Walk on the Wild SideEstimating the Global Magnitude of Visits to Protected Areas

Andrew Balmford, Jonathan Green, Michael A. Anderson, James Beresford, Charles Huang, Robin Naidoo, Matt Walpole, Andrea ManicaView original
OverviewBalancedalloy voice
Here’s a puzzle at the heart of conservation. Protected areas cover roughly one-eighth of Earth’s land and anchor everything from weekend hikes to once in a lifetime safaris. They’re where people go to feel small, grounded, and alive. Yet until very recently, we couldn’t answer a basic question: how many times do people actually visit them? That count isn’t trivia. It connects to human well-being, the tourism economies many regions depend on, and the political leverage conservation needs. Governments have even pledged, through targets like the Aichi Biodiversity Targets, to fold nature into national accounts. But if you don’t know how many visits parks get, it’s hard to argue for investing in them. Andrew Balmford, Matt Green, James Anderson, and colleagues decided to count globally and to do it in a way that could scale. They assembled visit records for a large, diverse set of terrestrial protected areas — five hundred fifty-six parks spread across fifty-one countries. Within that sample, the range was staggering: some parks saw no visitors in a year, whereas others topped ten million. The middle of that distribution, the median, sat at about twenty thousand three hundred thirty-three visits per park per year. That’s a lot of picnics and trail miles. It’s also just the starting point. Because the world isn’t homogeneous, they didn’t fit a single global curve. They split the planet into major regions and, within each, asked a simple question: given what we know about a park and its context, how many visits should we expect? Five ingredients went into the recipe. Park size, the number of people living nearby, how remote the park is, a bare-bones score for natural attractiveness, and national income. They explored each factor on its own, then built region-specific generalized linear models — the workhorses of applied statistics — to predict visit rates. To keep the math well-behaved, they log-transformed all five predictors. For visit rate, local population, and remoteness, they added one before taking logs so zeros didn’t break the calculation. Across regions, the resulting models together explained just over half — fifty-two point nine percent — of the global variation in visitation. Not perfect, but remarkably informative for such a messy human behavior. Then came the scale-up. The team took this regional modeling and applied it to the universe of terrestrial sites listed in the World Database on Protected Areas. They were careful about scope. Marine and Antarctic sites were out. So were the most strictly protected International Union for Conservation of Nature, or IUCN, Category I areas where tourism is typically discouraged. And because their sample didn’t include tiny parks under ten hectares — about the size of a few city blocks — they dropped those very small sites from the extrapolation, roughly forty thousand of them. That left ninety-four thousand two hundred thirty-eight terrestrial protected areas to estimate. A big number, and a big responsibility. They stress that data in the global database contain errors and that models have limited power, so you shouldn’t use these predictions to micromanage any one small country or site. But at continental scales, the patterns are robust and revealing. What did those patterns look like? First, the drivers weren’t uniform. In Europe, holding other factors constant, visit rates rose with the number of people living nearby — intuitive when you consider weekend nature trips out of dense cities. Remoteness depressed visits almost everywhere, except in Asia and Australasia where the relationship was more nuanced. Natural attractiveness, that simple proxy for scenic pull, mattered most clearly in North America and Latin America. And national income showed up as a consistent positive in most regions — richer countries tend to have more people with the means and leisure to visit parks. Layer all that together, and a clear rank emerges: the highest visit rates were estimated for North America, lower in Asia and Australasia, in Europe, and lowest in Africa and Latin America. Zooming in from regions to typical parks adds texture. Aggregate medians by country showed that parks in Africa tend to see around three thousand visits a year. In Latin America, it’s about four thousand. Then North America jumps off the page with a median near three hundred fifty thousand per park per year. That’s not a small gap; it’s a different order of magnitude. It reflects accessibility, wealth, infrastructure, and decades of a culture where going to a park is almost a weekly ritual. Now, stack the parks up and count visits by region. Europe’s protected areas pull in on the order of three point eight billion visits a year. North America’s total is about three point three billion. Together, that’s more than eighty percent of all terrestrial protected area recreation on Earth. Those two regions dominate the global ledger. Elsewhere, the totals are far smaller — not because people don’t love nature, but because