants that outcompete the natives and stall regeneration. Pulling them back is some of the least photogenic and most important work we do.

The main offenders

A few invaders cause most of the trouble here. African pasture grasses, brought in over generations to feed cattle, are the worst of them: dense, fast, and flammable, they carpet old clearings and choke out tree seedlings before they can get above the blades, and in the dry season they turn a recovering clearing into fuel for fire. Jackfruit, the jaqueira, is another: a generous fruit tree that unfortunately loves the Atlantic Forest a little too much, spreading into the understory and forming thickets where a diverse native community should be. Remnant pines and eucalyptus from old plantation stock seed themselves into open ground. And various ornamental and agricultural escapees turn up along the old roads and around former dwellings.

We treat these differently depending on how much havoc they cause and how entrenched they are. The grasses we suppress mechanically and by shading them out, cutting and smothering them so tree seedlings can get a head start, then letting the growing canopy finish the job, since most of these grasses cannot survive in shade. Woody invaders like jackfruit and stray pines we remove by hand and by saw, targeting seed-bearing adults first to stop the spread, then working through the younger ones over successive seasons. It is slow, repetitive work, and it is never truly finished, because seeds keep arriving from beyond our boundary. Invasive control is less a project with an end date than a permanent chore, like keeping a boat free of water.

We are honest about the limits here. We cannot and do not try to sterilize the reserve of every non-native plant, which would be impossible and, frankly, a poor use of scarce labor. We concentrate on the invaders that actively block forest recovery or raise fire risk, and we accept that a working, three-hundred-year-old farm will always carry some introduced species, the bananas most obviously among them. The banana grove is managed, contained, and part of how the reserve pays for itself; you can read how that side of the property works on the banana farm page. The line we hold is between what we cultivate on purpose and what escapes to smother the wild.

Corridors: why one big block beats many small ones

Sitting behind every decision we make is a single idea from ecology that is worth understanding, because it explains why our particular patch of land is worth protecting at all. A fragment of forest cut off from other forest slowly loses species even if not one tree inside it is ever felled. The populations trapped in a small island of trees are too small to stay healthy: they cannot bring in new blood from outside, they drift toward inbreeding, and a single bad year, a fire, a disease, a drought, can wipe a species out locally with no way for it to return. Isolation kills forests quietly, from the inside, long after the chainsaws have stopped.

The remedy is connectivity: keeping forest joined to forest so that animals and plants can move, breed across a wide range, and recolonize after a setback. This is why the arithmetic of our land matters so much. Our 100 owned acres are not a lonely dot. They join the adjoining 400 acres of protected parkland to make 500 acres of continuous forest, and that block in turn connects to the far larger protected forests of the Serra da Bocaina that stretch across the border country of Rio de Janeiro and Sao Paulo states. A jaguarundi or an ocelot on our land is not confined to our land; it is part of a population moving across a much bigger territory. That is the whole point.

Our restoration work is, in large part, corridor work. When we heal an old clearing or plant across a gap, we are usually trying to reconnect two pieces of forest that a past clearing separated, so that animals can pass through cover rather than break into the open. Even the farm and the built areas are managed with connectivity in mind: we keep hedgerows, stream-side strips, and living fences so that a bird or a small mammal can cross the human parts of the reserve under leaves rather than under open sky. A corridor does not have to be pristine forest to work. It just has to let life move.

The old roads as ready-made corridors

One happy accident of the reserve's history is that the ancient donkey-and-mule roads, built over centuries to move produce down the mountain, now serve double duty. They give us and our guests a way to move through the property on paths that already exist, so we are not cutting new trails through the forest. And because they follow the gentlest contours and run stream to stream, they line up naturally with the routes wildlife wants to use anyway. We manage the vegetation along them lightly, keeping them walkable without stripping the cover that makes them useful to animals. Three centuries of muleteers, without meaning to, laid out a network of shaded corridors we now simply maintain.

