Urban parks have long been celebrated as the green lungs of cities, promoted by planners and public health officials alike as instruments of preventive medicine. Yet for all the enthusiasm, city governments have rarely been able to answer a deceptively simple question with hard numbers: does a specific park, in a specific location, actually deliver enough health value to justify what it costs to build and maintain? A new study published in Nature Communications tackles that question head-on, using anonymized mobile-phone location data to measure how hundreds of thousands of people really use parks, and then converting that measured use into monetary estimates of preventive health benefit that can be weighed directly against land, construction and maintenance costs.
The research team, led by Hang Su and Wenjing Li, with co-authors Sheng Li, Shuisong Ke and Haoran Zhang, focused on Greater Tokyo, one of the largest and most densely populated metropolitan areas in the world. Tokyo offers an ideal natural laboratory for this kind of analysis because it contains an extraordinary range of park types packed into a dense urban fabric, from tiny neighborhood block parks squeezed between buildings to sprawling destination parks that draw visitors from across the region. The researchers assembled records for 4,139 parks across the metropolitan area, together with detailed information on land prices, population distribution and the costs of acquiring, building and maintaining green space, all expressed in 2019 Japanese yen.
The heart of the study is a dataset of remarkable scale: 1.79 million park visits made by roughly 320,000 users, reconstructed from anonymized mobile-phone location records. Rather than relying on surveys, self-reported exercise diaries or sparse manual counts, the team could observe where people actually traveled, how far they came from, and which parks they chose. This population-scale mobility approach sidesteps one of the most stubborn problems in urban health research, namely that people systematically overstate how much time they spend being active outdoors. By grounding the analysis in observed behavior, the researchers could estimate the walking-related physical activity that each park genuinely generates.
The technical logic of the framework proceeds in several steps. First, each park visit is linked back to the residential location of the visitor, allowing the team to map the catchment of every park in the dataset. Second, the walking activity associated with those visits is translated into preventive health benefits, using established relationships between physical activity and reduced disease risk. Third, those health benefits are monetized, so that the value of avoided illness can be expressed in the same currency as the costs of the park itself. Finally, the benefits are compared with land acquisition, construction and maintenance costs to compute a kind of return on investment for each individual park, something that has never before been possible at this level of granularity.
One of the clearest findings concerns distance. The preventive health benefits that a park generates decline steadily as residential distance increases, with the effect fading out within roughly seven kilometers of the park. In other words, parks are fundamentally local health infrastructure: the people who benefit most from a given patch of green are those who live close enough to walk to it regularly. Beyond that radius, a park may still serve recreational, cultural or ecological functions, but its contribution to walking-related preventive health diminishes to little. This distance-decay relationship gives planners a quantitative rule of thumb for thinking about where new green space will translate into measurable health activity.
Park type, by contrast, turned out to matter in a different way. Rather than changing how benefits decay with distance, the type of park mainly shifts the overall level of benefit it produces. Small block parks and large destination parks occupy different niches in the urban system, and the study shows that these niches carry very different economics. The returns on park investment proved to be highly uneven across the 4,139 sites, and the pattern of that unevenness is the study’s most striking result: the health returns are heavily concentrated among small parks.
Specifically, the researchers found that about 16 percent of block parks recover their costs within 20 years when only walking-related health benefits are counted. That is a remarkable figure, because it means a meaningful fraction of the smallest parks in Tokyo pay for themselves purely through the preventive health value of the walking they induce, before counting any of the other benefits that green space provides, such as cooling, stormwater management, biodiversity or social cohesion. Large destination parks, on the other hand, rarely recover their costs within 20 years on walking-related benefits alone. Their land footprints are enormous, and in a city as expensive as Tokyo, the opportunity cost of that land is staggering. This does not mean destination parks are worthless; it means their justification must rest on benefits beyond walking-related health, which the framework deliberately excludes to keep the accounting transparent.
The implications for planning practice are direct. Because the economics of small and large parks differ so sharply, the study’s planning estimates suggest different siting strategies for the two smallest park classes. Where a city can choose between scattering many small green spaces through dense residential neighborhoods or concentrating investment in one large regional attraction, the health-return calculus points toward the small, distributed option if preventive health is the goal. A block park embedded in a walkable neighborhood generates repeated, routine visits by nearby residents, and it is precisely this repetition that accumulates into disease-preventing physical activity. A destination park may generate spectacular single visits, but those visits are infrequent and draw from a population spread over many kilometers, diluting the per-resident health return.
What makes the framework genuinely useful, the authors argue, is its transparency. Every step of the calculation, from mobility data to health benefit to monetary value to cost comparison, is explicit and auditable. Cities facing tight budgets and soaring land prices can use the approach to screen proposed park investments, comparing projected walking-related health returns across candidate sites before committing scarce funds. The method does not pretend to capture the full value of green space, and the authors are careful to frame it as a screening tool rather than a comprehensive valuation. But by isolating one benefit stream and measuring it rigorously, the framework gives park advocates a number they can defend in a budget negotiation, and gives skeptics a clear account of what is and is not being counted.
The study also signals a broader shift in how cities can understand their own infrastructure. Mobile-phone data, collected ethically and anonymized at population scale, are turning everyday movement into a measurable signal of how the built environment shapes health. The same logic that reveals which parks pay off could, in principle, be extended to other preventive health assets, from walkable streets to sports facilities. For now, the Tokyo results stand as a proof of concept: with 1.79 million observed visits and 4,139 parks, the researchers have shown that the health value of urban green space is not an abstract aspiration but a quantifiable, park-by-park reality, and that in the dense fabric of a megacity, the humblest neighborhood parks may be among the best health investments a city can make.
Subject of Research: Using population-scale mobile-phone mobility data to quantify the preventive health returns of urban parks relative to their costs in Greater Tokyo
Article Title: Quantifying parks as preventive health assets using population-scale mobility data
Article References: Su, H., Li, W., Li, S., Ke, S., & Zhang, H. (2026). Quantifying parks as preventive health assets using population-scale mobility data. Nature Communications. https://doi.org/10.1038/s41467-026-78106-7
Image Credits: AI Generated
DOI: 10.1038/s41467-026-78106-7
Keywords: urban parks, preventive health, mobile phone data, Greater Tokyo, physical activity, health economics, urban planning, green space, cost-benefit analysis, mobility data, Nature Communications, public health
News Source: Denise Maddox. (October 10, 2026). Mobile Phone Data Reveal Which City Parks Actually Pay Off as Health Assets. Scienmag.
