Landscape intelligence for a territorial platform

Design Driven by Natural Forces

The building material in landscape architecture is living matter. Such building material is always alive, and it can be organic matter like plants, inorganic material which acts as alive, for example river sand moved by water, or inorganic matter like stone or concrete, which is a conglomerate of organic matter. The ecology of landscape systems is capable of adaptation and transformation through the dynamic interplay of growth and change. However, anthropogenic construction has historically concealed landscape systems by channelising waterways, manipulating the top ography, removing native vegetation and planting to extend cities and beautify urban environments. We are thus looking at how landscape systems can be reset, so they can deploy their way of operation at the territorial scale and use the intelligence imbedded in the landscape to make its own environments more resilient.

Landscape architecture was conceived of as a synthetic discipline, and one which can look beyond its immediate context and site. In the early days of the profession, agriculture, forestry, hydrologic engineering, bridges and roads were all within the range of landscape architects. The Back Bay Fens project in Boston (Frederick Law Olmsted) used various problems of urban life, such as sanitation, transportation, and bank stabilisation, as a catalyst for innovative design. Since the area was primarily not designed as a park, it was not manicured and maintained as a fixed design, so natural evolution could take place on its grounds. A project which was initially designed as a form of sanitation and transportation infrastructure is now seen as a natural public park with historic, as well as ecological, value. Following this promising start for the profession, there were almost two centuries of landscapes designed by landscape chief builders whose tendencies were to achieve equilibrium and the fixed state of sites.


Figures 1 and 2: Back Bay Fens by Frederick Law Olmsted1887 – History and present. What was once simply sanitation infrastructure turned into a naturalised park. Fig. 1 Courtesy of Harvard GSD/ LOC, accessed on 14/4/2018 at Fig. 2 Courtesyof The Cultural Landscape Foundation, accessed on 14/04/2018


Following a decline in this trend, the past two decades have increasingly witnessed a rebirth of landscape ideas and ecological thinking in discourses on society, culture, urbanism and design. Working concurrently with science, the idea of ecology has moved away from an equilibrium-based model which supported stability, certainty and order, to a continuously adaptive model of nature which favours more contemporary understandings of dynamic systemic change, and the related phenomena of adaptability, resilience and flexibility.

Together with other professions, built environment disciplines have had to accept the notion that we have entered the Anthropocene era, in which no region of the globe is left unaffected by human impact. Those forcesare now shaping a territory beyond the urban perimeters.The prevailing distinction between city and landscape no longer exists: That is whyevery site must be understood as the composition of forces which expand over the defined borders of the site to enter territories to lands beyond. Expanding human activity is causing global climate and ecological destabilisation, and has brought natural resources to the tipping point before the collapse of living ecosystems. We are beyond the point of sustaining things as they are, and are entering a time when we need to focus on the restoration of natural habitats, lost ecosystems, and extinct species to reframe our understanding of design on a territorial and geological scale, and source the landscape intelligence of sites to accelerate the regeneration process.

It is useful to acknowledge the understanding of both terms, “landscape” and “intelligence”, to attempt to define the fundamental shift in the term “landscape intelligence”. A general perception of landscape is a description of an expanse of scenery and surface, generally seen from a single viewpoint which may display and convey a narrative of nature as filtered through one person’s image and perceived with their senses. This is also how most anthropogenic landscapes are designed. The word “intelligence” is mostly connected to human cognitive and mental capabilities involved in calculating, reasoning, and so on (Columbia Encyclopedia, sixth edition, 2006).The Encyclopaedia Britannica provides a more comprehensive explanation of intelligence as an .. . ability to adapt effectively to the environment, either by making a change in oneself or by changing the environment or finding a new one . . . intelligence is not a single mental process, but rather a combination of many mental processes directed toward effective adaptation to the environment. (Encyclopaedia Britannica, 2006). The last description is closer to the capabilities which we want to recognise as landscape intelligence. When we describe landscape intelligence we understand landscape as territory. Territory is understood here as an expanse of land where the influence and complexity of networks engage fields beyond sites, national borders and continents. Despite the vast territorial engagement that is possible, the ingenuity of its adaptation techniques lies in the unseen, under layers and on a micro level, which really contains the information and properties which enable landscapes to adapt to new conditions.


Figure 3: River Aire during construction. Photo courtesy of Fabio Chironi and Superpositions


Figure 4: The river after some of the rhomboid shapes had been eroded. Photo courtesy of Fabio Chironi and Superpositions


5: The River Aire is finding its natural meanders. Photo courtesy of Fabio Chironi and Superpositions


With this idea of an open-ended design process, the Aire river project in Geneva by Superpositions uses the parameters of time, change, and chance to create conditions for landscape intelligence to take place. The competition brief was to break up the channel and re-naturalise the river bed. Instead of designing a conventional-looking river bed, a pattern is proposed which allows the river to carve its way through prepared mounds of sand. By understanding the landscape intelligence of water’s meandering forces, a rhomboid-shape pattern was proposed which opens a complex series of channels for the flows. The result is a process-based aesthetics which enables the site to constantly reinvent itself through the erosion of an initial design state, and respond to the influence of other natural forces.By working within a responsive territorial platform, design outcomes can adapt and be evaluated for specific environmental, social, and economic measurables in relation to a site’s goals and objectives.