Watching the forest: monitoring and camera traps

You cannot protect what you do not understand, and you cannot tell whether your work is succeeding without measuring it. Monitoring is the least visible part of conservation and one of the most important, because it is the difference between doing something and doing something that works. Our monitoring is modest and practical, matched to a small operation rather than a research institute, but it runs continuously and it shapes what we do next.

The simplest monitoring is the oldest: people who know the ground walking it regularly and paying attention. Staff and guides who are on the trails most days build up a detailed, running sense of the place, which trees are fruiting, where a stream has silted after a storm, where a new invasive has appeared, which corner of an old clearing is greening up and which is stalled. Much of what we know about the reserve comes not from instruments but from long familiarity, from people noticing change because they were there to see it. That kind of attention cannot be automated, and it is one of the quiet benefits of keeping the reserve inhabited rather than empty.

What camera traps are actually for

Camera traps, motion-triggered cameras left out in the forest to photograph whatever passes, are the tool people ask about most, and it is worth being clear-eyed about what they do and do not tell you. They are wonderful for confirming which shy, mostly nocturnal animals share the forest with us, the cats, the small carnivores, the ground birds and rodents that are almost never seen in daylight. They let us document wildlife using restored corridors, which is one of the most satisfying ways to check whether a corridor is doing its job. And over years they build a picture of which species are present and roughly how their activity shifts with the seasons.

What they are not is a census. A camera trap tells you that an animal passed a particular spot at a particular moment; it cannot, on its own, tell you how many of that animal live on the reserve, and we are careful never to dress up a handful of photographs as a population count. We would rather say plainly what we know, that a given species is present and using the forest, than invent numbers that sound impressive and mean nothing. The Serra da Bocaina forests are known to support an impressive cast of Atlantic Forest wildlife, and our continuous land is part of that larger system; the natural history of what lives here is covered on the Mata Atlantica biodiversity page. What monitoring adds is evidence, over time, that the system is holding and, in the restored areas, slowly filling back in.

Alongside the cameras we track simpler things that turn out to be powerful indicators. We note when key trees flower and fruit, since the timing of that, the phenology, is a sensitive early signal of a changing climate. We keep an eye on water: the flow of the springs and the clarity of the streams tell us, better than almost anything, whether the forest on the slopes above is doing its job of holding and filtering rain. Birds, being loud, visible, and fussy about habitat, are among the best barometers of forest health, and the dawn chorus, listened to year over year, says a great deal about whether the forest is thriving or thinning.

A camera trap tells you an animal passed; it does not tell you how many there are, and we refuse to pretend otherwise.

Keeping the forest standing: fire and poaching

Two threats can undo years of protection and restoration in a single season, and both come mostly from outside our boundary. Guarding against them is a permanent part of the model, not an occasional emergency.

Fire

Healthy, wet Atlantic Forest does not burn easily; it is one of the least fire-prone forests on Earth, which is exactly why fire is so damaging when it does get in. The plants and animals here did not evolve with regular fire and have few defenses against it. The danger is not the intact forest catching alight on its own but fire arriving from the edges, usually escaping from pasture burning on neighboring land in the dry months, or from a careless visitor, or from a burn set to clear ground getting away in a dry wind. Old, invasive-grass-choked clearings are the tinder that lets an edge fire climb into the trees.

Our defenses are unglamorous and effective. We keep firebreaks, cleared strips that deny a ground fire fuel, along the most exposed edges and around the built areas. We knock down the flammable exotic grasses in vulnerable clearings, which lowers the fuel load in the very places most likely to ignite. We raise our vigilance sharply through the dry season, when everyone on the property is watching for smoke and knows what to do at the first sign of it. We keep water and basic firefighting kit staged and ready. And we talk to our neighbors, because in this landscape fire safety is a shared, cross-boundary matter and good relations with the people around you are a genuine conservation asset. The best fire is the one that is caught in its first minutes, and a reserve full of people who know the ground catches fires early.

