City road networks grow like biological systems
An interesting new study has been published in the American Physical Society’s journal Physical Review Letters. The authors crunched data from some 300 cities around the world, including Delhi, looking at the physics behind urban street patterns. A short synopsis of their surprising results has been published in the New Scientist. An excerpt:
French and US physicists have shown that the road networks in cities evolve driven by a simple universal mechanism despite significant cultural and historical differences. The resulting patterns are much like the veins of a leaf.
The similarity that these road networks show is accounted for by a process that the authors describe as “local optimization” and results in networks with a consistent patterning. This suggests that top-down models of urban planning only go so far:
The study’s results might be important for understanding urban growth and “sprawl” says Barthélemy. More than half the world’s population lives in cities, a proportion that continues to increase.
“The approach could even help city planners to better predict how some street networks will evolve and to plan accordingly,” he adds.
Previous models of urban development assumed that efficient transport across the entire network motivated the system’s growth - as if planned from the top down. Focussing instead on the structure of local connections seems truer to real life, says Flammini.
Ok now everyone get over to Google Earth and see how well this works for your neighborhood…
Tags: hybridity, maps, networks, the city, urban theory
December 9th, 2008 at 10:43 pm
The biological model of urban structure would seem to apply to the oldest cities, particularly those which developed without much central planning over time. Clearly most American cities not on the east coast follow forms more dictated by the planning models of the day (grid iron) or the dictates of the elected government in response to zoning, etc. Nevertheless, the hierarchy of collector, subcollector, street, and alley does look a lot like arteries and capillaries. The question should be: does this “bioloigical” form minimize travel time between points and how can it respond to growth? Most of these older cities have absolutely horrible traffic. Does this suggest that the structure was developed when the population was 10,000 and was unable to evolve to a city of 1,000,000?
December 11th, 2008 at 9:35 am
An excellent followup: “The question should be: does this “biological” form minimize travel time between points and how can it respond to growth? Most of these older cities have absolutely horrible traffic. Does this suggest that the structure was developed when the population was 10,000 and was unable to evolve to a city of 1,000,000?”
Vehicular technology would have to be a factor here too. A road network that has “evolved biologically” to meet the needs of horse carts, pedestrians or streetcars will necessarily be a bad fit for SUVs and low-clearance buses. I suppose the suggestion is that some sort of deep evolutionary efficiency principle drives this process–in spite of, or maybe even beneath the urbanist models du jour and zoning committees–and that road networks, as a system, will necessarily adapt. Think of Boston’s Big Dig, to pick on one of your east coast examples. (Also a salutary reminder that evolution works very very slowly). What intrigues me is the question of agency: does the system reorganize itself? Do we reorganize it? Some hybrid solution where we as humans act according to deep evolutionary impulses that endow us with a sense of agency even as they more or less determine our planning behavior?
The other big question: what will happen when we finally invent personal jet packs?