A progression from this design stance is seen in the In-Situ Network project by Amphibious Lab and Linkscale, situated on the island in Kiribati in the Pacific Ocean. Palm tree branches are used as a simple measure to respond to the imminent threat of coastal erosion – due to rising waters and habitat destruction. Inserted into the sand, the spoon-shaped branches constitute an ideal barrier, causing sea currents to slow down and deposit material into the concave inner surface of the leaf branch. Sand mounds are thus created which gradually elevate coastline embankments, allowing aquatic plants such as mangroves to grow and secure the beach. Here the landscape intelligence is not used as a force to change the initial design state, but rather as a generator for sites to build and reassemble themselves.


Figure 6: Territorial generative production through a strategically inserted array that interacts with in-situ currents and orchestrates sedimentation patterns, enabling autonomous island nucleation. Images courtesy of Amphibious Lab and Linkscale.


Due to this responsive landscape platform being used within a digital three-dimensional environment, real-time data simulation can be explored to communicate analytical metrics in relative and visual measurements, thus reducing the abstraction of landscape performative benefits. Methods of representation are transformed to address landscape performance complexities by using dynamic visualisation techniques that can include analytical datascapes, possible future scenarios, and interactive models in response to spatial and temporal conditions. Territorial data modelling creates a responsive workflow by taking territorial data from deployed landscape receptors in networks, such as that monitoring sea currents, tidal fluctuations, hydrology, sand movement, vegetation, and climate – to devise a framework of constraints and opportunities on site. From these potential outputs specialist knowledge is used to draw decisions which are continuously reformed based on the ongoing territorial data outputs. Because the calculations are performed in a digital site model, the analytical output can respond to changing environmental conditions. By combining the quantitative territorial data with the digital model in a collective aggregation, not only can dynamic representations of existing conditions be extracted from the digital source model, but constantly reinvented futures can be projected, and from here more informed decisions can be made.


Figure 7: Simulation of coastline articulation processes. From minimal to maximal surface interface.
Figure 8: Territorial 3D printing with in-situ material and forces (water currents and tidal fluctuations). Generative production protocol for articulated island nucleation. Stages 1 to 4 for a resilient coastline (the In-Situ Network project, in development 2011-2013, Holcim Award Asia Pacific 2014) Both images courtesy of Amphibious Lab and Linkscale.


This last project reveals a new approach to landscape architecture. Every discipline that wants to stay relevant must constantly reinvent itself. Disciplinary realignments and alterations, scientific discoveries, and cultural shifts and disturbances help reimagine new positions in known design practices. In a hyper-connected world, the responsibility to do this doesn’t lie only on a single profession or discipline. There is instead an urgent need for multiple disciplines to come together and combine real-time information from the environment to form an alternative to the current decision-making system for both landscapes and anthropogenic environments.

We are past the point of a “multidisciplinary approach” being a collection of educated voices justifying their decision making based on their own isolated valuation systems and criteria. Architecture and other built environment professions can no longer define their parameters and responsiveness at the scale of an immediate site and context, but have to operate at the scale of expended territory, a broader site that is thickly overlaid with geological information, ecological systems, economic fluctuations, social and political demands, means of mobility and an ever-expanding built environment. This broadened field is based on the notion that the role of a future built environment specialist will be to synthesise across great extremes of scale and complexity using design thinking overlapped with environmental information, and so be able to make a design that fits the site.

A territorial platform is suggested here as a theoretical, educational and practical alternative to current design decision making, so the understanding of design for sites is reframed and constantly reinvented on a territorial and geological scale, incorporating the intelligence embedded in landscape systems. Organisation of this novel platform does not only demand expertise and knowledge from professionals, but must also make use of knowledge obtained with computation and artificial intelligence, combining facts, figures, and theories. Information from sensorial networks which will be deployed in territories will source and learn from the intelligence imbedded in landscapes and natural systems. A platform is needed to serve as an interface between the human specialist, artificial intelligence and territorial data. The proposed territorial platform goes beyond data visualisation and the stagnant projection of site phasing, where every site has a final state. It is rather an ongoing interface which showcases multiple concurring and ever-changing moving parts, with the stable points in time and space. These serve as a framework instead of just a form of how predictions will be made and scenarios of future outcomes derived. With time, change and chance as the environment’s main parameters, future territorial designers and specialists will define the levels of flexibility. Enhanced and sourced landscape intelligence will pave the way to a new processed-based aesthetic.


Ana Abram

Leave a Reply