Poaching and the wildlife trade

Poaching in the Atlantic Forest takes forms that surprise people who picture only big-game hunting. Yes, there is hunting of game animals like pacas and agoutis for meat, and it matters, both because it removes wildlife and because, as we have seen, those same animals are the forest's seed-planters. But two other kinds of poaching do quiet, serious damage here. One is the trapping of wild songbirds for the cage-bird trade, an old and stubborn practice across Brazil that strips the forest of some of its most beautiful and ecologically useful birds. The other is the cutting of the juçara palm for its heart, palmito, which kills the tree outright to harvest a single delicacy and has pushed this once-common palm into steep decline across its range.

Our response is the same one that works against every kind of encroachment: presence. Land that people live on, walk daily, and clearly care for is far harder and riskier to poach than an empty title on a map. Trails that are used are trails that are watched. Boundaries that people know and check do not quietly leak. We do not romanticize this; determined poaching is a real problem across the whole biome and no single reserve solves it alone. But a small, continuously inhabited property with eyes on the ground, connected to a much larger protected landscape, is exactly the kind of place where wildlife has a fighting chance. Keeping people here, gently and in small numbers, is itself an anti-poaching strategy.

The economics: how 25 chalets pay for a forest

Now the part that most conservation writing skips, which is how any of this gets paid for. Protecting and restoring forest costs real money, year in and year out: labor for invasive control and firebreaks, materials, monitoring, maintenance of trails and water systems, and simply the opportunity cost of not doing the more lucrative things this land could be turned to. Someone has to fund the quiet work, and wishing does not do it. Our answer is a small, deliberate tourism operation whose entire purpose is to make the forest worth more standing than cut.

The logic is straightforward. Land like this, in the mountains above a beautiful and increasingly popular coast, is under constant economic pressure to become something that pays: a cattle pasture, a logging lot, or, most tempting of all, a subdivision of holiday homes. Each of those uses would generate income by degrading or destroying the forest. Our model is designed to generate enough income to hold all of them off, by giving the intact forest a livelihood of its own. The 25 chalets are not the point of the reserve; they are the mechanism that lets the reserve exist without being carved up.

Why the cap is the whole point

The number that makes this conservation rather than development is the cap: one chalet per two acres across the 50-acre ecomix park, for a maximum of 25 chalets, permanently. That ceiling is what keeps the tourism small enough to fund the forest without becoming a threat to it. A resort would keep adding rooms until the income was maximized and the forest was an amenity squeezed between car parks. We have deliberately fixed the ceiling low and put it beyond the reach of future temptation, so that the pressure to grow, which destroys almost every well-meaning eco-project eventually, has nowhere to go. The scarcity is not a sales tactic. It is the load limit that keeps the model honest.

The chalets are built and run to keep their share of the impact as low as their number: solar power rather than a diesel line cut through the trees, spring water rather than drilled wells, composting rather than septic runoff toward the streams. That off-grid-minded design is not only about a light footprint; it also lowers the running costs, which means more of what guests pay goes to the forest rather than to fuel and infrastructure. The full picture of how the chalets are built and operated is on the ecomix park page.

The farm as a second leg

Tourism is not the only income, and we are glad it is not, because a reserve that depends on a single revenue stream is fragile. The three-hundred-year-old organic banana farm is a genuine agricultural business in its own right, and a good one: it diversifies the reserve's income, it keeps working knowledge of the land alive, and it connects the place to the region's real agricultural economy rather than turning it into a tourism bubble. An old, well-managed banana grove is also gentle on the slopes, holding soil and blending into the forest edge. The farm and the chalets together are more resilient than either alone, and they reinforce each other: the farm gives guests something real to eat and to see, and the guests give the farm a premium local market.

