But how much difference can urban agriculture make? And if every city followed York's example, how many people could be fed? A new study in Environmental Research Letters (ERL) provides some answers.

Locating every unused urban nook and cranny, and calculating how much food it could produce, is a herculean task. So instead, Federico Martellozzo from the University of Rome, Italy, and colleagues turned the problem on its head and asked what proportion of our cities would have to be set aside to provide urban-dwellers with all the vegetables they need.

"Given that cereals alone occupy much more land than all urban areas combined, it is clear that urban agriculture cannot literally feed the world," said Martellozzo, who was at McGill University in Canada when he carried out this research. "However, urban agriculture could help hundreds of thousands of people increase their intake of vegetables."

Martellozzo and his colleagues used an existing dataset based on MODIS satellite measurements from 2001 to estimate the maximum total cultivatable urban area available in each country across the world. This global dataset excludes very large open urban areas – like Hyde Park in London – but typically includes rooftops, backyards, parking spaces, city roads and small parks.

Next the researchers studied the actual vegetable consumption of each country alongside the recommended consumption, based on figures produced by the Food and Agriculture Organization and the World Health Organization of the United Nations. And finally, they calculated the percentage of cultivatable urban area necessary to meet those demands.

On average, the team found that one third of that maximum cultivatable urban area would be required to produce sufficient vegetables to feed all urban-dwellers. But hidden within this result were huge variations from country to country. In total, 11 of the 165 countries studied did not have enough cultivatable urban area to satisfy the actual vegetable consumption of their urban population. In contrast, 22 countries needed less than 10% of their cultivatable urban area to satisfy urban vegetable demand.

Countries with low urban density, such as Australia and Germany, tended to have the greatest potential, whereas countries with crowded cities like Bangladesh and Nigeria would not have enough urban area to devote to urban agriculture to meet either the actual or recommended vegetable consumption.

"Overall, the space required is regrettably the highest where urban agriculture is most needed – in more food-insecure countries," wrote Martellozzo and colleagues in ERL.

But the results also reveal the huge potential of urban agriculture. Although giant cities might find it difficult to make space for significant urban-agriculture projects, small cities have more opportunities. And roughly two-thirds of the world's urban land consists of cities less than 100 square kilometres in area.

"Implementing urban agriculture in these smaller cities will likely be easier, due to the lower urban density, e.g. more and larger backyards," said Martellozzo. Not only does planting a few peas or a row or two of spinach in idle urban spaces help to feed people, it has other benefits too. "Urban-agriculture harvests can be used to redistribute food to food-insecure people, and they can also make a difference by offering an income opportunity to [the] urban poor," said Jean-Sébastien Landry from McGill University.

Related links

Related stories