Akira Miyawaki, a Japanese botanist who died in 2021, spent decades studying which trees actually belonged on a given patch of land. He built his method by examining relict native vegetation preserved in the small forest groves surrounding Shinto shrines, known as chinju no mori, which had survived largely undisturbed for centuries and offered a rough template for what he called a site’s potential natural vegetation: the mature forest that would eventually establish itself there without human interference. From that research came a planting technique now applied at well over a thousand sites, first for Nippon Steel around its Ōita plant in the 1970s, and since then in urban lots, roadside strips, and degraded farmland across Japan and beyond.
The technique itself is straightforward to describe. Soil is prepared and enriched, then a dense, randomly arranged mix of native pioneer and secondary species is planted, typically three to five seedlings per square metre in temperate climates and considerably more in tropical settings. Packed that closely, and without the neat rows of conventional plantation forestry, the seedlings compete for light almost immediately, which pushes vertical growth and rapid canopy closure. Miyawaki received the Blue Planet Prize in 2006 for the work, and the method now circulates globally under his name, promoted heavily by organisations including Afforestt, SUGi, and various municipal and NGO “tiny forest” programmes.
The claim doing most of the marketing work
The figure attached to almost every popular description of the method is that Miyawaki forests grow roughly ten times faster than forests left to regenerate on their own, sometimes paired with a claim that they become thirty times denser. These numbers appear in project literature, news coverage, and fundraising material for tiny-forest initiatives from Massachusetts to Mysuru. What is less often reported is how thin the evidence behind them actually is once someone goes looking for it systematically.
A review published in the Journal of Applied Ecology, “Tiny forests, huge claims: the evidence gap behind the Miyawaki method for forest restoration,” examined 51 pieces of literature making claims about the method’s performance.
Only 41 per cent of that literature offered any quantitative data at all. Just 33 per cent used a control comparison, and only 14 per cent involved replication, which are basic requirements for drawing a reliable conclusion in restoration ecology, not unusual academic fussiness. The review’s authors traced most of the ten-times and thirty-times figures back to grey literature and promotional material from advocacy and implementation organisations, rather than to peer-reviewed studies designed to test the claim directly. Non-indexed sources, the review found, carried on average fifty per cent more unverified claims than peer-reviewed ones.
What the review actually found, claim by claim
The result is not a flat debunking. The review rated the ten-times faster growth claim as partially supported, which makes it the best-evidenced of the major claims attached to the method, not the weakest. The companion claim that Miyawaki forests reach full ecological maturity within twenty to thirty years, however, was rated as having null empirical evidence behind it: essentially no data either supporting or contradicting it. Claims about enhanced biodiversity, carbon sequestration, long-term self-sustainability without further management, and cost efficiency compared with other restoration approaches were all rated weak to null. On carbon specifically, only eight of the fifty-one papers attempted any claim about carbon storage at all, and just two measured it directly. Those two found no statistically significant advantage over conventional restoration methods once enough time had passed.
Karen Holl, a restoration ecologist at the University of California, Santa Cruz, who was not involved in the review, has separately described the method as impractical and cost-prohibitive at scale, citing the same shortage of rigorous, comparative long-term data.
This is not a new concern. A 2019 piece in JSTOR Daily raised similar questions about the evidentiary base years before this systematic review formalised them, and a 2011 field trial in the Mediterranean by Schirone and colleagues found sapling mortality of 61 to 84 per cent after twelve years at Miyawaki-style planting sites, a result that sits awkwardly next to the method’s reputation for rapid, reliable establishment.
Why the growth-rate claim is the one part that holds up better
It is worth separating out why the growth-rate figure fares better than the rest. Dense planting forcing seedlings to compete for light and grow vertically faster is mechanistically plausible and relatively easy to measure with a tape and a few years of monitoring, which is likely why more of the available literature manages to say something about it with at least partial support. Claims about biodiversity, soil carbon, and long-term self-sustainability require years of ecological monitoring across multiple dimensions, funding that most small-scale urban tiny-forest projects, run by municipalities or community groups rather than research institutions, simply do not have. The result is a lopsided evidence base: strong intuitive appeal and modest measurable data on growth speed, and comparatively little on almost everything the marketing claims after that.
The current wave of projects
None of this has slowed adoption. Massachusetts has seen a run of tiny-forest projects reported by regional outlets through 2026, including a Bridgeport, Connecticut expansion supported by a $400,000 state climate grant. In India, a municipal project in Mysuru has planted around 15,000 saplings using the method, and a “Million Miyawaki” campaign is under way across the Delhi-NCR region. The appeal is easy to understand: a small, visually dense patch of forest that appears within a few years on an otherwise unused urban lot is a compelling, photographable result, regardless of what the underlying ecological data can or cannot yet confirm.
What this means for a project considering the method
None of this suggests dense native planting is a bad idea, or that Miyawaki’s underlying insight about potential natural vegetation is wrong. The review’s authors are not arguing the method fails. They are arguing that the evidence available does not yet support the scale of the claims routinely made about it, and that a technique can be worth trying in a specific location while still being oversold in its marketing. For a city weighing a tiny-forest programme, the honest summary is closer to: seedlings planted densely probably will grow faster in the early years, the mature outcome twenty or thirty years out is largely unmeasured, and figures like “ten times faster” or “thirty times denser” belong to specific promotional sources rather than a settled scientific consensus.