parks are often farther from population centers, incomes are lower, and infrastructure is thinner. A country-level snapshot makes the scale concrete. The United States alone clears three billion protected area visits in a typical year. Let that land for a second: three billion. That’s roughly ten visits for every person in the country, per year, if you smeared them out evenly. Of course, they’re not even — it’s a mix of locals and visitors — but the magnitude tells you how woven parks are into daily life. Roll the whole world together and the estimate comes out at roughly eight billion visits per year to terrestrial protected areas. There’s uncertainty around that — the ninety-five percent confidence interval runs from five point four to eighteen point five billion — but even the conservative end is huge. And remember, this is model-based extrapolation from observed data, not a global turnstile count. If you normalize for the time people actually have to recreate, another contrast sharpens. When Balmford and colleagues scaled visits by population and non-workdays — the days people are free to go outside — Europe came out at roughly five protected area visits per one hundred non-work person-days. North America was about ten per one hundred. Other regions barely reached three-tenths of a visit per one hundred non-workdays. That’s a tidy way to see both preference and opportunity etched into the numbers. The economic translation is where the policy case snaps into focus. The team linked those regional visit totals to valuation studies to estimate two things: direct in-country expenditure — what visitors spend on fees, travel, accommodation, and food — and consumer surplus, which is economist-speak for the extra value people would have been willing to pay above what they actually did. Using medians from the literature, region by region, and multiplying by the modeled visits, they come to about six hundred billion US dollars per year in direct, within-country spending and another two hundred fifty billion US dollars per year in consumer surplus, both at twenty fourteen prices. For context, global travel and tourism spending of all kinds is about two thousand billion US dollars a year. And the money we actually spend to safeguard protected areas? Under ten billion US dollars annually. Put differently, we sit on a recreation economy measured in the hundreds of billions while investing pennies on the dollar to maintain the asset that generates it. Any honest accounting has to grapple with uncertainty, and the authors are explicit about it. Visit totals derived from regional models applied to a global database can be off by a factor of two or more. The World Database on Protected Areas has errors and gaps. And the economic values they draw from the literature vary by method and context. These estimates won’t tell you how many people went to a specific small park in a given year or what each family spent in the gift shop. They will tell you, with confidence, the shape of the mountain: vast visitation concentrated in wealthy regions, enormous aggregate value, and a yawning gap between what society gets from protected areas and what it pays to keep them protected. Two other caveats are important. First, these numbers cover terrestrial protected areas. Marine parks aren’t included here, so the global recreational footprint is larger still. Second, the models isolate recreation. They don’t count the many other services parks provide — clean water, pollination, carbon storage, cultural heritage — which push the true value of protected landscapes even higher. Recreation is just the most visible tip, now measured with global scope. And that brings us back to why counting matters. You can’t put a forest’s awe on a balance sheet, but visits are tangible in a way that sways budgets and ballots. They show up as jobs and taxes and trips you remember for decades. As Balmford and colleagues make clear, protected areas are not passive cost centers; they are active engines of well-being and economic activity. If governments are serious about integrating biodiversity into national accounts — that Aichi promise — then eight billion park visits and hundreds of billions in spending are the sort of line items that change minds. Where does this leave us? With an empirical baseline. We know, roughly, how much the world uses its parks for recreation, where those visits cluster, and the orders of magnitude of value tied to them. Perfect precision can come later, with better site-level data and improved global inventories. The signal is already loud enough to act on: invest in maintaining trails, staffing rangers, building access where it’s appropriate, and protecting the quiet that people travel to find. If there’s a forward look to indulge, it’s a humble one. Imagine pairing this kind of regional modeling with more timely, standardized visitation feeds — the passenger counts at park gates, anonymized mobility data where privacy allows, better mapping of trail networks — not to surveil, but to manage. To smooth crowding, to open opportunity where access is low, and to keep the promise of protected areas alive: places where nature thrives and people can go, again and again, to remember why that matters.

