The fragility of urban farmland — why beloved gardens end up as bulldozed lots

Los Angeles, 2006. The day of forced eviction came for South Central Farm — one of the largest urban farms in the US, where about 350 families grew vegetables on 14 acres. A famous actor occupied a tree in protest and it became global news, but after the city sold the land, the farm was razed. Soil and crops built over more than a decade, and a community of hundreds, were lost in a day.

The same thing, in different forms, happens worldwide. Manor Garden Allotments, evicted to build a London Olympics venue. Adam Purple's garden and the Esperanza Garden in New York, cleared for redevelopment. The Newham, Gorgie and Coventry city farms in Britain, closed by municipal budget cuts. These are not exceptional misfortunes but expressions of a structural fragility that urban farmland carries.

Starting an urban farm needs only enthusiasm and a little resource. Keeping it going for five, ten, thirty years is a wholly different difficulty. Many urban gardens exist on land temporarily borrowed from someone else, and the moment land values rise, a development plan moves, or budgets are cut, they are the first to be discarded as low-priority 'interim use.' This body of failure cases teaches a hard truth: unless you design how to secure the land from the very start, no garden — however beloved — can be protected.

The first type is forced eviction by development or public projects. South Central Farm was pushed out entirely when the city sold land it had once expropriated back to the former owner. In London, the roughly century-old Manor Garden Allotments were evicted to build the media centre for the 2012 Olympics. A relocation site was provided, but the soil and community the gardeners had built could not be restored.

New York is a trove of this type. Through the redevelopment waves of the 1980s and 90s, celebrated resident-led gardens — Adam Purple's Garden of Eden, the Esperanza Garden, the Chico Mendez Mural Garden — were cleared one after another. Britain's Old Tidemill Wildlife Garden and Grow Heathrow were likewise removed amid redevelopment and infrastructure plans. What they share is that the garden was treated as something 'temporarily there,' erased in front of a more profitable use.

The second type is a 'slow erosion' — unremarkable case by case, but dramatic in the statistics. Dobson and Edmondson's GIS analysis of historic maps across ten British cities (2020) revealed that allotment land had fallen about 65% from its mid-twentieth-century peak. And the most deprived areas were losing gardens eight times faster than the wealthiest — a quiet widening of inequality in access to places to grow food.

The same phenomenon is confirmed elsewhere. Spilková and Vágner's study of allotment loss in Prague (2016) documented how, under the conflicting pressures of urban planning and private development, planning policy failed to protect food-growing land and prioritised redevelopment over established urban agriculture. Santo and colleagues' review of land-access policy across 40 US cities (2024) likewise notes that most support skews to 'temporary use' rather than secure long-term tenure, rarely delivering the durable land access that farmers need.

The third type is a failure not of the garden itself but of the programs and budgets that sustained it. In Britain, city farms dependent on annual council grants closed one after another under austerity budgets: Newham City Farm, among London's oldest (closed 2022); Edinburgh's Gorgie City Farm, which drew ~200,000 visitors a year (liquidated 2019); Coventry City Farm, after 25 years (closed 2008). In Edmonton, Canada, a city-run garden program that donated its harvest to the food bank was cut in 2025 as a 'want rather than a need' amid budget reductions to limit a tax increase.

On the policy side, the failures continue. Celebrated food policy councils have dissolved — Portland/Multnomah (2012), and councils in Orange County, Denver and Washington DC lost their capacity to budget cuts. In the US, a USDA grant program supporting urban farmers' access to land and capital was terminated mid-project, and the Regional Food Business Centers were eliminated barely after launch. Detroit's municipal garden program 'Farm-A-Lot,' begun in 1975, dissolved when city finances cut off its public funding. Institutions are far harder to sustain than to create.

What the three types share is that the overwhelming majority of urban gardens 'use someone else's land under weak rights.' Free loans by an owner's goodwill, short leases, revocable licences to occupy — gardens rest on such insecure tenure. A study of Paris's urban gardens depicted a 'nomadic' state in which gardens placed on future building sites are forced to relocate each time a permit is revoked, exposing the structural instability of interim-use policy. Where land rights are weak, a garden — however beloved, however productive — is powerless before development.

This lens reveals the single greatest fork between success and failure. Efforts to raise a garden's value — building good soil, forming community, drawing visitors — ironically raise the land's value too, and can even strengthen the incentive to evict. Precisely for that reason, if you want a garden to last, you must work on 'how to firm up the land rights' before enriching the activities, or at least at the same time.

Behind the failures, mechanisms to protect farmland have also grown. In New York, when the city tried to auction 114 community-garden lots in 1999, residents' lawsuits, nonprofit buyouts and the state Attorney General's suit halted the auction, and a 2002 settlement permanently preserved many gardens. From that experience, community land trusts (CLTs) and land trusts that hold land off the market spread as a principal means of protecting gardens. Making land 'a shared commons' rather than 'someone's asset' is the surest bulwark against eviction.

Japan has its own kindred institutions. 'Productive green land' (seisan ryokuchi), introduced in the 1992 revision, uses urban planning to protect farmland inside urbanisation-promotion areas, securing continued farming in exchange for tax advantages. The Citizen Farm Development Promotion Act, the Specified Farmland Lending Act, and the 2018 Urban Farmland Leasing Facilitation Act likewise provide institutional footholds for keeping farmland in existence. That Japan passed the so-called '2022 problem' of expiring designations without major disruption owed much to these frameworks working.

The implications for practice are clear. First, firm up land rights as much as possible before starting — long contracts, permanent use of public land, nonprofit ownership, productive-green-land designation. Second, secure agreements with authorities and landowners in documents and institutions so they survive changes of staff or mayor. Third, if you depend on public support, diversify funding sources on the assumption that the budget could vanish in a single fiscal year. Love and productivity do not protect a garden. What protects it is land rights and the institutions behind them.