The Design Research and Academic Laboratory of ecoLogicStudio
< return to Latest Lab

PHYSAcityv1.0: BioCities / bio-computation / energy / landscape / urban designSeptember 29, 2014

Urbanism in the dust of the mining industry. Slime mold as a new urban planning model based on collective intelligence and embedded computation.

This research project is conducted within the UCL BIO Urban Design Lab and as part of the MArch Urban Design.   


The project emerges from a contemporary paradox of urban design: today, in the urban age, rather than about cities we shall talk of one large global urbansphere that wraps the world over and exchanges to an incredible rate with the natural biosphere. It has become almost impossible even to tell these two global systems apart. So we asked the following question: how can we design this urbanshpere so that it becomes more efficient, resilient and adaptive?  Can we re-conceive it as the bio-City of the future, in such a way that it can operate as a co-evolutionary system with the biosphere? And if so how? what could be a model for this biocity of the future and what form it may take?

It turns out we can find a model in the most unlikely of creatures; the slime mold, or in scientific terms, the Physarum Polycephalum.

This is one of the simplest creatures on earth, a protist, in fact it is a single cell; how can one cell be a blueprint for a city? Well this is not just any cell; within this cell we can find thousands, if not millions of nuclei; they float in a protoplasmic liquid. The whole thing is kept together but a membrane of actin. In the plasmodium phase, when enough nutrients are to be found, the membrane stretches to take virtually any shape. Not only, but the nuclei interact with each other and the environment via chemical reactions that generate gradients of pressure that in turn regulate the protoplasmic flows.

It can locally sense the presence and amount of nutrients in a given place react to it. The slime mold also leaves traces in the environment that constitute a form of spatial and distributed memory that the slime mold develops to optimise its behaviour. This is an example of embedded computation. Scientists discovered that this simplest of organists can perform, thanks to this distributed memory, extremely sophisticated tasks such as: network optimisations, nutrients regulation, events anticipation.

And this are precisely the kind of tasks that urban systems have to perform and optimise all the time. But the way slime mold does it is unlike any urban system as it is the product of emergent collective behaviour and distributes spatial memory rather than top down planning. So how can we explore and apply this behaviour in a real urban context?

For the purpose our research team developed a new kind of urban design apparatus that we tested on a prototypical scenario in the copper mining corridor in Arizona, US. The apparatus enables us to feed the slime mold with data feeds from the territory itself.  In one of these experiments, a slime mould is introduced onto a 3D-printed petri dish that reproduces the topography of the copper mining region near the city of Phoenix, Arizona. The slime mould is kept wet and food is dropped in the exact locations of present and future mining. The food is also coloured accordingly to the specific mineral content of the mine. As the mould expands to reach out for food it forms a network and begins dissolving and distributing the nutrients along it.

A high-resolution webcam captures the mould’s behaviour, morphology and the colour nuances at any moment in time; the resulting network is optimized quite like human made networks but the behaviour to achieve the solution is altogether different.

This behaviour is what we are interested in not the final solution. We than looked at ways of capturing this behaviour and plotting it back onto the territory; we developed hi-res images and a digital code able to recognise the morphology of the slime mold at any moment in time and to simulate the swarm of nuclei as they move along the network; our virtual agents would also leave traces as the real one do to create this form of spatial memory that enable the slime mould to compute, optimise and anticipate.

The emergent masterplan drawings appear fuzzy, their edges in continuous fluctuation. And as you zoom in you begin to discover more details and areas of more stable settlement. These representations are in constant progress and reveal a new urban morphogenesis that operates within the milieu or a specific environment.    

view related News
view related Lab
view related ecoLogic