Here’s a puzzle at the heart of conservation. Protected areas cover roughly one-eighth of Earth’s land and anchor everything from weekend hikes to once in a lifetime safaris. They’re where people go to feel small, grounded, and alive.

Yet until very recently, we couldn’t answer a basic question: how many times do people actually visit them? That count isn’t trivia. It connects to human well-being, the tourism economies many regions depend on, and the political leverage conservation needs.

Governments have even pledged, through targets like the Aichi Biodiversity Targets, to fold nature into national accounts. But if you don’t know how many visits parks get, it’s hard to argue for investing in them.

Andrew Balmford, Matt Green, James Anderson, and colleagues decided to count globally and to do it in a way that could scale. They assembled visit records for a large, diverse set of terrestrial protected areas — five hundred fifty-six parks spread across fifty-one countries. Within that sample, the range was staggering: some parks saw no visitors in a year, whereas others topped ten million.

The middle of that distribution, the median, sat at about twenty thousand three hundred thirty-three visits per park per year. That’s a lot of picnics and trail miles. It’s also just the starting point.

Because the world isn’t homogeneous, they didn’t fit a single global curve. They split the planet into major regions and, within each, asked a simple question: given what we know about a park and its context, how many visits should we expect? Five ingredients went into the recipe.

Park size, the number of people living nearby, how remote the park is, a bare-bones score for natural attractiveness, and national income. They explored each factor on its own, then built region-specific generalized linear models — the workhorses of applied statistics — to predict visit rates. To keep the math well-behaved, they log-transformed all five predictors.

For visit rate, local population, and remoteness, they added one before taking logs so zeros didn’t break the calculation. Across regions, the resulting models together explained just over half — fifty-two point nine percent — of the global variation in visitation. Not perfect, but remarkably informative for such a messy human behavior.

Then came the scale-up. The team took this regional modeling and applied it to the universe of terrestrial sites listed in the World Database on Protected Areas. They were careful about scope.

Marine and Antarctic sites were out. So were the most strictly protected International Union for Conservation of Nature, or IUCN, Category I areas where tourism is typically discouraged. And because their sample didn’t include tiny parks under ten hectares — about the size of a few city blocks — they dropped those very small sites from the extrapolation, roughly forty thousand of them.

That left ninety-four thousand two hundred thirty-eight terrestrial protected areas to estimate. A big number, and a big responsibility. They stress that data in the global database contain errors and that models have limited power, so you shouldn’t use these predictions to micromanage any one small country or site. But at continental scales, the patterns are robust and revealing.

What did those patterns look like? First, the drivers weren’t uniform. In Europe, holding other factors constant, visit rates rose with the number of people living nearby — intuitive when you consider weekend nature trips out of dense cities.

Remoteness depressed visits almost everywhere, except in Asia and Australasia where the relationship was more nuanced. Natural attractiveness, that simple proxy for scenic pull, mattered most clearly in North America and Latin America. And national income showed up as a consistent positive in most regions — richer countries tend to have more people with the means and leisure to visit parks.

Layer all that together, and a clear rank emerges: the highest visit rates were estimated for North America, lower in Asia and Australasia, in Europe, and lowest in Africa and Latin America.

Zooming in from regions to typical parks adds texture. Aggregate medians by country showed that parks in Africa tend to see around three thousand visits a year. In Latin America, it’s about four thousand.

Then North America jumps off the page with a median near three hundred fifty thousand per park per year. That’s not a small gap; it’s a different order of magnitude. It reflects accessibility, wealth, infrastructure, and decades of a culture where going to a park is almost a weekly ritual.

Now, stack the parks up and count visits by region. Europe’s protected areas pull in on the order of three point eight billion visits a year. North America’s total is about three point three billion.

Together, that’s more than eighty percent of all terrestrial protected area recreation on Earth. Those two regions dominate the global ledger. Elsewhere, the totals are far smaller — not because people don’t love nature, but because parks are often farther from population centers, incomes are lower, and infrastructure is thinner.