No invented numbers

You will notice we have not quoted you a budget, a percentage, or a return, and that is deliberate. Plenty of eco-tourism marketing throws around precise-sounding figures, so many trees planted, so much of every booking to conservation, that do not survive scrutiny. We would rather explain the structure honestly and let it stand on its logic than dress it in numbers we cannot defend. The structure is this: a hard-capped, low-impact tourism operation and a working organic farm together generate the income that pays for protecting and restoring 500 acres of continuous Atlantic Forest, and the cap ensures the tourism can never outgrow the forest it exists to serve. That is the model. If you want to see it in person, the stays and booking page is where a visit begins.

The honest trade-offs of bringing people here

It would be dishonest to describe a tourism-funded reserve without admitting that tourism itself carries a cost. Every guest who comes here has some impact. People need water, generate waste, and disturb wildlife simply by being present; trails see wear; the quiet the animals depend on is, at the margins, reduced by human activity. And there is the plain fact that most guests arrive by car and many by plane, which carries a carbon cost that no amount of on-site care fully erases. Pretending a visit is impact-free would be exactly the kind of greenwashing we are trying not to do.

So the honest framing is not that our tourism has no downside, but that it is the least-bad way we have found to keep this forest standing, and that we work continuously to keep its downsides small. The hard cap on chalets limits the total human load the land ever has to carry. The low-impact design limits each guest's footprint. Concentrating the built area and the traffic on a small, already-disturbed portion of the property keeps the wild majority undisturbed. And routing people onto the centuries-old road network, rather than cutting new trails, keeps human movement to paths the forest has already absorbed.

There is a subtler trade-off too, one worth naming: the risk that a place created to protect a forest slowly becomes a place that exists to serve visitors, with the forest reduced to scenery. We guard against that drift mainly through the cap and through keeping conservation, not occupancy, as the measure of success. The forest is the client. The chalets serve it, not the other way around, and the day that inverts is the day the model has failed. Holding that line is a choice we have to keep making, not a box we ticked once.

We also think the presence of guests, handled well, is a net positive beyond the money, and it would be false modesty to pretend otherwise. Visitors who walk this forest, hear it at dawn, drink water straight from a mountain spring, and see for themselves what an intact piece of the Mata Atlantica actually is, tend to leave caring about it, and about the wider fight to save what remains of this biome. Conservation depends, above all, on enough people believing a standing forest is worth more than a cleared one. A small number of people experiencing that directly, and carrying it home, is part of how the idea spreads. That is not a number we can put on a spreadsheet, but it is real.

What success looks like, and how long it takes

Restoration runs on a clock most human enterprises are not built for. A pasture can turn into a convincing young forest in a decade, but a young forest is not the same as the old-growth it is slowly working toward. The tall, slow, long-lived trees that define mature Atlantic Forest, and the animals that depend on the specific conditions those trees create, arrive over the course of many decades, not years. Some of what we plant and protect now will only become what it is meant to be long after everyone involved in planting it is gone. That is simply the timescale of forests, and accepting it is part of the work.

So we measure success in signs and directions rather than finish lines. Success is an old clearing that has closed over and no longer needs weeding, because the canopy now does the job. It is a stream that runs clear after a storm instead of brown, because the slope above it is holding. It is a corridor that camera traps show animals are using to cross from one block of forest to another. It is a dry season that passes without fire reaching the trees. It is a bird heard at dawn that was not heard here a few years ago. None of these make a headline. Together they are the whole game.

The most honest thing we can say about the model is that it is unfinished and always will be. Invasives will keep arriving. Fire risk will return every dry season. The economic pressure to do something more profitable with this land will never fully go away. The work of holding the line, and slowly gaining ground, does not end; it just gets handed on. What the 25 chalets and the old farm buy is the ability to keep doing it, indefinitely, without cutting the forest to survive. If that sounds modest, it is. Modest, durable, and honest is exactly what we are aiming for. If you would like to talk it through or come and see it, the contact page is the place to start.