A country-level snapshot makes the scale concrete. The United States alone clears three billion protected area visits in a typical year. Let that land for a second: three billion.

That’s roughly ten visits for every person in the country, per year, if you smeared them out evenly. Of course, they’re not even — it’s a mix of locals and visitors — but the magnitude tells you how woven parks are into daily life. Roll the whole world together and the estimate comes out at roughly eight billion visits per year to terrestrial protected areas.

There’s uncertainty around that — the ninety-five percent confidence interval runs from five point four to eighteen point five billion — but even the conservative end is huge. And remember, this is model-based extrapolation from observed data, not a global turnstile count.

If you normalize for the time people actually have to recreate, another contrast sharpens. When Balmford and colleagues scaled visits by population and non-workdays — the days people are free to go outside — Europe came out at roughly five protected area visits per one hundred non-work person-days. North America was about ten per one hundred.

Other regions barely reached three-tenths of a visit per one hundred non-workdays. That’s a tidy way to see both preference and opportunity etched into the numbers.

The economic translation is where the policy case snaps into focus. The team linked those regional visit totals to valuation studies to estimate two things: direct in-country expenditure — what visitors spend on fees, travel, accommodation, and food — and consumer surplus, which is economist-speak for the extra value people would have been willing to pay above what they actually did. Using medians from the literature, region by region, and multiplying by the modeled visits, they come to about six hundred billion US dollars per year in direct, within-country spending and another two hundred fifty billion US dollars per year in consumer surplus, both at twenty fourteen prices.

For context, global travel and tourism spending of all kinds is about two thousand billion US dollars a year. And the money we actually spend to safeguard protected areas? Under ten billion US dollars annually.

Put differently, we sit on a recreation economy measured in the hundreds of billions while investing pennies on the dollar to maintain the asset that generates it.

Any honest accounting has to grapple with uncertainty, and the authors are explicit about it. Visit totals derived from regional models applied to a global database can be off by a factor of two or more. The World Database on Protected Areas has errors and gaps.

And the economic values they draw from the literature vary by method and context. These estimates won’t tell you how many people went to a specific small park in a given year or what each family spent in the gift shop. They will tell you, with confidence, the shape of the mountain: vast visitation concentrated in wealthy regions, enormous aggregate value, and a yawning gap between what society gets from protected areas and what it pays to keep them protected.

Two other caveats are important. First, these numbers cover terrestrial protected areas. Marine parks aren’t included here, so the global recreational footprint is larger still.

Second, the models isolate recreation. They don’t count the many other services parks provide — clean water, pollination, carbon storage, cultural heritage — which push the true value of protected landscapes even higher. Recreation is just the most visible tip, now measured with global scope.

And that brings us back to why counting matters. You can’t put a forest’s awe on a balance sheet, but visits are tangible in a way that sways budgets and ballots. They show up as jobs and taxes and trips you remember for decades.

As Balmford and colleagues make clear, protected areas are not passive cost centers; they are active engines of well-being and economic activity. If governments are serious about integrating biodiversity into national accounts — that Aichi promise — then eight billion park visits and hundreds of billions in spending are the sort of line items that change minds.

Where does this leave us? With an empirical baseline. We know, roughly, how much the world uses its parks for recreation, where those visits cluster, and the orders of magnitude of value tied to them.

Perfect precision can come later, with better site-level data and improved global inventories. The signal is already loud enough to act on: invest in maintaining trails, staffing rangers, building access where it’s appropriate, and protecting the quiet that people travel to find.

If there’s a forward look to indulge, it’s a humble one. Imagine pairing this kind of regional modeling with more timely, standardized visitation feeds — the passenger counts at park gates, anonymized mobility data where privacy allows, better mapping of trail networks — not to surveil, but to manage. To smooth crowding, to open opportunity where access is low, and to keep the promise of protected areas alive: places where nature thrives and people can go, again and again, to remember why that matters.

More in Economics, Econometrics and Finance